ChangeLog revision c6e579a2
1commit a465588218c1643eedc35b3c24409cb775454eee
2Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3Date:   Fri Apr 5 15:50:06 2024 -0700
4
5    libX11 1.8.9
6    
7    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8
9commit 52a191ee096e93cf711f77b6a56c508b68afc0b6
10Author: Peter Hutterer <peter.hutterer@who-t.net>
11Date:   Fri Apr 5 13:15:32 2024 +1000
12
13    Revert "ximcp: Unmark to fabricate key events with XKeyEvent serial"
14    
15    This commit causes a regression, see #205, #206, #207, #208.
16    
17    This reverts commit 024d229fdf88a7755577b01b46af6ef908d599e0.
18
19commit 3ea9f4f769fecbbc8c5210c5833ad83ab8d9e97e
20Author: Peter Hutterer <peter.hutterer@who-t.net>
21Date:   Fri Apr 5 13:14:33 2024 +1000
22
23    Revert "imDefLkup: Commit first info in XimCommitInfo"
24    
25    This commit causes a regression, see #205, #206, #207, #208.
26    
27    This reverts commit 041b5291f0956c5cda5054be2981c0d02b009a4c.
28
29commit 9afd55ada5d7a3c1d3ed23ff8451629a1d989a7f
30Author: Alan Coopersmith <alan.coopersmith@oracle.com>
31Date:   Mon Mar 25 11:51:03 2024 -0700
32
33    xlibi18n: restore parse_line1 for WIN32 builds
34    
35    Accidentally removed by __UNIXOS2__ cleanup
36    Closes: #204
37    Fixes: 225a4bbb ("unifdef __UNIXOS2__")
38    
39    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
40
41commit e4927d0c4fbf950bc45fc27309290e1911e75f42
42Author: Alan Coopersmith <alan.coopersmith@oracle.com>
43Date:   Sun Mar 24 15:02:23 2024 -0700
44
45    libX11 1.8.8
46    
47    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
48
49commit 4ab58f26d388c82ffe46ef1c7dd1cee6259eec8a
50Author: Alan Coopersmith <alan.coopersmith@oracle.com>
51Date:   Wed Feb 21 18:23:36 2024 -0800
52
53    unifdef NULL_NOT_ZERO
54    
55    I can't find any evidence this was ever defined, should only have
56    been needed for odd-ball pre-C89 compilers.
57    
58    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
59
60commit 7bb2a5050a2f6049c92b301fae1cf89532b70060
61Author: Alan Coopersmith <alan.coopersmith@oracle.com>
62Date:   Wed Feb 21 18:18:46 2024 -0800
63
64    unifdef USL_SHAREDLIB
65    
66    I can't find any history of this being set in the imake or autoconf builds
67    
68    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
69
70commit 4400a68b3a5ecacd1f997d41a0572544fa8fe544
71Author: Alan Coopersmith <alan.coopersmith@oracle.com>
72Date:   Sat Feb 17 15:37:41 2024 -0800
73
74    unifdef Lynx
75    
76    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
77
78commit 1e56b27429667e772b15b0f39863a8467297062b
79Author: Alan Coopersmith <alan.coopersmith@oracle.com>
80Date:   Sun Feb 11 15:15:00 2024 -0800
81
82    unifdef __QNX__
83    
84    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
85
86commit ab0a3014822687da48cd3784253c34dea0a6ad2c
87Author: Alan Coopersmith <alan.coopersmith@oracle.com>
88Date:   Sun Feb 11 15:10:59 2024 -0800
89
90    unifdef __uxp__
91    
92    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
93
94commit 4ce3962b701c502acc96b6eaf104a5ffc317c5d7
95Author: Alan Coopersmith <alan.coopersmith@oracle.com>
96Date:   Sun Feb 11 15:07:15 2024 -0800
97
98    unifdef __vax__
99    
100    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
101
102commit 65a6f162fd9fee4e647ac49c546c5d6a1655c58c
103Author: Alan Coopersmith <alan.coopersmith@oracle.com>
104Date:   Sun Feb 11 15:03:19 2024 -0800
105
106    unifdef __sgi_not_xconsortium
107    
108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
109
110commit 3296d7b8d14a6c114489d40b76e8da339e11b25c
111Author: Alan Coopersmith <alan.coopersmith@oracle.com>
112Date:   Sun Feb 11 15:01:25 2024 -0800
113
114    unifdef __sgi
115    
116    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
117
118commit 4322fff7e125c7975dc1b83c433b904c7d3384b5
119Author: Alan Coopersmith <alan.coopersmith@oracle.com>
120Date:   Sun Feb 11 15:00:34 2024 -0800
121
122    unifdef sgi
123    
124    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
125
126commit 613d3624c2cc0ab2f18206e165a3e4a02e261148
127Author: Alan Coopersmith <alan.coopersmith@oracle.com>
128Date:   Sun Feb 11 14:59:28 2024 -0800
129
130    unifdef hpux
131    
132    Also removes shl_load() support, which was only buildable for HP-UX
133    
134    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
135
136commit 225a4bbbbd174e26cc408640e10f3c37f81f3a3f
137Author: Alan Coopersmith <alan.coopersmith@oracle.com>
138Date:   Sun Feb 11 14:56:22 2024 -0800
139
140    unifdef __UNIXOS2__
141    
142    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
143
144commit 0df284b450933e0239d1a85a7ff2d0c9d44d22af
145Author: Alan Coopersmith <alan.coopersmith@oracle.com>
146Date:   Sun Feb 11 14:50:29 2024 -0800
147
148    unifdef ultrix
149    
150    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
151
152commit c3f3eb1284c59305568288923810dc155df6025c
153Author: Alan Coopersmith <alan.coopersmith@oracle.com>
154Date:   Sun Feb 11 14:49:13 2024 -0800
155
156    unifdef AIXV3
157    
158    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
159
160commit b35344c9a74b478509e33384cc99bcc4fd092a2d
161Author: Alan Coopersmith <alan.coopersmith@oracle.com>
162Date:   Sun Feb 11 14:47:31 2024 -0800
163
164    unifdef __osf__
165    
166    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
167
168commit 041b5291f0956c5cda5054be2981c0d02b009a4c
169Author: Takao Fujiwara <tfujiwar@redhat.com>
170Date:   Wed Jan 31 20:27:57 2024 +0900
171
172    imDefLkup: Commit first info in XimCommitInfo
173    
174    Xic.private.proto.commit_info can receive multiple XimCommitInfo
175    when typing keys very quickly like an bar code scanner (or evemu-play)
176    and the first info in XimCommitInfo should be committed to keep
177    the typing key order.
178    
179    Fixes: #198
180
181commit 024d229fdf88a7755577b01b46af6ef908d599e0
182Author: Takao Fujiwara <tfujiwar@redhat.com>
183Date:   Wed Jan 31 20:26:40 2024 +0900
184
185    ximcp: Unmark to fabricate key events with XKeyEvent serial
186    
187    _XimProtoKeypressFilter() and _XimProtoKeyreleaseFilter() can
188    receive XKeyEvent from both the typing on the keyboard and the
189    callback of XIM_FORWARD_EVENT.
190    
191    If the filter functions unmark to fabricate XKeyEvent from the typing
192    on the keyboard during receiving XKeyEvent from the callback of
193    XIM_FORWARD_EVENT with typing keys very quickly likes an bar code
194    scanner (or evemu-play), XIM server cannot receive some key events and
195    it causes the key typing order to get scrambled.
196    
197    Now XIM client saves the serial in XKeyEvent and the filter functions
198    unmark to fabricate XKeyEvent from the callback of XIM_FORWARD_EVENT
199    only.
200    
201    Fixes: #198
202
203commit ae3eca18cec44a953789c7f77ffab888713ed132
204Author: Peter Hutterer <peter.hutterer@who-t.net>
205Date:   Thu Nov 30 15:05:51 2023 +1000
206
207    Fix _XkbReadGetDeviceInfoReply for nButtons == dev->buttons
208    
209    XkbGetDeviceInfo(dpy, XkbXI_ButtonActionsMask, 2, 0, 0) always returns
210    NULL because the number of buttons on the device equals (unsurpisingly)
211    the number of buttons requested (i.e. first + nBtns == dev->nbuttons).
212    
213    This currently causes it to bail out and return NULL.
214    
215    Fixes f293659d5a4024bda386305bb7ebeb4647c40934
216
217commit 0a951047f64e87c5ab328084310bf27bcd4964f3
218Author: Walter Harms <wharms@bfs.de>
219Date:   Mon Jan 8 17:18:19 2024 +0100
220
221    _XimProtoIMFree:no need to check arg for Xfree()
222    
223       Xfree() will happily ignore NULL, no need to check
224
225commit dce614623e1f069d79d33a233305e0754fa2fd48
226Author: Walter Harms <wharms@bfs.de>
227Date:   Mon Jan 8 17:01:44 2024 +0100
228
229    _XimEncodeString:no need to check arg for Xfree()
230    
231    Xfree() will happily ignore NULL, no need to check
232
233commit 4f78b615806fa1cfe2bdcf3f7c868b8a00dede77
234Author: Walter Harms <wharms@bfs.de>
235Date:   Mon Jan 8 16:50:52 2024 +0100
236
237    Fix XCreateIC() memory leak (Part 2)
238    
239    Direct leak of 12 byte(s) in 2 object(s) allocated from:
240        #0 0x7f4f25c3f7a7 in strdup (/usr/lib64/libasan.so.6+0x5c7a7)
241        #1 0x7f4f252ce6a1 in _XimEncodeString libX11-1.8.3/modules/im/ximcp/imRm.c:818
242        #2 0x7f4f252ce6a1 in _XimEncodeString libX11-1.8.3/modules/im/ximcp/imRm.c:807
243        #3 0x7f4f252d2f0f in _XimSetICValueData libX11-1.8.3/modules/im/ximcp/imRm.c:2912
244        #4 0x7f4f252b536a in _XimLocalCreateIC libX11-1.8.3/modules/im/ximcp/imLcIc.c:176
245        #5
246    
247     0x7f4f251f0105 in XCreateIC libX11-1.8.3/src/xlibi18n/ICWrap.c:251
248    
249    detected and fix by Patrick Lerda <patrick9876@free.fr>
250    applied with adjustment, do changes when OOM (unlikely but good practise)
251
252commit ed0b97e480d7d07c24719007273989b2eb3e4046
253Author: Walter Harms <wharms@bfs.de>
254Date:   Mon Jan 8 16:21:02 2024 +0100
255
256    _XimLocalDestroyIC:fix possible mem leak
257    
258    Adapted:
259    Fix XCreateIC() memory leak by Patrick Lerda <patrick9876@free.fr> Part 1
260
261commit 07978634b8f4f7d9537c7f46ddcd48f2ab3efcb8
262Author: Walter Harms <wharms@bfs.de>
263Date:   Mon Jan 8 15:16:11 2024 +0100
264
265    _XimLocalCreateIC: get rid of bzero
266
267commit 59c9a89e250d479334bdd4d800ca6cfa919b8fc3
268Author: Walter Harms <wharms@bfs.de>
269Date:   Mon Jan 8 15:09:49 2024 +0100
270
271    _XimLocalCreateIC: minor cleanup
272    
273    minor cleanup, no code change
274
275commit e5b14e59a16fa49c7e7af0acfc1e4311bebec536
276Author: Walter Harms <wharms@bfs.de>
277Date:   Mon Jan 8 13:35:28 2024 +0100
278
279    _XimLocalCreateIC:no need to check arg for Xfree()
280    
281    Xfree() will happily ignore NULL, no need to check
282
283commit e6310b5212f1dea184e1626ad5eeec167fcb41ac
284Author: Walter Harms <wharms@bfs.de>
285Date:   Mon Jan 8 13:06:03 2024 +0100
286
287    _XimLocalDestroyIC: no need to check arg for Xfree()
288    
289    Xfree() will happily ignore NULL, no need to check
290
291commit 1261802fb316f088504b99ffe80095d13825a85b
292Author: Walter Harms <wharms@bfs.de>
293Date:   Sun Jan 7 18:30:30 2024 +0100
294
295    fix table width
296    
297    the width of first column was to small and
298    caused a hyphenation. there is only one word,
299    fix for me.
300
301commit c745719e23af44a4b40ab4508447637b35d91a1e
302Author: Alan Coopersmith <alan.coopersmith@oracle.com>
303Date:   Tue Oct 3 09:00:01 2023 -0700
304
305    libX11 1.8.7
306    
307    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
308
309commit 7916869d16bdd115ac5be30a67c3749907aea6a0
310Author: Yair Mizrahi <yairm@jfrog.com>
311Date:   Thu Sep 7 16:15:32 2023 -0700
312
313    CVE-2023-43787: Integer overflow in XCreateImage() leading to a heap overflow
314    
315    When the format is `Pixmap` it calculates the size of the image data as:
316        ROUNDUP((bits_per_pixel * width), image->bitmap_pad);
317    There is no validation on the `width` of the image, and so this
318    calculation exceeds the capacity of a 4-byte integer, causing an overflow.
319    
320    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
321
322commit b4031fc023816aca07fbd592ed97010b9b48784b
323Author: Alan Coopersmith <alan.coopersmith@oracle.com>
324Date:   Thu Sep 7 16:12:27 2023 -0700
325
326    XCreatePixmap: trigger BadValue error for out-of-range dimensions
327    
328    The CreatePixmap request specifies height & width of the image as CARD16
329    (unsigned 16-bit integer), so if either is larger than that, set it to 0
330    so the X server returns a BadValue error as the protocol requires.
331    
332    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
333
334commit 73a37d5f2fcadd6540159b432a70d80f442ddf4a
335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
336Date:   Thu Sep 7 15:55:04 2023 -0700
337
338    XPutImage: clip images to maximum height & width allowed by protocol
339    
340    The PutImage request specifies height & width of the image as CARD16
341    (unsigned 16-bit integer), same as the maximum dimensions of an X11
342    Drawable, which the image is being copied to.
343    
344    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
345
346commit 204c3393c4c90a29ed6bef64e43849536e863a86
347Author: Alan Coopersmith <alan.coopersmith@oracle.com>
348Date:   Thu Sep 7 15:54:30 2023 -0700
349
350    CVE-2023-43786: stack exhaustion from infinite recursion in PutSubImage()
351    
352    When splitting a single line of pixels into chunks to send to the
353    X server, be sure to take into account the number of bits per pixel,
354    so we don't just loop forever trying to send more pixels than fit in
355    the given request size and not breaking them down into a small enough
356    chunk to fix.
357    
358    Fixes: "almost complete rewrite" (Dec. 12, 1987) from X11R2
359    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
360
361commit 6858d468d9ca55fb4c5fd70b223dbc78a3358a7f
362Author: Alan Coopersmith <alan.coopersmith@oracle.com>
363Date:   Sun Sep 17 14:19:40 2023 -0700
364
365    CVE-2023-43785: out-of-bounds memory access in _XkbReadKeySyms()
366    
367    Make sure we allocate enough memory in the first place, and
368    also handle error returns from _XkbReadBufferCopyKeySyms() when
369    it detects out-of-bounds issues.
370    
371    Reported-by: Gregory James DUCK <gjduck@gmail.com>
372    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
373
374commit 78b37accff1abbe713349d59fdefd963ffa04bbc
375Author: Alan Coopersmith <alan.coopersmith@oracle.com>
376Date:   Sun Sep 17 10:11:02 2023 -0700
377
378    XOpenDisplay: ensure each screen has a valid root_visual pointer
379    
380    Other code assumes this pointer cannot be NULL, so fail the connection
381    if a bug has caused the X server to give a non-existent visual ID for
382    the default visual of any screen.
383    
384    Reported-by: Gregory James DUCK <gjduck@gmail.com>
385    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
386
387commit 3962080b7846d79fd3cf10b16eddacf68eb321b7
388Author: Antti Savolainen <antti.savo@gmail.com>
389Date:   Tue Aug 8 11:37:03 2023 +0000
390
391    Add two compose sequences for dagger/obelisk symbols
392    
393    The dagger symbol has several modern uses such as marking someone as
394    dead or something as extinct. Historically it has been used to indicate
395    a footnote.
396
397commit 15bbc81cf6b7c76b8a59cab6d77c9aba88b8562d
398Author: Christopher Chavez <chrischavez@gmx.us>
399Date:   Sat Jul 29 16:58:05 2023 +0000
400
401    Xutil.h: use Bool for XEmptyRegion(), XEqualRegion()
402
403commit 5433cb93513c31e9cba2a447cf48d4b9aeb53285
404Author: Benno Schulenberg <bensberg@telfort.nl>
405Date:   Mon Jul 17 08:02:09 2023 +0200
406
407    nls: add two compose sequences for currency symbol `฿` (the Thai baht)
408    
409    The compose sequences are obvious: the capital B plus the vertical bar.
410    
411    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
412
413commit 2ff401a472c5f210844cc358d5a27bd340546796
414Author: Benno Schulenberg <bensberg@telfort.nl>
415Date:   Mon Jul 17 07:56:14 2023 +0200
416
417    nls: reshuffle the compose sequences for currencies into ascending order
418    
419    That is: sort them according to code point.
420
421commit c0e2d42d411be3245498ffe6484b35421f096c02
422Author: Benno Schulenberg <bensberg@telfort.nl>
423Date:   Mon Jul 17 07:27:14 2023 +0200
424
425    nls: add compose sequences for `₲` (guaraní), `₭` (kip), and `₮` (tugrik)
426    
427    When Ghana, Nigeria, Costa Rica and El Salvador have compose sequences
428    for their currency symbols (cedi: `₵`, naira: `₦`, colón: `₡`), then
429    Paraguay, Laos, and Mongolia deserve to have such sequences as well.
430    
431    The sequences should be obvious: the relevant capital letter of the
432    name of the currency (G, K, T) plus a vertical bar, a minus sign,
433    and an equals sign, respectively.
434    
435    Also add two sequences for `$` (the dollar sign), for consistency.
436    
437    Drop the tentative comments for drachma, penny, and austral, as those
438    currencies have been obsolete for more than twenty years.
439    
440    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
441
442commit d9fe5dc735c75398e36998c226d2418137229673
443Author: Benno Schulenberg <bensberg@telfort.nl>
444Date:   Mon Jul 17 07:06:51 2023 +0200
445
446    nls: change result of sequence `<C> <bar>` from `¢` (CENT) to `₵` (CEDI)
447    
448    As a cent is a small coin, it makes no sense to use an uppercase letter
449    to compose the `¢` symbol -- having four sequences with a lowercase `c`
450    plus a `bar` or a `slash` available for composing `¢` should be enough.
451    
452    Use the sequence `<C> <bar` (with the uppercase `C`) for composing `₵`
453    (the CEDI SIGN) instead.
454    
455    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
456
457commit 031e63060ed11eb8a876aac7515bc5a67371ed6b
458Author: Antti Savolainen <antti.savo@gmail.com>
459Date:   Thu Aug 17 18:25:20 2023 +0300
460
461    nls: add a compose sequence with double dead_tilde for ≈ (almost equal to)
462    
463    Some keyboards lack direct asciitilde and instead have only dead tilde.
464    This compose sequence allows those keyboards to type the almost equal
465    to sign.
466
467commit df2a18c4643bb4a50f2f795a75a905a1032a004a
468Author: Benno Schulenberg <bensberg@telfort.nl>
469Date:   Mon Jul 17 17:03:16 2023 +0200
470
471    nls: delete twenty seven untypable Greek compose sequences
472    
473    Only a few keyboard layouts contain the `dead_iota` keysym, and none
474    of those layouts contains the `acute` keysym, so compose sequences
475    that combine the two symbols cannot be typed and are thus useless.
476    
477    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
478
479commit 44f908d9283710ffc75b22d2ae7a8948119b3e61
480Author: Alan Coopersmith <alan.coopersmith@oracle.com>
481Date:   Sun Aug 20 15:11:04 2023 -0700
482
483    docs: fix names for XkbKeyKeyType & XkbKeyKeyTypeIndex
484    
485    Implementation had "Key" twice in these macro names,
486    but the docs had only listed it once.
487    
488    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
489
490commit 7b16c55063ba4eb258bf91c170e378c1e2d398bd
491Author: Alan Coopersmith <alan.coopersmith@oracle.com>
492Date:   Sun Aug 20 15:01:09 2023 -0700
493
494    docs: remove XkbKeySymsOffset
495    
496    This function was documented but never implemented.
497    
498    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
499
500commit 41a157dfeefe943f5a79948e4f9ef93473144fd8
501Author: Alan Coopersmith <alan.coopersmith@oracle.com>
502Date:   Sun Aug 20 14:52:18 2023 -0700
503
504    docs: remove XkbGetNameChanges
505    
506    This function was documented but never implemented.
507    
508    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
509
510commit 3442a8cee25dc0254ee69bcf83a89d287a86c97a
511Author: Alan Coopersmith <alan.coopersmith@oracle.com>
512Date:   Sun Aug 20 14:44:18 2023 -0700
513
514    docs: remove XkbChangeIndicators and Xkb{Get,Note}IndicatorChanges
515    
516    These functions were documented but never implemented.
517    
518    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
519
520commit b7a15d6c6efd0bfddc81cf8443073c4bc1c9772d
521Author: Alan Coopersmith <alan.coopersmith@oracle.com>
522Date:   Sun Aug 20 14:31:20 2023 -0700
523
524    docs: XkbSAActionSetCtrls is really named XkbActionSetCtrls
525    
526    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
527
528commit a1d8dd726757e00e3a774b92456f5c742492475b
529Author: Alan Coopersmith <alan.coopersmith@oracle.com>
530Date:   Sun Aug 20 14:18:32 2023 -0700
531
532    docs: remove XkbGetStickyKeysOptions & XkbSetStickyKeysOptions
533    
534    These functions were documented but never implemented.
535    
536    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
537
538commit bf13f197406fe745c5ff46a4748a41b6646aebf6
539Author: Alan Coopersmith <alan.coopersmith@oracle.com>
540Date:   Sun Aug 20 14:09:35 2023 -0700
541
542    docs: remove XkbGetSlowKeysDelay & XkbSetSlowKeysDelay
543    
544    These functions were documented but never implemented.
545    
546    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
547
548commit caab7321d2ec14ee776ec329ed2facf0816a3136
549Author: Alan Coopersmith <alan.coopersmith@oracle.com>
550Date:   Sun Aug 20 14:04:15 2023 -0700
551
552    docs: remove XkbGetAccessXTimeout & XkbSetAccessXTimeout
553    
554    These functions were documented but never implemented.
555    
556    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
557
558commit 843e89bd2467a58cb454705a09d095726b96d93f
559Author: Alan Coopersmith <alan.coopersmith@oracle.com>
560Date:   Sun Aug 20 12:58:47 2023 -0700
561
562    docs: remove XkbAllocDeviceLedInfo
563    
564    This function was documented but never implemented.
565    
566    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
567
568commit 2686dd7a9d6e8e02ce2868e446cdec11617ffab4
569Author: Alan Coopersmith <alan.coopersmith@oracle.com>
570Date:   Sun Aug 20 12:41:42 2023 -0700
571
572    docs: finish removing XkbGetBounceKeysDelay & XkbSetBounceKeysDelay
573    
574    These functions were documented but never implemented.
575    
576    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
577
578commit 188991fe7e41bbf2f384d0875ab2e25c936b0b41
579Author: G. Branden Robinson <g.branden.robinson@gmail.com>
580Date:   Sat Jul 29 10:10:00 2023 -0700
581
582    configure script reports stray, confusing "yes"
583    
584    Closes issue #194
585    
586    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
587
588commit 70f44d76fd8fad96f33ba5c80a394d68b77dda03
589Author: Benno Schulenberg <bensberg@telfort.nl>
590Date:   Sun Jul 16 17:08:52 2023 +0200
591
592    nls: add, correct, and normalize some comments of compose sequences
593    
594    Add comments for the Khmer digraphs, correct the comments for the
595    Arabic lam-alef decompositions, and normalize the comments for the
596    Breton digraphs and trigraphs.
597    
598    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
599
600commit 2a57c32e14779cf1c0c1d2da98a9cac21c00c095
601Author: Benno Schulenberg <bensberg@telfort.nl>
602Date:   Sun Jun 25 10:01:05 2023 +0200
603
604    nls: harmonize the comments for compose sequences with combining accents
605    
606    Replace the "WITH" with "plus" and lowercase the "AND" in the comments
607    for sequences with combining accents, to make it slightly clearer that
608    the resulting string consists of multiple code points.  Also, use the
609    word "COMBINING" in the names of the accents, so that these sequences
610    can be easily grepped, and drop the redundant word "ACCENT".
611    
612    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
613
614commit 757ab5ddd86f6cdce24b8ae2c43206b7beb7ab44
615Author: Alan Coopersmith <alan.coopersmith@oracle.com>
616Date:   Mon Aug 14 08:33:28 2023 -0700
617
618    Remove XkbSetBounceKeysDelay.man from list of man pages to build
619    
620    Fixes: f90e6bfa ("rm XkbSetBounceKeysDelay.man")
621    
622    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
623
624commit f90e6bface92fd192b26f03b7ae745661d9c2690
625Author: Walter Harms <wharms@bfs.de>
626Date:   Sun Aug 13 22:42:50 2023 +0200
627
628    rm XkbSetBounceKeysDelay.man
629    
630    there is no function XkbSetBounceKeysDelay().
631    It is listed in the specs but never implemented if someone
632    does it, feel free to get the file back.
633    
634    closes issue #105
635
636commit b2c3fb7b4dd4c4c7b9d28c7af7776cccc8e0a98b
637Author: Benno Schulenberg <bensberg@telfort.nl>
638Date:   Sat Jul 8 08:11:39 2023 +0200
639
640    nls: use the shorter and more consistent name for the dead capital schwa
641    
642    The `dead_SCHWA` name is available since xorgproto-2023.1, which was
643    released last month.
644
645commit e45dff5c2a51ad0392061208a8ad33211d952385
646Author: jmcwilliams403 <jmcwilliams403@gmail.com>
647Date:   Fri Jun 16 11:37:20 2023 -0400
648
649    NLS: Drop two math sequences, slightly clean up APL sequences.
650    
651    * Two compose sequences containing `leftshoe` and `rightshoe` are
652    dropped as no keyboard layout uses these keysyms.
653    * The compose sequences for `therefore` and `because` in the APL
654    block are moved to ascending Unicode order.
655    * The comments for the compose sequences for `U2299` are corrected to
656    more accurately reflect its sequence nodes and its real name in
657    Unicode.
658
659commit af2b6dfab1616dc85be9c9b196e4c56d00447851
660Author: Benno Schulenberg <bensberg@telfort.nl>
661Date:   Sun Jun 25 09:09:21 2023 +0200
662
663    nls: delete twenty eight compose sequences that cannot be typed
664    
665    The letters `Ǫ` (U+01EA, O with ogonek), `Ȩ` (U+0228, E with cedilla),
666    `Ȯ` (U+022E, O with dot above), `Ạ` (U+1EA0, A with dot below), and
667    their lowercase forms do not occur in any layout of xkeyboard-config,
668    meaning that the compose sequences that contained these letters could
669    not be typed.  Delete their dead weight.
670    
671    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
672
673commit ab0442d3fa835ce16559b29532ac7f674f8557f4
674Author: Benno Schulenberg <bensberg@telfort.nl>
675Date:   Sat Jun 24 09:46:07 2023 +0200
676
677    nls: remove four unobvious and redundant compose sequences for ¥ (yen)
678    
679    Composing ¥ with Y and a minus sign seems to have been added in analogy
680    to composing £ with L and a minus sign.  But ¥ clearly has a double line
681    through it, so using the equals sign for this is far more logical (and
682    those compose sequences of course exist).  Also, L plus an equals sign
683    produces ₤ (lira), not £ (sterling).  So... make these sequences more
684    consistent and allow composing ¥ only with Y/y plus an equals sign.
685    
686    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
687
688commit 933912480c08495428d45988eebb48fcbe62f14b
689Author: Benno Schulenberg <bensberg@telfort.nl>
690Date:   Sat Jun 24 10:02:45 2023 +0200
691
692    nls: order a few compose sequences left first, and in mirroring pairs
693
694commit 178c4fefd3b0b4be5e69274108b86a51f08ec3a6
695Author: Benno Schulenberg <bensberg@telfort.nl>
696Date:   Fri Jun 16 09:33:03 2023 +0200
697
698    nls: remove three duplicated combining acute accents
699    
700    For the Cyrillic YU with combining acute accent, the string between
701    the quotes contained two U+0301 code points, whereas one is enough,
702    like for all the neighboring strings.
703    
704    (This duplication was found by accident with `nano --mini --cons`.)
705
706commit 695e90ad26f632feb0f58ad94882fb3a263bf114
707Author: Alan Coopersmith <alan.coopersmith@oracle.com>
708Date:   Thu Jun 15 09:14:48 2023 -0700
709
710    libX11 1.8.6
711    
712    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
713
714commit 304a654a0d57bf0f00d8998185f0360332cfa36c
715Author: Alan Coopersmith <alan.coopersmith@oracle.com>
716Date:   Sat Jun 10 16:30:07 2023 -0700
717
718    InitExt.c: Add bounds checks for extension request, event, & error codes
719    
720    Fixes CVE-2023-3138: X servers could return values from XQueryExtension
721    that would cause Xlib to write entries out-of-bounds of the arrays to
722    store them, though this would only overwrite other parts of the Display
723    struct, not outside the bounds allocated for that structure.
724    
725    Reported-by: Gregory James DUCK <gjduck@gmail.com>
726    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
727
728commit 71b08b8af20474bb704a11affaa8ea39b06d5ddf
729Author: Alan Coopersmith <alan.coopersmith@oracle.com>
730Date:   Wed May 31 17:45:40 2023 -0700
731
732    libX11 1.8.5
733    
734    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
735
736commit 03ba0140940cc76524d83096a47309f5c398541f
737Author: Benno Schulenberg <bensberg@telfort.nl>
738Date:   Mon May 22 12:02:06 2023 +0200
739
740    nls: use a slash instead of a combining solidus in compose sequences
741    
742    First: combining diacritics like the combining long solidus (`U+0338`)
743    are not meant to be used in compose sequences.  Second: there is just
744    one layout in xkeyboard-config that contains the `U0338` character:
745    the deprecated/obsolete German T3 layout.  So, practically speaking,
746    these compose sequences with `U0338` were untypable.  So, use a slash
747    instead, that almost all layouts have.  This does require that the
748    sequence `<Compose> <less> <slash>` changes its meaning from backslash
749    to not-less-than (`≮`).  This seems like an acceptable sacrifice, as
750    the sequence `<Compose> <slash> <slash>` is a faster/easier sequence
751    for the backslash and most layouts contain a backslash already anyway.
752    
753    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
754
755commit 616701317e25b2c0f509f4d93af5dee6de09012d
756Author: Benno Schulenberg <bensberg@telfort.nl>
757Date:   Mon May 22 11:19:32 2023 +0200
758
759    nls: delete six compose sequences that cannot be typed
760    
761    None of the layouts in xkeyboard-config contain `Ohook` or `Uhook`
762    or `U+01D3` (U with caron) nor their lowercase forms, so these six
763    sequences cannot be typed on any known keyboard.
764    
765    (Also, the target characters of the sequences with hook and horn
766    seem to occur only in Vietnamese, where the horn is part of the
767    vowel and the hook is a tonal mark.  All Vietnamese layouts in
768    xkeyboard-config contain `Ohorn` and `Uhorn`, and then the five
769    tonal marks as dead keys, `dead_hook` among them.  It does not
770    make sense to want to apply a horn to a vowel that already has
771    a tonal mark, nor is it possible with the Vietnamese layouts.)
772    
773    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
774
775commit 35d9621d78cb196b4ccd3ee1baa1ce2a21e4674b
776Author: Benno Schulenberg <bensberg@telfort.nl>
777Date:   Sun May 21 16:17:32 2023 +0200
778
779    nls: sort three sequences alphabetically in their group, like all others
780    
781    Also, consistently form uppercase+lowercase pairs, like for most others.
782
783commit 208d550954c7266fa8093b02a2a97047e1478c00
784Author: Benno Schulenberg <bensberg@telfort.nl>
785Date:   Thu Apr 13 17:25:35 2023 +0200
786
787    nls: make the Amharic compose sequences use the dead-vowel symbols
788    
789    Last year, commit 488b156fe2 changed mistaken ASCII characters to the
790    code points that were actually used in the Ethiopian keyboard layout.
791    
792    But those code points were mistaken too: the intention was not to use
793    the code point U+FE68 in the layout but the raw symbol code 0xfe68 --
794    and so on for the other codes.  The symbols that these raw codes were
795    intended to refer to were `dead_a` and friends.  But these dead vowels
796    were added under different codes: 0xfe80 to 0xfe8b.
797    
798    So, correct the Amharic compose sequences to make use of the dead-vowel
799    symbols that were specifically added for them fifteen years ago.
800    
801    (As soon as this is merged, xkeyboard-config wil be changed too to use
802    the dead vowels in the Ethiopian layout.)
803    
804    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
805
806commit 96cbac89a02220eb21e13ca5fa7c8c5adf77028a
807Author: jmcwilliams403 <jmcwilliams403@gmail.com>
808Date:   Fri Mar 31 09:47:03 2023 -0400
809
810    nls: add compose sequence for capital schwa, and delete a deviant one
811    
812    This sequence has already been defined in the iso8859-9e compose
813    file meant for the Azerbaijani locale, yet for the longest time, only
814    the lowercase form has existed in the en_US.UTF-8 master compose
815    file. This patch adds one sequence for Latin Capital Letter Schwa for
816    the Azerbaijani language.
817    
818    Additionally, the sequence in the pt_BR.UTF-8 locale for capital
819    schwa has been removed as it uses two lowercase e's as part of its
820    sequence, which in practice would make typing both `Multi_key e e`
821    and `Multi_key E E` output capital schwa.
822
823commit a6528d92b3865adb4f9951df91a714f55f56d59e
824Author: Benno Schulenberg <bensberg@telfort.nl>
825Date:   Tue Apr 4 11:35:44 2023 +0200
826
827    XSetScreenSaver.man: restore the part that was accidentally snipped
828    
829    Four years ago, while removing private macros, commit fe46eb2871
830    unintentionally truncated the XSetScreenSaver manpage.  Restore
831    the snipped part (minus trailing spaces and a mistaken singular).
832    
833    This fixes issue #187.
834    
835    Reported-by: Sergio Gelato
836    
837    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
838
839commit eb166af8f9f4184108fe8e0611a09af03d12059e
840Author: jmcwilliams403 <jmcwilliams403@gmail.com>
841Date:   Fri Mar 24 08:37:32 2023 -0400
842
843    NLS: Remove strange dead_cedilla cedi sign sequences
844    
845    These sequences appear to be a leftover artifact of auto-generating
846    dead_currency sequences for cent and the cedi sign, where chaining
847    with ascii "c" vs "ccedilla" would output cent vs the cedi sign
848    respectively, but somehow this ended up creating strange sequences
849    with dead_cedilla followed by cent or the colon sign to output the
850    cedi sign, the latter sequence of which was already removed recently.
851    Additionally, sequences beginninng with dead_cedilla followed by
852    currency have been removed as they also appear to be related to these
853    artifacts. The two remaining sequences beginning with dead_currency
854    followed by dead_cedilla are maintained and moved to the stretch of
855    dead_currency sequences next to ccedilla, and sorted in order of
856    unicode decompositional complexity.
857
858commit 6e889f34679bb96f4d61ef069259c042cb4a9a59
859Author: Benno Schulenberg <bensberg@telfort.nl>
860Date:   Fri Mar 24 09:41:48 2023 +0100
861
862    nls: remove twenty two untypable Greek compose sequences
863    
864    None of the keyboard layouts in xkeyboard-config contains accented
865    Greek characters like `Greek_alphaaccent` or `Greek_iotadieresis`
866    (nor in their Unicode hexadecimal forms, like U03AC or 0x10003CA),
867    so none of these sequences could be typed on an existing layout.
868    
869    Also, the two sequences that started with `Greek_accentdieresis`
870    were plain wrong: making a spacing character into a dead key.
871    
872    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
873
874commit 98be4cb23a5eadc98f81f05c4696f687119152c8
875Author: Benno Schulenberg <bensberg@telfort.nl>
876Date:   Fri Mar 24 09:00:48 2023 +0100
877
878    nls: remove four hundred and sixty untypable Greek compose sequences
879    
880    None of the keyboard layouts in xkeyboard-config contains code points
881    in the range U+1F00..U+1FFF (the Greek Extended block), so none of the
882    removed sequences could be typed on an existing layout.
883    
884    (Of course it is possible that some people have their own private layout
885    that contains some of the Greek Extended characters, but if they do,
886    then they can also define private compose sequences, when needed.)
887    
888    Modern Greek does not use these characters with polytonic diacritics
889    anyway -- they are useful only for typing traditional Greek.
890    
891    The useless sequences were removed with this command:
892    
893      sed -i '/<U1F..>/d' nls/en_US.UTF-8/Compose.pre
894    
895    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
896
897commit ff8706a5eae25b8bafce300527079f68a201d27f
898Author: Benno Schulenberg <bensberg@telfort.nl>
899Date:   Wed Jun 15 11:54:54 2022 +0200
900
901    nls: delete compose sequences that pointlessly mix upper and lower case
902    
903    It's nice to have compose sequences for things like © and ®, but allowing
904    them to be composed also with letters of different case is unneeded.  It
905    is good enough to have sequences of only lowercase and only uppercase.
906    
907    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
908
909commit ca99e338a9b8aad300933b1336f9e3c091392213
910Author: jmcwilliams403 <jmcwilliams403@gmail.com>
911Date:   Sun Mar 12 15:13:27 2023 +0000
912
913    NLS: move dead-caron subscript compositions to the relevant Unicode block
914
915commit dac2985b347cebc0f631e43e41eaeddb68de1676
916Author: jmcwilliams403 <jmcwilliams403@gmail.com>
917Date:   Sun Mar 12 14:49:47 2023 +0000
918
919    NLS: implement the expansion of the six Breton N-graph keysyms
920    
921    These one-keysym-to-multiple-characters sequences will allow fully defining
922    the Breton keyboard layout, without needing custom compose sequences.
923
924commit e45ca7b41dcd3ace7681d6897505f85d374640f2
925Author: Alan Coopersmith <alan.coopersmith@oracle.com>
926Date:   Tue Mar 14 10:22:48 2023 -0700
927
928    gitlab CI: add workflow rules
929    
930    Needed due to the gitlab config changes described in:
931     https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/438
932    
933    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
934
935commit 72eeeb6dedc39f68608312eb0001f77fffccd499
936Author: jmcwilliams403 <jmcwilliams403@gmail.com>
937Date:   Sun Mar 5 13:33:38 2023 +0000
938
939    NLS: move long S compositions to respective blocks
940
941commit 8bbbed75a20556e41658103265974decee974645
942Author: Alan Coopersmith <alan.coopersmith@oracle.com>
943Date:   Sat Mar 4 10:44:45 2023 -0800
944
945    configure: Use LT_INIT from libtool 2 instead of deprecated AC_PROG_LIBTOOL
946    
947    AC_PROG_LIBTOOL was replaced by LT_INIT in libtool 2 in 2008,
948    so it's time to rely on it.
949    
950    Clears autoconf warnings:
951    
952    configure.ac:20: warning: The macro `AC_PROG_LIBTOOL' is obsolete.
953    configure.ac:20: You should run autoupdate.
954    m4/libtool.m4:100: AC_PROG_LIBTOOL is expanded from...
955    configure.ac:20: the top level
956    
957    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
958
959commit 53bf8584e8d7d5d4a4a8114bff26a6f631c7fac1
960Author: Alan Coopersmith <alan.coopersmith@oracle.com>
961Date:   Mon Feb 6 14:09:45 2023 -0800
962
963    configure: replace deprecated AC_HELP_STRING with AS_HELP_STRING
964    
965    This silences an autoconf warning.
966    
967    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
968
969commit 31a9451111fb1e3ae1199d44f94cd0b134ca455d
970Author: Alan Coopersmith <alan.coopersmith@oracle.com>
971Date:   Mon Feb 6 13:13:21 2023 -0800
972
973    configure: raise minimum autoconf requirement to 2.70
974    
975    Needed for builds on NetBSD to work correctly, since it depends on
976    AC_USE_SYSTEM_EXTENSIONS defining _OPENBSD_SOURCE to expose the
977    prototype for reallocarray() in the system headers.
978    
979    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
980
981commit 1ba02f878f83e3b42389547e01b24371f0a50e60
982Author: Alan Coopersmith <alan.coopersmith@oracle.com>
983Date:   Mon Feb 6 13:32:02 2023 -0800
984
985    gitlab CI: Add libtool to required packages
986    
987    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
988
989commit fc5ec8edda1efdf5d66db4e76796e3c407527b3b
990Author: Alan Coopersmith <alan.coopersmith@oracle.com>
991Date:   Sat Feb 4 12:02:29 2023 -0800
992
993    libX11 1.8.4
994    
995    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
996
997commit ba548ed1688d4d5a83913218155d0ddf616ff176
998Author: Alan Coopersmith <alan.coopersmith@oracle.com>
999Date:   Sun Jan 15 11:39:20 2023 -0800
1000
1001    configure: options summary should say "libX11", not just "X11"
1002    
1003    We're not configuring the entire window system here, just one library
1004    
1005    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1006
1007commit eb1c272ab5230d548077b9f59aca4b3457c3a8f8
1008Author: GaryOderNichts <garyodernichts@gmail.com>
1009Date:   Sat Dec 17 16:28:40 2022 +0100
1010
1011    Fix a9e845 and 797755 Allow X*IfEvent() to reenter libX11
1012
1013commit 0c7c4ce875ea65160705180ebb68a2c88afb678e
1014Author: Benno Schulenberg <bensberg@telfort.nl>
1015Date:   Thu Nov 3 17:25:29 2022 +0100
1016
1017    nls: add two compose sequences for capital B with stroke, for consistency
1018    
1019    Two compose sequences (with plain ASCII characters) existed for
1020    the lowercase b with stroke (ƀ) but not for the capital one (Ƀ).
1021    
1022    This addresses part of issue #166.
1023    
1024    Reported-by: J. McWilliams
1025    
1026    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1027
1028commit a80ef097e713aba0330b9cf9f3c1e309084b926f
1029Author: Benno Schulenberg <bensberg@telfort.nl>
1030Date:   Thu Nov 3 16:48:03 2022 +0100
1031
1032    nls: delete twelve compose sequences that cannot be typed
1033    
1034    There is not any keyboard layout (in xkeyboard-config) that contains
1035    `obelowdot` or `ubelowdot`, so having compose sequences that use these
1036    symbols is pointless.
1037    
1038    There are some layouts that contain `otilde` (and one that contains
1039    `utilde`), but those layouts then do not contain `dead_horn`, so
1040    the compose sequences that combine the two symbols are pointless.
1041    
1042    There are a few layouts that contain U+0256, U+025C, U+025F or U+0279,
1043    but those layouts do not contain `dead_hook`, so the compose sequences
1044    that combine the latter with one of the former are pointless.
1045    
1046    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1047
1048commit 078cc39f05ddfd6166762f3166129258ddd07dd3
1049Author: Benno Schulenberg <bensberg@telfort.nl>
1050Date:   Thu Nov 3 16:31:04 2022 +0100
1051
1052    nls: delete 88 compose sequences that have the accents in the wrong order
1053    
1054    The various compose sequences with the accents in their proper order
1055    (the highest placed accent first) continue to exist.
1056    
1057    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1058
1059commit aa5ef0ba80558f946871f574d38a15673e66221e
1060Author: Benno Schulenberg <bensberg@telfort.nl>
1061Date:   Thu Nov 3 16:27:51 2022 +0100
1062
1063    nls: move the compose sequences for Ș and Ț to the fitting Unicode block
1064    
1065    This addresses a side note of issue #166.
1066    
1067    Reported-by: J. McWilliams
1068    
1069    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1070
1071commit 82c25879b72f1ba7e9abe00e7bad55add33135cf
1072Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1073Date:   Wed Jan 4 11:25:48 2023 -0800
1074
1075    xlibi18n: Update Imake instructions to autoconf equivalent in comments
1076    
1077    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1078
1079commit 88399e01be679bfcc9a5e8922ffe2c47f0e56dee
1080Author: Yuxuan Shui <yshuiv7@gmail.com>
1081Date:   Tue Jan 3 15:09:28 2023 +0000
1082
1083    Revert "Update XPutBackEvent() to support clients that put back unpadded events"
1084    
1085    This reverts commit d6d6cba90215d323567fef13d6565756c9956f60.
1086    
1087    The reverted commit intended to fix the problem where an unpadded X
1088    event struct is passed into XPutBackEvent, by creating a padded struct
1089    with _XEventToWire and _XWireToEvent. However, _XWireToEvent updates the
1090    last sequence number in Display, which may cause xlib to complain about
1091    lost sequence numbers.
1092    
1093    IMO, the problem that commit tried to solve is a bug in the client
1094    library, and workaround it inside Xlib is bad practice, especially given
1095    the problem it caused. Plus, the offender cited in the original commit
1096    message, freeglut, has already fixed this problem.
1097    
1098    Fixes: #176 #174
1099    
1100    Signed-off-by: Yuxuan Shui <yshuiv7@gmail.com>
1101
1102commit 48487c4b9028e9eec5371742f51e1d64c2e17296
1103Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1104Date:   Thu Dec 15 14:52:20 2022 -0800
1105
1106    libX11 1.8.3
1107    
1108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1109
1110commit d6d6cba90215d323567fef13d6565756c9956f60
1111Author: Keith Packard <keithp@keithp.com>
1112Date:   Sun Dec 11 10:32:26 2022 -0800
1113
1114    Update XPutBackEvent() to support clients that put back unpadded events
1115    
1116    It seems to be common practice of some X11 clients to pass specific event
1117    types into APIs that take XEvent*.  For example, freeglut does:
1118    
1119       XConfigureEvent fakeEvent = {0};
1120       ...
1121       XPutBackEvent(fgDisplay.Display, (XEvent*)&fakeEvent);
1122    
1123    This can result in reads overflowing the input event when libX11 does:
1124    
1125       XEvent store = *event;
1126    
1127    =================================================================
1128    ==75304==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x00016ee4a8e8 at pc 0x000101c54d14 bp 0x00016ee4a0d0 sp 0x00016ee49888
1129    READ of size 192 at 0x00016ee4a8e8 thread T0
1130        #0 0x101c54d10 in __asan_memcpy+0x1a4 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x3cd10)
1131        #1 0x102848a18 in _XPutBackEvent PutBEvent.c:41
1132        #2 0x1028490a4 in XPutBackEvent PutBEvent.c:84
1133        #3 0x1013295c8 in fgOpenWindow freeglut_window.c:1178
1134        #4 0x101321984 in fgCreateWindow freeglut_structure.c:108
1135        #5 0x10132b138 in glutCreateWindow freeglut_window.c:1551
1136        #6 0x100fb7d94 in main+0x78 (checkeredTriangles:arm64+0x100003d94)
1137        #7 0x197de3e4c  (<unknown module>)
1138    
1139    Address 0x00016ee4a8e8 is located in stack of thread T0 at offset 840 in frame
1140        #0 0x1013282f8 in fgOpenWindow freeglut_window.c:1063
1141    
1142      This frame has 8 object(s):
1143        [32, 40) 'title.addr'
1144        [64, 176) 'winAttr' (line 1066)
1145        [208, 240) 'textProperty' (line 1067)
1146        [272, 352) 'sizeHints' (line 1068)
1147        [384, 440) 'wmHints' (line 1069)
1148        [480, 672) 'eventReturnBuffer' (line 1070)
1149        [736, 740) 'num_FBConfigs' (line 1072)
1150        [752, 840) 'fakeEvent' (line 1074) <== Memory access at offset 840 overflows this variable
1151    
1152    This change allows XPutBackEvent() to support such clients without
1153    risk of memory read overflow.
1154    
1155    Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1156    Tested-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1157
1158commit 496d9bfeedda89a2211ec665f9214c2693c022f4
1159Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1160Date:   Sat Dec 3 16:26:47 2022 -0800
1161
1162    ximcp: Address warning found by UBSan when growing an empty tree
1163    
1164    imLcPrs.c:681:52: runtime error: applying zero offset to null pointer
1165    
1166    Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
1167
1168commit 7f7bcd7b6f569e9f70e3ddd134924f178e2596b1
1169Author: Takao Fujiwara <tfujiwar@redhat.com>
1170Date:   Tue Dec 6 16:59:56 2022 +0000
1171
1172    nls: consecutive cs number in en_US.UTF-8/XLC_LOCALE
1173
1174commit ba095967349dff583bd0989d0ba78aa334e9bedf
1175Author: Ulrich Sibiller <uli42@gmx.de>
1176Date:   Wed Nov 30 23:47:29 2022 +0100
1177
1178    ChkIfEv.c: fix wrong handling of dpy->in_ifevent
1179    
1180    Is no longer a bool but a counter.
1181    
1182    Signed-off-by: Ulrich Sibiller <uli42@gmx.de>
1183
1184commit 70eaf1174e2809d1a6c43868d53f8cd844ef636a
1185Author: Ulrich Sibiller <uli42@gmx.de>
1186Date:   Wed Nov 30 23:42:02 2022 +0100
1187
1188    Indentation fixes around recent dpy->in_ifevent changes
1189    
1190    Use the same indentation as the surrounding code.
1191    
1192    Signed-off-by: Ulrich Sibiller <uli42@gmx.de>
1193    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1194
1195commit a9e845809bcaae22496bc8aa3ca252b410d5f39b
1196Author: Matthieu Herrb <matthieu@herrb.eu>
1197Date:   Fri Nov 11 18:55:23 2022 +0100
1198
1199    Fix 797755 Allow X*IfEvent() to reenter libX11
1200    
1201    - the activation logic is reversed
1202    - there is also _XInternalLockDisplay() that needs protection
1203    - I've found cases (in fvwm2) where the callback calls XCheckIfEvent()
1204      recursively. So the flag needs to be a counter.
1205    
1206    Reviewed-by: Adam Jackson <ajax@redhat.com>
1207
1208commit bccd787a565d3a88673bfc06574c1939f98d8d72
1209Author: Nia Alarie <nia@NetBSD.org>
1210Date:   Thu Nov 10 22:31:47 2022 +0100
1211
1212    Don't use pragma inside a function, it breaks compiling with older GCCs.
1213    
1214    XKBBind.c:230: error: #pragma GCC diagnostic not allowed inside functions
1215    
1216    Signed-off-by: Thomas Klausner <wiz@gatalith.at>
1217
1218commit 696d19d5db7bcb1c1f582c2b1846520e7e0870cb
1219Author: Oliver <halting@riseup.net>
1220Date:   Fri Nov 11 17:04:00 2022 +0000
1221
1222    Add XFreeThreads function.
1223
1224commit b4f24b272c6ef888b6fcfcf80670c196b2e8f755
1225Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1226Date:   Thu Nov 10 12:18:17 2022 -0800
1227
1228    libX11 1.8.2
1229    
1230    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1231
1232commit 77629ea5e79fa9b1596438bc36cfc7e0c631af43
1233Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1234Date:   Thu Nov 3 13:00:50 2022 -0700
1235
1236    README.md: Add 1.8.2 changes
1237    
1238    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1239
1240commit d1baf2ec4869dea3732d6fe13ba4935cb6caad01
1241Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1242Date:   Thu Nov 3 11:22:37 2022 -0700
1243
1244    Remove NEWS file which only covered 2006 & 2007 releases
1245    
1246    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1247
1248commit 1294bfa487bdecfd32621822922fec03203ee188
1249Author: Po Lu <luangruo@yahoo.com>
1250Date:   Mon Oct 3 08:44:59 2022 +0800
1251
1252    specs: document change in XIMPreeditCallbacks
1253    
1254    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1255
1256commit 1a7e54609ce4dc0f2d3b933502437a413cf3dea8
1257Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1258Date:   Thu Nov 3 14:44:22 2022 -0700
1259
1260    man pages: document XCloseIM frees its argument
1261    
1262    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1263
1264commit a04b84f0f7bb3a78088756759f4b9b8bb363251e
1265Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1266Date:   Thu Nov 3 20:31:31 2022 +0000
1267
1268    Copyright & license cleanup
1269
1270commit 5e41119d9611504ff7e61c6aa4da1df708096801
1271Author: Benno Schulenberg <bensberg@telfort.nl>
1272Date:   Mon Oct 31 12:42:51 2022 +0100
1273
1274    nls: reshuffle a few compose sequences, to have similar ones together
1275    
1276    It makes more sense to have similar sequences grouped together
1277    than to rigidly follow the order of ascending Unicode codes.
1278
1279commit 65d89342f59cfc2eac9bf8010076b937f22554e8
1280Author: Benno Schulenberg <bensberg@telfort.nl>
1281Date:   Mon Oct 31 12:50:11 2022 +0100
1282
1283    nls: remove two compose sequences that use deprecated symbols
1284    
1285    The last few occurrences of `leftcaret` and `rightcaret` were replaced
1286    with `less` and `greater` in xkeyboard-config half a year ago.
1287
1288commit 6baccbae5324e7342cb5176cae82f463a6b34eae
1289Author: Benno Schulenberg <bensberg@telfort.nl>
1290Date:   Mon Oct 31 12:40:12 2022 +0100
1291
1292    nls: let `<Multi_key> <minus> <underscore>` compose U+2212 (MINUS SIGN)
1293    
1294    There was not yet any way to compose the Unicode minus sign, U+2212.
1295
1296commit 79775575418fd6f8ee1c5e5bbe403df4606fb5b6
1297Author: Adam Jackson <ajax@redhat.com>
1298Date:   Fri Aug 5 15:19:08 2022 -0400
1299
1300    Allow X*IfEvent() to reenter libX11
1301    
1302    The documentation for this family of functions very clearly says not to
1303    call into xlib in your predicate function, but historically single
1304    threaded apps could get away with it just fine, and now that we've
1305    forced thread-safety on the world such apps will now deadlock instead.
1306    That's not an acceptable regression even if the app is technically
1307    broken. This has been reported with XFCE and FVWM, and Motif's
1308    cut-and-paste code has the same bug by inspection, so this does need to
1309    be addressed.
1310    
1311    This change nerfs LockDisplay/UnlockDisplay while inside the critical
1312    bit of an IfEvent function. This is still safe in the sense that the
1313    display remains locked and no other thread should be able to change it
1314    from under us, but the loop that scans the event queue might not be
1315    robust against it being modified as a side effect of protocol emitted by
1316    the predicate callback. But that's not new, non-XInitThreads'd xlib
1317    would have the same caveat.
1318    
1319    Closes: xorg/lib/libx11#157
1320
1321commit 0d1d65bdd98966f52bcac4077f94827b20b229dd
1322Author: Benno Schulenberg <bensberg@telfort.nl>
1323Date:   Fri Oct 21 12:18:25 2022 +0200
1324
1325    nls: change <Compose> <^> <-> to mean superscript minus instead of macron
1326    
1327    Several other `<Multi_key> <asciicircum> <symbol>` sequences
1328    produce the superscript equivalent of the given symbol.  So,
1329    let `<Multi_key> <asciicircum> <minus>` do the same.
1330    
1331    Also, add two other sequences for producing a plain macron,
1332    to compensate a bit the loss of the above sequence.
1333    
1334    Additionally, make `<Multi_key> <underscore> <minus>` produce
1335    a subscript minus, for consistency.
1336    
1337    This fixes issue #165.
1338    
1339    Requested-by: J. McWilliams
1340    
1341    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1342
1343commit 310898270787c1a3a6ab50221324f8654dddce8e
1344Author: Jan Tojnar <jtojnar@gmail.com>
1345Date:   Thu Oct 20 07:10:58 2022 +0000
1346
1347    nls: Map sr locales to sr_RS compose files
1348    
1349    Serbian used sr_YU (Yugoslavia) code in the past.
1350    Then it was succeeded by sr_CS (Serbia and Montenegro).
1351    Finally, it was split into sr_RS (Serbia) and sr_ME (Montenegro).
1352    
1353    https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd
1354    introduced the modern sr_RS and sr_ME codes.
1355    
1356    Next, https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/40761898692e5063957bfa2518cca3d35b2e354a
1357    added the Serbian compose table but only for the legacy sr_CS entry.
1358    
1359    https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/5cd60398b7787297008b13a848ed3cfbd7ef178d
1360    removed the legacy sr_CS entry, the only one pointing to the correct compose file.
1361    It also renamed the file to sr_RS, but did not update the compose mapping.
1362    
1363    Let’s point all Serbian locales to the Compose file again.
1364
1365commit a35d706cd8dc7b498bca2a3fc4a3732047439ed7
1366Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1367Date:   Fri Oct 14 13:40:37 2022 -0700
1368
1369    COPYING: remove notice for the removed UIThrStubs.c
1370    
1371    Fixes: 701e9e9a ("Use same pthread-stubs as libxcb")
1372    
1373    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1374
1375commit 3b1750e982fedce0d61bface68aadcd5b1ea1fae
1376Author: Benno Schulenberg <bensberg@telfort.nl>
1377Date:   Fri Oct 14 09:28:42 2022 +0200
1378
1379    nls: delete two compose sequences with an anomalous post-fixed breve
1380    
1381    Two years ago, commit b126bfd7fe allowed using also a lowercase `u`
1382    wherever an uppercase `U` was used to represent a breve.  But the
1383    commit should have limited itself to only the prefixed uses of `U`,
1384    as that is how most letters with a breve are composed.
1385    
1386    Also, group the two compose sequences with an uppercase post-fixed `U`
1387    together with the corresponding other post-fixed sequences.
1388    
1389    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1390
1391commit 701e9e9afb88bdc68c84a4611adc533400a3df36
1392Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1393Date:   Tue Oct 4 11:24:57 2022 -0700
1394
1395    Use same pthread-stubs as libxcb
1396    
1397    Avoid conflicts when libX11 calls libxcb and gets its pthread functions
1398    overriding our ancient stubs.
1399    
1400    v2: Keep linking with real threads libraries when thread safety
1401        constructor is enabled.
1402    
1403    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1404
1405commit 1d11822601fd24a396b354fa616b04ed3df8b4ef
1406Author: Thomas E. Dickey <dickey@invisible-island.net>
1407Date:   Tue Oct 4 18:26:17 2022 -0400
1408
1409    fix a memory leak in XRegisterIMInstantiateCallback
1410    
1411    Analysis:
1412    
1413        _XimRegisterIMInstantiateCallback() opens an XIM and closes it using
1414        the internal function pointers, but the internal close function does
1415        not free the pointer to the XIM (this would be done in XCloseIM()).
1416    
1417    Report/patch:
1418    
1419        Date: Mon, 03 Oct 2022 18:47:32 +0800
1420        From: Po Lu <luangruo@yahoo.com>
1421        To: xorg-devel@lists.x.org
1422        Subject: Re: Yet another leak in Xlib
1423    
1424        For reference, here's how I'm calling XRegisterIMInstantiateCallback:
1425    
1426        XSetLocaleModifiers ("");
1427        XRegisterIMInstantiateCallback (compositor.display,
1428                                        XrmGetDatabase (compositor.display),
1429                                        (char *) compositor.resource_name,
1430                                        (char *) compositor.app_name,
1431                                        IMInstantiateCallback, NULL);
1432    
1433        and XMODIFIERS is:
1434    
1435            @im=ibus
1436    
1437    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
1438
1439commit 1f8fd7ff1cf688ec1d3b34397c6d58110bb0cae5
1440Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1441Date:   Mon Sep 26 16:01:40 2022 -0700
1442
1443    If thread-safety-constructor is enabled, link against pthreads not stubs
1444    
1445    Only really makes a difference if pthreads is not in libc.
1446    
1447    Fixes: #162
1448    
1449    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1450    Reviewed-by: Adam Jackson <ajax@redhat.com>
1451
1452commit bfe5d2dd98f705438b0cdddbbf9c12882da89bae
1453Author: Thomas E. Dickey <dickey@invisible-island.net>
1454Date:   Sun Sep 11 11:44:41 2022 -0400
1455
1456    reduce compiler warnings with casts (no object change)
1457    
1458    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
1459
1460commit 926f2454bca9a7a7d1f4cf0648b0c193f5cef644
1461Author: Thomas E. Dickey <dickey@invisible-island.net>
1462Date:   Sun Sep 11 10:35:31 2022 -0400
1463
1464    use casts to reduce compiler warnings (no object change)
1465    
1466    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
1467
1468commit dc553ff908f10832c71dc2b5ba5af7e68589218e
1469Author: Mike FABIAN <mfabian@redhat.com>
1470Date:   Fri Jul 22 11:39:01 2022 +0200
1471
1472    Remove KOI8-R character set from en_US.UTF-8/XLC_LOCALE
1473    
1474    https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/master/src/xlibi18n/lcCT.c#L58
1475    has no escape sequence for KOI8-R.
1476    That makes Xutf8TextListToTextProperty() sometimes fail when trying to
1477    convert to COMPOUND_TEXT and when the KOI8-R charset is tried.
1478    
1479    This is the cause for this bug:
1480    
1481    https://github.com/ibus/ibus/issues/2422
1482    
1483    Removing the KOI8-R charset entry fixes the problem.
1484    
1485    This commit also fixes a few wrong texts in comments.
1486    
1487    Signed-off-by: Mike FABIAN <mfabian@redhat.com>
1488
1489commit 1c04ac514196db5e2255d99635fb6e3c36be330d
1490Author: Adam Sampson <ats@offog.org>
1491Date:   Sun Jul 4 23:02:04 2021 +0100
1492
1493    xkb: Fix off-by-one error in XKeycodeToKeysym
1494    
1495    The code here that made indexes greater than 3 refer to XKB symbol
1496    groups had an off-by-one error, so it would always leave out the symbol
1497    that should have been at index 4. Rewrite the code to fix this and
1498    simplify the logic a bit.
1499    
1500    Signed-off-by: Adam Sampson <ats@offog.org>
1501
1502commit ddaacd219609104f138006db6a8f708226db3874
1503Author: Thomas E. Dickey <dickey@invisible-island.net>
1504Date:   Fri Sep 9 20:17:35 2022 -0400
1505
1506    reduce compiler warnings for macros BufAlloc, Data and Data32 using casts
1507    
1508    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
1509
1510commit 62c3337d89d31e0d3ed807004e73ad711fad3342
1511Author: Kirill Chibisov <contact@kchibisov.com>
1512Date:   Thu Sep 8 22:50:30 2022 +0000
1513
1514    ximcp/imRm.c: allow XNSpotLocation with OnTheSpot
1515
1516commit 1272879074ad9612314858a75cf844783c1963b2
1517Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1518Date:   Thu Jul 28 17:30:21 2022 -0700
1519
1520    gitlab CI: stop requiring Signed-off-by in commits
1521    
1522    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1523
1524commit 38033b073f393564acf6a04501927002b2b4aa86
1525Author: David H. Gutteridge <david@gutteridge.ca>
1526Date:   Mon Jul 25 19:30:19 2022 -0400
1527
1528    configure.ac: report thread safety constructor build status
1529    
1530    Signed-off-by: David H. Gutteridge <david@gutteridge.ca>
1531
1532commit 9a90b3a63d02da92dc8692ed3f6269d3e0c7c8cc
1533Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1534Date:   Sun Jul 10 11:55:11 2022 -0700
1535
1536    doc: Document possible return values for XkbGetKeyboard()
1537    
1538    Fixes: #160
1539    
1540    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1541
1542commit 0beb0cf46ae48b0d4322ff8a0221f47bdb7d2296
1543Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1544Date:   Sun Jul 10 11:42:19 2022 -0700
1545
1546    xkb: set num_keys when filling in keys in _XkbReadGetNamesReply()
1547    
1548    Fixes: #160
1549    
1550    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1551
1552commit 487e9489d1e086cc62c53045206ab6d6d9d02031
1553Author: Jon Turney <jon.turney@dronecode.org.uk>
1554Date:   Wed Jun 22 16:15:37 2022 +0100
1555
1556    Fix Win32 build with -fno-common
1557    
1558    /work/xorg/lib/libX11/src/XlibInt.c:1968: multiple definition of `_Xdebug_p'; .libs/globals.o:globals.c:(.bss+0xc): first defined here
1559    
1560    Avoid redundant definition of _Xdebug_p in globals.c (which is under the
1561    influence of _Xdebug being #defined to _Xdebug_p.
1562    
1563    This appears to be an ancient hack to work around data exports resolving
1564    to the address of the import stub, not the import. (See [1]).
1565    
1566    (This is probably no longer needed or can be done in a better way, as
1567    per the discussion under --enable-auto-import in the ld manpage.)
1568    
1569    [1] https://cygwin.com/pipermail/cygwin-xfree/2001-May/004606.html
1570    
1571    Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
1572
1573commit 35ad99fad2f65ec2c69bd84bdeb8138329092b57
1574Author: Benno Schulenberg <bensberg@telfort.nl>
1575Date:   Fri Jul 15 11:29:46 2022 +0200
1576
1577    nls: add four sequences for the Samogitian E with dot above and macron
1578    
1579    These sequences each produce two code points: the E-with-dot-above and
1580    a combining macron.  The XIM input method is required for this to work.
1581    
1582    (Also add a missing comment for a Unicode block.)
1583    
1584    This fixes issue #54.
1585    
1586    Requested-by: Arns Udovīčė
1587    
1588    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1589
1590commit 2732dd20cc2694dd5c32fef292e1f48626011032
1591Author: Benno Schulenberg <bensberg@telfort.nl>
1592Date:   Wed Jul 6 16:44:50 2022 +0200
1593
1594    nls: delete eight sequences that pointlessly mix upper and lower case
1595    
1596    Typing a compose sequence requires some care -- surely the user is able
1597    to either keep holding the Shift key or not touch it at all while typing
1598    the sequence.  Also, compose sequences are not an infinite resource AND
1599    take up space and time -- defining redundant ones is a waste.
1600    
1601    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1602
1603commit ea3cc474e1acfed47b9fb856ea151008fa6cfd30
1604Author: Benno Schulenberg <bensberg@telfort.nl>
1605Date:   Wed Jul 6 16:36:48 2022 +0200
1606
1607    nls: add compose sequences for the double-struck capitals ℕ ℤ ℚ ℝ ℂ
1608    
1609    This allows the user to type the symbols for the five number systems.
1610    
1611    This fixes the reasonable part of issue #159.
1612    
1613    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1614
1615commit 95a03cde434c447fdb41d8cccbbdd41f20750f32
1616Author: Benno Schulenberg <bensberg@telfort.nl>
1617Date:   Thu Jul 14 12:00:45 2022 +0200
1618
1619    nls: delete some pointless and wrong compose sequences meant for Bépo
1620    
1621    Fourteen years ago, commit 7302984642 added some four hundred compose
1622    sequences for the benefit of the French Bépo layout.  But among these
1623    four hundred there are several that use symbols that are not available
1624    in the Bépo layout and are thus impossible to type.  Some of the used
1625    symbols, like Ahook, Ehook and Ohook, are not even present in *any*
1626    layout, making these sequences a dead weight in the Compose file.
1627    
1628    The Amacron and Omacron are available only in the Latvian, Hawaiian,
1629    and Maori layouts, and the Omacron also in the Silesian layout.  But
1630    the Latvian layouts and the Hawaiian do not contain any dead keys.
1631    Only in the Maori and Silesian layouts these sequences with Amacron
1632    and Omacron could be typed, but that was not why they were added.
1633    
1634    More importantly, as James Cloos noted in issue #54, sequences like
1635    `<dead_abovedot> <amacron>` for generating `ǡ` (that is: the macron
1636    above the dot) are questionable, as in compose sequences generally
1637    the first accent typed is the uppermost in the composed character.
1638    
1639    Reference:
1640      https://gitlab.freedesktop.org/xorg/lib/libx11/-/issues/54#note_17321
1641    
1642    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1643
1644commit 9702b836862da6838d725f555e7a1d83faf51740
1645Author: Matthieu Herrb <matthieu@herrb.eu>
1646Date:   Fri Jul 1 11:13:13 2022 +0200
1647
1648    Mention that the predicate function is called with the display lock
1649    
1650    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
1651
1652commit a5d00f889fadad3f5b11e53fd5c04cf5343fa8a2
1653Author: Benno Schulenberg <bensberg@telfort.nl>
1654Date:   Fri Jun 24 16:20:52 2022 +0200
1655
1656    docs: replace three placeholders with something that makes sense
1657    
1658    Also, uncapitalize two arguments, to match the style of all others.
1659    
1660    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1661
1662commit 9ff8502040e94cb172426d59df722cefa752a401
1663Author: Benno Schulenberg <bensberg@telfort.nl>
1664Date:   Fri Jun 24 16:07:22 2022 +0200
1665
1666    docs: remove the unsightly dashes from the overviews of arguments
1667    
1668    Seventeen months ago, commits 78027fdb7a and 4f15cfc645 removed
1669    these dashes from two of the man pages.  Now, remove them all.
1670    They are unhelpful and just make one wonder why they are there
1671    (probably to function as improvised bullet points).
1672    
1673    Also remove four leading spaces and a trailing comma.
1674    
1675    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1676
1677commit a7968c47e35703446d67d070234fd9adf07f34b0
1678Author: Benno Schulenberg <bensberg@telfort.nl>
1679Date:   Thu Jun 23 15:48:05 2022 +0200
1680
1681    docs: hard-wrap some items so that Table 2 fits within 80 characters
1682    
1683    (Table 1 hard-wraps the first-column items in the same way.)
1684    
1685    Also, correct the formatting of the subsequent paragraph.
1686    
1687    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1688
1689commit c601c779ca8f46dc4335180ec2829851fb8bbc13
1690Author: Benno Schulenberg <bensberg@telfort.nl>
1691Date:   Wed Jun 15 10:55:38 2022 +0200
1692
1693    nls: delete compose sequences that mix top-row digits with numpad digits
1694    
1695    Compose sequences for circled numbers, like ⑫  or ㉑, are nice to have,
1696    but allowing them to be composed by typing one digit on the top row and
1697    the other on the numerical keypad (or the other way around) is over the
1698    top.  Remove these absurd sequences.  Keep only the sequences where both
1699    digits are either on the top row or on the numerical keypad.
1700    
1701    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1702
1703commit 3a30ada60c5217ada37b143b541c8e6f6284c7fa
1704Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1705Date:   Wed Jun 8 17:30:14 2022 -0700
1706
1707    libX11 1.8.1
1708    
1709    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1710
1711commit 70f7403fd3bf362fc50b8de4a2e26300f757c68e
1712Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1713Date:   Tue May 3 14:45:05 2022 -0700
1714
1715    configure.ac: Fix --enable-thread-safety-constructor
1716    
1717    Prior to this, --enable-thread-safety-constructor would disable it,
1718    while --disable worked as expected, and no option left it enabled as
1719    expected by default.  This also fixes the --help text to be displayed.
1720    
1721    Fixes: afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9
1722    Reported-by: @igor.v.kovalenko
1723    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1724    Reviewed-by: Adam Jackson <ajax@redhat.com>
1725
1726commit d0da5a1e0fb214ff911ecff86d35f6673ff70023
1727Author: Adam Jackson <ajax@redhat.com>
1728Date:   Fri Apr 29 11:27:44 2022 -0400
1729
1730    libX11 1.8
1731    
1732    Minor number bump to reflect the thread safety default change.
1733    
1734    Signed-off-by: Adam Jackson <ajax@redhat.com>
1735
1736commit afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9
1737Author: Adam Jackson <ajax@redhat.com>
1738Date:   Tue Mar 22 18:24:29 2022 -0400
1739
1740    global: call XInitThreads() from the library's constructor
1741    
1742    There is really no point in not being thread safe, I measured, all you
1743    can see happen is noop performance gets like twice as slow and you have
1744    thread safety bugs. And we're using xcb as the transport which means we
1745    should expect threads in our clients anyway. Just do it.
1746    
1747    This assumes your compiler understands __attribute__((constructor)). If
1748    this is not your compiler, you can disable this with the appropriate
1749    configure flag, but be aware you're asking for bugs.
1750    
1751    Signed-off-by: Adam Jackson <ajax@redhat.com>
1752
1753commit 1ab48f3cc966751d86e434808180beb2dc76c4fa
1754Author: Benno Schulenberg <bensberg@telfort.nl>
1755Date:   Sun Apr 24 15:06:45 2022 +0200
1756
1757    docs: add release notes for versions 1.7.3, 1.7.4, and 1.7.5
1758    
1759    They were forgotten when those releases were made.
1760    
1761    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1762
1763commit 35d1513bc7fdd3ac6f5807feb601efc34ac19163
1764Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1765Date:   Sat Apr 16 13:04:23 2022 -0700
1766
1767    XkbOpenDisplay: Add _Xconst qualifier to display name argument
1768    
1769    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1770
1771commit c88ceaad4aaa0439e5cba13c1f5737ad54566e0b
1772Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1773Date:   Sat Apr 16 12:59:39 2022 -0700
1774
1775    XKBgeom.h: Add _Xconst qualifier to char * arguments in functions
1776    
1777    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1778
1779commit 0ce4425826074b12a58ced432f9d6fb346474737
1780Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1781Date:   Sat Apr 16 12:20:51 2022 -0700
1782
1783    XKBgeom.h: Remove XKB_IN_SERVER section
1784    
1785    The Xserver made its own copy of this file in 2008, and the API's are
1786    no longer the same between the server and client forks.
1787    
1788    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1789
1790commit be8c01946ab1a1d4b79fdc5358541d630f14dc0a
1791Author: Benno Schulenberg <bensberg@telfort.nl>
1792Date:   Wed Apr 20 09:29:09 2022 +0200
1793
1794    lcCT: use the correct index for checking the second byte
1795    
1796    (Not that it will make any difference, as the checking of these
1797    high bits looks like an excess of precaution.)
1798    
1799    This fixes issue #134.
1800    
1801    Reported-by: Rafał Mikrut
1802    
1803    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1804
1805commit 488b156fe2cc8aca6946a49236ec7b7698fceda4
1806Author: Benno Schulenberg <bensberg@telfort.nl>
1807Date:   Fri Feb 25 16:35:46 2022 +0100
1808
1809    nls: make the Ethiopian compose sequences actually work
1810    
1811    In the Ethiopian keyboard layout, the dead vowel keys do not produce <e>
1812    and <u> and so on, but instead produce <U+FE69> and <U+FE75> and so on,
1813    so the compose sequences should use those latter code points.
1814    
1815    Also, include the basic compose sequences from en_US.UTF-8/Compose,
1816    so that, when switching to a different layout in the Ethiopian locale,
1817    all the usual compose sequences work too.
1818    
1819    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1820
1821commit e6f98a3dbd82928e10adaf9d034dc574af1de6e5
1822Author: Benno Schulenberg <bensberg@telfort.nl>
1823Date:   Mon Apr 18 17:24:55 2022 +0200
1824
1825    nls: add Multi-key sequences for abovedot, diaeresis, and ogonek
1826    
1827    These accents by themselves could only be produced when one had a
1828    dead key for them, not with the help of the Multi key.
1829    
1830    [Note that the sequences <dead_acute> <space> for apostrophe (')
1831    and <dead_diaeresis> <space> for double quote (") are anomalies,
1832    as normally <dead_accent> <space> produces the accent itself.]
1833    
1834    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1835
1836commit d323704e7c57f9ac7fdfea3dc9fa73bf943e72d3
1837Author: Benno Schulenberg <bensberg@telfort.nl>
1838Date:   Mon Apr 18 16:57:35 2022 +0200
1839
1840    nls: remove eight pointless self-producing compose sequences
1841    
1842    Compose sequences are meant to produce certain symbols by combining
1843    certain different symbols, not to produce a symbol with the help of
1844    the symbol itself.
1845    
1846    This fixes issue #59.
1847    
1848    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1849
1850commit 3b62c0322c112c03597d850197dbce22d1830bb7
1851Author: Benno Schulenberg <bensberg@telfort.nl>
1852Date:   Fri Feb 25 14:12:02 2022 +0100
1853
1854    nls: remove misplaced sequences from Ethiopian and Greek compose files
1855    
1856    The Khmer digraphs and Arabic ligatures have nothing to do with
1857    Amharic or Greek.
1858    
1859    (Also rewrap a comment and correct two others.)
1860    
1861    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1862
1863commit 479af78f2e5caaf9b98a35ae5ff4ccdd2f040289
1864Author: Benno Schulenberg <bensberg@telfort.nl>
1865Date:   Fri Feb 25 14:05:35 2022 +0100
1866
1867    nls: remove redundant Khmer digraphs, already defined in included file
1868    
1869    This fixes issue #160.
1870    
1871    Reported-by: Mike Fabian
1872    
1873    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1874
1875commit 6d7d08726f4b0c517041842b27cd7e66e8f371eb
1876Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1877Date:   Sun Apr 10 14:51:55 2022 -0700
1878
1879    XkbUpdateKeyTypeVirtualMods: always initialize mask
1880    
1881    XkbVirtualModsToReal should only fail to set mask if the server does
1882    not support XKB, but it still made Oracle Parfait complain:
1883    
1884    Error: Uninitialised memory
1885       Uninitialised memory variable [uninitialised-mem-var] (CWE 457):
1886          Possible access to uninitialised memory referenced by variable 'mask'
1887            at line 863 of lib/libX11/src/xkb/XKBMisc.c in function 'XkbUpdateKeyTypeVirtualMods'.
1888            Path in callee avoiding write at line 862
1889              mask allocated at line 860
1890    
1891    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1892
1893commit 9ac6859c20be2fc5e70c2908de60c6e466ec04e1
1894Author: Matt Turner <mattst88@gmail.com>
1895Date:   Sun Apr 3 10:15:30 2022 -0700
1896
1897    libX11 1.7.5
1898    
1899    Signed-off-by: Matt Turner <mattst88@gmail.com>
1900
1901commit 76d1cc3c1ce943c6ff81dc8c62a1d1b30fabf02e
1902Author: Simon McVittie <smcv@debian.org>
1903Date:   Sun Apr 3 14:23:36 2022 +0100
1904
1905    Don't try to destroy NULL condition variables
1906    
1907    This avoids a segfault during error-unwinding if an invalid display name
1908    is passed to XOpenDisplay().
1909    
1910    Fixes: 8a368d80 "Fix two memory leaks in _XFreeX11XCBStructure()"
1911    Resolves: #155
1912    Signed-off-by: Simon McVittie <smcv@debian.org>
1913
1914commit 80b30d12519ac54fc5b2b05500521d9c8269be9c
1915Author: Matt Turner <mattst88@gmail.com>
1916Date:   Wed Mar 30 13:54:13 2022 -0700
1917
1918    libX11 1.7.4
1919    
1920    Signed-off-by: Matt Turner <mattst88@gmail.com>
1921
1922commit 8a368d808fec166b5fb3dfe6312aab22c7ee20af
1923Author: Hodong <hodong@yozmos.com>
1924Date:   Thu Jan 20 00:57:41 2022 +0900
1925
1926    Fix two memory leaks in _XFreeX11XCBStructure()
1927    
1928    Even when XCloseDisplay() was called, some memory was leaked.
1929    
1930    XCloseDisplay() calls _XFreeDisplayStructure(), which calls
1931    _XFreeX11XCBStructure().
1932    
1933    However, _XFreeX11XCBStructure() did not destroy the condition variables,
1934    resulting in the leaking of some 40 bytes.
1935    
1936    Signed-off-by: Hodong <hodong@yozmos.com>
1937
1938commit da97120f2322882a17f14e5d7da00c1e772679e8
1939Author: Adam Jackson <ajax@redhat.com>
1940Date:   Tue Mar 22 14:32:49 2022 -0400
1941
1942    xcb: Clarify the XInitThreads error message
1943    
1944    XInitThreads has been called if _Xglobal_lock != NULL, we may as well
1945    check that before printing a misleading error message.
1946    
1947    Signed-off-by: Adam Jackson <ajax@redhat.com>
1948
1949commit 257927c51b08242aa5bf239346717fc817b2b286
1950Author: Giovanni Mascellani <gmascellani@codeweavers.com>
1951Date:   Fri Feb 4 10:49:25 2022 +0100
1952
1953    xcb_io: Allow jumps backwards when widening the request number.
1954    
1955    Request numbers are not always seen in the numeric order by widen(),
1956    for example due to Mesa directly calling _XError(). When this happens,
1957    widen() adds 2^32 to the reported widened number, triggering failed
1958    assertions and bad behavior.
1959    
1960    With this commit, wrapping of the lower dword is detected in a more
1961    robust way, by requiring that a skip of at least 2^31 is seen.
1962    
1963    This fixes issue #152.
1964    
1965    Signed-off-by: Giovanni Mascellani <gmascellani@codeweavers.com>
1966
1967commit 918063298cb893bee98040c9dca45ccdb2864773
1968Author: Benno Schulenberg <bensberg@telfort.nl>
1969Date:   Thu Mar 3 12:00:45 2022 +0100
1970
1971    nls: add a compose sequence for the peace symbol
1972    
1973    As the thing in the circle looks a bit like an upside-down Y,
1974    use <O> <Y> as the sequence, similar to <O> <A> for anarchism.
1975    
1976    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1977
1978commit 4f9eb32709dca705f02ca771bbd47d6c08aba5e0
1979Author: Benno Schulenberg <bensberg@telfort.nl>
1980Date:   Fri Feb 25 12:45:06 2022 +0100
1981
1982    nls: remove obsolete locale aliases with uppercase in the language code
1983    
1984    (These aliases may have been useful on some systems twenty years ago,
1985    but nowadays all should have settled on lowercase language code plus
1986    uppercase country code.)
1987    
1988    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1989
1990commit ba0d3b437deb2212bc7fc7c6f1c603fa6ced443c
1991Author: Benno Schulenberg <bensberg@telfort.nl>
1992Date:   Fri Feb 25 12:35:28 2022 +0100
1993
1994    nls: map eo.UTF-8 directly to en_US.UTF-8 instead of using a relay alias
1995    
1996    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1997
1998commit cfddb79981f23650feb0f9a0f4442c9cad584b33
1999Author: Benno Schulenberg <bensberg@telfort.nl>
2000Date:   Fri Feb 25 12:12:05 2022 +0100
2001
2002    nls: remove the mistaken eo_EO names -- EO is not a valid country code
2003    
2004    (Debian has had several of these lines commented out for years.)
2005    
2006    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2007
2008commit eb5885245a1ba6955d058a03e31894c23b9f5490
2009Author: iyzana <iyzana@protonmail.com>
2010Date:   Mon Feb 21 13:20:23 2022 +0100
2011
2012    nls: add locale alias from eo.UTF-8 to eo_XX.UTF-8
2013    
2014    There is no matching locale in the locale.dir and compose.dir files for
2015    the locale eo.UTF-8. Setting it as the locale resulted in compose files
2016    not being loaded.
2017    
2018    Signed-off-by: iyzana <iyzana@protonmail.com>
2019
2020commit 24df23dff9d8c1cf5e918a5f040b852bdf2cfb84
2021Author: Benno Schulenberg <bensberg@telfort.nl>
2022Date:   Thu Feb 3 17:19:05 2022 +0100
2023
2024    nls: group the compose sequences for tilde, degree, and cedilla together
2025    
2026    Also, correct the comment for the double quote character ("), and move
2027    the soft hyphen and the interrobangs to a special punctuation group.
2028    
2029    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2030
2031commit c0191f020f94907770937bc8ae67ea8acd7d63be
2032Author: Benno Schulenberg <bensberg@telfort.nl>
2033Date:   Thu Feb 3 16:33:05 2022 +0100
2034
2035    nls: use shape-based compose sequences for hammer-and-sickle instead
2036    
2037    Using a misspelled abbreviation (CCCP) of an obsolete country's name
2038    to compose a fairly general communist symbol does not seem right.
2039    
2040        https://en.wikipedia.org/wiki/Hammer_and_sickle
2041    
2042    Many compose sequences instead combine characters that together look
2043    similar to the target character.  Do the same here: question mark (?)
2044    plus backslash (\) look a bit like ☭ .
2045    
2046    In the bargain, this fixes issue #63.
2047    
2048    Reported-by: Marc Mezzarobba
2049    
2050    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2051
2052commit 582dc6f89e1f9288710a55cb2b8fbf2af99d7616
2053Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2054Date:   Sun Feb 20 11:49:42 2022 -0800
2055
2056    gitlab CI: enable Static Application Security Testing (SAST)
2057    
2058    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2059
2060commit 336e8f72559b489396c5cd503c4586ef800c2a8f
2061Author: Benno Schulenberg <bensberg@telfort.nl>
2062Date:   Sun Feb 20 15:06:02 2022 +0100
2063
2064    nls: add compose sequences for leftward, upward and downward double arrow
2065    
2066    These sequences follow the logic of: the arrow shaft (=) first, the
2067    arrow head (< or > or ^ or v) second.  This is so because the sequence
2068    <= is already taken for ≤, and there is no obvious ASCII character for
2069    a double vertical line, so we have to make do with a horizontal one.
2070    
2071    Since commit 6101b967b6 from eight years ago, there is a compose sequence
2072    for the rightward double arrow, but not for the other three directions.
2073    
2074    This fixes issue #138.
2075    
2076    Requested-by: Mélanie Chauvel
2077    
2078    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2079
2080commit 623b77d4f30b47258a40f89262e5aa5d25e95fa7
2081Author: Benno Schulenberg <bensberg@telfort.nl>
2082Date:   Mon Feb 14 11:33:25 2022 +0100
2083
2084    imDefLkup: verify that a pointer isn't NULL before using it
2085    
2086    It is possible for _XimICOfXICID() to return NULL, so it is necessary
2087    to check this isn't actually the case before dereferencing the pointer.
2088    All other callers of _XimICOfXICID() do this check too.
2089    
2090    (The check itself is ugly, but it follows the style of the code in the
2091    rest of the module.)
2092    
2093    Fixes issue #45.
2094    
2095    Reported-by: Bhavi Dhingra
2096    
2097    Original-patch-by: Bhavi Dhingra
2098    
2099    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2100
2101commit b83938f7f97aa6210b00835002e6e7d614b2374a
2102Author: Christopher Chavez <chrischavez@gmx.us>
2103Date:   Mon Feb 14 01:00:45 2022 +0000
2104
2105    Xlib.h: fix spelling in comment
2106    
2107    Signed-off-by: Christopher Chavez <chrischavez@gmx.us>
2108
2109commit 4791055682552b4cc8e775479700fb06980d0b59
2110Author: Christopher Chavez <chrischavez@gmx.us>
2111Date:   Sun Feb 13 05:26:14 2022 +0000
2112
2113    Xlib.h: spelling fix in comment
2114
2115commit 9304e645bdcdd3a71ff449a1157c4815e6304d7a
2116Author: Denis Drakhnia <numas13@gmail.com>
2117Date:   Wed Feb 2 11:54:43 2022 +0200
2118
2119    nls: Add ru_UA.utf8 locale alias.
2120    
2121    Signed-off-by: Denis Drakhnia <numas13@gmail.com>
2122
2123commit 9ef2551f0c201f9163b66d17189a98076b5be07a
2124Author: Benno Schulenberg <bensberg@telfort.nl>
2125Date:   Wed Feb 2 16:50:35 2022 +0100
2126
2127    nls: do some manual adjustment of whitespace in the compose sequences
2128    
2129    This achieves a more consistent vertical alignment (per group) of the
2130    target character and the trailing comments.
2131    
2132    (Most blocks were aligned in some fashion, except the "Greek Extended"
2133    block, which has lines of greatly varying lengths.)
2134    
2135    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2136
2137commit 0dad289ead6eb3412027c9365192441cc539d747
2138Author: Benno Schulenberg <bensberg@telfort.nl>
2139Date:   Wed Feb 2 14:08:22 2022 +0100
2140
2141    nls: among the compose sequences, replace all runs of spaces with tabs
2142    
2143    It was done with this script:
2144    
2145        cat nls/en_US.UTF-8/Compose.pre |
2146        sed 's!\(^<.\{6,13\}>\)\s*:!\1\t\t\t\t:!' |
2147        sed 's!\(^<.\{14,21\}>\)\s*:!\1\t\t\t:!' |
2148        sed 's!\(^<.\{22,29\}>\)\s*:!\1\t\t:!' |
2149        sed 's!\(^<.\{30,37\}>\)\s*:!\1\t:!' |
2150        sed 's!\(: ".*"\)\s*!\1\t!' |
2151        sed 's!\(\s*# \)! # !' > trimmed &&
2152        mv  trimmed  nls/en_US.UTF-8/Compose.pre
2153    
2154    This saves 37 kilobytes of whitespace.
2155    
2156    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2157
2158commit fc79e39c236501357e1fdbe8964882864eeef5d3
2159Author: Benno Schulenberg <bensberg@telfort.nl>
2160Date:   Wed Feb 2 14:06:29 2022 +0100
2161
2162    docs: add missing word, wrap line, fix typo, and trim an excess space
2163    
2164    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2165
2166commit 93446455c46dfc019984ab822ffa662cf6e2949c
2167Author: Benno Schulenberg <bensberg@telfort.nl>
2168Date:   Thu Jan 27 15:01:18 2022 +0100
2169
2170    KeyBind: reshuffle two ifs into the general order of ascending codes
2171    
2172    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2173
2174commit 93f15f5f29378ec4cc1dd011b155104e9a52b058
2175Author: Benno Schulenberg <bensberg@telfort.nl>
2176Date:   Thu Jan 27 14:27:41 2022 +0100
2177
2178    KeyBind: slightly speed up some case conversions by adding missing returns
2179    
2180    This avoids pointlessly tumbling through several more ifs when the
2181    conversion has aleady been done.
2182    
2183    Also remove two redundant conditions (as lower codes have already
2184    been handled) and fold two other conditions together.
2185    
2186    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2187
2188commit 317c7abbcc848011393d1336b9d020af8cabb000
2189Author: Benno Schulenberg <bensberg@telfort.nl>
2190Date:   Thu Jan 27 13:54:29 2022 +0100
2191
2192    KeyBind: update the Greek case-conversion tables to Unicode Data 14.0
2193    
2194    A few symbols, like Heta and Sampi and dotted lunate Sigma, have been
2195    assigned code points since Unicode Data 4.0, and need their entries.
2196    
2197    This fixes issue #132.
2198    
2199    Reported-by: Ray Vine
2200    
2201    Original-patch-by: Ray Vine
2202    
2203    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2204
2205commit 402b843fa72166d9ceefd887f5cea4728cbc6995
2206Author: Benno Schulenberg <bensberg@telfort.nl>
2207Date:   Wed Jan 26 17:20:28 2022 +0100
2208
2209    remove a commented-out code fragment, and remove a stray blank line
2210    
2211    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2212
2213commit fbbc3cf8b8b03f74f778772d4703fc4fec242282
2214Author: Benno Schulenberg <bensberg@telfort.nl>
2215Date:   Wed Jan 26 16:55:50 2022 +0100
2216
2217    XKBCvt: remove an if that never gets triggered
2218    
2219    The 'if' at the beginning of _XkbHandleSpecialSym() allows only symbols
2220    from the numeric keypad and some control symbols to pass -- XK_hyphen
2221    is not among them, so the check for XK_hyphen in the later 'if' will
2222    always be false.
2223    
2224    (The corresponding conversion in _XTranslateKeySym() in KeyBind.c was
2225    removed in March 1994.)
2226    
2227    (Also, several keyboard layouts nowadays contain the 'hyphen' symbol,
2228    allowing the user to enter soft hyphens into a document.  So we really
2229    don't want to remap this symbol.)
2230    
2231    Fixes issue #48.
2232    
2233    Reported-by: Bhavi Dhingra
2234    
2235    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2236
2237commit 6786744f636b3fc545157a13cb2f6169b5be9fa5
2238Author: Benno Schulenberg <bensberg@telfort.nl>
2239Date:   Sun Jan 23 17:12:03 2022 +0100
2240
2241    nls: add the es_CU locale name, for Cuba
2242    
2243    It was added to glibc ten years ago.
2244    
2245    This fixes issue #57.
2246    
2247    Reported-by: Dominique Michel
2248    
2249    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2250
2251commit d241d8af7f24e95949c2cf7ce791e35df7665e98
2252Author: Benno Schulenberg <bensberg@telfort.nl>
2253Date:   Sun Jan 23 16:44:03 2022 +0100
2254
2255    nls: remove some obsolete Norwegian and Yugoslavian locale names
2256    
2257    They have not existed in glibc for more than fifteen years.
2258    
2259    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2260
2261commit 32dd7c03e92e01015744867b9b3148c250cc159a
2262Author: Benno Schulenberg <bensberg@telfort.nl>
2263Date:   Sat Jan 22 17:56:32 2022 +0100
2264
2265    nls: put some entries in their alphabetical order, and adjust alignment
2266    
2267    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2268
2269commit 46a066f25103bb909259baea25de8dd5c4a2b338
2270Author: Benno Schulenberg <bensberg@telfort.nl>
2271Date:   Sat Jan 22 18:00:20 2022 +0100
2272
2273    nls: fix a typo (be_BG => bg_BG) and drop an obsolete comment
2274    
2275    The typo was only partially fixed in commit 22a5255b80 sixteen years ago,
2276    and the internet shows that there was at least one user frustrated that
2277    bg_BG.utf8 didn't work but bg_BG.UTF-8 did.
2278    
2279    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2280
2281commit 84efc0aa598c8f7d6dd9a49e7450e85c8bc4773b
2282Author: Benno Schulenberg <bensberg@telfort.nl>
2283Date:   Wed Jan 19 15:32:15 2022 +0100
2284
2285    nls: remove the compose sequences with the deviant KP_Space symbol
2286    
2287    Using this KP_Space symbol as a stand-in for the symbol 2 doesn't make
2288    sense.  It looks like a mistake, or as if someone had a broken keyboard
2289    and used KP_Space as a substitute for 2.
2290    
2291    Also, no keyboard layout in the last fifteen years has contained the
2292    KP_Space symbol, so I don't see how anyone could type it.
2293    
2294    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2295
2296commit b76e661ca223489da9027b0ab1ca1586efd0a280
2297Author: Benno Schulenberg <bensberg@telfort.nl>
2298Date:   Wed Jan 19 13:49:56 2022 +0100
2299
2300    nls: move the block of Jamo to its logical place (ascending Unicodes)
2301    
2302    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2303
2304commit 21f82fb9c69f8fdbf0bdc1aac6f8c11c2c6d55d1
2305Author: Benno Schulenberg <bensberg@telfort.nl>
2306Date:   Wed Jan 19 13:40:56 2022 +0100
2307
2308    nls: delete some ineffective compose sequences (that were commented out)
2309    
2310    They have never worked; there is no point in keeping them in the file.
2311    
2312    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2313
2314commit 6cc31eafa8095d2117ff1906d7c14c1c01173248
2315Author: Benno Schulenberg <bensberg@telfort.nl>
2316Date:   Wed Jan 19 12:18:21 2022 +0100
2317
2318    nls: group some compose sequences better, and add some comments
2319    
2320    Also, align a few trailing comments more consistently, and change
2321    some comments so they will be excluded from the installed file --
2322    there is no need for those comments there.
2323    
2324    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2325
2326commit d60ede7843612f3c2967f83d629bf8272ba09ff1
2327Author: ArenaL5 <arenal5@protonmail.com>
2328Date:   Thu Jan 13 01:13:56 2022 +0100
2329
2330    nls: add a compose sequence for the copyleft symbol
2331    
2332    Probably not used anywhere yet, let alone seriously, as stated in
2333    https://en.wikipedia.org/wiki/Copyleft#Symbol
2334    
2335    Signed-off-by: ArenaL5 <arenal5@protonmail.com>
2336
2337commit df4dcdf9125c595f6df3a302356647cb8df38553
2338Author: Benno Schulenberg <bensberg@telfort.nl>
2339Date:   Mon Jan 17 17:31:08 2022 +0100
2340
2341    nls: remove redundant compose sequences with deprecated Hebrew symbols
2342    
2343    The compose sequences with the valid Hebrew key symbols are right there.
2344    
2345    Also, no keyboard layout in the past seventeen years has contained these
2346    deprecated symbols on any key.
2347    
2348    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2349
2350commit 4268472bd11a07a799716fb5833ea46cad3dd479
2351Author: Benno Schulenberg <bensberg@telfort.nl>
2352Date:   Sun Jan 16 17:51:29 2022 +0100
2353
2354    nls: remove redundant compose sequences with the deprecated 'underbar'
2355    
2356    The key symbol 'underbar' is equivalent to 'underscore' -- the mirror
2357    compose sequences with the latter symbol continue to exist.
2358    
2359    Also, no keyboard layout in the past fifteen years has contained the
2360    symbol 'underbar' on any key -- except a Telugu layout since a few
2361    years, but that will be corrected soon.
2362    
2363    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2364
2365commit 86e9749ef37189486c205ee7f77e43fed655ac70
2366Author: ArenaL5 <arenal5@protonmail.com>
2367Date:   Sat Jan 15 16:20:56 2022 +0100
2368
2369    nls: add a compose sequence for the middle dot (·) using dead_circumflex
2370    
2371    This makes the middle dot compose sequences consistent with others that
2372    begin with <dead_circumflex> and <Multi_key> <asciicircum>, e.g.
2373    superscripted 3 (³)
2374    
2375    Signed-off-by: ArenaL5 <arenal5@protonmail.com>
2376
2377commit 1adb531cd5be2b6489c2a8106e850e7418fae056
2378Author: Benno Schulenberg <bensberg@telfort.nl>
2379Date:   Fri Jan 14 13:06:10 2022 +0100
2380
2381    docs: unwrap the comments of structure elements in the man pages
2382    
2383    It doesn't matter that these lines get wider than 80 columns, as it's
2384    easy enough to make the pager scroll a bit to the right, or to have a
2385    terminal that is a 100 or more columns wide.  It looks better to not
2386    hard-wrap these comments.
2387    
2388    Also, fix the layout of two table elements.
2389    
2390    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2391
2392commit da330a1644f10f9669c8dd9ee06489c40f9f0bd1
2393Author: Benno Schulenberg <bensberg@telfort.nl>
2394Date:   Fri Jan 14 12:33:23 2022 +0100
2395
2396    docs: unwrap some mistakenly wrapped lines, and correct a comment
2397    
2398    The wrapped lines messed up the table.  Also, add some blank lines
2399    to make the table more readable.
2400    
2401    Elsewhere, add spaces around an operator (for consistency),
2402    and correct a copy-paste error.
2403    
2404    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2405
2406commit b9e0346abebb0c9e5c80c780bba4372419e63bde
2407Author: Benno Schulenberg <bensberg@telfort.nl>
2408Date:   Fri Jan 14 15:57:56 2022 +0100
2409
2410    XKBMAlloc: use the correct first-key variable when extending a key range
2411    
2412    When extending the range of changed virtual-modifier keys, one should
2413    pass the two variables relevant to that range.
2414    
2415    These two mistakes were found by grepping for 'modmap_key' and then
2416    looking for asymmetries between the handling of the plain modmap thing
2417    and the virtual modmap thing.  Inspired by issue #76.
2418    
2419    (I suspect these mistakes were at the base of some intractable bugs
2420    reported against xkeyboard-config where virtual modifiers refused
2421    to work and had to be worked around by making other changes too.)
2422    
2423    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2424
2425commit f15d2a01f75304bc2d9dac109d9a5a47d1ce28e0
2426Author: Benno Schulenberg <bensberg@telfort.nl>
2427Date:   Wed Jan 12 11:34:23 2022 +0100
2428
2429    lcUTF8: remove two unneeded ifs and two elses that are never reached
2430    
2431    When there is a break at the end of the while loop, it means that the
2432    loop will finish the first time that _XlcGetCharSetWithSide() returns
2433    a non-NULL result, which means that the first '(last_charset == NULL)'
2434    will always be true and the else part is redundant.
2435    
2436    Fixes issue #46.
2437    
2438    Reported-by: Bhavi Dhingra
2439    
2440    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2441
2442commit 033f37d507801645a8251fd33c25ebd145d22cce
2443Author: Antti Savolainen <antti.savo@gmail.com>
2444Date:   Wed Jan 12 16:10:21 2022 +0000
2445
2446    nls: add two compose sequences for an upwards arrow using dead_circumflex
2447    
2448    Signed-off-by: Antti Savolainen <antti.savo@gmail.com>
2449
2450commit b7d724df615311c35418afc733e1fcc68c740aba
2451Author: Hodong <hodong@nimfsoft.com>
2452Date:   Fri Jan 7 10:41:15 2022 +0000
2453
2454    specs/XIM: fix a missing "EXT" in the documentation
2455    
2456    There was an "EXT" in the original text, but it seems to be missing.
2457    
2458    See: https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/4e66da0783b2e5e3b288aaecd3c89396ed425c20/specs/XIM/xim.ms#L693
2459    
2460    Signed-off-by: Hodong Kim <hodong@nimfsoft.com>
2461
2462commit 4c96f3567a8d045ee57b886fddc9618b71282530
2463Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2464Date:   Thu Dec 9 19:51:52 2021 -0800
2465
2466    libX11 1.7.3.1
2467    
2468    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2469
2470commit c1710589bb00d6f442a383fc41f0d6496b436c88
2471Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2472Date:   Tue Dec 7 08:29:43 2021 -0800
2473
2474    specs/XIM: Fix "encording" typos
2475    
2476    Fixes: #145
2477    
2478    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2479
2480commit e30771b4bbdd84fc7ff64a1cda483a46edb5707b
2481Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2482Date:   Mon Dec 6 17:04:14 2021 -0800
2483
2484    libX11 1.7.3
2485    
2486    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2487
2488commit 934417b1e89fe715b9cf6c0d62a3867fa1f5bccb
2489Author: Liang Chang <l-chang@users.sourceforge.net>
2490Date:   Tue Sep 7 18:54:10 2021 +0800
2491
2492    Make Xutf8DrawString works correctly with ISO10646 fonts.
2493    
2494    Signed-off-by: Liang Chang <l-chang@users.sourceforge.net>
2495
2496commit 3ad36c5d3ddca3e4c97faaf87b72a53954d1f473
2497Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2498Date:   Thu Dec 2 17:48:55 2021 -0800
2499
2500    Build xz tarballs instead of bzip2
2501    
2502    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2503
2504commit 93a050c3ad2d2264d3880db3791387b1a9bf2e9e
2505Author: Tatu Frisk <tatu.frisk@ge.com>
2506Date:   Tue Mar 14 14:41:27 2017 +0200
2507
2508    Fix hanging issue in _XReply
2509    
2510    Assume event queue is empty if another thread is blocking waiting for event.
2511    
2512    If one thread was blocking waiting for an event and another thread sent a
2513    reply to the X server, both threads got blocked until an event was
2514    received.
2515    
2516    Signed-off-by: Tatu Frisk <tatu.frisk@ge.com>
2517    Signed-off-by: Jose Alarcon <jose.alarcon@ge.com>
2518
2519commit 8382253010267b93c9aeffc769e3f5e43d698094
2520Author: Matthieu Herrb <matthieu@herrb.eu>
2521Date:   Mon Aug 30 21:47:09 2021 +0200
2522
2523    Avoid NULL pointer deref. Fixes issue #47.
2524    
2525    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2526
2527commit 58d7b540c453601faf942c6f19f8277a5c5f78cc
2528Author: Peter Hutterer <peter.hutterer@who-t.net>
2529Date:   Wed Dec 1 14:31:16 2021 +1000
2530
2531    gitlab CI: update to use the ci-fairy image
2532    
2533    Let's use the pre-prepared image instead of building our own.
2534    
2535    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2536
2537commit 150154c994829f9f3734a560f99f70b21d86b49c
2538Author: Hodong <hodong@nimfsoft.com>
2539Date:   Sun Nov 14 06:36:42 2021 +0000
2540
2541    Fix a typo
2542
2543commit 2356e59ff24f8d1b25cdc4dffc5171c65dc2b86e
2544Author: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
2545Date:   Sun Sep 5 10:25:38 2021 +0300
2546
2547    libX11: do not crash in GetResReq() macro
2548    
2549    When _XGetRequest() detects that requested length exceeds remaining display
2550    output buffer capacity it would return NULL. GetResReq() macro obtains "req"
2551    pointer from a call to _XGetRequest() and then proceeds to assign request id
2552    through "req" pointer which leads to NULL pointer dereference in this case.
2553    
2554    Fix this by checking if "req" is valid before assigning request id.
2555    
2556    Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
2557
2558commit e92efc63acd7b377faa9e534f4bf52aaa86be2a9
2559Author: Peter Hutterer <peter.hutterer@who-t.net>
2560Date:   Tue Jul 27 11:46:19 2021 +1000
2561
2562    makekeys: handle the new _EVDEVK xorgproto symbols
2563    
2564    These keys are all defined through a macro in the form:
2565       #define XF86XK_BrightnessAuto                _EVDEVK(0x0F4)
2566    
2567    The _EVDEVK macro is simply an offset of 0x10081000.
2568    Let's parse these lines correctly so those keysyms end up in our
2569    hashtables.
2570    
2571    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2572
2573commit d01d23374107f6fc55511f02559cf75be7bdf448
2574Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
2575Date:   Wed Jun 16 12:17:04 2021 +0100
2576
2577    Avoid undefined behaviour after realloc()
2578    
2579    Adding the offset between the realloc result and the old allocation to
2580    update pointers into the new allocation is undefined behaviour: the
2581    old pointers are no longer valid after realloc() according to the C
2582    standard. While this works on almost all architectures and compilers,
2583    it causes  problems on architectures that track pointer bounds (e.g.
2584    CHERI or Arm's Morello): the value_list pointers will still have the
2585    bounds of the previous allocation and therefore any dereference will
2586    result in a run-time trap.
2587    
2588    I found this due to a crash (dereferencing an invalid capability) while
2589    trying to run `xev` over SSH on a CHERI-RISC-V system. With these two
2590    realloc changes, and https://gitlab.freedesktop.org/xorg/proto/xorgproto/-/merge_requests/41
2591    I am able to succesfully run `xev` compiled for CHERI-RISC-V.
2592    
2593    Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
2594
2595commit 1c845834a32421abb59408aee259ac0bef054d30
2596Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
2597Date:   Tue Jun 15 08:46:53 2021 +0100
2598
2599    Fix cross-compilation from macOS
2600    
2601    We can't use `LC_CTYPE=C sed` there since /usr/bin/sed is not compatible
2602    with the expressions in nls/ (`sed: RE error: illegal byte sequence`).
2603    To fix this use $(SED) instead which autotools will set to a GNU
2604    version of sed (usually /usr/local/bin/gsed) on macOS.
2605    
2606    Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
2607
2608commit 401f58f8ba258d4e7ce56a8f756595b72e544c15
2609Author: Lev Korol <epicatsupercell@gmail.com>
2610Date:   Sun Apr 11 21:57:45 2021 +0300
2611
2612    nls: Add en_IL locale
2613    
2614    Signed-off-by: Lev Korol <epicatsupercell@gmail.com>
2615
2616commit f906fe8e9769e4313294b68e61c402610ade69da
2617Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2618Date:   Sat Jun 5 09:12:13 2021 -0700
2619
2620    libX11 1.7.2
2621    
2622    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2623
2624commit 51b73ac0acda65005c8a9f17ca4ea7281b00ca84
2625Author: Tobias Stoeckmann <tobias@stoeckmann.org>
2626Date:   Sun May 23 15:05:17 2021 +0200
2627
2628    Protect against overly long strings
2629    
2630    Checking against upper limit of USHRT_MAX must happen before truncating
2631    size_t to int. On 64 bit systems with strings larger than 2 GB this
2632    could otherwise lead to negative ints or ints smaller than USHRT_MAX.
2633    
2634    In XParseColor this could lead to out of boundary access with strings
2635    starting with a # (color sequence). A modulo 12 operation is performed
2636    to validate the string length, but with an overflown length, the for
2637    loop would eventually read behind terminating '\0' character.
2638    
2639    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
2640
2641commit ab2f59530b16bdfbf023b8e025c7c8aba3b6fd0c
2642Author: Matthieu Herrb <matthieu@herrb.eu>
2643Date:   Sun May 30 16:33:48 2021 +0200
2644
2645    Check for NULL strings before getting their lengths
2646    
2647    Problem reported by Karsten Trulsen
2648    
2649    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2650
2651commit a8216e8653a246db664e1443ecf707b27de65067
2652Author: Matthieu Herrb <matthieu@herrb.eu>
2653Date:   Sat May 22 17:35:47 2021 +0200
2654
2655    include <limits.h> always, not if HAVE_CONFIG_H is set.
2656    
2657    via Matthew Green from the NetBSD project
2658    
2659    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2660
2661commit 6953a586df4819143c4d55e011b3a5e5377981b8
2662Author: Matthieu Herrb <matthieu.herrb@laas.fr>
2663Date:   Tue May 18 15:27:58 2021 +0200
2664
2665    Version 1.7.1
2666    
2667    Release notes in README.md, version bump in configure.ac
2668
2669commit 8d2e02ae650f00c4a53deb625211a0527126c605
2670Author: Matthieu Herrb <matthieu@herrb.eu>
2671Date:   Fri Feb 19 15:30:39 2021 +0100
2672
2673    Reject string longer than USHRT_MAX before sending them on the wire
2674    
2675    The X protocol uses CARD16 values to represent the length so
2676    this would overflow.
2677    
2678    CVE-2021-31535
2679    
2680    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2681
2682commit 838ea5a5a0267c25b20c095c9a70684edeeefba4
2683Author: Gaurav Ujjwal <gujjwal00@gmail.com>
2684Date:   Sun May 9 05:04:36 2021 +0000
2685
2686    Fix out-of-bound access in KeySymToUcs4()
2687    
2688    Array `keysym_to_unicode_590_5fe` is only valid for range  [0x590, 0x5fe] but current lower-bound is checked against 0x589.
2689    
2690    So invalid values from 0x58a to 0x58f are being allowed by current check.
2691    
2692    If any of these invalid value is passed as `keysym`,    `keysym - 0x590` would underflow.
2693    
2694    Signed-off-by: Gaurav Ujjwal <gujjwal00@gmail.com>
2695
2696commit 32491b02c76ac22f99dc433a32466e22bdc181a7
2697Author: Christopher Chavez <chrischavez@gmx.us>
2698Date:   Mon May 3 15:46:44 2021 +0000
2699
2700    Xlib.h: spelling fix in comment
2701
2702commit 5faa8dc0b364f55f19034c370436074646d9ad54
2703Author: Benno Schulenberg <bensberg@telfort.nl>
2704Date:   Tue Nov 24 18:54:24 2020 +0100
2705
2706    nls: adjust three comments about the APL compose sequences
2707    
2708    Commit 0bbc0d5e605e (from eight years ago) removed the lines that two
2709    of these comments referred to.  Without those lines, the comments don't
2710    make sense any more.  Reword and shorten them.
2711    
2712    Also reword a comment about two sequences that don't work.
2713    
2714    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2715
2716commit b126bfd7fe990b8c1df088f55c31d9b3fd440cde
2717Author: Benno Schulenberg <bensberg@telfort.nl>
2718Date:   Tue Nov 24 12:23:32 2020 +0100
2719
2720    nls: allow composing all breved letters also with a lowercase "u"
2721    
2722    The letters ă and ŭ can already be composed with "u a" and "u u", but
2723    ĕ, ğ, ĭ, and ŏ can be composed only with an uppercase U.  Emancipate
2724    the latter four and understand also a lowercase "u" to mean 'breve'.
2725    
2726    (Yesterday I needed ğ and was annoyed that "u g" did not work.)
2727    
2728    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2729
2730commit 78027fdb7aa07621a3b7cdf37050bae39012f2fb
2731Author: Walter Harms <wharms@bfs.de>
2732Date:   Sat Nov 28 21:05:33 2020 +0100
2733
2734    fix same roff code
2735    
2736    Signed-off-by: Walter Harms <wharms@bfs.de>
2737
2738commit 4f15cfc64548098fef3ddd24f7359b76cf1bbda7
2739Author: Walter Harms <wharms@bfs.de>
2740Date:   Sat Nov 28 20:56:35 2020 +0100
2741
2742    Fix some roff code add see also
2743    
2744    Signed-off-by: Walter Harms <wharms@bfs.de>
2745
2746commit daa106927f78e6c5c9c36558a4b951d7cb58a8d1
2747Author: Walter Harms <wharms@bfs.de>
2748Date:   Sat Nov 28 17:49:25 2020 +0100
2749
2750    fix broken nroff coding for code comments
2751    
2752    the comments /* */ are code as /\(**  */ that does not work.
2753    the coding in other X11 man pages is /\&* */ so we do the same here.
2754
2755commit 7bdeae239b17db002964ac73d06f43f216d1cf86
2756Author: Walter Harms <wharms@bfs.de>
2757Date:   Fri Nov 27 22:43:21 2020 +0100
2758
2759    FIX: warning: macro `hN' not defined
2760    
2761    Signed-off-by: Walter Harms <wharms@bfs.de>
2762
2763commit b7ec67d3ea78ab68857328d8f72d923324e68cc2
2764Author: Walter Harms <wharms@bfs.de>
2765Date:   Fri Nov 27 22:26:15 2020 +0100
2766
2767    FIX: warning: macro `IN' not defined
2768    
2769    just remove an other dead macro use.
2770
2771commit 7ca3ceb9175a829a40c347cb3b2e98341ec5a2df
2772Author: Walter Harms <wharms@bfs.de>
2773Date:   Fri Nov 27 21:58:04 2020 +0100
2774
2775    fix warning: macro `s' not defined
2776    
2777    this is caused by bad nroff coding, fix some more issues on the fly
2778
2779commit 14fb4e535626ba934cad5bc4308f511524cd1103
2780Author: Walter Harms <wharms@bfs.de>
2781Date:   Fri Nov 27 20:04:22 2020 +0100
2782
2783    FIX: warning: macro `hN' not defined
2784    
2785    this was found by checking man pages with
2786     groff -t -mandoc -Z  -wmac -Tutf8 $FILE >/dev/null
2787    
2788    In most cases .hN could be replaced with .BR
2789    
2790    Signed-off-by: Walter Harms <wharms@bfs.de>
2791
2792commit cb03da4447aa53e8e5382f124917d89b50148036
2793Author: Walter Harms <wharms@bfs.de>
2794Date:   Fri Nov 27 19:00:00 2020 +0100
2795
2796    FIX: warning: macro `Pn' not defined
2797    
2798    The missing macro is found via:
2799    roff -t -mandoc -Z  -wmac -Tutf8 XAnyEvent.man >/dev/null
2800    
2801    To fix the problem the macro is replaced with .RB.
2802    
2803    Signed-off-by: Walter Harms <wharms@bfs.de>
2804
2805commit cc9f8878f2cbe17c7b4035b4ff4352b52ece38e0
2806Author: Benno Schulenberg <bensberg@telfort.nl>
2807Date:   Wed Nov 25 16:58:27 2020 +0100
2808
2809    nls: add 'C.utf8' as an alias for 'en_US.UTF-8'
2810    
2811    The normal form is 'C.UTF-8', but 'C.utf8' has been seen in the wild.
2812    
2813    Fixes #102.
2814    
2815    Reported-by: Tomas Korbar
2816    
2817    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2818
2819commit ca8115186f810eccb7d86b0979980eff3ba95f0b
2820Author: Keith Packard <keithp@keithp.com>
2821Date:   Thu Nov 19 11:30:41 2020 -0800
2822
2823    Version 1.7.0
2824    
2825    Release notes in README.md, version bump in configure.ac
2826    
2827    Signed-off-by: Keith Packard <keithp@keithp.com>
2828
2829commit 325ac177fb6d38f4c7689d9ce059792eaef388ef
2830Author: Keith Packard <keithp@keithp.com>
2831Date:   Mon Nov 16 15:48:40 2020 -0800
2832
2833    Update libX11 .so version to 6.4.0
2834    
2835    There have been API additions since the previous release, so the
2836    .so version needs to change.
2837    
2838    Signed-off-by: Keith Packard <keithp@keithp.com>
2839
2840commit 103e2e11519984aa5746c06e7d90d1f5bc8174a3
2841Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2842Date:   Sat Oct 31 09:24:01 2020 -0700
2843
2844    Don't leave dangling pointers in Free functions
2845    
2846    While these are mostly called during teardown of larger structures
2847    that are about to themselves be freed, there's no guarantee that
2848    will always be the case, so try to be safer here.
2849    
2850    [ This bug was found by the Parfait 4.0 bug checking tool.
2851      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
2852    
2853    v2: Deduplicate & simplify pointer clearing in _XFreeEventCookies
2854        as suggested by @keithp
2855    
2856    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2857
2858commit 7d70e302218e2ec8518548ddcdc02d828e7889af
2859Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2860Date:   Sat Oct 31 08:56:34 2020 -0700
2861
2862    XimProto.h: wrap XIM_HEADER_SIZE definition in parens
2863    
2864    Resolves parfait warning of potential macro misinterpretation if
2865    expanded in the midst of other arithmetic operations with higher
2866    precedence.
2867    
2868    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2869
2870commit a3c0b5dbd6b12ae64bc78b11795647a7f6df0c7a
2871Author: Keith Packard <keithp@keithp.com>
2872Date:   Tue Nov 17 13:15:18 2020 -0800
2873
2874    Copy locale modifiers when creating XimInstCallback [v2]
2875    
2876    Locale modifiers may be freed whenever XSetLocaleModifiers gets
2877    called, even if the locale hasn't changed. This means that we cannot
2878    save a pointer to those modifiers in the XimInstCallback record and
2879    must, instead, make a copy of them instead.
2880    
2881    This fixes a problem uncovered when running wish under libasan as
2882    follows (on current Debian unstable):
2883    
2884            $ LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libasan.so.6 wish
2885    
2886    Reported-by: Vittorio Zecca <zeccav@gmail.com>
2887    Signed-off-by: Keith Packard <keithp@keithp.com>
2888    
2889    v2:
2890            Remove incorrect 'else' token found by @alanc
2891
2892commit 960e2e0cfac12c3477c672d0d40818a0dc74aca5
2893Author: Peter Hutterer <peter.hutterer@who-t.net>
2894Date:   Mon Nov 16 09:34:57 2020 +1000
2895
2896    gitlab CI: add a basic build test
2897    
2898    Using Arch as base distribution here because we can expect our dependencies to
2899    be up-to-date. We rely on the Arch for our dependencies rather than building
2900    those from git (notably: xorg-macros, xtrans and libxcb).
2901    
2902    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2903
2904commit dbb55e1a5e82870466b095097d9e46046680ec25
2905Author: Frediano Ziglio <fziglio@redhat.com>
2906Date:   Wed Jan 29 09:06:54 2020 +0000
2907
2908    Fix poll_for_response race condition
2909    
2910    In poll_for_response is it possible that event replies are skipped
2911    and a more up to date message reply is returned.
2912    This will cause next poll_for_event call to fail aborting the program.
2913    
2914    This was proved using some slow ssh tunnel or using some program
2915    to slow down server replies (I used a combination of xtrace and strace).
2916    
2917    How the race happens:
2918    - program enters into poll_for_response;
2919    - poll_for_event is called but the server didn't still send the reply;
2920    - pending_requests is not NULL because we send a request (see call
2921      to  append_pending_request in _XSend);
2922    - xcb_poll_for_reply64 is called from poll_for_response;
2923    - xcb_poll_for_reply64 will read from server, at this point
2924      server reply with an event (say sequence N) and the reply to our
2925      last request (say sequence N+1);
2926    - xcb_poll_for_reply64 returns the reply for the request we asked;
2927    - last_request_read is set to N+1 sequence in poll_for_response;
2928    - poll_for_response returns the response to the request;
2929    - poll_for_event is called (for instance from another poll_for_response);
2930    - event with sequence N is retrieved;
2931    - the N sequence is widen, however, as the "new" number computed from
2932      last_request_read is less than N the number is widened to N + 2^32
2933      (assuming last_request_read is still contained in 32 bit);
2934    - poll_for_event enters the nested if statement as req is NULL;
2935    - we compare the widen N (which now does not fit into 32 bit) with
2936      request (which fits into 32 bit) hitting the throw_thread_fail_assert.
2937    
2938    To avoid the race condition and to avoid the sequence to go back
2939    I check again for new events after getting the response and
2940    return this last event if present saving the reply to return it
2941    later.
2942    
2943    To test the race and the fix it's helpful to add a delay (I used a
2944    "usleep(5000)") before calling xcb_poll_for_reply64.
2945    
2946    Original patch written by Frediano Ziglio, see
2947    https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/34
2948    
2949    Reworked primarily for readability by Peter Hutterer, see
2950    https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/53
2951    
2952    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2953
2954commit 30ccef3a48029bf4fc31d4abda2d2778d0ad6277
2955Author: Keith Packard <keithp@keithp.com>
2956Date:   Sat Nov 7 22:22:47 2020 -0800
2957
2958    Avoid recursing through _XError due to sequence adjustment
2959    
2960    This patch is based on research done by Dmitry Osipenko to uncover the
2961    cause of a large class of Xlib lockups.
2962    
2963    _XError must unlock and re-lock the display around the call to the
2964    user error handler function. When re-locking the display, two
2965    functions are called to ensure that the display is ready to generate a request:
2966    
2967        _XIDHandler(dpy);
2968        _XSeqSyncFunction(dpy);
2969    
2970    The first ensures that there is at least one XID available to use
2971    (possibly calling _xcb_generate_id to do so). The second makes sure a
2972    reply is received at least every 65535 requests to keep sequence
2973    numbers in sync (possibly generating a GetInputFocus request and
2974    synchronously awaiting the reply).
2975    
2976    If the second of these does generate a GetInputFocus request and wait
2977    for the reply, then a pending error will cause recursion into _XError,
2978    which deadlocks the display.
2979    
2980    One seemingly easy fix is to have _XError avoid those calls by
2981    invoking InternalLockDisplay instead of LockDisplay. That function
2982    does everything that LockDisplay does *except* call those final two
2983    functions which may end up receiving an error.
2984    
2985    However, that doesn't protect the system from applications which call
2986    some legal Xlib function from within their error handler. Any Xlib
2987    function which cannot generate protocol or wait for events is valid,
2988    including many which invoke LockDisplay.
2989    
2990    What we need to do is make LockDisplay skip these two function calls
2991    precisely when it is called from within the _XError context for the
2992    same display.
2993    
2994    This patch accomplishes this by creating a list of threads in the
2995    display which are in _XError, and then having LockDisplay check the
2996    current thread against those list elements.
2997    
2998    Inspired-by: Dmitry Osipenko <digetx@gmail.com>
2999    Signed-off-by: Keith Packard <keithp@keithp.com>
3000    Tested-by: Dmitry Osipenko <digetx@gmail.com>
3001    Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
3002
3003commit c9c4d6efbf92ab51695e2e740319503221d68eed
3004Author: Benno Schulenberg <bensberg@telfort.nl>
3005Date:   Fri Nov 13 11:34:04 2020 +0100
3006
3007    nls: improve the comments for the Serbian compose sequences
3008    
3009    Also put an extra space before the lone combining characters
3010    so they have some room to breathe.
3011    
3012    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3013
3014commit 3497dae3d9c52a252110b2a0983b82000be0fe8e
3015Author: Benno Schulenberg <bensberg@telfort.nl>
3016Date:   Fri Nov 13 11:23:09 2020 +0100
3017
3018    nls: remove mistaken Serbian compose sequences with combining diacritics
3019    
3020    Combining characters are not dead keys -- they have an immediate effect
3021    and combine with the preceding character.  So they cannot be used in
3022    compose sequences.
3023    
3024    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3025
3026commit 5cd60398b7787297008b13a848ed3cfbd7ef178d
3027Author: Benno Schulenberg <bensberg@telfort.nl>
3028Date:   Mon Nov 2 16:34:39 2020 +0100
3029
3030    nls: rename the obsolete sr_CS locale to sr_RS
3031    
3032    Fixes #107, for the most part.
3033    
3034    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3035
3036commit 7cf1729590534e257b851899cf03421b65c538e3
3037Author: Keith Packard <keithp@keithp.com>
3038Date:   Sat Nov 7 17:02:56 2020 -0800
3039
3040    Don't attempt to unload locale context data
3041    
3042    Most locale context users call _XlcCurrentLC, which returns a pointer
3043    which never needs to be passed to _XCloseLC, meaning it has unbounded
3044    lifetime, so that locale data can never be freed.
3045    
3046    Remove all reference counting and just leave all locales that were
3047    ever used in memory.
3048    
3049    Signed-off-by: Keith Packard <keithp@keithp.com>
3050    Acked-by: Martin Peres <martin.peres@mupuf.org>
3051
3052commit 87a8fd8051b823636c3c4f3c37c4cfd11fdb880d
3053Author: Jacek Caban <jacek@codeweavers.com>
3054Date:   Mon Aug 14 19:20:19 2017 +0200
3055
3056    Don't cache last lcd in _XlcCurrentLC.
3057    
3058    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678
3059    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538
3060    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088
3061    
3062    The way it's currently cached is not thread safe. As long as locale doesn't change, the same object is reused anyway.
3063    
3064    Signed-off-by: Jacek Caban <jacek@codeweavers.com>
3065    Signed-off-by: Keith Packard <keithp@keithp.com>
3066    Acked-by: Martin Peres <martin.peres@mupuf.org>
3067
3068commit 7a38c09b82ee2522d987c3da4af8af186e562c61
3069Author: Keith Packard <keithp@keithp.com>
3070Date:   Sat Nov 7 16:59:37 2020 -0800
3071
3072    Don't cache converters in _Xlcmbtowc and Xlcwctomb
3073    
3074    These functions were caching encoding conversion functions in static
3075    variables which is not thread safe. Let the conversion loader do its
3076    job and cache locale to converters there. It's less efficient, but
3077    it's also (now) thread safe.
3078    
3079    Signed-off-by: Keith Packard <keithp@keithp.com>
3080    Acked-by: Martin Peres <martin.peres@mupuf.org>
3081
3082commit b52ba5c209165fc2c533b77b86147b31763299f3
3083Author: Jacek Caban <jacek@codeweavers.com>
3084Date:   Mon Aug 14 19:20:20 2017 +0200
3085
3086    Make conv_list thread safe.
3087    
3088    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678
3089    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538
3090    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088
3091    Signed-off-by: Jacek Caban <jacek@codeweavers.com>
3092    Signed-off-by: Keith Packard <keithp@keithp.com>
3093    Acked-by: Martin Peres <martin.peres@mupuf.org>
3094
3095commit 8b5ce3a9b1219c81711944880e5a1e1cfad3d956
3096Author: Antti Savolainen <antti.savo@gmail.com>
3097Date:   Wed Sep 16 07:47:26 2020 +0000
3098
3099    Added DOWNWARD ARROW support
3100
3101commit 7138826dccba9d538dfb142ce9af1a40d64ea849
3102Author: Antti Savolainen <antti.savo@gmail.com>
3103Date:   Tue Sep 8 00:54:23 2020 +0300
3104
3105    Added support for UPWARDS ARROW and ALMOST EQUAL TO
3106    
3107    Signed-off-by: Antti Savolainen <antti.savo@gmail.com>
3108
3109commit 56d59299f7433cdeb88bac970b06eb7df44df5bb
3110Author: Jonathan Belsewir <jonbel@jonasund.de>
3111Date:   Thu Nov 5 22:57:50 2020 +0000
3112
3113    Add less and greater to characters that are difficult to access on some keyboards
3114
3115commit 5caf45f8d3bebad5b72ac88545adc156ccfd754d
3116Author: Benno Schulenberg <bensberg@telfort.nl>
3117Date:   Fri Oct 23 17:12:22 2020 +0200
3118
3119    nls: Allow all letters with a caron to be composed also with v.
3120    
3121    It was strange that the accented letters Ž and ž can be composed with
3122    sequences that start with "v" ("v Z" and "v z"), but not Č and č and
3123    Š and š (and other letters with a caron).  For these letters, compose
3124    sequences that start with a "c" had to be used, which was frustrating
3125    because it is hard to remember that "c" stands for "caron", AND the
3126    graphically more obvious "v" is right next to it.
3127    
3128    (Unfortunately, the sequence "v l" is already taken for vertical line.
3129    Maybe the compose sequences for vertical line could be reduced to just
3130    "V L" and "L V"?)
3131    
3132    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3133
3134commit 62de4ac9025e2cb3f62a890241c06ebc75acce62
3135Author: Benno Schulenberg <bensberg@telfort.nl>
3136Date:   Sun Nov 1 20:23:55 2020 +0100
3137
3138    nls: vertically align the target column of aliases
3139    
3140    Also improve the grammar of the initial comment.
3141    
3142    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3143
3144commit f5b26a16ef959fc6860e9cf58b97a042600900b4
3145Author: Benno Schulenberg <bensberg@telfort.nl>
3146Date:   Sun Nov 1 20:17:05 2020 +0100
3147
3148    nls: remove some twenty aliases that have been obsolete for fifteen years
3149    
3150    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3151
3152commit 3e61f3b49660e081108f06bc514366f1784cb2e0
3153Author: Benno Schulenberg <bensberg@telfort.nl>
3154Date:   Sun Nov 1 19:50:56 2020 +0100
3155
3156    nls: remove country-specific aliases for two constructed languages
3157    
3158    These artificial languages are meant to be international and are
3159    thus not specific to any country.  If one would want to support
3160    aliases like ia_FR or ia_CH, then one would also have to support
3161    ia_AU, ia_DE, ia_ES, et cetera, et cetera.  That would be silly.
3162    
3163    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3164
3165commit 3f9c43adde22bb2ab4bef5837bc8401af9bb537c
3166Author: Benno Schulenberg <bensberg@telfort.nl>
3167Date:   Sun Nov 1 17:28:24 2020 +0100
3168
3169    nls: remove four aliases where the name is identical to the definition
3170    
3171    They were found with:
3172    
3173        while read one two; do
3174            if [[ $one == $two: ]]; then echo $two; fi;
3175        done  <nls/locale.alias.pre
3176    
3177    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
3178
3179commit 9f9c5365816bdd036fd80d54b22e86764ea4b7a7
3180Author: Carlos Garnacho <carlosg@gnome.org>
3181Date:   Fri Jun 14 17:55:14 2019 +0200
3182
3183    Add XSetIOErrorExitHandler() function
3184    
3185    This function complements XSetIOErrorHandler(), allowing to override
3186    the default behavior that trusts on I/O errors never coming back
3187    (i.e. exit()ing the process).
3188    
3189    This is meant as a mechanism for Wayland compositors (that are too
3190    a X11 client + compositing manager) to unfasten seatbelts and jump
3191    through the car window. It might get lucky and land on a stack of
3192    pillows.
3193    
3194    In consequence, some functions labeled as _X_NORETURN can as a
3195    matter of fact return. So those hints were removed.
3196    
3197    Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
3198    Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
3199
3200commit 4cb758019e374fa7c022fe79d28444e13441717b
3201Author: Carlos Garnacho <carlosg@gnome.org>
3202Date:   Fri Jun 14 17:54:47 2019 +0200
3203
3204    Prepare for _XIOError() possibly returning
3205    
3206    Ensure current state is cut short on _XIOError(), possible reentrancy
3207    should be skipped through the XlibDisplayIOError flag checks.
3208    
3209    Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
3210    Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
3211
3212commit a37986f7a7d4ae9b8292b132dbb0da2885ea919c
3213Author: Ross Burton <ross.burton@intel.com>
3214Date:   Wed Jan 22 17:11:23 2020 +0000
3215
3216    cpprules.in: squash whitespace in generated files
3217    
3218    CPP is used to generate files, but as cpp reads files from the build host the
3219    output has a number of blank lines at the beginning which varies depending on
3220    what GCC and friends is used.
3221    
3222    Pathalogical example:
3223    
3224     $ cpp -undef -traditional /dev/null
3225     # 1 "/dev/null"
3226     # 1 "<built-in>"
3227     # 1 "<command-line>"
3228     # 31 "<command-line>"
3229     # 1 "/usr/include/stdc-predef.h" 1 3 4
3230    
3231     # 17 "/usr/include/stdc-predef.h" 3 4
3232    
3233     [ 40 blank line ]
3234    
3235     # 32 "<command-line>" 2
3236     # 1 "/dev/null"
3237    
3238    So depending on the content of stdc-predef.h and what other headers CPP will
3239    load, the amount of whitespace in the generates files varies. This can result in
3240    differences in reproducible environments, and file conflicts in multilib
3241    environments.
3242    
3243    As whitespace is irrelevant to these machine-readable files, extend the sed to
3244    just delete blank lines.
3245
3246commit 16192ce2a7e462e09f95aa45ce9eaa180e483c43
3247Author: Carmina16 <mistresssilvara@hotmail.com>
3248Date:   Sun Jun 16 02:54:25 2019 +0000
3249
3250    Adding ia and ie locales
3251
3252commit caa71668af7fd3ebdd56353c8f0ab90824773969
3253Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3254Date:   Mon Sep 28 16:16:07 2020 -0700
3255
3256    Use __builtin_popcountl if available to replace Ones() in GetImage.c
3257    
3258    If the compiler knows of a better algorithm for counting the number of
3259    bits set in a word for the target CPU, let it use that, instead of the
3260    classic algorithm optimized for PDP-6.
3261    
3262    Based on libXext commit 490a25e6f8a4d2482af4364c700b68ad11a4d10b
3263    
3264    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3265
3266commit 1a9269a95c353e3de9f4d33bab9f1363d84e79db
3267Author: Krzesimir Nowak <qdlacz@gmail.com>
3268Date:   Mon Jun 1 12:58:30 2020 +0000
3269
3270    Fix leaking modifiers string.
3271    
3272    Reported by valgrind:
3273    
3274    ```
3275    ==118175== 17 bytes in 1 blocks are definitely lost in loss record 13 of 1,675
3276    ==118175==    at 0x483A809: malloc (vg_replace_malloc.c:307)
3277    ==118175==    by 0x5CD1B46: _XlcDefaultMapModifiers (in /usr/lib64/libX11.so.6.3.0)
3278    ==118175==    by 0x5CD1F1A: XSetLocaleModifiers (in /usr/lib64/libX11.so.6.3.0)
3279    ==118175==    by 0x496841C: X11_InitKeyboard (SDL_x11keyboard.c:324)
3280    ==118175==    by 0x496F0CA: X11_VideoInit (SDL_x11video.c:455)
3281    ==118175==    by 0x494747B: SDL_VideoInit_REAL (SDL_video.c:532)
3282    ==118175==    by 0x489E886: SDL_InitSubSystem_REAL (SDL.c:206)
3283    ==118175==    by 0x402634: main (fade.cc:35)
3284    ```
3285
3286commit d127217f26df1bf7566c1f372d8b5329a06754ea
3287Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3288Date:   Sun Aug 16 17:03:24 2020 -0700
3289
3290    Exempt XLookupKeysym from XKeycodeToKeysym deprecation warnings
3291    
3292    Gets rid of:
3293    src/xkb/XKBBind.c: In function ‘XLookupKeysym’:
3294    src/xkb/XKBBind.c:234:5: warning: ‘XKeycodeToKeysym’ is deprecated
3295    [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wdeprecated-declarations-Wdeprecated-declarations]
3296      234 |     return XKeycodeToKeysym(dpy, event->keycode, col);
3297          |     ^~~~~~
3298    src/xkb/XKBBind.c:96:1: note: declared here
3299       96 | XKeycodeToKeysym(Display *dpy,
3300          | ^~~~~~~~~~~~~~~~
3301    
3302    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3303
3304commit a64eb5f478c6ca84ef3457a01f9a7c3db8ecf1e5
3305Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3306Date:   Sun Aug 16 16:50:43 2020 -0700
3307
3308    lcUTF8: Increase string buffer size from 20 to 32 bytes
3309    
3310    While we don't expect large enough ints to need it, we don't
3311    enforce a maximum size, so gcc assumes the worst and warns:
3312    
3313    ../../../src/xlibi18n/lcUTF8.c: In function ‘create_tofontcs_conv’:
3314    ../../../src/xlibi18n/lcUTF8.c:1736:34: warning: ‘.charset.name’ directive output may be truncated writing 13 bytes into a region of size between 8 and 17 [-Wformat-truncation=]
3315     1736 |  snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
3316          |                                  ^~~~~~~~~~~~~
3317    ../../../src/xlibi18n/lcUTF8.c:1736:2: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20
3318     1736 |  snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
3319          |  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3320    ../../../src/xlibi18n/lcUTF8.c:1739:46: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
3321     1739 |      snprintf(buf, sizeof(buf), "fs%d.charset", i);
3322          |                                              ^
3323    ../../../src/xlibi18n/lcUTF8.c:1739:6: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20
3324     1739 |      snprintf(buf, sizeof(buf), "fs%d.charset", i);
3325          |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3326    ../../../src/xlibi18n/lcUTF8.c:1754:41: warning: ‘.charset.name’ directive output may be truncated writing 13 bytes into a region of size between 8 and 17 [-Wformat-truncation=]
3327     1754 |         snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
3328          |                                         ^~~~~~~~~~~~~
3329    ../../../src/xlibi18n/lcUTF8.c:1754:9: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20
3330     1754 |         snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
3331          |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3332    ../../../src/xlibi18n/lcUTF8.c:1757:53: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
3333     1757 |             snprintf(buf, sizeof(buf), "fs%d.charset", i);
3334          |                                                     ^
3335    ../../../src/xlibi18n/lcUTF8.c:1757:13: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20
3336     1757 |             snprintf(buf, sizeof(buf), "fs%d.charset", i);
3337          |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3338    
3339    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3340
3341commit 54925250ad9c825bea671dcade5c913bad88abee
3342Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3343Date:   Sun Aug 16 16:33:59 2020 -0700
3344
3345    i18n: use memcpy instead of strncpy on unterminated char arrays
3346    
3347    Avoids gcc warnings that we're using strncpy wrong to copy a known-length
3348    set of characters without a terminating '\0' to a buffer whose length we
3349    are checking separately.  (Should also be imperceptibly faster since we
3350    no longer check if each byte is '\0' when we already know it won't be.)
3351    
3352    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3353
3354commit 8754868dbbbc5fa096282df0504aa7880488eeb9
3355Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3356Date:   Sun Aug 16 16:02:01 2020 -0700
3357
3358    Mark FreeBlueTblElements label unused in LRGB.c
3359    
3360    Quiets gcc 10.2 warning of:
3361    src/xcms/LRGB.c: In function ‘LINEAR_RGB_InitSCCData’:
3362    src/xcms/LRGB.c:798:1: warning: label ‘FreeBlueTblElements’ defined
3363     but not used
3364    [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wunused-label-Wunused-label]
3365      798 | FreeBlueTblElements:
3366          | ^~~~~~~~~~~~~~~~~~~
3367    
3368    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3369
3370commit 52b0cd67073f2f4665884b5a7decfedeb3b7ca1e
3371Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3372Date:   Sun Aug 16 14:54:47 2020 -0700
3373
3374    Raise minimum required xproto version to 7.0.25 (released 2013-11-23)
3375    
3376    Allows us to depend on _X_COLD directly instead of having to check for it.
3377    
3378    (Since we also use _X_UNUSED, 7.0.22 or later was implicitly required
3379     already but not checked for.)
3380    
3381    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3382
3383commit deafb99fd8faf8b044eaee13d072440ce4db76af
3384Author: Maya Rashish <maya@NetBSD.org>
3385Date:   Sun Aug 23 21:19:33 2020 +0300
3386
3387    Avoid the use of "register" keyword in public headers.
3388    
3389    This causes issues when compiling code for C++17.
3390
3391commit 505420662577749e36640db48f6b6b9ae0236e09
3392Author: Matthieu Herrb <matthieu@herrb.eu>
3393Date:   Mon Aug 24 15:42:25 2020 +0200
3394
3395    libX11 1.6.12
3396    
3397    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3398
3399commit acdaaadcb3d85c61fd43669fc5dddf0f8c3f911d
3400Author: Matthieu Herrb <matthieu@herrb.eu>
3401Date:   Thu Aug 13 18:02:58 2020 +0200
3402
3403    Fix an integer overflow in init_om()
3404    
3405    CVE-2020-14363
3406    
3407    This can lead to a double free later, as reported by Jayden Rivers.
3408    
3409    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3410
3411commit d15c24c8b44be5e4054c8ecd0ff9dcf2c8e18e5b
3412Author: Niclas Zeising <zeising@daemonic.se>
3413Date:   Tue Aug 11 13:44:38 2020 +0200
3414
3415    Fix input clients connecting to server
3416    
3417    Fix a bug where some input clients can't connect to the input server.
3418    This fixes #117.
3419    
3420    FreeBSD bugzilla reference:
3421    https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248549
3422    
3423    Signed-off-by: Niclas Zeising <zeising@daemonic.se>
3424
3425commit 6dd618e5ea8e7558dd8e99a2777ab7b69211789c
3426Author: Maya Rashish <maya@NetBSD.org>
3427Date:   Sat Aug 15 00:48:56 2020 +0300
3428
3429    Avoid the use of "register" keyword in XkbTranslateKeySym.
3430    
3431    This causes issues when compiling code for C++17.
3432    While here, make function prototype match the header with regards
3433    to removal of another register keyword.
3434
3435commit 780d222343098b4bfe3ec529b4a0dc2aadea19bf
3436Author: Christopher Chavez <chrischavez@gmx.us>
3437Date:   Mon Aug 10 17:08:39 2020 +0000
3438
3439    Fix typo GCCLipYOrigin -> GCClipYOrigin in XCreateGC() manpage
3440
3441commit 29a8251a456bb7dc10146b4713f499a624d6dacd
3442Author: Felix Yan <felixonmars@archlinux.org>
3443Date:   Thu Aug 6 16:03:38 2020 +0000
3444
3445    Correct a typo in GetStCmap.c
3446
3447commit 0d8f038db83f68724a84ef9d4ccd0d5e33ec6a70
3448Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3449Date:   Thu Aug 6 08:07:57 2020 -0700
3450
3451    libX11 1.6.11
3452    
3453    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3454
3455commit 93fce3f4e79cbc737d6468a4f68ba3de1b83953b
3456Author: Yichao Yu <yyc1992@gmail.com>
3457Date:   Sun Aug 2 13:43:58 2020 -0400
3458
3459    Fix size calculation in `_XimAttributeToValue`.
3460    
3461    The check here guards the read below.
3462    For `XimType_XIMStyles`, these are `num` of `CARD32` and for `XimType_XIMHotKeyTriggers`
3463    these are `num` of `XIMTRIGGERKEY` ref[1] which is defined as 3 x `CARD32`.
3464    (There are data after the `XIMTRIGGERKEY` according to the spec but they are not read by this
3465    function and doesn't need to be checked.)
3466    
3467    The old code here used the native datatype size instead of the wire protocol size causing
3468    the check to always fail.
3469    
3470    Also fix the size calculation for the header (size). It is 2 x CARD16 for both types
3471    despite the unused `CARD16` for `XimType_XIMStyles`.
3472    
3473    [1] https://www.x.org/releases/X11R7.6/doc/libX11/specs/XIM/xim.html#Input_Method_Styles
3474    
3475    This fixes a regression caused by 388b303c62aa35a245f1704211a023440ad2c488 in 1.6.10.
3476    
3477    Fix #116
3478
3479commit 9949364ea761ab7efa2a9930ec7718f966a740c0
3480Author: Matthieu Herrb <matthieu@herrb.eu>
3481Date:   Thu Jul 30 21:36:22 2020 +0200
3482
3483    libX11 1.6.10
3484    
3485    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3486
3487commit 1703b9f3435079d3c6021e1ee2ec34fd4978103d
3488Author: Matthieu Herrb <matthieu@herrb.eu>
3489Date:   Fri Jul 24 21:09:10 2020 +0200
3490
3491    Change the data_len parameter of _XimAttributeToValue() to CARD16
3492    
3493    It's coming from a length in the protocol (unsigned) and passed
3494    to functions that expect unsigned int parameters (_XCopyToArg()
3495    and memcpy()).
3496    
3497    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3498    Reviewed-by: Todd Carson <toc@daybefore.net>
3499
3500commit 1a566c9e00e5f35c1f9e7f3d741a02e5170852b2
3501Author: Todd Carson <tc@daybefore.net>
3502Date:   Fri Jul 24 20:59:32 2020 +0200
3503
3504    Zero out buffers in functions
3505    
3506    It looks like uninitialized stack or heap memory can leak
3507    out via padding bytes.
3508    
3509    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3510    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3511
3512commit 2fcfcc49f3b1be854bb9085993a01d17c62acf60
3513Author: Todd Carson <tc@daybefore.net>
3514Date:   Fri Jul 24 20:22:44 2020 +0200
3515
3516    Fix more unchecked lengths
3517    
3518    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3519    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3520
3521commit 388b303c62aa35a245f1704211a023440ad2c488
3522Author: Todd Carson <tc@daybefore.net>
3523Date:   Fri Jul 24 19:36:51 2020 +0200
3524
3525    fix integer overflows in _XimAttributeToValue()
3526    
3527    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3528    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3529
3530commit 0e6561efcfaa0ae7b5c74eac7e064b76d687544e
3531Author: Todd Carson <tc@daybefore.net>
3532Date:   Fri Jul 24 19:33:30 2020 +0200
3533
3534    Fix signed length values in _XimGetAttributeID()
3535    
3536    The lengths are unsigned according to the specification. Passing
3537    negative values can lead to data corruption.
3538    
3539    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3540    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3541
3542commit 2b7598221d87049d03e9a95fcb541c37c8728184
3543Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3544Date:   Wed Jul 22 15:19:58 2020 -0700
3545
3546    Fix spelling/wording issues
3547    
3548    Found by using:
3549        codespell --builtin clear,rare,usage,informal,code,names
3550    
3551    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3552
3553commit f9e3378ebcfc79ff00591b8617a68e97580a741e
3554Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
3555Date:   Sun Jul 19 22:19:24 2020 +0200
3556
3557    xlibi18n: Fix converting horizline keysyms to UCS
3558    
3559    U+23BA - U+23BD are meant to represent the scan lines, and U+2500 is
3560    unified with scan line 5.
3561    
3562    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3563
3564commit 4385a84c4a79c1b60bf12839899a1544198de975
3565Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
3566Date:   Thu Jun 4 02:19:13 2020 +0200
3567
3568    Braille: Fix default lookup
3569    
3570    cbb59d172 ('Braille: Fix typing quickly') broke the default lookup that
3571    translates Braille keysym patterns to Braille Unicode patterns since it
3572    rightfully clears brl_committing, but then we do not have it any more to
3573    fill brl_committed.
3574    
3575    This change saves the committed pattern so we can return it in the
3576    default lookup.
3577    
3578    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3579
3580commit fb65b67169105c82c243f265e0f993cf72efdacd
3581Author: Marko Myllynen <myllynen@redhat.com>
3582Date:   Fri Feb 28 13:59:36 2020 +0200
3583
3584    Update Finnish compose sequences for SFS 5966:2019 standard
3585    
3586    U0219, U0218, U021B, U021A are now part of the standard.
3587    
3588    Fixes #70 which was printed incorrectly in the earlier standard version.
3589
3590commit 07724ce046348a21cc82a874853b14b9a285df75
3591Author: Tristan Miller <psychonaut@nothingisreal.com>
3592Date:   Wed Sep 25 13:34:01 2019 +0200
3593
3594    Add acute accent compose sequences for Russian vowels.
3595    
3596    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/104
3597    
3598    Signed-off-by: Tristan Miller <psychonaut@nothingisreal.com>
3599
3600commit 3ec9264d01a3a66c0fa0145a7c1db06a64be1507
3601Author: Thomas E. Dickey <dickey@invisible-island.net>
3602Date:   Sat Apr 25 14:40:54 2020 -0400
3603
3604    document an ancient performance problem (see Debian #954845).
3605    
3606    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3607
3608commit 791c93816ce0520d54925b0b3ccbc29823370041
3609Author: Thomas E. Dickey <dickey@invisible-island.net>
3610Date:   Sat Apr 25 14:22:11 2020 -0400
3611
3612    update/correct function prototypes in manpages to match headers
3613    
3614    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3615
3616commit 844271305a28e87031562dfeb43defbea02960a2
3617Author: Thomas E. Dickey <dickey@invisible-island.net>
3618Date:   Tue Apr 21 20:53:35 2020 -0400
3619
3620    reduce gcc-normal warnings using casts (no object change)
3621    
3622    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3623
3624commit 53b59ddfc114a55d26e9717a1b4822f1b18c40ca
3625Author: Thomas E. Dickey <dickey@invisible-island.net>
3626Date:   Tue Apr 21 18:13:22 2020 -0400
3627
3628    reduce gcc-normal warnings using casts (no object change)
3629    
3630    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3631
3632commit 68c72a7341b114277ab232f2499ee3bd035af8a0
3633Author: Adam Jackson <ajax@redhat.com>
3634Date:   Wed Dec 11 11:53:11 2019 -0500
3635
3636    Fix XTS regression in XCopyColormapAndFree
3637    
3638    XCopyColormapAndFree/5 threw an assertion:
3639    
3640        520|4 5 00014017 1 2|Assertion XCopyColormapAndFree-5.(A)
3641        520|4 5 00014017 1 3|When a colourmap argument does not name a valid colourmap,
3642        520|4 5 00014017 1 4|then a BadColor error occurs.
3643        520|4 5 00014017 1 5|METH: Create a bad colourmap by creating and freeing a colourmap.
3644        520|4 5 00014017 1 6|METH: Call test function using bad colourmap as the colourmap argument.
3645        520|4 5 00014017 1 7|METH: Verify that a BadColor error occurs.
3646        520|4 5 00014017 1 8|unexpected signal 6 (SIGABRT) received
3647        220|4 5 2 15:05:53|UNRESOLVED
3648        410|4 5 1 15:05:53|IC End
3649        510|4|system 0: Abandoning testset: caught unexpected signal 11 (SIGSEGV)
3650    
3651    More specifically:
3652    
3653        lt-XCopyColormapAndFree: xcb_io.c:533: _XAllocID: Assertion `ret != inval_id' failed.
3654    
3655    This bug was introduced (by following my advice, d'oh) in:
3656    
3657        commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d
3658        Author: Tapani Pälli <tapani.palli@intel.com>
3659        Date:   Mon May 13 08:29:49 2019 +0300
3660    
3661            Protect colormap add/removal with display lock
3662    
3663    In that patch we moved the call to _XcmsCopyCmapRecAndFree inside the
3664    display lock. The problem is said routine has side effects, including
3665    trying to implicitly create a colormap in some cases. Since we don't run
3666    the XID handler until SyncHandle() we would see inconsistent internal
3667    xlib state, triggering the above assert.
3668    
3669    Fix this by dropping and re-taking the display lock before calling into
3670    XCMS.
3671    
3672    Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
3673
3674commit 7f46a38139f66fda734f3a6c445b84ea89c8f310
3675Author: Alex Henrie <alexhenrie24@gmail.com>
3676Date:   Wed Jan 1 14:54:06 2020 -0700
3677
3678    Handle small final sigma in XConvertCase
3679    
3680    lowercase: GREEK SMALL LETTER FINAL SIGMA (U+03C2)
3681    uppercase: GREEK CAPITAL LETTER SIGMA (U+03A3)
3682    
3683    This mapping was correct in UCSConvertCase, but the "legacy" mapping
3684    must also be correct for Caps Lock to work with the final sigma key.
3685    
3686    https://gitlab.freedesktop.org/xorg/lib/libx11/issues/5
3687    
3688    Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
3689
3690commit a48787d3d155c8a09cead345f5055da30c1db537
3691Author: Peter Hutterer <peter.hutterer@who-t.net>
3692Date:   Wed Dec 11 14:12:27 2019 +1000
3693
3694    Handle ssharp in XConvertCase()
3695    
3696    lowercase: LATIN SMALL LETTER SHARP S (U+00DF)
3697    uppercase: LATIN CAPITAL LETTER SHARP S (U+1E9E)
3698    
3699    The uppercase sharp s (XK_ssharp) is a relatively recent addition to unicode
3700    but was added to the relevant keyboard layouts in xkeyboard-config-2.25
3701    (d1411e5e95c)
3702    https://gitlab.freedesktop.org/xkeyboard-config/xkeyboard-config/issues/144
3703    
3704    Alas, the CapsLock behavior was broken on the finnish layout (maybe others).
3705    This was due to xkbcomp using XConvertCase() to determine whether a key
3706    requires the type FOUR_LEVEL_ALPHABETIC or FOUR_LEVEL_SEMIALPHABETIC.
3707    
3708    Let's make this function return the right lower/upper symbols for the sharp s
3709    and hope that the world won't get any worse because of it.
3710    
3711    https://gitlab.freedesktop.org/xorg/lib/libx11/issues/110
3712    
3713    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3714
3715commit b8766a438461e20b5c6f7a6111b703d4cb71501f
3716Author: Raul Fernandes <rgfernandes@gmail.com>
3717Date:   Wed Oct 9 18:39:26 2019 +0000
3718
3719    Use memcmp and memcpy
3720
3721commit db7cca17ad7807e92a928da9d4c68a00f4836da2
3722Author: Adam Jackson <ajax@redhat.com>
3723Date:   Wed Oct 9 13:41:29 2019 -0400
3724
3725    libX11 1.6.9
3726    
3727    Signed-off-by: Adam Jackson <ajax@redhat.com>
3728
3729commit f5ba2c632a4c84279bdd48c687271292f5dc4f87
3730Author: Dmitry Osipenko <digetx@gmail.com>
3731Date:   Sat May 18 19:10:32 2019 +0300
3732
3733    Fix lockup in _XReply() caused by recursive synchronization
3734    
3735    This patch is based on a suggestion made by Uli Schlachter in a comment
3736    to the bug report https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93.
3737    
3738    Explanation of the bug (given by Uli Schlachter as well):
3739    
3740    An error was received and handled. Since there was an error callback set,
3741    Xlib unlocks the display, runs the error callback, and then locks the display
3742    again. This goes through _XLockDisplay and then calls _XSeqSyncFunction.
3743    On this "lock the thing"-path, Xlib notices that sequence numbers are close to
3744    wrap-around and tries to send a GetInputFocus request. However, the earlier
3745    calls already registered themselves as "we are handling replies/errors, do
3746    not interfere!" and so the code here waits for "that other thread" to be done
3747    before it continues. Only that there is no other thread, but it is this thread
3748    itself and thus a deadlock follows.
3749    
3750    The bug is relatively easy to reproduce on any desktop environment by
3751    using actively a touchscreen input that supports multitouch, i.e. practically
3752    all mobile devices are affected.
3753    
3754    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93
3755    Suggested-by: Uli Schlachter <psychon@znc.in>
3756    Tested-by: Dmitry Osipenko <digetx@gmail.com>
3757    Reported-by: Dmitry Osipenko <digetx@gmail.com>
3758    Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
3759
3760commit 1f1ca0863fd81fd0538b104eb9d9841ddbc7c905
3761Author: Adam Jackson <ajax@redhat.com>
3762Date:   Fri Sep 13 16:44:32 2019 -0400
3763
3764    xkb: Provide <X11/extensions/XKBgeom.h> ourselves
3765    
3766    This header refers to libX11 types, it doesn't belong in xorgproto and
3767    has been moved to legacy in 2019.2. Supply it ourselves.
3768
3769commit edc7680ed5a03cedb5facf14693823455e12c29c
3770Author: Ross Burton <ross.burton@intel.com>
3771Date:   Tue Aug 6 14:53:43 2019 +0100
3772
3773    src/util/Makefile: explicitly reset LINK to not use libtool
3774    
3775    Simply looking at libtool redefines LINK globally to use libtool, which when
3776    you're trying to cross-compile to Windows can cause complications.
3777    
3778    As in src/util/ we're simply building a small binary for the build host, reset
3779    LINK to the automake default so that the traditional compile/link steps occur
3780    without libtool.
3781    
3782    Also remove -all-static from LDFLAGS as that is a libtool-specific argument
3783    intended to solve this problem.
3784    
3785    Closes: #100
3786    Signed-off-by: Ross Burton <ross.burton@intel.com>
3787
3788commit f111bf3ac091a3a6d1f094172621d9d033c3833c
3789Author: Thomas E. Dickey <dickey@invisible-island.net>
3790Date:   Tue Jul 30 16:53:27 2019 -0400
3791
3792    improve some formatting
3793    
3794    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3795
3796commit 1d0a6fe9126c7ce61c6648cabf43b2de4d6d3ba9
3797Author: Thomas E. Dickey <dickey@invisible-island.net>
3798Date:   Tue Jul 30 15:49:09 2019 -0400
3799
3800    fix a substitution error from recent commit, e.g, "s/^\.EE/XDe/"
3801    
3802    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3803
3804commit 8fe8aafe7e9beffaff0460fe08354dfcde1efe98
3805Author: Thomas E. Dickey <dickey@invisible-island.net>
3806Date:   Tue Jul 30 05:07:40 2019 -0400
3807
3808    split lines at sentence endings
3809    
3810    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3811
3812commit 6f2cbcea5d6dcd1df7ef309481cad1663284c061
3813Author: Thomas E. Dickey <dickey@invisible-island.net>
3814Date:   Tue Jul 30 04:52:25 2019 -0400
3815
3816    trim trailing whitespace from manpages
3817    
3818    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3819
3820commit 4abf482d8e6b0b1ae84e0f528e71066afd6d18b5
3821Author: Thomas E. Dickey <dickey@invisible-island.net>
3822Date:   Tue Jul 30 04:36:24 2019 -0400
3823
3824    another fake-quote fix
3825    
3826    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3827
3828commit c0f0651f2b4dfa9fef86d1374689220125631993
3829Author: Thomas E. Dickey <dickey@invisible-island.net>
3830Date:   Tue Jul 30 04:34:54 2019 -0400
3831
3832    the last commit overlooked some fake-quote pairs
3833    
3834    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3835
3836commit ef31143c9decda54b3ab2937d3300fcf5b493fd6
3837Author: Walter Harms <wharms@bfs.de>
3838Date:   Mon Jul 29 18:50:19 2019 +0200
3839
3840    fix ``fake quotes'' in text
3841    
3842    to remove the fake quotes replace them with propper
3843    predefined macros \*(lq and \*(rq. this will allow
3844    nroff to choose the propper characters when using ps etc.
3845    
3846    Signed-off-by: Walter Harms <wharms@bfs.de>
3847
3848commit 705027ea9b17d6552388d2067736ab4b0bf9dfb5
3849Author: Walter Harms <wharms@bfs.de>
3850Date:   Mon Jul 29 18:27:12 2019 +0200
3851
3852    get rid of ``fake quotes''
3853    
3854    these fake quotes are inside a comment and can be replaced with
3855    some double quotes
3856    
3857    Signed-off-by: Walter Harms <wharms@bfs.de>
3858
3859commit fe46eb287106ad05538e896783a5cf877a4138df
3860Author: Walter Harms <wharms@bfs.de>
3861Date:   Mon Jul 29 17:57:02 2019 +0200
3862
3863    remove all private macro defines
3864    
3865    after converting everything to st. man page macros there is
3866    no need to maintain X11 private nroff macros, so remove them.
3867    
3868    Signed-off-by: Walter Harms <wharms@bfs.de>
3869
3870commit b7f09ae19c9a098ecd545b4d4c45a6afe39a8764
3871Author: Walter Harms <wharms@bfs.de>
3872Date:   Mon Jul 29 17:47:35 2019 +0200
3873
3874    Replace home-grown .Ds .De macro with man page .EX/.EE macro
3875    
3876    Same pages use the man page .EX/.EE macro. Replace all occurences
3877    of .De/.Ds with the std. macros to make the code better to maintain.
3878    
3879    Signed-off-by: Walter Harms <wharms@bfs.de>
3880
3881commit 951e772bc8d89bc62c4bbac14adc5d4f5027419f
3882Author: Walter Harms <wharms@bfs.de>
3883Date:   Wed Jul 3 18:23:51 2019 +0200
3884
3885    fix TBL format
3886    
3887    Signed-off-by: Walter Harms <wharms@bfs.de>
3888
3889commit 4eb167f5aca4cca05a230c728ffdb4fd382a4e1e
3890Author: Walter Harms <wharms@bfs.de>
3891Date:   Wed Jul 3 18:18:09 2019 +0200
3892
3893    replace home grown .ZN with std, .B and .BR
3894    
3895    Replace the home grown macro .ZN with std. macros
3896    from man macro paket. So we can get rid of the
3897    definition an get a clean header.
3898    
3899    Signed-off-by: Walter Harms <wharms@bfs.de>
3900
3901commit 08e6b36292ebf01ba2dcc4a48ec371c6f5d13458
3902Author: Walter Harms <wharms@bfs.de>
3903Date:   Sat Jun 22 18:33:48 2019 +0200
3904
3905    remove in-text macros
3906    
3907    remove in-text macros an fix a few typos on the way.
3908    hopefuly that will make maintenance more easy.
3909    
3910    Signed-off-by: Walter Harms <wharms@bfs.de>
3911
3912commit cb0e894fe9927b07bd9d97c067fd0d007accb105
3913Author: Walter Harms <wharms@bfs.de>
3914Date:   Fri Jun 21 18:01:16 2019 +0200
3915
3916    note that we can handle kbd==NULL
3917    
3918    Signed-off-by: Walter Harms <wharms@bfs.de>
3919
3920commit 0327c427d62f671eced067c6d9b69f4e216a8cac
3921Author: Adam Jackson <ajax@redhat.com>
3922Date:   Mon Jun 17 13:36:08 2019 -0400
3923
3924    makekeys: Detach ourselves from X headers entirely
3925    
3926    Subsequent to a121b7b0c210efe10bf93453b29050282324c906 we are no longer
3927    building makekeys with enough -I/foo/bar to find the X11 headers, so if
3928    they're not in a system include path, things fail. Since this utility is
3929    only needed at build time, there's no real reason to demand the X
3930    headers be installed for both the build and target machines if cross-
3931    compiling, we can just assume a vaguely ANSI environment instead.
3932    
3933    Tested-by: Niclas Zeising <zeising@daemonic.se>
3934    Reviewed-by: Keith Packard <keithp@keithp.com>
3935    Reviewed-by: Matt Turner <mattst88@gmail.com>
3936
3937commit 57f0d421a1320090d3a4948181fe228d86cc8977
3938Author: Matt Turner <mattst88@gmail.com>
3939Date:   Mon Jun 17 10:45:45 2019 -0400
3940
3941    libX11 1.6.8
3942    
3943    Signed-off-by: Matt Turner <mattst88@gmail.com>
3944
3945commit 5464b302bc0c4a055d3eac06c6ab15e209ff7668
3946Author: Matt Turner <mattst88@gmail.com>
3947Date:   Sun Jun 9 14:00:15 2019 -0700
3948
3949    Use AC_SYS_LARGEFILE
3950    
3951    ... and include config.h in makekeys.c to get the definition of
3952    _FILE_OFFSET_BITS. Without it, libX11 can fail to build on a file
3953    system with 64-bit inode numbers.
3954    
3955    Bug: https://bugs.gentoo.org/550502
3956    Bug: https://bugs.gentoo.org/616140
3957    Signed-off-by: Matt Turner <mattst88@gmail.com>
3958
3959commit 33b9148a833ab2ea46e44871da3c031bf1b78f77
3960Author: Ran Benita <ran234@gmail.com>
3961Date:   Thu Feb 13 20:55:08 2014 +0200
3962
3963    Compose.man: fix escaped hexadecimal char description
3964    
3965    The man page says:
3966        Strings may be direct text encoded in the locale for which the
3967        compose file is to be used, or an escaped octal or hexadecimal
3968        character code.   Octal codes are specified as "\123" and
3969        hexadecimal codes as "\0x123a".
3970    
3971    But the grammar in the parser and the implementation say:
3972        ESCAPED_CHAR  ::= ('\\' | '\"' | OCTAL | HEX )
3973        HEX           ::= '\' (x|X) HEX_CHAR [HEX_CHAR]]
3974        HEX_CHAR      ::= (0|1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|a|b|c|d|e|f)
3975    
3976    So "\0x123a" -> "\x3a".
3977    
3978    Signed-off-by: Ran Benita <ran234@gmail.com>
3979    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3980
3981commit d9b2cc35edf01e4f751a9ccb5ac185745184cec0
3982Author: ojab <ojab@ojab.ru>
3983Date:   Mon Apr 24 20:12:17 2017 +0000
3984
3985    Compose sequences for rouble sign
3986    
3987    Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout.
3988    
3989    Signed-off-by: Slava Kardakov <ojab@ojab.ru>
3990    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3991
3992commit 3f211616f6414bd9958566c0ca137e87a2b6df95
3993Author: Pavel Labath <pavelo@centrum.sk>
3994Date:   Fri Nov 30 20:45:04 2012 +0100
3995
3996    Fix a leak in XCreateFontSet
3997    
3998    a simple snippet like XFreeFontSet(d, XCreateFontSet(d, ...)) will generate lots of memory leaks,
3999    as evidenced by the following valgrind output:
4000    ==983== HEAP SUMMARY:
4001    ==983==     in use at exit: 39,409 bytes in 341 blocks
4002    ==983==   total heap usage: 4,795 allocs, 4,454 frees, 489,086 bytes allocated
4003    ==983==
4004    ==983== 1,688 (136 direct, 1,552 indirect) bytes in 1 blocks are definitely lost in loss record
4005    40 of 46
4006    ==983==    at 0x4C2B042: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
4007    ==983==    by 0x56D5A93: add_codeset.clone.9 (in /usr/lib64/libX11.so.6.3.0)
4008    ==983==    by 0x56D5FE0: load_generic (in /usr/lib64/libX11.so.6.3.0)
4009    ==983==    by 0x56D7612: initialize (in /usr/lib64/libX11.so.6.3.0)
4010    ==983==    by 0x56D7E75: _XlcCreateLC (in /usr/lib64/libX11.so.6.3.0)
4011    ==983==    by 0x56F9A5F: _XlcUtf8Loader (in /usr/lib64/libX11.so.6.3.0)
4012    ==983==    by 0x56DF815: _XOpenLC (in /usr/lib64/libX11.so.6.3.0)
4013    ==983==    by 0x56B255A: XOpenOM (in /usr/lib64/libX11.so.6.3.0)
4014    ==983==    by 0x56A665A: XCreateFontSet (in /usr/lib64/libX11.so.6.3.0)
4015    ==983==    by 0x4FCA80: conky::x11_output::create_gc() (x11.cc:746)
4016    ==983==    by 0x4FC3B4: conky::x11_output::use_own_window() (x11.cc:602)
4017    ==983==    by 0x4FAD42: conky::priv::own_window_setting::set(bool const&, bool) (x11.cc:92)
4018    ==983==
4019    ==983== LEAK SUMMARY:
4020    ==983==    definitely lost: 136 bytes in 1 blocks
4021    ==983==    indirectly lost: 1,552 bytes in 34 blocks
4022    ==983==      possibly lost: 0 bytes in 0 blocks
4023    ==983==    still reachable: 37,721 bytes in 306 blocks
4024    ==983==         suppressed: 0 bytes in 0 blocks
4025    
4026    This patch makes the leak dissappear (Well, at least the "definitely lost part". The "still
4027    reachable" thingy remains). After some analysis, I've discovered that the XLCd structure is
4028    destroyed improperly. The "constructor" is in lcGeneric.c, but the structure is destroyed using
4029    code from lcPublic.c. I've found that changing the destructor call to _XlcDestroyLC executes the
4030    correct code path, and I'm pretty sure this is correct (the object was constructed using
4031    _XlcCreateLC, it make sense to destroy it using its conterpart).
4032    
4033    So far I haven't observed any strange behaviour on my system caused by this change (although, I'm
4034    not sure, how many programs actually use this function).
4035    
4036    Signed-off-by: Pavel Labath <pavelo@centrum.sk>
4037    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4038
4039commit 4645e219133458781e3fb48eaea6a74cccb1b9aa
4040Author: Jon Turney <jon.turney@dronecode.org.uk>
4041Date:   Tue Apr 30 14:39:06 2019 +0100
4042
4043    Avoid using libtool wrapper for makekeys
4044    
4045    For Windows targets, libtool uses a wrapper executable, not a wrapper
4046    script (see [1]), which it compiles with the host compiler.  This
4047    doesn't work when cross-compiling.
4048    
4049    Since we don't actually need to link with anything, use the libtool flag
4050    -all-static to tell it to stay completely out of this.
4051    
4052    [1] https://www.gnu.org/software/libtool/manual/html_node/Wrapper-executables.html
4053
4054commit 6886d9ba06826785a8ccb312c04ea82b91bb6a25
4055Author: Jon Turney <jon.turney@dronecode.org.uk>
4056Date:   Wed Apr 24 12:50:41 2019 +0100
4057
4058    Use EXEEXT_FOR_BUILD for makekeys
4059    
4060    Use EXEXT_FOR_BUILD, to fix cross-compiling where EXEEXT differs from
4061    EXEEXT_FOR_BUILD, such as when building for Windows from unix.
4062    
4063    (Note: As written, this assumes EXEEXT_FOR_BUILD is always empty when
4064    cross-compiling.  There could be some elaborate autodetection for
4065    EXEXT_FOR_BUILD, but for the moment, if you are cross-compiling from
4066    Windows to Unix, you'll need to set EXEEXT_FOR_BUILD explicity...)
4067
4068commit a121b7b0c210efe10bf93453b29050282324c906
4069Author: Jon Turney <jon.turney@dronecode.org.uk>
4070Date:   Tue Apr 30 14:30:41 2019 +0100
4071
4072    Remove makekeys dependency on X headers
4073    
4074    This is the patch from https://bugs.freedesktop.org/show_bug.cgi?id=6669
4075    by Pierre Ossman, reworked for master.
4076    
4077    Avoid using LIBS (which are for host, but we don't need) and rewrite
4078    makekeys slightly to avoid needing to include any X headers, which
4079    avoids potentially having -I with host paths in CFLAGS, which can cause
4080    standard headers e.g. stdio.h for the host to also be used, which can
4081    break things...
4082
4083commit cbb59d1727cb1bfcff91b54fd55f4139e360b7ab
4084Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
4085Date:   Fri Jun 7 23:35:33 2019 +0200
4086
4087    Braille: Fix typing quickly
4088    
4089    On finishing releasing Braille keys, we should clear the just-commited
4090    pattern, to reset the state to initial state, and avoid having to wait for
4091    0.3s before typing the next pattern.
4092    
4093    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
4094    Tested-by: Jean-Philippe Mengual <jpmengual@hypra.fr>
4095
4096commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d
4097Author: Tapani Pälli <tapani.palli@intel.com>
4098Date:   Mon May 13 08:29:49 2019 +0300
4099
4100    Protect colormap add/removal with display lock
4101    
4102    This fixes a bug where concurrent threads call XCreateColormap and
4103    XFreeColormap corrupting a linked list where colormap structures
4104    are stored.
4105    
4106    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/94
4107    
4108    v2: handle XCopyColormapAndFree also (Adam Jackson)
4109    
4110    Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
4111
4112commit fea80d03c381a4de379863c0d41f6d4c00f15e1e
4113Author: Jon Turney <jon.turney@dronecode.org.uk>
4114Date:   Wed Mar 13 13:12:20 2019 +0000
4115
4116    Add autoconf checks for headers we include for FIONREAD
4117    
4118    Add autoconf checks for the extra headers we include to define FIONREAD.
4119    This needs sys/socket.h on Cygwin, and none of the alternatives on
4120    Windows.
4121
4122commit 87c77a1e6d7034536e9d25ce24a667ebf53486a7
4123Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org>
4124Date:   Mon Mar 11 14:19:52 2019 +0000
4125
4126    locale.alias.pre - add the de_IT locale
4127    
4128    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4129
4130commit 568f77f414072506bbe48646637c872987c97e63
4131Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org>
4132Date:   Mon Mar 11 14:17:10 2019 +0000
4133
4134    locale.dir.pre - add the de_IT locale
4135    
4136    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4137
4138commit 38c0c92feaef7dc4a72920b1dd79872e6578086c
4139Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org>
4140Date:   Mon Mar 11 14:12:07 2019 +0000
4141
4142    compose.dir.pre - add de_IT locale
4143    
4144    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4145
4146commit 49f835611ba6ab82e68b7b851507931c142db7a5
4147Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org>
4148Date:   Tue Jan 29 15:34:34 2019 +0000
4149
4150    Update locale.alias.pre - introduce the de_LI aliases (equivalent to de_CH)
4151    
4152    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4153
4154commit ed669234d0cd5933618aa5fd2da697b59845b9be
4155Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org>
4156Date:   Tue Jan 29 15:28:18 2019 +0000
4157
4158    locale.dir.pre - fix the de_LI latin15 locale
4159    
4160    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4161
4162commit 82b5c1454a49333b4d94984d40db79cff386d37f
4163Author: Josh Triplett <josh@joshtriplett.org>
4164Date:   Sun Feb 3 08:31:41 2019 +0100
4165
4166    compose.dir.pre: Give the C.UTF-8 locale a full set of UTF-8 compose sequences
4167    
4168    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
4169
4170commit 603c3ab25a692194805fd4b0c62afac8093897fe
4171Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4172Date:   Sat Mar 9 20:58:07 2019 -0800
4173
4174    Add missing close paren to XrmGetStringDatabase synopsis in man page
4175    
4176    Fixes: commit 0642e99ac05afb0c
4177    
4178    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4179
4180commit 6590cc41e3c5c5c836a49a2c82774d7cc9ab3c85
4181Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4182Date:   Sat Feb 23 11:12:27 2019 -0800
4183
4184    Include reallocarray.h in tarballs so "make distcheck" passes
4185    
4186    Fixes: commit bcf7b5aa06c23aee
4187    
4188    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4189
4190commit a4a7df06351ce9565d298a84547ff71f2d220fbf
4191Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4192Date:   Sat Feb 23 11:04:41 2019 -0800
4193
4194    XlibInt.c: include headers needed for ioctl(...FIONREAD...) on Solaris
4195    
4196    Fixes: commit 5538b3e4ae6dee
4197    
4198    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4199
4200commit 38ec04de7e537dde6b7b1eaca46b1a24fcdbc044
4201Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4202Date:   Sat Feb 23 10:48:26 2019 -0800
4203
4204    specs/XKB: Fix misprint in documentation of XkbFreeGeomOverlayKeys()
4205    
4206    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/86
4207    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23550
4208    
4209    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4210
4211commit 74f2f26811ae82b15dede31504b0eb541f7a3293
4212Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4213Date:   Sat Feb 23 10:46:05 2019 -0800
4214
4215    specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayKeys()
4216    
4217    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/85
4218    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23549
4219    
4220    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4221
4222commit 66b039a41690f076e962ac6bd46cca017360a7d1
4223Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4224Date:   Sat Feb 23 10:44:15 2019 -0800
4225
4226    specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayRows()
4227    
4228    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/84
4229    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23548
4230    
4231    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4232
4233commit cdea0e840315171f8390adc63a2397ddf111eea4
4234Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4235Date:   Sat Feb 23 10:38:08 2019 -0800
4236
4237    specs/XKB: Fix misprint in documentation of XkbGetNamedGeometry()
4238    
4239    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/82
4240    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23520
4241    
4242    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4243
4244commit c72d17ada98723cd9e03ddf06680a0e54b2c3089
4245Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4246Date:   Sat Feb 23 10:30:32 2019 -0800
4247
4248    XkbAllocGeomOverlayKey man page should be named XkbAllocGeomOverlayKeys
4249    
4250    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4251
4252commit 05002b8aef9544c8545e0823a02cf14f335d2b68
4253Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4254Date:   Sun Feb 17 15:13:25 2019 -0800
4255
4256    Fix man page typo: XkbAllBellNotifyMask -> XkbAllBellEventsMask
4257    
4258    Reported-by: Daniel Hahler <git@thequod.de>
4259    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4260
4261commit 5538b3e4ae6dee32c47db9dfc85b07bbe7b90f6c
4262Author: Adam Jackson <ajax@redhat.com>
4263Date:   Fri Mar 24 11:07:36 2017 -0400
4264
4265    _XDefaultIOError: Do better at detecting explicit shutdown
4266    
4267    Currently, when the X server crashes or a client is disconnected with
4268    XKillClient, you get a somewhat confusing error message from libX11
4269    along the lines of:
4270    
4271    XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
4272          after 98 requests (40 known processed) with 0 events remaining.
4273    
4274    What's happening here is the previous recvmsg has thrown EAGAIN, since
4275    the socket is non-blocking. In this case, check whether the socket has
4276    any more data to read, and if not treat it like EPIPE.
4277    
4278    Signed-off-by: Adam Jackson <ajax@redhat.com>
4279
4280commit 6d2cde9633b5ee020cb60caea1cf61e090b86dd2
4281Author: Adam Jackson <ajax@redhat.com>
4282Date:   Fri Mar 24 11:07:35 2017 -0400
4283
4284    _XDefaultIOError: Reformat to be less ugly
4285    
4286    Signed-off-by: Adam Jackson <ajax@redhat.com>
4287    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4288
4289commit 003e30a66a249f5c70b30d1c187385124cd4cdad
4290Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4291Date:   Tue Jan 1 14:34:04 2019 -0800
4292
4293    Avoid use-after-free in _XimProtoSetIMValues()
4294    
4295    Fixes gitlab issue #49
4296    
4297    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4298
4299commit 30656fd66ad47aeaca0bfc237a818f189554f07e
4300Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4301Date:   Tue Jan 1 12:21:04 2019 -0800
4302
4303    Fix implicit conversion warnings in _XlcCreateDefaultCharSet
4304    
4305    lcCharSet.c:187:50: warning: implicit conversion changes signedness:
4306          'int' to 'unsigned long' [-Wsign-conversion]
4307        tmp = Xmalloc(name_len + 1 + ct_sequence_len + 1);
4308              ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
4309    ../../include/X11/Xlibint.h:453:32: note: expanded from macro 'Xmalloc'
4310                           ~~~~~~  ^~~~
4311    lcCharSet.c:192:31: warning: implicit conversion changes signedness:
4312          'int' to 'unsigned long' [-Wsign-conversion]
4313        memcpy(tmp, name, name_len+1);
4314        ~~~~~~            ~~~~~~~~^~
4315    lcCharSet.c:216:45: warning: implicit conversion changes signedness:
4316          'int' to 'unsigned long' [-Wsign-conversion]
4317        memcpy(tmp, ct_sequence, ct_sequence_len+1);
4318        ~~~~~~                   ~~~~~~~~~~~~~~~^~
4319    lcCharSet.c:183:16: warning: implicit conversion loses integer precision:
4320          'unsigned long' to 'int' [-Wshorten-64-to-32]
4321        name_len = strlen(name);
4322                 ~ ^~~~~~~~~~~~
4323    lcCharSet.c:184:23: warning: implicit conversion loses integer precision:
4324          'unsigned long' to 'int' [-Wshorten-64-to-32]
4325        ct_sequence_len = strlen(ct_sequence);
4326                        ~ ^~~~~~~~~~~~~~~~~~~
4327    lcCharSet.c:198:37: warning: implicit conversion loses integer precision:
4328         'long' to 'unsigned int' [-Wshorten-64-to-32]
4329            unsigned int length = colon - charset->name;
4330                         ~~~~~~   ~~~~~~^~~~~~~~~~~~~~~
4331    
4332    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4333
4334commit 2e6300905784fa8adfcea019262ea58b3e1ed4ad
4335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4336Date:   Tue Jan 1 11:41:47 2019 -0800
4337
4338    Remove no-longer-used name variable in _XGetAtomName
4339    
4340    Fixes gcc warning:
4341    GetAtomNm.c: In function ‘_XGetAtomName’:
4342    GetAtomNm.c:39:11: warning: unused variable ‘name’ [-Wunused-variable]
4343         char *name;
4344               ^~~~
4345    
4346    Introduced by commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96
4347    
4348    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4349
4350commit f077871c2a12cf098b0c50572e112c6dc5e04de0
4351Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4352Date:   Sat Dec 29 11:39:29 2018 -0800
4353
4354    Fix out-of-tree builds after commit 8a62e26515e528bb47917ec4a55cc2b492419eeb
4355    
4356    In-tree builds found reallocarray.h in $(top_builddir)/src but the
4357    out-of-tree build didn't find it at all.
4358    
4359    Reported-by: Emmanuele Bassi <ebassi@gmail.com> from GNOME continuous integration pipeline
4360    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4361
4362commit ba7f7cc783672784bebc6e69349b55c6c75c1973
4363Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4364Date:   Sun Sep 27 18:35:27 2015 -0700
4365
4366    Convert src/xlibi18n to use reallocarray()
4367    
4368    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4369
4370commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96
4371Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4372Date:   Sat Sep 26 11:12:47 2015 -0700
4373
4374    Replace Xmalloc+strcpy pairs with strdup calls
4375    
4376    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4377
4378commit 258a8ced681dc1bc50396be7439fce23f9807e2a
4379Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4380Date:   Sun Jul 19 11:58:50 2015 -0700
4381
4382    Replace _XkbTypedRealloc & _XkbClearElems with new _XkbResizeArray
4383    
4384    Makes resizing & clearing more consistent and gets rid of some weird
4385    quirks like always subtracting 1 from the size passed to _XkbClearElems
4386    so it could always add 1 to the size passed in.
4387    
4388    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4389
4390commit f94273e2f4284040306b888e5ca0381ee00591af
4391Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4392Date:   Sun Jul 19 09:47:29 2015 -0700
4393
4394    Add some missing allocation failure checks in src/xcms
4395    
4396    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4397
4398commit 76850471b4a5a6d30f5660234e87ec2772e7b426
4399Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4400Date:   Sun Jul 19 09:33:21 2015 -0700
4401
4402    Convert src/xcms to use reallocarray()
4403    
4404    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4405
4406commit 8a62e26515e528bb47917ec4a55cc2b492419eeb
4407Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4408Date:   Sun Jul 19 09:24:02 2015 -0700
4409
4410    Convert src/XKB to use reallocarray()
4411    
4412    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4413
4414commit 6d1dc1f6169ebf0ba71785d461bd98129c65c862
4415Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4416Date:   Thu Jun 4 23:30:43 2015 -0700
4417
4418    Convert main src directory to use reallocarray()
4419
4420commit bcf7b5aa06c23aee00af7999b58cb96a8571ac42
4421Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4422Date:   Thu Jun 4 23:18:23 2015 -0700
4423
4424    Import reallocarray() from OpenBSD
4425    
4426    Wrapper for realloc() that checks for overflow when multiplying
4427    arguments together, so we don't have to add overflow checks to
4428    every single call.  For documentation on usage, see:
4429    http://www.openbsd.org/cgi-bin/man.cgi/OpenBSD-current/man3/calloc.3
4430    
4431    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4432
4433commit 9bdfe9c9af11d77d66e5ff651e82b20e695cb460
4434Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4435Date:   Mon Nov 19 21:26:23 2018 -0800
4436
4437    Update README for gitlab migration
4438    
4439    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4440
4441commit c8bcc19970a2cc4ad337de1338c63a07325eefcd
4442Author: Albert Astals Cid <aacid@kde.org>
4443Date:   Fri Nov 9 22:24:52 2018 +0100
4444
4445    locale.dir: Add ast_ES.UTF-8
4446    
4447    With this patch xev properly reports
4448       XLookupString gives 2 bytes: (c2 b4) "´"
4449    for the dead_acute key when using the Asturian locale
4450
4451commit 117bb310b44ce03802c91a70949eb58dcc702a3c
4452Author: Albert Astals Cid <aacid@kde.org>
4453Date:   Fri Nov 9 22:18:43 2018 +0100
4454
4455    compose.dir: Add ast_ES.UTF-8
4456    
4457    With this patch one can properly type dead keys like á when using the
4458    Asturian locale
4459
4460commit d7e22252d68258a63ca537827c99d0ac0bc4b9d8
4461Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4462Date:   Sat Nov 10 13:30:47 2018 -0800
4463
4464    Remove documentation of obsolete B16 & B32 tags in specs/libX11
4465    
4466    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4467
4468commit f3c978476e0be6813268af494efb7ac507451116
4469Author: Matt Turner <mattst88@gmail.com>
4470Date:   Tue Oct 9 10:26:05 2018 -0400
4471
4472    libX11 1.6.7
4473    
4474    Signed-off-by: Matt Turner <mattst88@gmail.com>
4475
4476commit 823a0f8a820247b6c1e092f679b49cbdc2ea5c95
4477Author: Michel Dänzer <michel.daenzer@amd.com>
4478Date:   Fri Sep 28 17:24:17 2018 +0200
4479
4480    poll_for_event: Allow using xcb_poll_for_queued_event
4481    
4482    It avoids reading from the display connection again in cases where that
4483    was already done.
4484    
4485    Suggested-by: Uli Schlachter <psychon@znc.in>
4486    Reviewed-by: Uli Schlachter <psychon@znc.in>
4487
4488commit 406afe4b0f1b655c0db19bbc9a0c48da9a46acf5
4489Author: Michel Dänzer <michel.daenzer@amd.com>
4490Date:   Tue Sep 25 17:10:58 2018 +0200
4491
4492    poll_for_response: Call poll_for_event again if xcb_poll_for_reply fails
4493    
4494    If xcb_poll_for_reply fails to find a reply, poll_for_response would
4495    always return NULL. However, xcb_poll_for_reply may have read events
4496    from the display connection while looking for a reply. In that case,
4497    returning NULL from poll_for_response is wrong and can result in the
4498    client hanging, e.g. because it returns to waiting for the display
4499    connection file descriptor becoming readable after XPending incorrectly
4500    returned 0 pending events.
4501    
4502    The solution is to call poll_for_event again after xcb_poll_for_reply
4503    returned 0. This will return the first of any events read by
4504    xcb_poll_for_reply.
4505    
4506    Fixes issue #79.
4507    
4508    Reported-by: Yuxuan Shui <yshuiv7@gmail.com>
4509    Bugzilla: https://bugs.freedesktop.org/108008
4510    Bugzilla: https://bugs.freedesktop.org/107992
4511    Reviewed-by: Adam Jackson <ajax@redhat.com>
4512
4513commit d0416863d5bf75af54ce81f6c30d4c1476b5e04f
4514Author: Bhavi Dhingra <b.dhingra@samsung.com>
4515Date:   Mon Sep 28 08:33:40 2015 +0000
4516
4517    XcmsLookupColor: fully initialize XColor structs passed to _XColor_to_XcmsRGB
4518    
4519    Fixes https://gitlab.freedesktop.org/xorg/lib/libx11/issues/44
4520    aka https://bugs.freedesktop.org/show_bug.cgi?id=92154
4521    
4522    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4523    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4524
4525commit 733f64bfeb311c1d040b2f751bfdef9c9d0f89ef
4526Author: Matthieu Herrb <matthieu@herrb.eu>
4527Date:   Tue Aug 21 16:54:50 2018 +0200
4528
4529    libX11 1.6.6
4530    
4531    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
4532
4533commit 173704243fbcbea0b6de2068b4fb403c7408fb54
4534Author: Matthieu Herrb <matthieu@herrb.eu>
4535Date:   Tue Aug 21 16:53:40 2018 +0200
4536
4537    Remove statement with no effect.
4538    
4539    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
4540
4541commit e83722768fd5c467ef61fa159e8c6278770b45c2
4542Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4543Date:   Fri Jul 27 16:38:00 2018 +0200
4544
4545    Fixed crash on invalid reply (CVE-2018-14598).
4546    
4547    If the server sends a reply in which even the first string would
4548    overflow the transmitted bytes, list[0] (or flist[0]) will be set to
4549    NULL and a count of 0 is returned.
4550    
4551    If the resulting list is freed with XFreeExtensionList or
4552    XFreeFontPath later on, the first Xfree call:
4553    
4554        Xfree (list[0]-1)
4555     turns into
4556        Xfree (NULL-1)
4557    
4558    which will most likely trigger a segmentation fault.
4559    
4560    I have modified the code to return NULL if the first string would
4561    overflow, thus protecting the freeing functions later on.
4562    
4563    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4564
4565commit dbf72805fd9d7b1846fe9a11b46f3994bfc27fea
4566Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4567Date:   Fri Jul 27 16:37:17 2018 +0200
4568
4569    Fixed out of boundary write (CVE-2018-14600).
4570    
4571    The length value is interpreted as signed char on many systems
4572    (depending on default signedness of char), which can lead to an out of
4573    boundary write up to 128 bytes in front of the allocated storage, but
4574    limited to NUL byte(s).
4575    
4576    Casting the length value to unsigned char fixes the problem and allows
4577    string values with up to 255 characters.
4578    
4579    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4580
4581commit b469da1430cdcee06e31c6251b83aede072a1ff0
4582Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4583Date:   Fri Jul 27 16:36:34 2018 +0200
4584
4585    Fixed off-by-one writes (CVE-2018-14599).
4586    
4587    The functions XGetFontPath, XListExtensions, and XListFonts are
4588    vulnerable to an off-by-one override on malicious server responses.
4589    
4590    The server replies consist of chunks consisting of a length byte
4591    followed by actual string, which is not NUL-terminated.
4592    
4593    While parsing the response, the length byte is overridden with '\0',
4594    thus the memory area can be used as storage of C strings later on. To
4595    be able to NUL-terminate the last string, the buffer is reserved with
4596    an additional byte of space.
4597    
4598    For a boundary check, the variable chend (end of ch) was introduced,
4599    pointing at the end of the buffer which ch initially points to.
4600    Unfortunately there is a difference in handling "the end of ch".
4601    
4602    While chend points at the first byte that must not be written to,
4603    the for-loop uses chend as the last byte that can be written to.
4604    
4605    Therefore, an off-by-one can occur.
4606    
4607    I have refactored the code so chend actually points to the last byte
4608    that can be written to without an out of boundary access. As it is not
4609    possible to achieve "ch + length < chend" and "ch + length + 1 > chend"
4610    with the corrected chend meaning, I removed the inner if-check.
4611    
4612    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4613
4614commit d81da209fd4d0c2c9ad0596a8078e58864479d0d
4615Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4616Date:   Tue Jul 3 22:31:37 2018 +0200
4617
4618    Validation of server response in XListHosts.
4619    
4620    If a server sends an incorrect length in its response, a client is prone
4621    to perform an out of boundary read while processing the data.
4622    
4623    The length field of xHostEntry is used to specify the amount of bytes
4624    used to represent the address. It is 16 bit, which means that it is not
4625    possible to perform an arbitrary memory access, but it might be enough
4626    to read sensitive information, e.g. malloc-related pointers and offsets.
4627    
4628    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4629    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4630
4631commit b676e62377483df77bcb6472d26b24f901323fa9
4632Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
4633Date:   Wed Jun 13 15:46:58 2018 +0200
4634
4635    XkbOpenDisplay.3: fix typo
4636    
4637    XkbOpenDisplay returns a pointer to Display, not a Display.
4638    
4639    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
4640
4641commit 9c5845ff0dc080ff10bd68af4fc40fcd805728ca
4642Author: Martin Natano <natano@natano.net>
4643Date:   Sat Oct 8 19:57:50 2016 +0200
4644
4645    Don't rebuild ks_tables.h if nothing changed.
4646    
4647    ks_tables.h is always considered out of date due to the forced rebuild
4648    of the makekeys util. This means the file is also rebuilt during 'make
4649    install', which is usually performed as root, which can to lead
4650    permission problems later on.
4651    
4652    Signed-off-by: Martin Natano <natano@natano.net>
4653    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4654
4655commit 796f754cba6d75b676a0fc39b97802198fceda4f
4656Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4657Date:   Sat May 5 14:43:30 2018 -0700
4658
4659    Change fall through comment in lcDB.c to match gcc's requirements
4660    
4661    Needs to match one of the regexps shown under
4662    https://gcc.gnu.org/onlinedocs/gcc-7.3.0/gcc/Warning-Options.html#index-Wimplicit-fallthrough
4663    
4664    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4665
4666commit 82ca6308757126fa7ffc6588f1e5d8e3be04251b
4667Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4668Date:   Sat Mar 24 19:45:14 2018 -0700
4669
4670    Use size_t for buffer sizes in SetHints.c
4671    
4672    These variables store values returned from strlen() as a size_t
4673    and are passed to Xmalloc, which expects a size_t, so lets stop
4674    converting back and forth to int along the way.
4675    
4676    Reported by: Konstantin SKliarov
4677    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4678    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4679
4680commit 50a5a98984097d614227d22a49265e12b172cac7
4681Author: Bhavi Dhingra <b.dhingra@samsung.com>
4682Date:   Tue Jul 5 11:37:50 2016 +0530
4683
4684    Fix possible memory leak in cmsProp.c:140
4685    
4686    https://bugs.freedesktop.org/show_bug.cgi?id=96814
4687    
4688    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4689    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4690
4691commit a9dafdd57c71473fa3a2ec4887e973e4e9876d83
4692Author: Michal Srb <msrb@suse.com>
4693Date:   Thu Mar 15 09:50:58 2018 +0100
4694
4695    Use flexible array member instead of fake size.
4696    
4697    The _XimCacheStruct structure is followed in memory by two strings containing
4698    fname and encoding. The memory was accessed using the last member of the
4699    structure `char fname[1]`. That is a lie, prohibits us from using sizeof and
4700    confuses checkers. Lets declare it properly as a flexible array, so compilers
4701    don't complain about writing past that array. As bonus we can replace the
4702    XOffsetOf with regular sizeof.
4703    
4704    Fixes GCC8 error:
4705      In function 'strcpy',
4706          inlined from '_XimWriteCachedDefaultTree' at imLcIm.c:479:5,
4707          inlined from '_XimCreateDefaultTree' at imLcIm.c:616:2,
4708          inlined from '_XimLocalOpenIM' at imLcIm.c:700:5:
4709      /usr/include/bits/string_fortified.h:90:10: error: '__builtin_strcpy'
4710      forming offset 2 is out of the bounds [0, 1] [-Werror=array-bounds]
4711         return __builtin___strcpy_chk (__dest, __src, __bos (__dest));
4712    
4713    Caused by this line seemingly writing past the fname[1] array:
4714      imLcIm.c:479:  strcpy (m->fname+strlen(name)+1, encoding);
4715    
4716    Reviewed-by: Keith Packard <keithp@keithp.com>
4717    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4718
4719commit 34f4464f69a4d6ff0d1042500a62f9a5ed7f3647
4720Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4721Date:   Tue Mar 6 11:42:27 2018 -0800
4722
4723    If XGetImage fails to create image, don't dereference it to bounds check
4724    
4725    Reported by gcc 7.3:
4726    
4727    GetImage.c:110:25: warning: potential null pointer dereference [-Wnull-dereference]
4728      if (planes < 1 || image->height < 1 || image->bytes_per_line < 1 ||
4729                        ~~~~~^~~~~~~~
4730    
4731    Introduced by 8ea762f94f4c942d898fdeb590a1630c83235c17 in Xlib 1.6.4
4732    
4733    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4734    Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
4735
4736commit e835a9dcc3362b5e92893be756dd7ae361e64ced
4737Author: wharms <wharms@bfs.de>
4738Date:   Sun Sep 3 14:17:45 2017 +0200
4739
4740    silence gcc warning assignment discards 'const' qualifier from pointer target type
4741
4742commit 36a1ac0253fea82ff79cc52ba56c5691cfd07a3b
4743Author: wharms <wharms@bfs.de>
4744Date:   Sun Aug 20 21:51:57 2017 +0200
4745
4746    remove empty line
4747
4748commit e02dfe54f32b4165351d2712a9d2e0584906a3ce
4749Author: wharms <wharms@bfs.de>
4750Date:   Sun Aug 20 21:50:33 2017 +0200
4751
4752    add _X_UNUSED to avoid unused variable warnings
4753
4754commit 2911c39cecd63ed3747072a5eeeb9eedffc881e9
4755Author: walter harms <wharms@bfs.de>
4756Date:   Sat Jun 4 17:19:59 2016 +0200
4757
4758    Fixes: warning: variable 'req' set but not,used
4759    
4760    Fixes: warning: variable 'req' set but not used [-Wunused-but-set-variable]
4761           by marking req _X_UNUSED
4762            Solution was discussed on xorg-devel ML
4763           Peter Hutter, Alan Coopersmith
4764            Re: [PATCH libX11 3/5] fix: warning: pointer targets in passing argument 2 of '_XSend' differ in signedness [-Wpointer-sign]
4765    
4766    Signed-off-by: harms wharms@bfs.de
4767
4768commit bf82ec0402479fd8399d69e7d62fc17d7956699a
4769Author: walter harms <wharms@bfs.de>
4770Date:   Sat Jun 4 17:22:07 2016 +0200
4771
4772    mark _XDefaultIOError as no_return
4773    
4774    mark _XDefaultIOError as no_return. No one comes back from exit() ...
4775    
4776    Signed-off-by: harms wharms@bfs.de
4777
4778commit 9abe8380074edea3ac4e72466ec47e921ca05e47
4779Author: walter harms <wharms@bfs.de>
4780Date:   Sat Jun 4 17:21:52 2016 +0200
4781
4782    no need to check XFree arguments
4783    
4784    You can save a bit of code. The is no need to check XFree arguments bring  free_fontdataOM in line with other free function and check for NULL arg
4785    
4786    Signed-off-by: harms wharms@bfs.de
4787
4788commit 433477fcb7e07d0c26a22ba78aae88827ed1f440
4789Author: walter harms <wharms@bfs.de>
4790Date:   Tue Apr 26 16:32:20 2016 +0200
4791
4792    fix memleak in error path
4793    
4794    free all mem on error
4795    Signed-off-by: walter harms <wharms@bfs.de>
4796
4797commit ed9f0d34abc645eee56e21863f23acb4d0bb8e9a
4798Author: walter harms <wharms@bfs.de>
4799Date:   Tue Apr 26 17:58:16 2016 +0200
4800
4801    fix memleak in error path
4802    
4803    V2: remove unneeded NULL (reported by eric.engestrom@imgtec.com)
4804    
4805    fix mem leak in error path
4806    Signed-off-by: walter harms <wharms@bfs.de>
4807
4808commit 7c78fc57693afa94cf26170f0f6276e3b7374ed0
4809Author: walter harms <wharms@bfs.de>
4810Date:   Tue Apr 26 16:34:11 2016 +0200
4811
4812    no need to check args for Xfree()
4813    
4814    simplify code
4815    
4816    Signed-off-by: walter harms <wharms@bfs.de>
4817
4818commit c1c14af441ae73d1a8e67a971fafcf967e45ac48
4819Author: walter harms <wharms@bfs.de>
4820Date:   Tue Apr 26 16:23:46 2016 +0200
4821
4822    remove stray extern
4823    
4824    remove stray extern
4825    
4826    Signed-off-by: walter harms <wharms@bfs.de>
4827
4828commit 714921f041a245dc5f37a689268b584226a2ccb9
4829Author: walter harms <wharms@bfs.de>
4830Date:   Mon Apr 11 18:26:52 2016 +0200
4831
4832    no need to check argument for _XkbFree()
4833    
4834    simplify code by removing unneeded checks
4835    
4836    Signed-off-by: walter harms <wharms@bfs.de>
4837
4838commit d02c2466f65063a03c97dbcee05071c12a3676e6
4839Author: walter harms <wharms@bfs.de>
4840Date:   Mon Apr 11 18:22:38 2016 +0200
4841
4842    fix more shadow warning
4843    
4844    Signed-off-by: walter harms <wharms@bfs.de>
4845
4846commit 0355c3926d5372f9762f235071dbd94a89bbbdad
4847Author: walter harms <wharms@bfs.de>
4848Date:   Thu Mar 31 19:16:33 2016 +0200
4849
4850    fix shadow char_size
4851    
4852    Signed-off-by: walter harms <wharms@bfs.de>
4853
4854commit 916dffadf052135df3398651be873c353da629e1
4855Author: walter harms <wharms@bfs.de>
4856Date:   Thu Mar 31 19:14:32 2016 +0200
4857
4858    remove argument check for free() adjust one inden
4859    
4860    Signed-off-by: walter harms <wharms@bfs.de>
4861
4862commit 6ec901ebca3fea6a762e22090dc35b1b90911133
4863Author: walter harms <wharms@bfs.de>
4864Date:   Thu Mar 31 19:12:17 2016 +0200
4865
4866    _XIOError(dpy); will never return so remore dead
4867    
4868    Signed-off-by: walter harms <wharms@bfs.de>
4869
4870commit 83107a677b2ed458e4d62ea4a601e8181d3683d8
4871Author: walter harms <wharms@bfs.de>
4872Date:   Thu Mar 31 19:10:49 2016 +0200
4873
4874    fix shadow warning
4875    
4876    Signed-off-by: walter harms <wharms@bfs.de>
4877
4878commit 4fe66b1c5112b07bd09e28bbc021911d08a9621f
4879Author: Ryan C. Gordon <icculus@icculus.org>
4880Date:   Wed Aug 2 02:41:03 2017 -0400
4881
4882    Valgrind fix for XStoreColor and XStoreColors.
4883    
4884    If the "pad" field isn't set, Valgrind will report it as uninitialized
4885    memory accesses when the struct is copied into the Display's send buffer.
4886    
4887    In practice, this is (probably) harmless, but Valgrind is correct in
4888    believing it's a bug.
4889    
4890    https://bugs.freedesktop.org/attachment.cgi?id=133189
4891    
4892    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4893    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4894
4895commit 7d2010fec25c2f52b873ad0572479eb43128b038
4896Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4897Date:   Fri Apr 7 00:13:03 2017 -0700
4898
4899    Improve table formatting in XkbChangeControls & XkbKeyNumGroups man pages
4900    
4901    Includes fix for Solaris Bug 24564279: "XkbKeyNumGroups.3x11 man page
4902    contains some malformed text" caused by extra whitespace after .TE macros
4903    
4904    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4905
4906commit b856d5d929047d1ea169814d56e43784ea404c83
4907Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4908Date:   Thu Mar 16 00:05:49 2017 -0700
4909
4910    Clarify state parameter to XkbSetNamedDeviceIndicator
4911    
4912    Checking a Bool to see if it's NULL does not work well in C.
4913    Also reported in https://bugs.freedesktop.org/show_bug.cgi?id=251
4914    
4915    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4916    Reviewed-by: Adam Jackson <ajax@redhat.com>
4917
4918commit c6dadd4cebd994aafb37a58b3adbaa82507c2d18
4919Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4920Date:   Wed Mar 15 23:50:26 2017 -0700
4921
4922    Make Xkb{Get,Set}NamedIndicator spec & manpages match code
4923    
4924    The XKB Library spec and the man pages for XkbGetNamedIndicator &
4925    XkbSetNamedIndicator included a device_spec argument neither function
4926    takes, and do not include the XkbGetNamedDeviceIndicator &
4927    XkbSetNamedDeviceIndicator variants that do take it (along with two
4928    other arguments).
4929    
4930    This updates them to match the interfaces the code has provided for
4931    decades.
4932    
4933    This has been reported multiple times, so this fixes:
4934    https://bugs.freedesktop.org/show_bug.cgi?id=251
4935    https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=729812
4936    Sun Bug 4528016 XkbSetNamedIndicator & XkbGetNamedIndicator man pages are wrong
4937      (filed: alan.coopersmith@sun.com 2001-11-15 - now aka Oracle bug 15087506)
4938    X.Org Group Defect Id #9418
4939    
4940    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4941    Reviewed-by: Adam Jackson <ajax@redhat.com>
4942
4943commit 2d20890e7ffd3ee88a9ceb25cdd2ac1fe7aaceb6
4944Author: Arthur Huillet <ahuillet@nvidia.com>
4945Date:   Wed Feb 1 15:02:41 2017 +0100
4946
4947    _XDefaultError: set XlibDisplayIOError flag before calling exit
4948    
4949    _XReply isn't reentrant, and it can lead to deadlocks when the default error
4950    handler is called: _XDefaultError calls exit(1). It is called indirectly by
4951    _XReply when a X protocol error comes in that isn't filtered/handled by an
4952    extension or the application. This means that if the application (or one of its
4953    loaded shared libraries such as the NVIDIA OpenGL driver) has registered any
4954    _fini destructor, _fini will get called while still on the call stack of
4955    _XReply. If the destructor interacts with the X server and calls _XReply, it
4956    will hit a deadlock, looping on the following in _XReply:
4957    
4958        ConditionWait(dpy, dpy->xcb->reply_notify);
4959    
4960    It is legal for an application to make Xlib calls during _fini, and that is
4961    useful for an OpenGL driver to avoid resource leaks on the X server side, for
4962    example in the dlopen/dlclose case. However, the driver can not readily tell
4963    whether its _fini is being called because Xlib called exit, or for another
4964    reason (dlclose), so it is hard to cleanly work around this issue in the driver.
4965    
4966    This change makes it so _XReply effectively becomes a no-op when called after
4967    _XDefaultError was called, as though an XIOError had happened. The dpy
4968    connection isn't broken at that point, but any call to _XReply is going to hang.
4969    This is a bit of a kludge, because the more correct solution would be to make
4970    _XReply reentrant, maybe by broadcasting the reply_notify condition before
4971    calling the default error handler. However, such a change would carry a grater
4972    risk of introducing regressions in Xlib.
4973    
4974    This change will drop some valid requests on the floor, but this should not
4975    matter, as it will only do so in the case where the application is dying: X will
4976    clean up after it once exit() is done running. There is the case of
4977    XSetCloseDownMode(RETAIN_PERMANENT), but an application using that and wishing
4978    to clean up resources in _fini would currently be hitting a deadlock, which is
4979    hardly a better situation.
4980    
4981    Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
4982    Reviewed-by: Jamey Sharp <jamey@minilop.net>
4983
4984commit 42f4d7af9cf6d1dbfa575552e057328b054a20c9
4985Author: Matt Turner <mattst88@gmail.com>
4986Date:   Sat Feb 25 21:54:22 2017 -0800
4987
4988    libX11 1.6.5
4989    
4990    Signed-off-by: Matt Turner <mattst88@gmail.com>
4991
4992commit a0da5835e8078445947c828fe2d86c1a31439012
4993Author: Adam Jackson <ajax@redhat.com>
4994Date:   Tue Feb 14 15:33:29 2017 -0500
4995
4996    Revert "Compose sequences for rouble sign"
4997    
4998    This reverts commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0.
4999    
5000    Reported to break 'make check':
5001    
5002    https://lists.freedesktop.org/archives/xorg-devel/2017-February/052720.html
5003
5004commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0
5005Author: Mihail Konev <k.mvc@ya.ru>
5006Date:   Fri Feb 10 18:48:18 2017 +0500
5007
5008    Compose sequences for rouble sign
5009    
5010    Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout.
5011    Also add Cyrillic sequences for hryvnia sign.
5012    
5013    Submitted-by: Victor V. Kustov <coyote@bks.tv>
5014    Reviewed-by: Victor V. Kustov <coyote@bks.tv>
5015    Signed-off-by: Mihail Konev <k.mvc@ya.ru>
5016
5017commit 23d9623c661694aba8cf1e8f277dffa7a86cf065
5018Author: Petr Písař <petr.pisar@atlas.cz>
5019Date:   Sun Oct 30 12:49:11 2016 +0100
5020
5021    Revert cs_CZ.UTF-8 XLC_LOCALE to en_US.UTF-8
5022    
5023    The cs_CZ.UTF-8/XLC_LOCALE is an empty file leading to unsupported cs_CZ.UTF-8
5024    locale and reporting this error:
5025    
5026    Warning: locale not supported by Xlib, locale set to C
5027    
5028    Therefore this patch reverts to the en_US.UTF-8 definition file that was used
5029    before. This patch also deduplicates the cs_CZ.UTF-8 entry.
5030    
5031    <https://bugs.freedesktop.org/show_bug.cgi?id=98219>
5032    
5033    This reverts commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51
5034    
5035    Signed-off-by: Julien Cristau <jcristau@debian.org>
5036
5037commit 71b0929ebc1f0f877f63e3f6de260f529daa6c69
5038Author: Mihail Konev <k.mvc@ya.ru>
5039Date:   Thu Jan 26 13:52:49 2017 +1000
5040
5041    autogen: add default patch prefix
5042    
5043    Signed-off-by: Mihail Konev <k.mvc@ya.ru>
5044
5045commit 2979011bc170c55894b9185b26376f8efc6db7d4
5046Author: Emil Velikov <emil.l.velikov@gmail.com>
5047Date:   Mon Mar 9 12:00:52 2015 +0000
5048
5049    autogen.sh: use quoted string variables
5050    
5051    Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent
5052    fall-outs, when they contain space.
5053    
5054    Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
5055    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5056    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5057
5058commit 4a0082a1b6affa65d38294f0e13511525cd8ad15
5059Author: Peter Hutterer <peter.hutterer@who-t.net>
5060Date:   Tue Jan 24 10:32:07 2017 +1000
5061
5062    autogen.sh: use exec instead of waiting for configure to finish
5063    
5064    Syncs the invocation of configure with the one from the server.
5065    
5066    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5067    Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
5068
5069commit c74b070f2712c95f0db7c320a10232b0e5c83049
5070Author: Julien Cristau <jcristau@debian.org>
5071Date:   Sat Jan 7 16:20:31 2017 +0100
5072
5073    Fix wrong Xfree in XListFonts failure path
5074    
5075    'ch' gets moved inside the allocated buffer as we're looping through
5076    fonts, so keep a reference to the start of the buffer so we can pass
5077    that to Xfree in the failure case.
5078    
5079    Fixes: commit 20a3f99eba5001925b8b313da3accb7900eb1927 "Plug a memory leak"
5080    
5081    Signed-off-by: Julien Cristau <jcristau@debian.org>
5082    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5083    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5084
5085commit 663f47075fe67bab4f99bc5d186c49175d4a4334
5086Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5087Date:   Sun Jan 1 20:57:58 2017 -0800
5088
5089    specs/libX11: Update Portability Considerations for the 21st century
5090    
5091    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5092    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5093
5094commit 28f4b989b52fcf45c7e75a878d8d9c2583cd6b3c
5095Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5096Date:   Sun Jan 1 20:39:53 2017 -0800
5097
5098    specs/libX11: Fix broken synopsis for Data/Data16/Data32
5099    
5100    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5101
5102commit 382561951e3460b09c21a1a23748cde0315fbb19
5103Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5104Date:   Sun Jan 1 20:31:54 2017 -0800
5105
5106    specs/libX11: Add missing parameter types for XGetWindowProperty()
5107    
5108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5109
5110commit 2beaecdb66965b861d6b790d151ba947f65f0a22
5111Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5112Date:   Sun Jan 1 20:16:08 2017 -0800
5113
5114    specs/libX11: Make paramdef spacing more consistent
5115    
5116    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5117
5118commit 4c436c6c14cfd4c397b011563bf13c2872861133
5119Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5120Date:   Sun Jan 1 20:02:53 2017 -0800
5121
5122    specs/libX11: Fix paramdef entries listing multiple parameters
5123    
5124    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5125
5126commit 1728b1a8a4718315da8e5c9cbc2c04bb75c74107
5127Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5128Date:   Sun Jan 1 19:38:40 2017 -0800
5129
5130    specs/libX11: More synopsis fixes
5131    
5132    Mostly transforming macro definitions and functions taking void arguments
5133    from undecorated <para> tags to use <funcsynopsis> tags to get decorations.
5134    
5135    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5136
5137commit f0dc83db7e3a3d4a76c0f9d24763b80f01c893a9
5138Author: Lucien Gentis <lucien.gentis@waika9.com>
5139Date:   Wed Jun 1 17:02:47 2016 +0200
5140
5141    Typos in "Xlib - C Language X Interface" document - Chapter 02
5142    
5143    This patch fixes typos and lack of tags in "Xlib - C Language X Interface" document - Chapter 02.
5144    
5145    Signed-off-by: Lucien Gentis <lucien.gentis@waika9.com>
5146    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5147    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5148
5149commit 20a3f99eba5001925b8b313da3accb7900eb1927
5150Author: Emilio Pozuelo Monfort <pochu@debian.org>
5151Date:   Tue Oct 25 21:30:15 2016 +0200
5152
5153    Plug a memory leak
5154    
5155    This was introduced in 8ea762f.
5156    
5157    Reported-by: Julien Cristau <jcristau@debian.org>
5158    Signed-off-by: Emilio Pozuelo Monfort <pochu@debian.org>
5159    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5160
5161commit 8f349feac24aacc958bd816afcc52380764e3d92
5162Author: Matthieu Herrb <matthieu.herrb@laas.fr>
5163Date:   Tue Oct 4 21:01:39 2016 +0200
5164
5165    libX11 1.6.4
5166    
5167    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
5168
5169commit 8ea762f94f4c942d898fdeb590a1630c83235c17
5170Author: Tobias Stoeckmann <tobias@stoeckmann.org>
5171Date:   Sun Sep 25 21:25:25 2016 +0200
5172
5173    Validation of server responses in XGetImage()
5174    
5175    Check if enough bytes were received for specified image type and
5176    geometry. Otherwise GetPixel and other functions could trigger an
5177    out of boundary read later on.
5178    
5179    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
5180    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5181
5182commit 8c29f1607a31dac0911e45a0dd3d74173822b3c9
5183Author: Tobias Stoeckmann <tobias@stoeckmann.org>
5184Date:   Sun Sep 25 21:22:57 2016 +0200
5185
5186    The validation of server responses avoids out of boundary accesses.
5187    
5188    v2: FontNames.c  return a NULL list whenever a single
5189    length field from the server is incohent.
5190    
5191    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
5192    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5193
5194commit 78851f6a03130e3c720b60c3cbf96f8eb216d741
5195Author: walter harms <wharms@bfs.de>
5196Date:   Mon Aug 15 19:18:14 2016 +0200
5197
5198    XFree will accept NULL as argument
5199    
5200    since Xfree is a define for free():
5201      Xlibint.h:#define Xfree(ptr) free((ptr))
5202    
5203    Xfree will accept NULL and do nothing.
5204    
5205    Signed-off-by: Hans de Goede <hdegoede@redhat.com>
5206
5207commit 83adf3d1e3d0d6602244381334f75c216da4ab6e
5208Author: Matthew D. Fuller <fullermd@over-yonder.net>
5209Date:   Sat Jun 4 11:24:01 2016 -0500
5210
5211    Fixup param specification for XChangeProperty()
5212    
5213    Signed-off-by: Matthew D. Fuller <fullermd@over-yonder.net>
5214    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5215
5216commit 3129c757f9da8586ab8b8654a56c8f687cc9ef5c
5217Author: Mats Blakstad <mats.gbproject@gmail.com>
5218Date:   Sun Feb 28 13:22:03 2016 -0500
5219
5220     New compose keys for local languages in Togo
5221    
5222    Signed-off-by: James Cloos <cloos@jhcloos.com>
5223
5224commit e1011b9e2f6c82255959cf3cc1d8cda402ded0a9
5225Author: Daniel Albers <daniel@lbe.rs>
5226Date:   Wed Mar 9 14:35:48 2016 +0100
5227
5228    Add Compose sequence for U+1F4A9.
5229    
5230    Signed-off-by: Daniel Albers <daniel@lbe.rs>
5231
5232commit 6d7bb040c928485f2557c2c914b95cffb2354179
5233Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5234Date:   Sat Feb 6 14:18:32 2016 -0800
5235
5236    xcms: use size_t for pointer offsets passed to strncmp
5237    
5238    instead of converting to int and back
5239    
5240    Fixes clang warnings of the form:
5241    HVC.c:190:43: warning: implicit conversion changes signedness: 'int' to
5242          'unsigned long' [-Wsign-conversion]
5243              if (strncmp(spec, _XcmsTekHVC_prefix, n) != 0) {
5244                  ~~~~~~~
5245    
5246    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5247
5248commit a9266804eed38a83897ab5f0f9f8a8ab82a98882
5249Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5250Date:   Sat Feb 6 13:32:44 2016 -0800
5251
5252    xcms: use unsigned indexes when looping through unsigned values
5253    
5254    Clears many gcc warnings of the form:
5255    
5256    uvY.c: In function ‘XcmsCIEuvYToCIEXYZ’:
5257    uvY.c:263:19: warning: comparison between signed and unsigned integer
5258      expressions [-Wsign-compare]
5259         for (i = 0; i < nColors; i++, pColor++) {
5260                       ^
5261    
5262    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5263
5264commit 0ee0d383b4488b7b90d8bd50b75c371e0dc0d397
5265Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5266Date:   Sat Feb 6 13:01:25 2016 -0800
5267
5268    xcms: use size_t for strlen/sizeof values instead of converting to int & back
5269    
5270    Fixes gcc warnings of the form:
5271    
5272    IdOfPr.c: In function ‘XcmsFormatOfPrefix’:
5273    IdOfPr.c:69:32: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
5274         if ((len = strlen(prefix)) >= sizeof(string_buf)) {
5275                                    ^
5276    IdOfPr.c:83:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
5277       if (len >= sizeof(string_buf)) Xfree(string_lowered);
5278               ^
5279    IdOfPr.c:97:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
5280       if (len >= sizeof(string_buf)) Xfree(string_lowered);
5281               ^
5282    IdOfPr.c:104:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
5283         if (len >= sizeof(string_buf)) Xfree(string_lowered);
5284                 ^
5285    
5286    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5287
5288commit 4de6ed3e7b1833c52c9d58ab74d59d57ca2a9f0d
5289Author: Dominik Muth <nxdomainuser-muth@yahoo.com>
5290Date:   Thu Mar 26 07:52:58 2015 +0100
5291
5292    Xlib.h: Fix macros imitating C functions.
5293    
5294    The basic rule "put parantheses around macro parameters" should be
5295    observed where possible. Otherwise code like
5296    
5297        ConnectionNumber(foo = bar);
5298    
5299    fails to compile. (It obviously passes if ConnectionNumber is a C
5300    function.) There are several other macros amended for the same reason.
5301    
5302    This bug appeared while building http://ioccc.org/1993/cmills.c, so
5303    historically it was not present.
5304    
5305    Signed-off-by: Dominik Muth <muth@nxdomain.no-ip.biz>
5306    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5307    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5308
5309commit 3706b0f2b14cc97578a6bee620266edca2722ebf
5310Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5311Date:   Fri Nov 15 18:03:25 2013 -0800
5312
5313    Don't need to link libX11-xcb against libX11
5314    
5315    libX11-xcb only accesses data structures defined in X11 headers,
5316    it doesn't call any functions or reference any global variables
5317    in libX11 itself.  (Seems to have been left from previous XCL
5318    implementation.)
5319    
5320    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5321
5322commit eddf1bbd18872b286a9f939140f0cd9ba4e93804
5323Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5324Date:   Fri Jan 22 11:44:25 2016 -0800
5325
5326    Stop checking for preferred order of local transports
5327    
5328    Removes --with-local-transport-order=... flag to configure.
5329    
5330    Code which used this ordered list was removed in commit 15e5eaf6289
5331    which outsourced X11 connection handling & authentication to libxcb.
5332    
5333    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5334    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5335
5336commit 1a66c1e964ff8d11382313404f48b5a3d5ed8be8
5337Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5338Date:   Fri Jan 22 09:39:28 2016 -0800
5339
5340    Stop checking XTRANS_SECURE_RPC_FLAGS since we no longer use them
5341    
5342    Removes --enable-secure-rpc & --disable-secure-rpc flags to configure
5343    
5344    Code that used SECURE_RPC definitions was removed in commit 15e5eaf6289
5345    which outsourced X11 connection handling & authentication to libxcb.
5346    
5347    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5348    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5349
5350commit 7eb724dc24505f1591ef32620fa63f079b540646
5351Author: Olivier Fourdan <ofourdan@redhat.com>
5352Date:   Thu Jan 21 11:54:19 2016 +0100
5353
5354    XKB: fix XkbGetKeyboardByName with Xming server
5355    
5356    XkbGetKeyboardByName relies on flags to read the data from the server.
5357    
5358    If the X server sends us the wrong flags or if a subreply is smaller
5359    than it should be, XkbGetKeyboardByName will not read all the available
5360    data and leave data in the buffer, which will cause the next _XReply()
5361    to fail with:
5362    
5363    [xcb] Extra reply data still left in queue
5364    [xcb] This is most likely caused by a broken X extension library
5365    [xcb] Aborting, sorry about that.
5366    xcb_io.c:576: _XReply: Assertion `!xcb_xlib_extra_reply_data_left' failed.
5367    Aborted
5368    
5369    Check if there is some extra data left at the end of
5370    XkbGetKeyboardByName() and discard that data if any is found.
5371    
5372    Many thanks to Peter Hutterer <peter.hutterer@who-t.net> for finding the
5373    root cause of the issue and Adam Jackson <ajax@redhat.com> for helping
5374    with the analysis!
5375    
5376    Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
5377    Reviewed-by: Daniel Stone <daniels@collabora.com>
5378    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5379    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5380
5381commit 43ba0a68d3d17b496ec1f48d44921122ddd7d7d9
5382Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5383Date:   Sat Dec 19 18:03:41 2015 -0800
5384
5385    lcPubWrap: replace malloc(strlen) + strcpy with strdup
5386    
5387    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5388    Reviewed-by: Adam Jackson <ajax@redhat.com>
5389
5390commit 6fc95cb12b70c5a67cb4fc5e5749f9f1ec741e2a
5391Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5392Date:   Sat Dec 19 10:21:04 2015 -0800
5393
5394    XlcDL.c: reduce code duplication
5395    
5396    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5397    Reviewed-by: Adam Jackson <ajax@redhat.com>
5398
5399commit f7ecc0856be58608881d2086954cb71857ad64e1
5400Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5401Date:   Sat Dec 19 10:19:25 2015 -0800
5402
5403    XlcDL.c: replace strcpy+strcat sequences with snprintf
5404    
5405    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5406    Reviewed-by: Adam Jackson <ajax@redhat.com>
5407
5408commit 522989b34398bd6a6ea144c4af0ba69d6dc4faea
5409Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5410Date:   Sat Dec 19 10:05:42 2015 -0800
5411
5412    XDefaultOMIF: Remove comments referring to ancient Sun bug ids
5413    
5414    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5415    Reviewed-by: Adam Jackson <ajax@redhat.com>
5416
5417commit b738a104ae80e4270dd1d215ad0c6a80016982c2
5418Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5419Date:   Sat Dec 19 10:00:22 2015 -0800
5420
5421    XDefaultOMIF: additional code simplification
5422    
5423    Don't need to test for a case that we already returned for, don't need
5424    to store a count that will only ever be 1 if we didn't return, don't
5425    need to increment pointers to allow storing more than one item when we
5426    can only ever possibly do one.
5427    
5428    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5429    Reviewed-by: Adam Jackson <ajax@redhat.com>
5430
5431commit 31011cf100419269eae7409581c784638be503cf
5432Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5433Date:   Sat Dec 19 09:46:31 2015 -0800
5434
5435    XDefaultOMIF: replace strlen+Xmalloc+strcpy with strdup
5436    
5437    Code seems to have been originally written to handle appending multiple
5438    strings, but only ever operates on a single string.
5439    
5440    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5441    Reviewed-by: Adam Jackson <ajax@redhat.com>
5442
5443commit c27c46d5e22bbf60fb5608eaabe584b7fdeb0b09
5444Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5445Date:   Sat Dec 19 09:20:55 2015 -0800
5446
5447    Use strdup instead of Xmalloc+strcpy in _XDefaultOpenIM
5448    
5449    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5450    Reviewed-by: Adam Jackson <ajax@redhat.com>
5451
5452commit 4359dfabc04af082872d2bc2d5b52e26d6d93290
5453Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5454Date:   Fri Dec 4 22:20:53 2015 -0800
5455
5456    Delete #if 0 hunks of code
5457    
5458    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5459
5460commit a2f9dfac286f37e54eb47d4736cc3f0150224a84
5461Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5462Date:   Thu Dec 3 23:38:07 2015 -0800
5463
5464    Bug 93183: _XDefaultOpenIM memory leaks in out-of-memory error paths
5465    
5466    Rework code to store allocations directly into XIM struct instead of
5467    temporary local variables, so we can use _XCloseIM to unwind instead
5468    of duplicating it, and consistently jump to error handler on failure,
5469    instead of sometimes leaking and sometimes freeing.
5470    
5471    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93183
5472    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5473
5474commit 07a97b3944467dce085a1efd24706cc851d2caf2
5475Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5476Date:   Thu Dec 3 23:19:48 2015 -0800
5477
5478    Bug 93184: read_EncodingInfo invalid free
5479    
5480    Free the correct bits of memory if we run out and need to unwind
5481    
5482    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93184
5483    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5484
5485commit 11118e9eb3705fcbe42b6a68d4a8aa86ab0211f1
5486Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5487Date:   Sat Nov 28 13:18:11 2015 -0800
5488
5489    Remove unused definition of XCONN_CHECK_FREQ
5490    
5491    The only use of XCONN_CHECK_FREQ was removed in commit 15e5eaf62897b3179
5492    when we dropped the old Xlib connection handling in favor of xcb's.
5493    
5494    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5495    Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
5496
5497commit 5f0da8311a61498edf073cc877f5b467bfd5f863
5498Author: James Cloos <cloos@jhcloos.com>
5499Date:   Thu Dec 3 18:24:44 2015 -0500
5500
5501    Fix another missing update in cf4d5989383a
5502    
5503    Reported in:
5504    
5505       https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7
5506    
5507    Signed-off-by: James Cloos <cloos@jhcloos.com>
5508
5509commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51
5510Author: James Cloos <cloos@jhcloos.com>
5511Date:   Thu Dec 3 18:15:40 2015 -0500
5512
5513    Fix missing update in cf4d5989383a
5514    
5515    Reported in:
5516    
5517       https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7
5518    
5519    Signed-off-by: James Cloos <cloos@jhcloos.com>
5520
5521commit dbcb847a08c44d99e4e1de2ba777d63238fb0e03
5522Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5523Date:   Sun Sep 27 18:38:32 2015 -0700
5524
5525    Get rid of some extraneous ; at the end of C source lines
5526    
5527    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5528    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
5529
5530commit 121a1bad334459f66f78bfca6df53dc841cf97f8
5531Author: Gunnar Hjalmarsson <gunnarhj@ubuntu.com>
5532Date:   Wed Sep 23 11:44:55 2015 -0400
5533
5534    Add compose file for pt_PT similar to pt_BR
5535    
5536    This is a forward of the Ubuntu bug https://launchpad.net/bugs/518056
5537    
5538    One of the conclusions from the discussion on that bug report, which
5539    basically is about typing the ccedilla character easily on a non-
5540    Portuguese keyboard, is that X11 should include a compose file for
5541    pt_PT.UTF-8 similar to the file for pt_BR.UTF-8.
5542    
5543    FDO bug: https://bugs.freedesktop.org/show_bug.cgi?id=90300
5544    
5545    Signed-off-by: Gunnar Hjalmarsson <gunnarhj@ubuntu.com>
5546    Signed-off-by: James Cloos <cloos@jhcloos.com>
5547
5548commit 3f41d8a7f82eb5ffbd5c5d36472cf7043186b904
5549Author: Julien Cristau <jcristau@debian.org>
5550Date:   Fri May 1 13:50:15 2015 +0200
5551
5552    Mark _XNextRequest as hidden
5553    
5554    It's only used inside XNextRequest(), so doesn't need to be exported.
5555    
5556    Signed-off-by: Julien Cristau <jcristau@debian.org>
5557    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5558
5559commit a72d2d06c002b644b7040a0a9936c8525e092ba8
5560Author: Christian Linhart <chris@demorecorder.com>
5561Date:   Mon Sep 7 17:17:32 2015 +0200
5562
5563    fix for Xlib 32-bit request number issues
5564    
5565    Make use of the new 64-bit sequence number API in XCB 1.11.1 to avoid
5566    the 32-bit sequence number wrap in libX11.
5567    
5568    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71338
5569    Signed-off-by: Christian Linhart <chris@demorecorder.com>
5570    Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
5571    Reviewed-by: Adam Jackson <ajax@redhat.com>
5572
5573commit 58af066a764305c506efea7065ef7679369a1a98
5574Author: Thomas Klausner <wiz@NetBSD.org>
5575Date:   Sun Jul 19 10:23:21 2015 +0200
5576
5577    Ignore test-driver (used by newer autoconf).
5578    
5579    Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
5580    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5581
5582commit 80b9a346b9ba200fa4652560282e80d249519287
5583Author: Thomas Klausner <wiz@NetBSD.org>
5584Date:   Sun Jul 19 10:22:45 2015 +0200
5585
5586    Do not return() after exit().
5587    
5588    Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
5589    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5590
5591commit c827edcd1c4a7f920aa25208083b5b58d60d2b44
5592Author: Ross Burton <ross.burton@intel.com>
5593Date:   Mon May 18 14:49:01 2015 +0100
5594
5595    Add missing NULL checks to ICWrap
5596    
5597    ICWrap.c dereferences the xim parameter passed in from client code without a
5598    NULL check.  I have seen mplayer trigger this resulting in a segfault.  In this
5599    case mplayer had called XOpenIM and NULL was returned which was later passed
5600    into XCreateIC.
5601    
5602    Patch originally by Drew Moseley <drew_moseley@mentor.com>.
5603    
5604    Signed-off-by: Ross Burton <ross.burton@intel.com>
5605    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5606    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5607
5608commit 26e0d2de294f8adf1ce65f1dbff0b59af41a00b9
5609Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5610Date:   Thu Jun 4 20:51:17 2015 -0700
5611
5612    Replace Xmalloc+memset pairs with Xcalloc calls
5613    
5614    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5615
5616commit f0286b2770ece10aef5e2e8c004260217f12fd25
5617Author: Bhavi Dhingra <b.dhingra@samsung.com>
5618Date:   Thu Jun 4 19:07:12 2015 -0700
5619
5620    omGeneric.c: Correct the parameter usage of sizeof
5621    
5622    Incorrect parameter usage with sizeof. Earlier passed argument FontData
5623    will be 4 bytes always as its a pointer hence the change is needed and
5624    FontDataRec should be used for memset.
5625    
5626    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5627    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5628
5629commit 47da70d75f9e48e800719c0db752f9ccd2d77aea
5630Author: Peter Hutterer <peter.hutterer@who-t.net>
5631Date:   Tue May 19 12:30:22 2015 +1000
5632
5633    Fix three "use of uninitialized variable" coverity warnings
5634    
5635    False positive, if rlen/nbytes are unset we quit early before using it. Still,
5636    initialize it so we don't have to deal with these warnings again.
5637    
5638    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5639    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
5640    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5641
5642commit 19a30f17f30e9ae9641a7c0634fc52134208b060
5643Author: Peter Hutterer <peter.hutterer@who-t.net>
5644Date:   Mon May 18 07:56:22 2015 +1000
5645
5646    Fix an indentation issue
5647    
5648    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5649    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
5650    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5651
5652commit 013ccece124b990217ad3bcf2c41688e8fda1df8
5653Author: Peter Hutterer <peter.hutterer@who-t.net>
5654Date:   Mon May 18 07:55:17 2015 +1000
5655
5656    Fix potential memory leak
5657    
5658    If we hit the depth limit, filename leaks. Move the depth check up before we
5659    allocate filename.
5660    Introduced in 226622349a4b1e16064649d4444a34fb4be4f464.
5661    
5662    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5663    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
5664    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5665
5666commit d3415d1f052530760b4617db45affcb984cfe35c
5667Author: Mike FABIAN <mfabian@redhat.com>
5668Date:   Mon Apr 20 17:59:30 2015 +0200
5669
5670    Fix spelling mistake introduced by 748d47e69f5c12d8557d56a8a8ec166588da7b93
5671    
5672    Sorry, my patch to fix the spelling mistakes in the ks_IN and sd_IN
5673    locales fixed it only partly, I introduced a new spelling mistake
5674    in the sd_IN locales. This patch fixes this.
5675    
5676    Signed-off-by: James Cloos <cloos@jhcloos.com>
5677
5678commit 748d47e69f5c12d8557d56a8a8ec166588da7b93
5679Author: Mike FABIAN <mfabian@redhat.com>
5680Date:   Wed Feb 19 11:46:45 2014 +0100
5681
5682    fix spelling mistakes in ks_IN and sd_IN devanagari locales
5683    
5684    The codeset must be *before* the modifier.
5685    
5686    See also: http://pubs.opengroup.org/onlinepubs/7908799/xbd/envvar.html
5687    
5688    opengroup> The syntax for these environment variables is thus defined as:
5689    opengroup>
5690    opengroup> [language[_territory][.codeset][@modifier]]
5691    
5692    Signed-off-by: James Cloos <cloos@jhcloos.com>
5693
5694commit c64fe5553aa4738f9d1d74a795f5651fbb7b1b09
5695Author: Mike FABIAN <mfabian@redhat.com>
5696Date:   Wed Feb 19 11:50:55 2014 +0100
5697
5698    add be_BY.UTF-8@latin and sr_RS.UTF-8@latin to locale.dir
5699    
5700    See also: https://bugzilla.redhat.com/show_bug.cgi?id=1066910
5701    
5702    If these are not in locale.dir,
5703    
5704        $ LANG=sr_RS.UTF-8@latin xterm
5705    
5706    and
5707    
5708        $ LANG=sr_RS@latin xterm
5709    
5710    give the warning:
5711    
5712        Warning: locale not supported by Xlib, locale set to C
5713    
5714    and some programs (like xmms) fail to find translations for Serbian
5715    in Latin because of this.
5716    
5717    Signed-off-by: James Cloos <cloos@jhcloos.com>
5718
5719commit c85be01b006126c4407eebd1eb6e01a17312b7b4
5720Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5721Date:   Sun Mar 22 16:46:45 2015 -0700
5722
5723    Move Compose \ o / to be with other emoji compose sequences
5724    
5725    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5726
5727commit 5a499ca7b064bf7e6a4fcc169f22862dce0c60c5
5728Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5729Date:   Mon Mar 9 15:28:29 2015 -0700
5730
5731    libX11 1.6.3
5732    
5733    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5734
5735commit c8e19b393defd53f046ddc2da3a16881221b3c34
5736Author: Joonas Javanainen <joonas.javanainen@gmail.com>
5737Date:   Thu Feb 5 17:31:04 2015 +0200
5738
5739    Fix XErrorEvent struct field order in man page
5740    
5741    In the man page the field "resourceid" was in a different place than
5742    in the actual struct layout in Xlib.h
5743    
5744    Signed-off-by: Joonas Javanainen <joonas.javanainen@gmail.com>
5745    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5746
5747commit ddf3b09bb262d01b56fbaade421ac85b0e60a69f
5748Author: Ran Benita <ran234@gmail.com>
5749Date:   Tue Feb 3 13:23:50 2015 -0500
5750
5751    compose: fix the description of modifiers in compose sequences
5752    
5753    The Compose format has a feature which allows specifying certain
5754    modifiers must (or must not) be present with a given keysym in the
5755    sequence.
5756    
5757    The grammar in imLcPrs.c and the Compose man page both do not match what
5758    the code actually does (see the handling of the variables
5759    `modifier_mask` and `modifier` in parseline() in imLcPrs.c, which are
5760    eventually matched as `ev->state & modifier_mask == modifier`).
5761    
5762    Also explicitly list the accepted modifier names, since they are
5763    not standard (e.g. "Ctrl" instead of "Control").
5764    
5765    Signed-off-by: Ran Benita <ran234@gmail.com>
5766    Signed-off-by: James Cloos <cloos@jhcloos.com>
5767
5768commit 129f13f385c50e3d8b53ea7441b17386b0f36aeb
5769Author: Ran Benita <ran234@gmail.com>
5770Date:   Tue Feb 3 13:23:49 2015 -0500
5771
5772    xkb: fix misleading comment about consumed modifiers
5773    
5774    In the spec and the man page the `mods_rtrn` argument is described as
5775    "backfilled with unconsumed modifiers" but actually it is backfilled
5776    with the *consumed* modifiers. This is also mentioned a few lines below
5777    in each case.
5778    
5779    Signed-off-by: Ran Benita <ran234@gmail.com>
5780    Signed-off-by: James Cloos <cloos@jhcloos.com>
5781
5782commit 446f5f7f41317a85a0cd0efa5e6a1b37bc99fba2
5783Author: Ingo Schwarze <schwarze@usta.de>
5784Date:   Tue Dec 9 10:44:13 2014 +0100
5785
5786    Fix pasto in XkbGetKeyBehaviors(3) manual
5787    
5788    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
5789    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5790
5791commit f3831dde6972e4da9e018c6a5f4013d8756a5e78
5792Author: Benno Schulenberg <bensberg@justemail.net>
5793Date:   Sun Nov 23 21:35:36 2014 +0100
5794
5795    nls: Sorting compose sequences rigorously in mirroring pairs, as is custom.
5796    
5797    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5798
5799commit a51681b60c84109fe19f5d449e13080522499324
5800Author: Benno Schulenberg <bensberg@justemail.net>
5801Date:   Thu Sep 19 16:42:01 2013 +0200
5802
5803    nls: Remove a duplicate locale name, and sort some others more strictly.
5804    
5805    Also improve the grammar of the initial comment.
5806    
5807    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5808    Signed-off-by: James Cloos <cloos@jhcloos.com>
5809
5810commit 426b7f850f5376db96a4b12420ee141603fcc3cd
5811Author: Benno Schulenberg <bensberg@justemail.net>
5812Date:   Fri Sep 20 12:37:29 2013 +0200
5813
5814    nls: Add a comment to the block of accented Hebrew letters.
5815    
5816    And align them in a nicer manner.
5817    
5818    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5819    Signed-off-by: James Cloos <cloos@jhcloos.com>
5820
5821commit 7474c6f1ee78dd097b1d0b4c7e3e4ea41317e335
5822Author: Benno Schulenberg <bensberg@justemail.net>
5823Date:   Fri Sep 20 12:05:09 2013 +0200
5824
5825    nls: Add one lowercase compose variant for ®, to mirror those for ©.
5826    
5827    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5828    Signed-off-by: James Cloos <cloos@jhcloos.com>
5829
5830commit 18dcd13514fa538afefa78c93523d9dbd4688e74
5831Author: Benno Schulenberg <bensberg@justemail.net>
5832Date:   Sat Sep 7 20:10:43 2013 +0200
5833
5834    nls: Adding the visual composing characters to two comment lines.
5835    
5836    The lines around them also contain the characters in the comments.
5837    
5838    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5839    Signed-off-by: James Cloos <cloos@jhcloos.com>
5840
5841commit 18d8307575af748913d5da17e3de45da2a22ede4
5842Author: Benno Schulenberg <bensberg@justemail.net>
5843Date:   Sat Sep 7 20:09:32 2013 +0200
5844
5845    nls: Grouping a lone superscript minus together with its mates.
5846    
5847    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5848    Signed-off-by: James Cloos <cloos@jhcloos.com>
5849
5850commit 33301cc45e6a2b8aa841ed6325547af970f8c4db
5851Author: Benno Schulenberg <bensberg@justemail.net>
5852Date:   Sat Sep 7 19:53:38 2013 +0200
5853
5854    nls: Grouping the compose sequences for Dstroke/dstroke together.
5855    
5856    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5857    Signed-off-by: James Cloos <cloos@jhcloos.com>
5858
5859commit 78fe1becb6c18fa33c0d5f04005b45d856f8952a
5860Author: Gioele Barabucci <gioele@svario.it>
5861Date:   Sun Sep 21 23:05:55 2014 +0200
5862
5863    Add compose sequence for U+20B9 INDIAN RUPEE SIGN
5864    
5865    The compose sequence for the new Indian Rupee sign is modelled after
5866    the sequence for the Euro sign.
5867    
5868    Signed-off-by: Gioele Barabucci <gioele@svario.it>
5869    Signed-off-by: James Cloos <cloos@jhcloos.com>
5870
5871commit 6101b967b641355dd863fd1ce52c6a7d58bcbe68
5872Author: Gabriel Souza Franco <gabrielfrancosouza@gmail.com>
5873Date:   Thu Jul 31 22:23:28 2014 -0300
5874
5875    Add double-arrow compose sequence
5876    
5877    Signed-off-by: Gabriel Souza Franco <gabrielfrancosouza@gmail.com>
5878    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5879    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5880
5881commit 368a6401c6a3275d3497fec38a3dcbc38cd9df60
5882Author: James Cloos <cloos@jhcloos.com>
5883Date:   Fri Aug 1 18:30:42 2014 -0400
5884
5885    Add cs_CZ.UTF-8 locale to configure.ac
5886    
5887    Commit cf4d5989383a should have included this.
5888    
5889    Reported-by: Colin Harrison <colin.harrison@virgin.net>
5890    Signed-off-by: James Cloos <cloos@jhcloos.com>
5891
5892commit cf4d5989383acc4ed1b7eebadde9f380f2129766
5893Author: James Cloos <cloos@jhcloos.com>
5894Date:   Tue Jul 29 15:02:56 2014 -0400
5895
5896    Add nls for cs_CZ.UTF-8
5897    
5898    Based on the iso8859-2 compose, and a bug report by Vladimír Marek,
5899    override the en_US.UTF-8 use of <dead_caron> <u> to enter »ǔ« instead
5900    to enter »ů«, and likewise for the majuscule, for the Czech locale.
5901    
5902    This evidently is the norm for Czech keyboards.
5903    
5904    Fixes bz#81875.
5905    
5906    The XI18N_OBJS and XLC_LOCALE.pre files are empty, as they are for
5907    several other locales.  That may require an update.
5908    
5909    Reported-by:  Vladimír Marek <vlmarek@volny.cz>
5910    Signed-off-by: James Cloos <cloos@jhcloos.com>
5911
5912commit d9f569572bd14db31921471e7b877523b5cf1e4c
5913Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5914Date:   Sat Jul 26 12:17:47 2014 -0700
5915
5916    Fix source paths for out-of-tree lintlib builds
5917    
5918    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5919
5920commit 0885cad1e4a9ed57266582be320be55259c881bf
5921Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5922Date:   Sun Jul 20 09:21:20 2014 -0700
5923
5924    specs/XKB: Markup fractions as <{super,sub}script> instead of <emphasis>
5925    
5926    Matches the way they were styled in original doc, before conversion
5927    to DocBook.
5928    
5929    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5930
5931commit d0a9e9d56bb003315787201ee525b4d00fd54e06
5932Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5933Date:   Sat Jul 19 23:55:47 2014 -0700
5934
5935    specs/XKB: acknowledge my contributions
5936    
5937    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5938
5939commit 9c0be82017f513e2eb63d59b095f1cf1955f2e2b
5940Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5941Date:   Sat Jul 19 23:53:48 2014 -0700
5942
5943    specs/XKB: Trim leading spaces off text lines
5944    
5945    perl -i -p -e 's{/\*(\S)}{/* \1}g;' *.xml
5946    
5947    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5948
5949commit 75b0b10990f38d966c6fcc821bf15e58c5a90c91
5950Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5951Date:   Sat Jul 19 20:30:55 2014 -0700
5952
5953    specs/XKB: Add olinks to libX11 for "X Library Functions Affected by Xkb"
5954    
5955    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5956
5957commit 5009621799444e9d1d284719f871d00be13e7330
5958Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5959Date:   Sat Jul 19 12:34:28 2014 -0700
5960
5961    specs/XKB: Fix miscelleanous typos & spelling errors
5962    
5963    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5964
5965commit ac219bf1522a592bd3e3283b1a6ea3dfd2c3c48c
5966Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5967Date:   Sat Jul 19 00:46:41 2014 -0700
5968
5969    specs/XKB: add some more indexterms by hand
5970    
5971    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5972
5973commit 5c3aa4c69e65ecf2e56d5e26f3833fb5d31973c2
5974Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5975Date:   Sat Jul 19 00:48:02 2014 -0700
5976
5977    specs/XKB: fixup various formatting issues in <programlisting>s
5978    
5979    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5980
5981commit 70c648ff852fd9bc784967cfc77ea70bd7f14c8d
5982Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5983Date:   Fri Jul 18 23:56:29 2014 -0700
5984
5985    specs/XKB: fixup various formatting issues in <literallayout>s
5986    
5987    Including translating some that are really just lists into
5988    <simplelist> markup.
5989    
5990    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5991
5992commit 94b56774784ac00b9db02403aecea10bb0814c10
5993Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5994Date:   Fri Jul 18 23:18:52 2014 -0700
5995
5996    specs/XKB: add some more links by hand
5997    
5998    random bits where a link looked handy
5999    
6000    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6001
6002commit 7a15a934cdb07ed1b991bd0ef633f32ee00b1833
6003Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6004Date:   Fri Jul 18 23:16:31 2014 -0700
6005
6006    specs/XKB: add links for terms in definition list under figure 1.1
6007    
6008    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6009
6010commit 92b86fc3c50fbb7ab2e36af10a2fb3fe6284f58c
6011Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6012Date:   Fri Jul 18 22:52:16 2014 -0700
6013
6014    specs/XKB: add links to more tables listing section references
6015    
6016    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6017
6018commit 06a4483a05053b4f8d8c0d4cc0513c68ea912676
6019Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6020Date:   Fri Jul 18 22:49:34 2014 -0700
6021
6022    specs/XKB: Table 4.1: remove page numbers & unnecessary para tags
6023    
6024    Page numbers refer to old doc format, didn't translate to new one
6025    
6026    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6027
6028commit 61bd55c6d1ab211b89d604bd05555dc417f6e53d
6029Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6030Date:   Sat Jul 19 11:55:51 2014 -0700
6031
6032    specs/XKB: remove unwanted white space around C -&gt; struct references
6033    
6034    perl -i -0 -p -e 's{\s*-&gt;\s*}{-&gt;}g' *xml
6035    
6036    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6037
6038commit 2be0cc0b2abbcc98cfd150210dea415a04787251
6039Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6040Date:   Fri Jul 18 22:26:20 2014 -0700
6041
6042    specs/XKB: replace -&gt; with &rarr; when used as arrow, not in C structs
6043    
6044    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6045
6046commit bf1f3d6f6f995303624679ae546f507c70967dc0
6047Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6048Date:   Fri Jul 18 22:21:10 2014 -0700
6049
6050    specs/XKB: replace =&gt; with &rArr; for double arrows
6051    
6052    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6053
6054commit c7ee427fc0a72abd4a4f147ab16a5d1128a6a2ba
6055Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6056Date:   Sat Jul 19 11:33:35 2014 -0700
6057
6058    specs/XKB: Markup some ranges with &ndash; instead of -
6059
6060commit ec4075303c6c0d1d64bfe378e585968f9a137da7
6061Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6062Date:   Fri Jul 18 22:10:45 2014 -0700
6063
6064    specs/XKB: Markup some subtractions with &minus; instead of -
6065    
6066    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6067
6068commit 441a267e461132a38abed205245f028686526f1d
6069Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6070Date:   Fri Jul 18 22:04:40 2014 -0700
6071
6072    specs/XKB: make sure all files have DOCTYPEs so standard entities work
6073    
6074    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6075
6076commit 88fd70bee410f290b4f540405fdc7ecd85c26f25
6077Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6078Date:   Fri Jul 18 21:55:41 2014 -0700
6079
6080    specs/XKB: Markup quoted terms as <quote> instead of with ""
6081    
6082    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6083
6084commit 3576587ff10334a8f48c34b4fe5b7e829dec9a1b
6085Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6086Date:   Sat Jul 19 09:01:26 2014 -0700
6087
6088    specs/XKB: fixup newlines between tags and punctuation
6089    
6090    Get rid of unwanted whitespace before punctuation by moving them to the
6091    lines with the tags, instead of before & after.
6092    
6093    perl -i -0 -p -e 's{\>\s*\n([\.,;:])}{>\1\n}g' *xml
6094    
6095    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6096
6097commit 6d5ec492cd28c206423337f926503349702af5a6
6098Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6099Date:   Fri Jul 18 21:29:33 2014 -0700
6100
6101    specs/XKB: fixup newlines between tags and parens
6102    
6103    Get rid of unwanted whitespace inside parens by moving them to the
6104    lines with the tags, instead of before & after.
6105    
6106    perl -i -0 -p \
6107         -e 's{(?<!--) \(\s*\n\<}{\n(<}g;' \
6108         -e 's{\>\s*\n\)([\.,;]?)(?! [^\n]*--)}{>)\1\n}g' *xml
6109    
6110    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6111
6112commit 59d688f4c787250e0b401a92b1db0437d8c60f2d
6113Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6114Date:   Fri Jul 18 21:09:24 2014 -0700
6115
6116    specs/XKB: Markup key terms as <firstterm> instead of <emphasis>
6117    
6118    Also add <indexterm> entries for most of them, to make their definitions
6119    or introductions easy to find from the index.
6120    
6121    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6122
6123commit 861f3087ee0f501362a67501f384c2ca4c7bfe73
6124Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6125Date:   Thu Jul 10 20:00:53 2014 -0700
6126
6127    specs/XKB: Manual fixup of type markup
6128    
6129    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6130
6131commit 6b96259dabe52701fd1bcaa0625b574180c4e769
6132Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6133Date:   Thu Jul 10 15:00:30 2014 -0700
6134
6135    specs/XKB: Manual fixup of parameter markup
6136    
6137    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6138
6139commit 5526dce6812a84102f556fdde8f2b52b21c8bcdc
6140Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6141Date:   Thu Jul 10 13:40:40 2014 -0700
6142
6143    specs/XKB: Manual fixup of struct name/field markup
6144    
6145    Handles typos that caused the scripts to miss matches, misnamed structs, etc.
6146    
6147    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6148
6149commit c36ee1a4db4e7876526190b8ab6b0da5867f76f7
6150Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6151Date:   Wed Jul 9 23:26:37 2014 -0700
6152
6153    specs/XKB: Manual fixup of symbol name markup
6154    
6155    Handles typos that caused the scripts to miss matches, misnamed masks, etc.
6156    
6157    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6158
6159commit eb1453a0c69606b8af96b90ddccf1b93a069fb35
6160Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6161Date:   Wed Jul 9 00:03:23 2014 -0700
6162
6163    specs/XKB: Manual fixup of function name markup
6164    
6165    Handles typos that caused the scripts to miss matches, functions like
6166    malloc & free from other libraries, function name patterns, etc.
6167    
6168    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6169
6170commit 252d99c87b60ac6f3f2b36f292f9b3880daabe26
6171Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6172Date:   Thu Jul 10 19:42:31 2014 -0700
6173
6174    specs/XKB: Markup protocol requests as <systemitem> instead of <emphasis>
6175    
6176    No great fit in DocBook, so follow what we used in Xlib spec.
6177    
6178    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6179
6180commit ed60df10aad15057577d5714c955d22d2a446e51
6181Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6182Date:   Thu Jul 10 15:45:57 2014 -0700
6183
6184    specs/XKB: Markup keyboard keys as <keycap> instead of <emphasis>
6185    
6186    Also uses <guilabel> for LED names/labels, for lack of a better fit
6187    in DocBook.
6188    
6189    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6190
6191commit f57b91ee497414083cc1bf481d28eb9ad9f965fb
6192Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6193Date:   Thu Jul 10 15:41:19 2014 -0700
6194
6195    specs/XKB: Markup characters & strings as <literal> instead of <emphasis>
6196    
6197    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6198
6199commit cfd4279c9b59d7e87c9f7c67692c87973adb7667
6200Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6201Date:   Thu Jul 10 11:08:14 2014 -0700
6202
6203    specs/XKB: Markup structs as <struct{name,field}> instead of <emphasis>
6204    
6205    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6206
6207commit a014bb4cc013b0d1b76524b7868e860c7e7ebc79
6208Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6209Date:   Wed Jul 9 23:30:54 2014 -0700
6210
6211    specs/XKB: Markup symbol names in table entries too
6212    
6213    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6214
6215commit 907f7ad7a7a977bf4f19daa3143d47c4d07ca33d
6216Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6217Date:   Tue Jul 8 23:35:48 2014 -0700
6218
6219    specs/XKB: Markup *Ptr as <type> instead of <emphasis>
6220    
6221    perl -i -p -e 's{<emphasis>(\w*Ptr)</emphasis>}{<type>\1</type>}g' *xml
6222    
6223    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6224
6225commit 9f6c00629fd4b713082cc11f9150f7aafd272c89
6226Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6227Date:   Tue Jul 8 23:34:29 2014 -0700
6228
6229    specs/XKB: Markup *Rec as <structname> instead of <emphasis>
6230    
6231    perl -i -p -e \
6232     's{<emphasis>(\w*Rec)</emphasis>}{<structname>\1</structname>}g' *xml
6233    
6234    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6235
6236commit 83839e37802b8b752c77859a95de60ad757feb79
6237Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6238Date:   Tue Jul 8 23:29:49 2014 -0700
6239
6240    specs/XKB: Markup XKB macros as <symbol> instead of <emphasis>
6241    
6242    Performed via:
6243     perl -n -e 'printf "s{<emphasis>\\s*%s\\s*</emphasis>}{<symbol>%s</symbol>};\n", $1, $1 if m{^#define\s+([^\s\(]*)}' \
6244      /usr/include/X11/extensions/XK*h /usr/include/X11/XKBlib.h \
6245      | sort -u > xkb-defines.pl
6246     perl -i -p -f xkb-defines.pl *.xml
6247    
6248    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6249
6250commit 9e397ed37ce4cc70621de347de3a795df88d4506
6251Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6252Date:   Tue Jul 8 22:53:12 2014 -0700
6253
6254    specs/XKB: Use ° instead of o for degrees.
6255    
6256    Conversion from framemaker turned superscripted "o" into plain "o".
6257    
6258    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6259
6260commit 3b8364c21f5119105a2c14ae8cc75a11494cb7a7
6261Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6262Date:   Tue Jul 8 22:46:03 2014 -0700
6263
6264    specs/XKB: Markup keysyms as <keysym> instead of <emphasis>
6265    
6266    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6267
6268commit 33bef065683c8f910f3722730503c0c0699ee8be
6269Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6270Date:   Tue Jul 8 22:26:56 2014 -0700
6271
6272    specs/XKB: Markup NULL as <symbol> instead of <emphasis>
6273    
6274    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6275
6276commit f10aa1e09468bd28454d85ac8ab55d9dc7178029
6277Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6278Date:   Tue Jul 8 22:22:29 2014 -0700
6279
6280    specs/XKB: Markup function args as <parameter> instead of <emphasis>
6281    
6282    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6283
6284commit 79ba58cceb69521fcb313c69233cf93a9fcb177f
6285Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6286Date:   Mon Jul 7 23:56:13 2014 -0700
6287
6288    specs/XKB: Convert remaining error names to errorname tags
6289    
6290    Most were caught by applying libX11 lists, but BadKeyboard & XKB*_Bad*
6291    are XKB-specific.   (Plus some were badly split across tag boundaries.)
6292    
6293    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6294
6295commit 42b2f5388c399949ece377f9cc9c479c06964972
6296Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6297Date:   Mon Jul 7 23:16:56 2014 -0700
6298
6299    specs/XKB: manually fixup some more emphasis tagging mismatches
6300    
6301    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6302
6303commit 3b2f47d44a55d93c65455ff183f3b47da04b1de1
6304Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6305Date:   Mon Jul 7 23:02:31 2014 -0700
6306
6307    specs/XKB: re-normalize <emphasis> layout in xml files
6308    
6309    Same script as before, just with <!-- xref --> comments out of the way
6310    now.
6311    
6312    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6313
6314commit ce95f152eda509263874c53fb7c6b4a6bdab2c29
6315Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6316Date:   Mon Jul 7 22:59:25 2014 -0700
6317
6318    specs/XKB: Remove remaining xref comments
6319    
6320    All the places marked by these have been turned into the appropriate
6321    link, xref, or olink tags now.
6322    
6323    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6324
6325commit 6590b66e19af8dff68888ac403ac82f5d585b4e2
6326Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6327Date:   Mon Jul 7 22:56:36 2014 -0700
6328
6329    specs/XKB: make olinks to xkbproto for references in section 10.13
6330    
6331    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6332
6333commit fcda446877a62e7443d7bc704ba3610e90d1e755
6334Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6335Date:   Mon Jul 7 22:22:28 2014 -0700
6336
6337    specs/XKB: Add <figure> tags and make Figure references link to them
6338    
6339    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6340
6341commit 087a2338476719e340dc3d5af0df6fdc4a26ce7a
6342Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6343Date:   Mon Jul 7 21:30:01 2014 -0700
6344
6345    specs/XKB: Turn Table references into links
6346    
6347    Adds id attributes to all table tags so we can link to them
6348    
6349    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6350
6351commit 135fa07b74cb50172c6a75768d499cd87ddb336e
6352Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6353Date:   Sun Jul 6 21:29:59 2014 -0700
6354
6355    specs/XKB: Turn section references into xref links
6356    
6357    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6358
6359commit 53e931d79926af8a3996253efd8b5f6c21d9e5d7
6360Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6361Date:   Sun Jul 6 21:17:18 2014 -0700
6362
6363    specs/XKB: Turn Chapter references into xref links
6364    
6365    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6366
6367commit b7f00ce5bcb0c00696bb82503ab548e14f04d17d
6368Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6369Date:   Sun Jul 6 20:40:18 2014 -0700
6370
6371    specs/XKB: Apply <emphasis> to semantic tag transformations from Xlib spec
6372    
6373    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6374
6375commit b00a7ddff2744238fbfe31c2298b02028a45a6ff
6376Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6377Date:   Sun Jul 6 20:34:51 2014 -0700
6378
6379    specs/XKB: Markup function names as <function> instead of <emphasis>
6380    
6381    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6382
6383commit bfbb58b7679221cb5c9212665209ea9099ad079a
6384Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6385Date:   Sun Jul 6 20:25:46 2014 -0700
6386
6387    specs/XKB: normalize <emphasis> layout in xml files
6388    
6389    - Stop placing <emphasis> on empty space, commas, and periods.
6390    - Move periods & commas after closing </emphasis> tag
6391    - move <emphasis> open & close tags to same line, instead of mirroring
6392      nroff layout.
6393    
6394    Simplifies automating further transformations of these tags.
6395    
6396    Performed via:
6397     perl -i -0 -p \
6398        -e 's{<emphasis>(\s*)</emphasis>}{}msg;' \
6399        -e 's{<emphasis>([\s\.,]*)</emphasis>\s*}{\1}msg;' \
6400        -e 's{\n([\.,])\s*}{\1\n}msg;' \
6401        -e 's{([^\.])([\.,])\s*</emphasis>}{\1</emphasis>\2}msg;' \
6402        -e 's{\s*<emphasis>\n\s*}{\n<emphasis>}msg;' *xml
6403    
6404    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6405
6406commit b16ee69a0103109a661a88140a1765dcd7bda634
6407Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6408Date:   Sun Jul 6 15:57:41 2014 -0700
6409
6410    specs/XKB: Convert to funcsynopsis+variablelist instead of informaltable
6411    
6412    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6413
6414commit b41d43d4cf0c0a1a049a171ee8cf6fd8a3ee4335
6415Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6416Date:   Sun Jul 6 19:29:38 2014 -0700
6417
6418    specs/XKB: Add index
6419    
6420    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6421
6422commit 72ae1d793be078db521dda60af578ece71f364de
6423Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6424Date:   Sun Jul 6 13:21:40 2014 -0700
6425
6426    specs/XKB: Fix various markup issues in functiondecl tables
6427    
6428    - Merge some functionargdecl entries incorrectly split across rows
6429    - Add missing parameter name markup to some functionargdecls
6430    - Add missing function prototype markup to a functiondecl
6431    - Remove stray emphasis tags in a functiondecl
6432    
6433    Allows them to correctly convert to funcsynopsis markup in next step.
6434    
6435    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6436
6437commit 9fdb973012de80ac60dbc59c39162f4e839fc5a4
6438Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6439Date:   Sun Jul 6 12:15:23 2014 -0700
6440
6441    specs/XKB: Convert header filenames to filename tags
6442    
6443    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6444
6445commit 5525e8433f93bce464412f27cffa203ea628f368
6446Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6447Date:   Tue Jul 8 00:05:32 2014 -0700
6448
6449    specs/libX11: disengender a user reference
6450    
6451    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6452
6453commit d8679eae9317b389ad4acb0430360ee0663e2af8
6454Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6455Date:   Fri Jul 11 18:41:42 2014 -0700
6456
6457    specs/libX11: Correct value of IconicState to match Xutil.h
6458    
6459    Xutil.h has always had a value of 3 for IconicState, since 2 was
6460    previously used for the long-obsolete ZoomState, so make the spec
6461    match what programs have used for decades.
6462    
6463    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6464    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
6465
6466commit 7ce2b0f12a48fb832f457cbafb0e1144ef557f9a
6467Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6468Date:   Fri Jul 11 10:34:08 2014 -0700
6469
6470    Use C99 named initializers to fill in events passed to XSendEvent
6471    
6472    Forces compiler to zero-fill unset fields in the struct (fixing bug 81236)
6473    and allows optimizer to order field initialization to best fit cache layout
6474    or other considerations.
6475    
6476    Before & after output of gcc -S on AMD64 shows insertion of "rep stosq"
6477    instructions to rapidly zero-fill structs.
6478    
6479    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6480    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
6481
6482commit 169805e1dc8743b37b00e24cf3a5eb8748f733ad
6483Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6484Date:   Sun Jul 6 11:13:49 2014 -0700
6485
6486    Fix validation of ctrls parameter to XkbGetPerClientControls()
6487    
6488    Nothing in the XKB spec states that the memory pointed to by ctrls has to
6489    be initialized to any given value when passed to the function, only that
6490    it is set by the function to the values returned by the X server:
6491    http://www.x.org/releases/X11R7.7/doc/libX11/XKB/xkblib.html#The_Miscellaneous_Per_client_Controls
6492    
6493    The check for the incoming value seems to be copied from
6494    XkbSetPerClientControls without explanation.
6495    
6496    Instead change it to checking if ctrls is non-NULL, since there's no
6497    point asking the X server to return a value the caller won't even see.
6498    
6499    Found while investigating report from cppcheck-1.65:
6500    [src/xkb/XKB.c:699] -> [src/xkb/XKB.c:719]: (warning) Possible null pointer
6501     dereference: ctrls - otherwise it is redundant to check it against null.
6502    
6503    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6504
6505commit 1e362fac92c6688fb42b195ccad16d7a337a34c1
6506Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6507Date:   Sun Jul 6 10:54:57 2014 -0700
6508
6509    Fix map->num_types check in XkbAddKeyType()
6510    
6511    Check is intended to ensure we allocate at least XkbNumRequiredTypes
6512    in map, but was accidentally marked with a ! causing the wrong check.
6513    
6514    Reported-by: Harms <wharms@bfs,de>
6515    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
6516    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6517
6518commit ff9a5c199251a84fa59d14fd48dadb3f8920b54b
6519Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6520Date:   Sun Jul 6 15:08:21 2014 -0700
6521
6522    specs/libX11: Add missing spaces to 'unsignedint' & 'unsignedlong' types
6523    
6524    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6525
6526commit a06ea86773568926c36ae650b188fc818d540db7
6527Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6528Date:   Sun Jul 6 15:04:27 2014 -0700
6529
6530    specs/libX11: Fix height & width in parameter lists to be two separate entries
6531    
6532    "unsigned int width, unsigned int height", not a single parameter "height"
6533    of type "unsignedintwidth,".
6534    
6535    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6536
6537commit e4db5e503682b3304fe82e4b17b419a8e0f0a9f2
6538Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6539Date:   Sun Jul 6 14:38:10 2014 -0700
6540
6541    specs/libX11: Fix x & y in parameter lists to be two separate parameters
6542    
6543    "int x, int y" not a single parameter y of type "intx"
6544    
6545    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6546
6547commit aa8bda0db2c6d82515b90ceb4a7d6403e38895e9
6548Author: walter harms <wharms@bfs.de>
6549Date:   Sat Jun 7 12:03:17 2014 +0200
6550
6551    lcDefConv.c: fix use before check
6552    
6553    * Do not use variables before checked for NULL.
6554    * remove some superfluid spaces (Mark Kettenis)
6555    
6556    Signed-off-by: Harms <wharms@bfs,de>
6557    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6558    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6559
6560commit c0670e5d3ae330e611ecb05303d579a4f8a3d114
6561Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6562Date:   Mon Jun 16 15:16:10 2014 -0700
6563
6564    Start adding Unicode 7.0 support to compose table
6565    
6566    New characters defined in http://www.unicode.org/charts/PDF/U1F300.pdf
6567    
6568    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6569
6570commit a4679baaa18142576d42d423afe816447f08336c
6571Author: walter harms <wharms@bfs.de>
6572Date:   Sat Jun 7 11:54:34 2014 +0200
6573
6574    rm redundant null checks
6575    
6576    remove more redundant NULL checks
6577    note that _XkbFree() is really Xfree()
6578    
6579    Signed-off-by: Harms <wharms@bfs,de>
6580    Reviewed-by: Rémi Cardona <remi@gentoo.org>
6581    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6582
6583commit 602d7f5030fe93b2fe7f29fb7310deb6f50cb6df
6584Author: walter harms <wharms@bfs.de>
6585Date:   Sat Jun 7 15:17:27 2014 +0200
6586
6587    libX11: rm redundante NULL checks
6588    
6589    This patch removes the last remaining  NULL checks for Xfree()
6590    
6591    Signed-off-by: Harms <wharms@bfs,de>
6592    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
6593    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6594
6595commit d81fed46144d089bdfa1d916a28dffc9ebffe1e4
6596Author: walter harms <wharms@bfs.de>
6597Date:   Fri Jun 6 22:53:05 2014 +0200
6598
6599    Remove more redundant null checks before Xfree()
6600    
6601    Signed-off-by: Harms <wharms@bfs,de>
6602    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6603    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6604
6605commit 0b7fd7dbec136bae317bd9a329309eaa089beee3
6606Author: walter harms <wharms@bfs.de>
6607Date:   Thu Jun 5 18:37:40 2014 +0200
6608
6609    Remove redundant null checks before free
6610    
6611    This patch removes some redundant null checks before free.
6612    It should not change the code otherwise. Be aware that this
6613    is only the first series.
6614    
6615    Signed-off-by: Harms <wharms@bfs,de>
6616    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6617    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6618
6619commit 7d452fad5068ba92b964e92bc46708046f4044aa
6620Author: walter harms <wharms@bfs.de>
6621Date:   Wed Jun 4 17:12:31 2014 +0200
6622
6623    libX11/lcUTF8.c fix: dereferenced before check
6624    
6625    * Do not use variables before checked for NULL.
6626    
6627    Signed-off-by: Harms <wharms@bfs,de>
6628    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6629    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6630
6631commit a3808f51517a720e7ff738208af60865779dd6ef
6632Author: walter harms <wharms@bfs.de>
6633Date:   Wed Jun 4 17:10:20 2014 +0200
6634
6635    libX11/XKBNames.c fix: dereferenced before check
6636    
6637    * Do not use variables before checked for NULL.
6638    
6639    Signed-off-by: Harms <wharms@bfs,de>
6640    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6641    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6642
6643commit b3c9f6a17e430aabe16aecbe097f7312a0f6ff78
6644Author: walter harms <wharms@bfs.de>
6645Date:   Wed Jun 4 17:08:12 2014 +0200
6646
6647    libX11/lcGenConv.c fix: dereferenced before check
6648    
6649    * Do not use variables before checked for NULL.
6650    
6651    Signed-off-by: Harms <wharms@bfs,de>
6652    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6653    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6654
6655commit c6bc28d433243c32b3f74955f1478451b4fd27b5
6656Author: Andreas Schwab <schwab@linux-m68k.org>
6657Date:   Sun Jan 19 16:59:13 2014 +0100
6658
6659    Restore lost tabs in sed commands
6660    
6661    Signed-off-by: Andreas Schwab <schwab@linux-m68k.org>
6662    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6663
6664commit e3dc0d17339e61eaf0b51b8907510984e3bf23cb
6665Author: Benno Schulenberg <bensberg@justemail.net>
6666Date:   Thu Sep 19 11:01:11 2013 +0200
6667
6668    nls: Transform Brazilian compose file to an include plus three overrides.
6669    
6670    Signed-off-by: James Cloos <cloos@jhcloos.com>
6671
6672commit 16c87dda4da2271aaecc5d8b6fe6ecd072cc584c
6673Author: Benno Schulenberg <bensberg@justemail.net>
6674Date:   Fri Sep 6 12:10:01 2013 +0200
6675
6676    nls: Adding the missing compose sequences with <comma> for O with ogonek.
6677    
6678    All other vowels with ogoneks can already be composed using <comma>.
6679    
6680    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6681    Signed-off-by: James Cloos <cloos@jhcloos.com>
6682
6683commit 655b60f48376069750b151c46da836fdd411c83b
6684Author: Benno Schulenberg <bensberg@justemail.net>
6685Date:   Fri Sep 6 12:04:24 2013 +0200
6686
6687    nls: Adding the missing postfix sequences for composing vowels with ogoneks.
6688    
6689    It existed for lowercase i, but not for uppercase I nor the other vowels.
6690    
6691    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6692    Signed-off-by: James Cloos <cloos@jhcloos.com>
6693
6694commit 7f8f9a36ef901f31279c385caf960a22daeb33fe
6695Author: Owen W. Taylor <otaylor@fishsoup.net>
6696Date:   Fri May 9 18:21:05 2014 -0400
6697
6698    Fix XNextRequest() after direct usage of XCB
6699    
6700    When XCB owns the X socket, dpy->request is not updated, so
6701    NextRequest() and XNextRequest() return the wrong value. There's
6702    nothing we can do to fix NextRequest() while retaining ABI compat,
6703    but change XNextRequest() to grab the socket back from XCB,
6704    updating dpy->request.
6705    
6706    Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
6707    Reviewed-by: Uli Schlachter <psychon@znc.in>
6708    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6709
6710commit 0f9e734ea96556fe750a4baf354d42d5a87bcd14
6711Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6712Date:   Sun May 4 11:54:59 2014 -0700
6713
6714    Add missing .TE tags to end tables in Xkb man pages
6715    
6716    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6717
6718commit 280274e5292e013b43e552274111fab434f5ed4e
6719Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6720Date:   Sun May 4 11:23:17 2014 -0700
6721
6722    XCreateGC.man: simplify table to work with Solaris tbl
6723    
6724    Having every table cell be a text diversion (T{...T}) was too much for
6725    Solaris tbl to handle, and thus "man XCreateGC" would print the error
6726    /usr/man/man3x11/XCreateGC.3x11: line 402: Too many text block diversions
6727    tbl quits
6728    and not display the table of mask bits or any text in the man page after
6729    that table.   Since the #define column doesn't need special handling,
6730    making it not use text diversions brings the table under the tbl limit.
6731    
6732    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6733
6734commit 93bb325a54025dd15f4744abce54b358960420f9
6735Author: James Cloos <cloos@jhcloos.com>
6736Date:   Tue May 20 17:30:10 2014 -0400
6737
6738    Revert "nls: Adding compose sequences (with <parenleft> first) that GTK also has."
6739    
6740    Parenleft is already in use for sequences of the form <(> <letter> <)>
6741    to generate circled letters.
6742    
6743    Eg, <Multikey> <parenleft> <a> <parenright> generates ⓐ.
6744    
6745    This reverts commit f020235f4bd91fb6eade82f8c9f7b85a57981768.
6746    
6747    Signed-off-by: James Cloos <cloos@jhcloos.com>
6748
6749commit 060707851be918f2f507a26d17b016f764ddf2b4
6750Author: Benno Schulenberg <bensberg@justemail.net>
6751Date:   Fri Sep 6 11:42:59 2013 +0200
6752
6753    nls: Adding accessible compose sequences for Ș and Ț (with comma below).
6754    
6755    Compose sequences with <dead_belowcomma> exist, but very few keyboard
6756    layouts contain that symbol.  So a more usual character is needed to be
6757    able to easily compose Ș, ș, Ț and ț.  The semicolon is normally only
6758    used for composing letters with ogoneks -- but only vowels take ogoneks,
6759    so the character is free for consonants, and thus <semicolon> is used
6760    here to compose commas below.  It is somewhat fitting, because on most
6761    Romanian keyboards the Ș is placed on that key, and the Ț next to it.
6762    
6763    (Oh -- the more obvious sequences with <comma> were already taken for
6764    composing S and T with cedillas.)
6765    
6766    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6767    Signed-off-by: James Cloos <cloos@jhcloos.com>
6768
6769commit ca435c2f753aa2961fb35ac448cdb2cc77112755
6770Author: Benno Schulenberg <bensberg@justemail.net>
6771Date:   Tue Sep 3 21:10:33 2013 +0200
6772
6773    nls: Ordering some compose sequences in a more customary way.
6774    
6775    The custom seems to be: pairing the ones that have only the sequence of
6776    two keys reversed, and putting the one with the diacritic first first.
6777    
6778    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6779    Signed-off-by: James Cloos <cloos@jhcloos.com>
6780
6781commit f020235f4bd91fb6eade82f8c9f7b85a57981768
6782Author: Benno Schulenberg <bensberg@justemail.net>
6783Date:   Tue Sep 3 20:44:42 2013 +0200
6784
6785    nls: Adding compose sequences (with <parenleft> first) that GTK also has.
6786    
6787    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6788    Signed-off-by: James Cloos <cloos@jhcloos.com>
6789
6790commit bda0b3b5bd19154206dc40166364e73d4b6b1374
6791Author: Benno Schulenberg <bensberg@justemail.net>
6792Date:   Tue Sep 3 11:14:16 2013 +0200
6793
6794    nls: Allowing Romanian Ă and ă to be composed also with lowercase <u>.
6795    
6796    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6797    Signed-off-by: James Cloos <cloos@jhcloos.com>
6798
6799commit 8be4610939b833587954957f5963eb4191b43d19
6800Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6801Date:   Thu Mar 13 23:22:48 2014 -0700
6802
6803    Fix "follwing" typo in en_US.UTF-8/Compose comment
6804    
6805    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6806
6807commit 6f30e9034f29c3ae6ad7e617b3d5e903aa107b6a
6808Author: Reuben Thomas <rrt@sc3d.org>
6809Date:   Mon Jan 27 14:18:24 2014 +0000
6810
6811    en_US.UTF-8/Compose: Fix apparent copy-paste-o, changing capital to small A.
6812    
6813    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6814
6815commit d6bd988bc00494914b38b95ee5df77ac4f32f19f
6816Author: Peter Hutterer <peter.hutterer@who-t.net>
6817Date:   Mon Mar 3 12:38:48 2014 +1000
6818
6819    man: fix man page for XkbGetMap
6820    
6821    Returned structure must be freed with XkbFreeKeyboard().
6822    
6823    Reported-by: Morten Bøgeskov <mb@dbc.dk>
6824    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6825
6826commit a6dcf2201a05adbff54122df05a1e6325936abb6
6827Author: Ran Benita <ran234@gmail.com>
6828Date:   Tue Feb 11 13:26:16 2014 +0200
6829
6830    Remove dead USE_OWN_COMPOSE-protected code
6831    
6832    The build doesn't provide any way to define this option. It also refers
6833    to files (imComp.h) and functions (e.g. XimCompInitTables(),
6834    XimCompProcessSym()) which are not found anywhere, and the ordinary
6835    Compose implementation in xim doesn't use any of it.
6836    
6837    Signed-off-by: Ran Benita <ran234@gmail.com>
6838    Reviewed-by: Julien Cristau <jcristau@debian.org>
6839    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6840
6841commit b64bee2ddb7b96f00713a8b8435f11ad9ac1c9e0
6842Author: Ran Benita <ran234@gmail.com>
6843Date:   Sun Feb 16 15:24:58 2014 +0200
6844
6845    nls: always use XCOMM instead of # for comments in Compose.pre files
6846    
6847    Lines starting with # are considered as preprocessor directives in the
6848    .pre files.
6849    
6850    Fixes warnings like:
6851    <stdin>:3:0: error: invalid preprocessing directive #Khmer
6852    
6853    Signed-off-by: Ran Benita <ran234@gmail.com>
6854    Signed-off-by: James Cloos <cloos@jhcloos.com>
6855
6856commit 470e2289a3ebc59c5a35e54e1adeb0f261d5bf88
6857Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6858Date:   Thu Feb 6 13:48:08 2014 -0800
6859
6860    Fix typos in Xrm.c comments
6861    
6862    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6863
6864commit aacf95dacc7c598e7297894580d4d655593813b2
6865Author: Marko Myllynen <myllynen@redhat.com>
6866Date:   Mon Jan 13 16:43:18 2014 +0200
6867
6868    Annotate Finnish Compose map with Unicode code points
6869    
6870    Scripted annotation, no functional changes.
6871    
6872    Signed-off-by: James Cloos <cloos@jhcloos.com>
6873
6874commit 20fdccd81b54678376d49e00edfebbbe94951f07
6875Author: Teemu Likonen <tlikonen@iki.fi>
6876Date:   Fri Jul 20 19:21:04 2012 +0300
6877
6878    Fix "RING ABOVE" key in the Finnish compose file
6879    
6880    The Finnish keyboard standard defines that <dead_abovering> <space> must
6881    insert the character U+02DA RING ABOVE. Currently the Finnish Compose
6882    file inserts U+00B0 DEGREE SIGN even though the line's comment says
6883    "RING ABOVE". This commit changes the character to U+02DA RING ABOVE.
6884    
6885    Signed-off-by: Teemu Likonen <tlikonen@iki.fi>
6886    Signed-off-by: James Cloos <cloos@jhcloos.com>
6887
6888commit 8757e2ac8e04f2932ff437127f3e2ae9ac20c1d7
6889Author: Ran Benita <ran234@gmail.com>
6890Date:   Wed Jan 29 02:11:47 2014 +0200
6891
6892    nls: remove duplicate 'ohorn' and 'uhorn' compose sequences
6893    
6894    Since <Ohorn> == <U01A0> and <ohorn> == <U01A1>, when translated to
6895    keysyms:
6896    
6897     #define XK_Ohorn                      0x10001a0  /* U+01A0 LATIN CAPITAL LETTER O WITH HORN */
6898     #define XK_ohorn                      0x10001a1  /* U+01A1 LATIN SMALL LETTER O WITH HORN */
6899    
6900    (and similarly for uhorn), there is no need to have both names. Remove
6901    the unicode literal ones.
6902    
6903    Signed-off-by: Ran Benita <ran234@gmail.com>
6904    Signed-off-by: James Cloos <cloos@jhcloos.com>
6905
6906commit b98998cb3bea7cb3005f2e9d5bc5332d14b1d5d3
6907Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6908Date:   Sat Jan 18 22:35:18 2014 -0800
6909
6910    Add RANDR 1.4 requests & events to XErrorDB
6911    
6912    Checked against randrproto.txt & randr.h
6913    
6914    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6915
6916commit 321392ded15a7ee9d177d4ebe8846336ba76741c
6917Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6918Date:   Fri Jan 3 20:04:33 2014 -0800
6919
6920    Remove unused ETEST & ESZTEST macros from XlibInt.c
6921    
6922    Left behind when 15e5eaf62897 removed support for building without XCB.
6923    
6924    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6925    Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
6926
6927commit 1ffc0c5503d4f419fdbc765243832a53491bf5bc
6928Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6929Date:   Thu Dec 26 09:26:13 2013 -0800
6930
6931    _XkbReadGeomOverlay: check for NULL first, then use pointer
6932    
6933    Flagged by cppcheck 1.62:
6934    [lib/libX11/src/xkb/XKBGeom.c:479] -> [lib/libX11/src/xkb/XKBGeom.c:480]:
6935     (warning) Possible null pointer dereference: row - otherwise it is
6936     redundant to check it against null.
6937    
6938    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6939
6940commit ddf5f130cc29bb3bf8b9c757dcbac31bc56e9379
6941Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6942Date:   Thu Dec 26 09:22:49 2013 -0800
6943
6944    XkbSelectEventDetails: remove unnecessary assignments
6945    
6946    clear & selectAll are set to 0 already a few lines earlier,
6947    affectWhich is set to XkbMapNotifyMask a few lines later.
6948    None are used between the other assignments and the removed ones.
6949    
6950    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6951
6952commit 3d69b0a83e62f8f6fbdd952fc49cdbdf8825e1e6
6953Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6954Date:   Thu Oct 7 18:46:08 2010 +0100
6955
6956    Don't try so hard to find a matching font with the given encoding
6957    
6958    See http://sourceware.org/bugzilla/show_bug.cgi?id=10948
6959    
6960    Currently, if the locale is UTF-8, no CJK fonts are installed, and someone
6961    does XCreateFontSet() with a font name of "*", we end up asking the server
6962    to list the (non-existent) fonts 11 times for each CJK encoding, which can
6963    take a while.
6964    
6965    A * wildcard can match multiple components in a XLFD name in XListFonts(),
6966    so there's no need to try adding more than one to get a match.
6967    
6968    We do try once with a leading '*-' in case the fontname isn't a full
6969    well-formed XLFD name, maybe even that isn't needed?
6970    
6971    (See also http://invisible-island.net/xterm/xterm.faq.html#slow_menus)
6972    
6973    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6974    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6975
6976commit 7e163300735d4bcd3386b86eec112acdad139c59
6977Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6978Date:   Mon Dec 2 21:51:27 2013 -0800
6979
6980    unifdef -UISC
6981    
6982    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6983
6984commit 1e43c262d13cab2b759665f9f13bdedbc7afbfd4
6985Author: Benno Schulenberg <bensberg@justemail.net>
6986Date:   Thu Sep 19 13:20:05 2013 +0200
6987
6988    nls: Fix transposed locale identifier for Khmer.
6989    
6990    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6991    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6992    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6993
6994commit 0e45f64766c0557c8e99a979c70ca6f55664dae7
6995Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6996Date:   Sat Nov 16 20:21:54 2013 -0800
6997
6998    Drop X_LOCALE fallback for OS'es without setlocale()
6999    
7000    C89 or bust!   This was documented as being needed for "only Lynx,
7001    Linux-libc5, OS/2" and has never been enabled in modular builds,
7002    since none of those platforms have had anyone step up to add support
7003    since the X11R7 conversion to autotools.
7004    
7005    Mostly performed with unifdef -UX_LOCALE, followed by removal of files
7006    left without any purpose, and manual cleanup of remaining references.
7007    
7008    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7009    Reviewed-by: Adam Jackson <ajax@redhat.com>
7010
7011commit 6cb02b166361200da35ba14f52cd9aaa493eb0ea
7012Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7013Date:   Wed Oct 23 10:37:53 2013 -0700
7014
7015    Xcms file parsing should not require the impossible to succeed
7016    
7017    The field2 helper function, to split lines from Xcms.txt files into
7018    two tab delimited fields, contained a check:
7019    
7020        if ((*pBuf != '\n') || (*pBuf != '\0')) {
7021            return(XcmsFailure);
7022    
7023    which would cause it to return failure unless *pBuf had a value that
7024    was simultaneously equal to both \n & \0, and no one wants to live in
7025    a world where that could ever be true.
7026    
7027    This has gone unnoticed since 1991, since this only caused lines
7028    in Xcms.txt that started with whitespace to be rejected, but now
7029    gcc -Wlogicalop has brought it to our attention, and
7030    https://bugs.freedesktop.org/show_bug.cgi?id=70803 was filed.
7031    
7032    Now that we see it, and cannot unsee it, we change it to use the
7033    same logic as the check at other points in this function, to return
7034    failure only if we hit \n or \0 before we find the first non-whitespace
7035    character, so that lines starting with whitespace will have the space
7036    skipped over to get to the color name to be defined.
7037    
7038    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7039    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
7040
7041commit 18a5278b008e9faa59b346fcab18a8d74b875fda
7042Author: Gaetan Nadon <memsize@videotron.ca>
7043Date:   Sat Sep 28 17:33:52 2013 -0400
7044
7045    makekeys: don't need to use target-specific CFLAGS
7046    
7047    It's the only thing built in that directory, so we can use AM_CFLAGS
7048    and AM_CPPFLAGS as usual.
7049    
7050    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7051    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
7052
7053commit 5dcb40f28d59587597d2ff6e6ac64c71cfe6ff7b
7054Author: James Cloos <cloos@jhcloos.com>
7055Date:   Tue Sep 17 12:50:42 2013 -0400
7056
7057    nls/en_US.UTF-8/Compose.pre: Fix typo.
7058    
7059    Fix typo added in 215ce6a67863, s/actute/acute/.
7060    
7061    Fixes bug #69476.  Reported by Jean Krohn.
7062    
7063    Signed-off-by: James Cloos <cloos@jhcloos.com>
7064
7065commit cb107760df33ffc8630677e66e2e50aa37950a5c
7066Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7067Date:   Sun Sep 8 18:37:01 2013 -0700
7068
7069    libX11 1.6.2
7070
7071commit 215ce6a67863de7acfd6dd3562b4fd97ef87b411
7072Author: Benno Schulenberg <bensberg@justemail.net>
7073Date:   Sun Sep 1 12:38:30 2013 +0200
7074
7075    nls: Adding more accessible compose sequences for J́ and j́.
7076    
7077    Few keyboards have an <acute> key, so this adds the much more
7078    accessible and usual compose sequences with <apostrophe>, ánd
7079    the most comfortable ones with <dead_acute>.
7080    
7081    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
7082    Signed-off-by: James Cloos <cloos@jhcloos.com>
7083
7084commit e9b14d10d0258bfcc273ff8bc84cd349dccda62c
7085Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7086Date:   Sat Aug 24 17:27:43 2013 -0700
7087
7088    Bug 68413 - [Bisected]Error in `xterm': realloc(): invalid next size
7089    
7090    Pass *new* size to realloc, not old size.
7091    
7092    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7093
7094commit c2b8e30790c21d6386767265263b3294ce1b1f9a
7095Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7096Date:   Fri Aug 16 21:04:02 2013 -0700
7097
7098    Stop checking for HAVE_DIX_CONFIG_H on the client side
7099    
7100    Leftover from when these XKB files were shared with the server sources
7101    and could be compiled in either the client or server, with the different
7102    autoconf config files in each.
7103    
7104    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7105
7106commit 84276609b2f0aec74fb464c428c7db5714b0fcfc
7107Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7108Date:   Fri Aug 16 18:27:28 2013 -0700
7109
7110    Rearrange some variable declarations & initializations in XKB
7111    
7112    Little things noticed during XKB restyling that seemed to make the
7113    code easier to read.
7114    
7115    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7116
7117commit b90b7e859cf45ec76921fa21bbfc1f3840d6e8d1
7118Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7119Date:   Sun Aug 11 13:29:33 2013 -0700
7120
7121    Reindent XKB code to X.Org standard style
7122    
7123    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7124
7125commit c0a0f78eb49c2e4ad956209de77475c85b9314ea
7126Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7127Date:   Fri Aug 16 18:14:14 2013 -0700
7128
7129    Fix overflow checks in _XkbReadKeySyms when key_sym_map is already created
7130    
7131    We were checking to make sure that the largest keysym value was within
7132    the range of the allocated buffer, but checking against different limits
7133    in the not-yet-allocated vs. the already-allocated branches.
7134    
7135    The check should be the same in both, and reflect the size used for the
7136    allocation, which is based on the maximum key code value, so we move it
7137    to be a common check, before we branch, instead of duplicating in each
7138    branch.
7139    
7140    map->key_sym_map is an array of XkbSymMapRec structs, [0..max_key_code]
7141    map->syms is the array for which num_syms is recorded, hence is not the
7142    right value to check for ensuring our key_sym_map accesses are in range.
7143    
7144    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7145    Reported-by: Barry Kauler <bkauler@gmail.com>
7146    Tested-by: Barry Kauler <bkauler@gmail.com>
7147
7148commit bea6cbd027973142fc64532274e1d16861b47190
7149Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7150Date:   Sun Aug 11 17:02:21 2013 -0700
7151
7152    Remove long unused src/udcInf.c
7153    
7154    I can find no record of what this file was for.  Neither the X11R6.8.2
7155    monolith Imakefile nor any modular release Makefile.am have ever built
7156    it and nothing else references it.
7157    
7158    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7159
7160commit bf3501e0395abe890acfea98fdd9f50a6966f118
7161Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7162Date:   Sun Aug 11 00:07:33 2013 -0700
7163
7164    Remove unnecessary casts of pointers to (char *) in calls to Xfree()
7165    
7166    Left one cast behind that is necessary to change from const char *
7167    to char * in src/xlibi18n/lcCharSet.c.
7168    
7169    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7170
7171commit 6ead9dd92ab90aabd9f0e328d59597e6b5bc09d3
7172Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7173Date:   Sat Aug 10 23:57:55 2013 -0700
7174
7175    Don't cast sizeof() results to unsigned when passing to Xmalloc/Xcalloc
7176    
7177    sizeof() returns size_t, malloc() & calloc() expect sizes in size_t,
7178    don't strip down to unsigned int and re-expand unnecessarily.
7179    
7180    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7181
7182commit 25a7a329def672fc8d26078538173777850c6390
7183Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7184Date:   Sat Aug 10 23:51:08 2013 -0700
7185
7186    Remove even more casts of return values from Xmalloc/Xrealloc
7187    
7188    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7189
7190commit f8fa16092a148b74ca35b4beb182053352606f2f
7191Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7192Date:   Sat Aug 10 23:05:13 2013 -0700
7193
7194    xlibi18n: fix argsize argument to _XlcParsePath
7195    
7196    The array is defined as having NUM_LOCALEDIR entries, so use that
7197    instead of hardcoded 256 value (the other two calls already did this).
7198    
7199    Reported by parfait:
7200       Buffer overflow (CWE 120): In pointer dereference of argv[argc] with index argc
7201          Pointer size is 64 elements (of 8 bytes each), index is 255
7202            at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'.
7203            called at line 178 in function '_XlcParsePath' with argv = argv.
7204            called at line 722 in function '_XlcLocaleLibDirName' with argv = args, argsize = 256.
7205            at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'.
7206            called at line 178 in function '_XlcParsePath' with argv = argv.
7207            called at line 638 in function '_XlcLocaleDirName' with argv = args, argsize = 256.
7208    
7209    [ This bug was found by the Parfait 1.2.0 bug checking tool.
7210      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
7211    
7212    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7213
7214commit e7d46c6452c0b90fd66ae9f538546b968e0dd608
7215Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7216Date:   Sat Aug 10 22:32:42 2013 -0700
7217
7218    i18n modules: Fix some const cast warnings
7219    
7220    imRm.c: In function '_XimSetICMode':
7221    imRm.c:2419:37: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7222    imRm.c:2420:30: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7223    
7224    lcGenConv.c: In function 'byteM_parse_codeset':
7225    lcGenConv.c:345:13: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7226    
7227    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7228
7229commit cbd86eccf175dc82a5cbcea54c8bd21ce18b70c0
7230Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7231Date:   Sat Aug 10 22:21:54 2013 -0700
7232
7233    xlibi18n: Fix a bunch of const cast warnings
7234    
7235    Add const qualifiers to casts where needed, remove other casts that
7236    are no longer needed.
7237    
7238    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7239
7240commit eb3676113fc2dd0f34d92b89beb81b3f61569aa1
7241Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7242Date:   Sat Aug 10 22:18:00 2013 -0700
7243
7244    Fix const handling in XSetLocaleModifiers
7245    
7246    Instead of reusing the input parameter to store the output, make a
7247    result variable instead, so that there's less const confusion.
7248    
7249    Fixes gcc warnings:
7250    lcWrap.c: In function 'XSetLocaleModifiers':
7251    lcWrap.c:87:18: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7252    lcWrap.c:91:25: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7253    lcWrap.c:93:12: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7254    
7255    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7256
7257commit 8ebbffa98563960910152e4f2e31cb032375d871
7258Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7259Date:   Sat Aug 10 21:46:37 2013 -0700
7260
7261    Constify lc_name argument to _XlcLocaleDirName() & _XlcLocaleLibDirName()
7262    
7263    Makes code considerably less crufty and clears gcc warnings:
7264    XlcDL.c: In function '_XlcDynamicLoad':
7265    XlcDL.c:384:44: warning: cast discards '__attribute__((const))' qualifier
7266     from pointer target type [-Wcast-qual]
7267    XlcDL.c:386:51: warning: cast discards '__attribute__((const))' qualifier
7268     from pointer target type [-Wcast-qual]
7269    
7270    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7271
7272commit 07e4e864107b38c2f393564fdacc90f4e858f23f
7273Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7274Date:   Sat Aug 10 13:37:53 2013 -0700
7275
7276    init_om: remove unneeded extra copy of string to local buffer
7277    
7278    Strings from the supported_charset_list[] were being copied one by
7279    one to a stack buffer, and then strdup called on that buffer.
7280    
7281    Instead, just strdup the original string, without the local copy,
7282    and use a more traditional for loop, so it's easier to figure out
7283    what the code is doing (cleaning up a gcc const-cast warning in
7284    the process).
7285    
7286    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7287    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7288
7289commit 1cec14dad904ba21a861f4af131be5982ecb83dd
7290Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7291Date:   Sat Aug 10 12:34:53 2013 -0700
7292
7293    Delete unused XKB_INSURE_SIZE macro from XKBlibint.h
7294    
7295    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7296    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7297
7298commit 5f32182c7c4045540ff3833c48ee24a3a25726e2
7299Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7300Date:   Sat Aug 10 12:19:17 2013 -0700
7301
7302    miRegionOp(): ensure region size is not updated if realloc fails
7303    
7304    This function performs operations on a region, and when finished,
7305    checks to see if it should compact the rectangle list.  If the
7306    number of rectangles for which memory is allocated in the list is
7307    more than twice the number used, it tries to shrink.   realloc()
7308    should not fail in this case, but if it does, might as well keep
7309    the correct value for the number of allocated rectangles, so we
7310    don't try to grow it unnecessarily later if adding to the region.
7311    
7312    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7313
7314commit bd2a0b5a187798bb2e2f05dc5062ca79e37075dd
7315Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7316Date:   Sat Aug 10 12:19:17 2013 -0700
7317
7318    miRegionCopy(): handle realloc failure better
7319    
7320    Zero out the region size when freeing the region so callers don't think
7321    there's anything there.   (Pointer is already set to NULL from the realloc
7322    result itself.)  Return 0 to the callers, and have them cascade that back
7323    to their callers to indicate failure, instead of their usual return value
7324    of 1 on success.
7325    
7326    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7327
7328commit 5dc8b5385d513bbda88697c2372db750d23f46d4
7329Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7330Date:   Sat Aug 10 11:27:22 2013 -0700
7331
7332    Avoid memory leak/corruption if realloc fails in Xregion.h:MEMCHECK macro
7333    
7334    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7335
7336commit 453c4ee436ef32d91501d7736d7a91c1aeafc565
7337Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7338Date:   Sat Aug 10 12:07:51 2013 -0700
7339
7340    Avoid memory leak/corruption if realloc fails in imLcPrs.c:parseline()
7341    
7342    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7343
7344commit b3fea74ec5b7d4f83755a52a8d49c564b71c6d12
7345Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7346Date:   Sat Aug 10 12:30:39 2013 -0700
7347
7348    lcDB.c: ensure buffer size is updated correctly if realloc fails
7349    
7350    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7351
7352commit 43bb822c714a73c3b2d15e621ffb3333cd10da8c
7353Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7354Date:   Sat Aug 10 11:07:47 2013 -0700
7355
7356    Avoid memory leak/corruption if realloc fails in XlcDL.c:resolve_object()
7357    
7358    Previously, if realloc failed to increase the size, we'd still
7359    record that we had allocated the larger size, but the pointer
7360    to it would be NULL, causing future calls to be broken, and the
7361    previous allocation to be lost/leaked.
7362    
7363    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7364    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7365
7366commit 5d47a39978e92bb34ec928b1b15d71c0c2434870
7367Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7368Date:   Fri Aug 9 23:33:03 2013 -0700
7369
7370    omGeneric.c: convert sprintf calls to snprintf
7371    
7372    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7373    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7374
7375commit 88a27a2aa9b7d35cb79b16334ea3413e572b724a
7376Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7377Date:   Fri Aug 9 23:30:30 2013 -0700
7378
7379    ximcp/imRm.c: convert sprintf calls to snprintf
7380    
7381    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7382    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7383
7384commit 4fadae243fb485628c9a137f5da3489ed6214b21
7385Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7386Date:   Fri Aug 9 23:02:12 2013 -0700
7387
7388    xlibi18n: convert sprintf calls to snprintf
7389    
7390    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7391    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7392
7393commit 36a7edf0e5edfc5ef4ff2c3a8b4fa3dc4796e854
7394Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7395Date:   Fri Aug 9 23:02:12 2013 -0700
7396
7397    lcfile: skip over any null entries in args list
7398    
7399    Previous code seemed to assume that printf("%s", NULL) would result
7400    in a 0-length string, not "(null)" or similar, but since there's no
7401    point looking for files in "(null)/filepath...", instead we just
7402    skip over NULL entries in search paths when generating file names.
7403    
7404    In the *DirName() functions, this effectively just moves the "bail on
7405    NULL in arg[i]" check up from the later code that assigned it to targetdir
7406    and then bailed if that was NULL.
7407    
7408    Not sure how there ever could be a NULL in arg[i], given the current
7409    implementation of XlcParsePath, but it's easy enough to check once and
7410    reject up front instead of on every reference.
7411    
7412    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7413
7414commit ee0824f24392d5ca3d5fd5f5ed8d78c0d892f7c0
7415Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7416Date:   Fri Aug 9 22:00:09 2013 -0700
7417
7418    Fix file leak on malloc error in XlcDL.c:resolve_object()
7419    
7420    File Leak: Leaked File fp
7421       at line 219 of lib/libX11/src/xlibi18n/XlcDL.c in function 'resolve_object'.
7422          fp initialized at line 198 with fopen
7423    
7424    [ This bug was found by the Parfait 1.2.0 bug checking tool.
7425      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
7426    
7427    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7428    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7429
7430commit 9b291044a240e5b9b031ed814e0c84e53a1c3084
7431Author: Julien Cristau <jcristau@debian.org>
7432Date:   Sat Jun 15 18:02:21 2013 +0200
7433
7434    Add missing locales to configure.ac
7435    
7436    Commits 40761898692e5063957bfa2518cca3d35b2e354a and
7437    f198c6aa98f88ff285d903175a3c4c0fd33a4575 added two new locales
7438    (sr_CS.UTF-8 and km_KH.UTF-8), but didn't list them in configure.ac,
7439    meaning they're not included in tarballs.
7440    
7441    Signed-off-by: Julien Cristau <jcristau@debian.org>
7442    Reviewed-by: James Cloos <cloos@jhcloos.com>
7443
7444commit 8f58e54a5f46c3cd4897a23b89950f4800ae38d4
7445Author: ISHIKAWA,chiaki <ishikawa@yk.rim.or.jp>
7446Date:   Tue Dec 18 15:28:05 2012 +0000
7447
7448    Fix bogus timestamp generated by XIM
7449    
7450    Fix bogus timestamp generted by XIM due to uninitialized
7451    data field. Also set appropriate serial, too.
7452    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39367
7453    
7454    Signed-off-by: Chiaki ISHIKAWA <ishikawa@yk.rim.or.jp>
7455    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7456
7457commit e7fd6f0eda57300df4d6b695b7064610ca5dec57
7458Author: Egbert Eich <eich@freedesktop.org>
7459Date:   Thu Jun 16 18:47:49 2011 +0200
7460
7461    XIM: Fix sync problem on focus change.
7462    
7463    XSetICFocus() and XUnsetICFocus() are both asynchronous events.
7464    This is a pretty stupid idea: those functions may undo certain
7465    settings on the client side for which requests from the server
7466    may still be in the queue unprocessed. Thus things may be set
7467    in the wrong order ie instead of set -> unest it will be unset -> set.
7468    Moreover there is no way for either the client or the server to
7469    cause the event queue to be flushed - which is pretty bad as
7470    XIM is bidirectional.
7471    The scenario is as follows:
7472    Two ICs are created:
7473            ic1 = XCreateIC(im,
7474                XNInputStyle, XIMPreeditCallbacks | XIMStatusCallbacks,
7475                XNClientWindow, window,
7476                XNPreeditAttributes, preedit_attr,
7477                XNStatusAttributes, status_attr,
7478                NULL);
7479            ic2 = XCreateIC(im, XNInputStyle,
7480                           XIMPreeditNothing | XIMStatusNothing,
7481                           XNClientWindow, window, NULL);
7482    Then the focus is removed from ic2:
7483            XUnsetICFocus(ic2);
7484    If SCIM is used as the input server it will send a bunch of requests
7485    following an XCreateIC(). One of the requests registers a key release
7486    filter. XUnsetICFocus() unsets both key press and release filters.
7487    Since it is asynchronous, the input server requests to register key
7488    press and release filters may not have been processed, when XUnsetICFocus()
7489    is called. Since there is no explicite way for client programs to enforce
7490    the request queue to be flushed explicitely before an X[Set/Unset]ICFocus()
7491    call it would be safest to make those two calls synchronous in the sense
7492    that they ensure the request queue has been handled before they execute.
7493    The easiest way to do this from Xlib is thru a call to XGetICValues()
7494    which sends a request to the server and subsequently reads the queue
7495    from the server to the client. This will cause all outstanding requests
7496    in the queue to be read and handled.
7497    This is an ugly hack and this could be fixed directly in the client,
7498    however it seems to be easier to fix Xlib than to fix numerous clients.
7499    This problem arose since there is no well documented way how to handle
7500    and synchronize XIM requests and not all input servers send requests
7501    when an IC is created.
7502    This has been discussed extensively in:
7503     https://bugzilla.novell.com/show_bug.cgi?id=221326
7504    
7505    Signed-off-by: Egbert Eich <eich@freedesktop.org>
7506
7507commit 26ec7d3821bc19debc73c8c3e42e6e33ef6f856e
7508Author: Egbert Eich <eich@freedesktop.org>
7509Date:   Thu Jun 16 17:28:39 2011 +0200
7510
7511    XIM: Fix race on focus change: set 'FABRICATED' only when keyev filters in place.
7512    
7513    When synthesized key events are sent on commit XIM sets the 'fabricated'
7514    flag so that the keypress handler knows that these were not real events.
7515    This also happens when committing due to the loss of focus. However in this
7516    case the keypress/release filters which consume and unset this flag are no
7517    longer in the filter chain.
7518    So the flag is erronously set when a real keyboard event is received after
7519    focus has been regained. So the first event is wrongly treated as a
7520    fabricated key in the keypress handler which will at the same time reset
7521    the flag so the second key event is treated correctly.
7522    This fix only sets the flag when at least one of the keyboard filters is in
7523    place.
7524    How to reproduce this bug: run scim, choose a Japanese input method start
7525    two instances of xterm: start typing in one xterm (this should pop up an
7526    IM window). Without comitting (hitting 'enter') move focus to the other
7527    xterm, then move focus back. Start typing again. The first character will
7528    be committed immediately without popping up an input window.
7529    With this fix this behavior is gone.
7530    
7531    See also: https://bugzilla.novell.com/show_bug.cgi?id=239698
7532    
7533    Signed-off-by: Egbert Eich <eich@freedesktop.org>
7534
7535commit 44f84223f5e2dd46883fcbd352af2798bfa9aeb6
7536Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7537Date:   Mon Jul 29 21:29:49 2013 -0700
7538
7539    libX11 1.6.1
7540    
7541    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7542
7543commit d19cfaca15826943d3c637ef7fa5db0a23d5feed
7544Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7545Date:   Sat Jul 27 12:19:00 2013 -0700
7546
7547    Fix undefined XCMSDIR error when building lint library
7548    
7549    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7550
7551commit 3083cd43d7dcd59da587975e7cadda681cd8a103
7552Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7553Date:   Sat Jul 27 00:36:08 2013 -0700
7554
7555    Add ku_TR.UTF-8 (Kurdish language, Turkey region) to compose/locale.dir
7556    
7557    Upstreaming from changes originally integrated into OpenSolaris
7558    under Sun bug id 6882572.
7559    
7560    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7561    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
7562
7563commit 208e586c808e88a2ee819e4450dc27f557afc2bf
7564Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7565Date:   Sat Jul 27 01:03:18 2013 -0700
7566
7567    omGeneric: remove space between struct name & member name
7568    
7569    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7570
7571commit 7db74514e454d3fc4ff70aa08ddac66bfffda4dd
7572Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7573Date:   Tue Jul 23 22:18:46 2013 -0700
7574
7575    Refactor common code from XAddHost & XRemoveHost into single function
7576    
7577    On the Xlib side, the only real difference is the mode flag we send
7578    to the server with the address, so just make that an argument to the
7579    function with the common code for packing the address into the request.
7580    
7581    (Aside from labels, gcc 4.7.2 generates identical code before & after
7582     this change due to inlining, verified via diff of gcc -S output.)
7583    
7584    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7585
7586commit 3292195a64a9ce4f0d27134cd544651ec647e728
7587Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7588Date:   Tue Jul 23 22:11:34 2013 -0700
7589
7590    XSetModifierMapping: Use Data instead of GetReqExtra
7591    
7592    Handle arbitrary length data in the same fashion as other calls,
7593    avoiding need to ensure it fits all in the request buffer.
7594    
7595    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7596
7597commit feb131b18aee31c2c125dc3275b0260940245882
7598Author: Kees Cook <kees@outflux.net>
7599Date:   Sun Jun 9 11:13:43 2013 -0700
7600
7601    libX11: check "req" when calling GetReqExtra
7602    
7603    This fixes the two callers of GetReqExtra to check that "req" is non-NULL
7604    to avoid crashing now that GetReqExtra does internal bounds-checking on
7605    the resulting buffer sizes.
7606    
7607    Additionally updates comment describing return values to use names
7608    instead of only literal values.
7609    
7610    Signed-off-by: Kees Cook <kees@outflux.net>
7611    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7612    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7613
7614commit 54540d7cba0c2bfe9176221c7bca910058d304df
7615Author: Kees Cook <kees@outflux.net>
7616Date:   Sun Jun 9 11:13:42 2013 -0700
7617
7618    libX11: check size of GetReqExtra after XFlush
7619    
7620    Two users of GetReqExtra pass arbitrarily sized allocations from the
7621    caller (ModMap and Host). Adjust _XGetRequest() (called by the GetReqExtra
7622    macro) to double-check the requested length and invalidate "req" when
7623    this happens. Users of GetReqExtra passing lengths greater than the Xlib
7624    buffer size (normally 16K) must check "req" and fail gracefully instead
7625    of crashing.
7626    
7627    Any callers of GetReqExtra that do not check "req" for NULL
7628    will experience this change, in the pathological case, as a NULL
7629    dereference instead of a buffer overflow. This is an improvement, but
7630    the documentation for GetReqExtra has been updated to reflect the need
7631    to check the value of "req" after the call.
7632    
7633    Bug that manifested the problem:
7634    https://bugs.launchpad.net/ubuntu/+source/x11-xserver-utils/+bug/792628
7635    
7636    Signed-off-by: Kees Cook <kees@outflux.net>
7637    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7638    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7639
7640commit 24d3ee0d08f24e23c91d55702f010f73d7b908e5
7641Author: Thomas Klausner <wiz@NetBSD.org>
7642Date:   Tue Jun 25 22:35:29 2013 +0200
7643
7644    Tighten out-of-range comparisons.
7645    
7646    [For all of these, LONG_MAX was the correct value to prevent overflows
7647     for the recent CVEs.   Lowering to INT_MAX catches buggy replies from
7648     the server that 32-bit clients would reject but 64-bit would accept,
7649     so we catch bugs sooner, and really, no sane & working server should
7650     ever report more than 2gb of extension names, font path entries,
7651     key modifier maps, etc.  -alan- ]
7652    
7653    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7654    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7655
7656commit 6d926088d80a08e13e6d6c4ff207b81ad52e667f
7657Author: Thomas Klausner <wiz@NetBSD.org>
7658Date:   Tue Jun 25 18:34:32 2013 +0200
7659
7660    Fix out-of-range comparison in _XF86BigfontQueryFont
7661    
7662    clang complained (correctly):
7663    warning: comparison of constant 768614336404564650 with expression
7664    of type 'CARD32' (aka 'unsigned int') is always true
7665    [-Wtautological-constant-out-of-range-compare]
7666    
7667    [While LONG_MAX is correct, since it's used in size_t math, the
7668     numbers have to be limited to 32-bit range to be usable by 32-bit
7669     clients, and values beyond that range are far more likely to be
7670     bugs in the data from the server than valid numbers of characters
7671     in a font.   -alan- ]
7672    
7673    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7674    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7675
7676commit 383e2b0d029482a0f4c39fe00e15397538576fc1
7677Author: Thomas Klausner <wiz@NetBSD.org>
7678Date:   Tue Jun 25 18:33:56 2013 +0200
7679
7680    Check for symbol existence with #ifdef, not #if
7681    
7682    Reviewed-by: Jamey Sharp <jamey@minilop.net>
7683    
7684    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7685
7686commit 9955d1c8de994a90fe7f2e3187e7362611d7d265
7687Author: Thomas Klausner <wiz@NetBSD.org>
7688Date:   Tue Jun 25 18:33:07 2013 +0200
7689
7690    Use newer callback-based API for XIM.
7691    
7692    Let libX11 load and make available the newer (X11R6) callback-based
7693    API for XIM (expected by emacs).
7694    
7695    This patch updates the files to match the other nls/ files.
7696    
7697    Patch from Ian D. Leroux <idleroux@fastmail.fm> on pkgsrc-users@NetBSD.org
7698    following a hint by Nhat Minh Lê <nhat.minh.le@gmail.com>.
7699    
7700    Reviewed-by: James Cloos <cloos@jhcloos.com>
7701    
7702    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7703
7704commit a17ceb7100bd36c2db210473ee701deb5d515731
7705Author: Thomas Klausner <wiz@NetBSD.org>
7706Date:   Tue Jun 25 18:31:32 2013 +0200
7707
7708    Stop truncating source to destination length if it is larger.
7709    
7710    It seems useless to do that since the code tests for both source
7711    length and destination to be non-zero. This fixes a cut'n'paste
7712    problem in xterm where the paste length was limited to 1024 (BUFSIZ)
7713    in button.c.
7714    
7715    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7716
7717commit a336db9a0add3ae0783dda6e52459236622a12af
7718Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7719Date:   Mon Jun 24 23:02:05 2013 -0700
7720
7721    Require ANSI C89 pre-processor, drop pre-C89 token pasting support
7722    
7723    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7724
7725commit 9dfb0f3c0a761590bcdc1f3396b1e064da4e18e8
7726Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7727Date:   Fri Jun 7 11:30:11 2013 -0700
7728
7729    troff macro expansion in specs/libX11
7730    
7731    Many of the custom nroff macros (.ds <macro> <contents>) were left
7732    unsubstituted in the nroff->docbook conversion.   This substitution
7733    is now performed, via the following perl script:
7734    
7735    #! /usr/bin/perl -w -i
7736    
7737    use Text::Wrap;
7738    
7739    while ($_ = <>) {
7740        while ($_ =~ m/\((\w+)\b/g) {
7741            my $m = $1;
7742            if (exists $macro{$m}) {
7743                $_ =~ s/\($m/$macro{$m}/;
7744                $_ = wrap('', '', $_);
7745                $_ =~ s/[ \t]+$//;
7746            }
7747        }
7748    
7749        if ($_ =~ /\<!-- .ds (\w+) (.*) -->/) {
7750            my ($m, $s) = ($1, $2);
7751            $macro{$m} = $s;
7752            while ($macro{$m} =~ /\\\s*$/) {
7753                $macro{$m} =~ s/\\\s*$//ms;
7754                $macro{$m} .= <>;
7755                chomp($macro{$m});
7756            }
7757            $macro{$m} =~ s/\\ / /g;
7758        } else {
7759            print $_;
7760        }
7761    }
7762    
7763    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7764
7765commit 20c17bd9ebf767a24643279e45866dddcb57b5ce
7766Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7767Date:   Fri Jun 7 09:27:26 2013 -0700
7768
7769    specs/libX11: correct prototype for XListPixmapFormats/XImageByteOrder
7770    
7771    The XListPixmapFormats arguments was being shown with XImageByteOrder's
7772    name and return types.   Appears to have been a glitch in the nroff ->
7773    docbook conversion.
7774    
7775    Reported-by: ZHANG Zhaolong <zhangzl2013@126.com>
7776    Reviewed-by: Jamey Sharp <jamey@minilop.net>
7777    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7778
7779commit ed3d830243c8a0eefaf24e15b11823272ffe5049
7780Author: Thomas Klausner <wiz@NetBSD.org>
7781Date:   Sun Jun 2 20:49:55 2013 +0200
7782
7783    Deal with the limited range of VAX floating point numbers when compiling for VAX.
7784    
7785    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7786
7787commit 0a48235d63639fb917c44d27c86e928e79fbac66
7788Author: Eric S. Raymond <esr@thyrsus.com>
7789Date:   Thu Jun 6 16:43:56 2013 -0400
7790
7791    Remove call to undefined macro.
7792
7793commit 9e4719b9b719f2f8d255f6778e2e8c1809e32599
7794Author: Eric S. Raymond <esr@thyrsus.com>
7795Date:   Thu Jun 6 16:42:20 2013 -0400
7796
7797    Remove call to undefined macro.
7798
7799commit 8496122eb00ce6cd5d2308ee54f64b68c378e455
7800Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7801Date:   Mon Jun 3 20:06:43 2013 -0700
7802
7803    Update README to reflect where to find the Xlib specs now
7804    
7805    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7806
7807commit 655d631e86c95b14888758b27ed2836ca3e3ce86
7808Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7809Date:   Mon Jun 3 19:21:06 2013 -0700
7810
7811    libX11 1.6.0
7812    
7813    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7814
7815commit 4a89b7ea908554628f374537a79931c8006a2de3
7816Author: Thomas Klausner <wiz@NetBSD.org>
7817Date:   Sun Jun 2 11:49:54 2013 -0700
7818
7819    cmsMath.c: Add missing stdio header for printf(3) in DEBUG build.
7820    
7821    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7822
7823commit 96dcf747f13b26b8e4d17b1bc8605d933c3e1dc6
7824Author: Thomas Klausner <wiz@NetBSD.org>
7825Date:   Sun Jun 2 20:49:48 2013 +0200
7826
7827    XCreateGC man page: Avoid .TS H and .TH macros
7828    
7829    Avoid .TS H and .TH for now as it doesn't alter the output in this case,
7830    and improve the output with mandoc(1).
7831    
7832    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7833
7834commit 95a388158c9d73df7d24016d6a3d61506d7d53a4
7835Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7836Date:   Thu May 23 19:43:35 2013 -0700
7837
7838    libX11 1.5.99.902 (1.6 RC2)
7839    
7840    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7841
7842commit a3bdd2b090915fe0163b062f0e6576fe05dd332e
7843Author: Julien Cristau <jcristau@debian.org>
7844Date:   Thu May 23 20:39:46 2013 +0200
7845
7846    xkb: fix off-by-one in _XkbReadGetNamesReply and _XkbReadVirtualModMap
7847    
7848    The size of the arrays is max_key_code + 1.  This makes these functions
7849    consistent with the other checks added for CVE-2013-1997.
7850    
7851    Also check the XkbGetNames reply when names->keys was just allocated.
7852    
7853    Signed-off-by: Julien Cristau <jcristau@debian.org>
7854    Tested-by: Colin Walters <walters@verbum.org>
7855    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7856
7857commit 7e30056e78e4b7979ff47f102e00327617266019
7858Author: Niveditha Rau <Niveditha.Rau@Oracle.COM>
7859Date:   Fri May 17 15:26:21 2013 -0700
7860
7861    Make sure internal headers include required headers
7862    
7863    Fixes builds with Solaris Studio 12.3 when lint is enabled, since it no
7864    longer ignores *.h files, but complains when they reference undefined
7865    typedefs or macros.
7866    
7867    Signed-off-by: Niveditha Rau <Niveditha.Rau@Oracle.COM>
7868    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7869
7870commit 2820100bf8ba130b94253f415e7fa5ac28bb2037
7871Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7872Date:   Thu May 16 23:05:36 2013 -0700
7873
7874    Free fs->properties in _XF86BigfontQueryFont overflow error path
7875    
7876    Fixes small memory leak introduced in commit 5669a22081
7877    
7878    Reported-by: Julien Cristau <jcristau@debian.org>
7879    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7880
7881commit 3131740513133a9ff7cb12123d29ceb18584fc38
7882Author: Matthieu Herrb <matthieu.herrb@laas.fr>
7883Date:   Wed May 8 19:33:09 2013 +0200
7884
7885    XListFontsWithInfo: Re-decrement flist[0] before calling free() on it.
7886    
7887    Freeing a pointer that wasn't returned by malloc() is undefined
7888    behavior and produces an error with OpenBSD's implementation.
7889    
7890    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7891    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7892    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7893
7894commit 3fe4bea086149f06a142a8f1d575f627ec1e22c7
7895Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7896Date:   Fri Apr 19 14:30:40 2013 -0700
7897
7898    Give GNU & Solaris Studio compilers hints about XEatData branches
7899    
7900    Try to offset the cost of all the recent checks we've added by giving
7901    the compiler a hint that the branches that involve us eating data
7902    are less likely to be used than the ones that process it.
7903    
7904    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7905
7906commit e1b457beb8d4e831ef44279dada6c475cb955738
7907Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7908Date:   Sun Mar 31 12:22:35 2013 -0700
7909
7910    _XkbReadGetMapReply: reject maxKeyCodes smaller than the minKeyCode
7911    
7912    Various other bounds checks in the code assume this is true, so
7913    enforce it when we first get the data from the X server.
7914    
7915    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7916
7917commit 12ad4c6432496897ff000eb7cfecd0fb4b290331
7918Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7919Date:   Sat Mar 16 10:03:13 2013 -0700
7920
7921    Use calloc in XOpenDisplay to initialize structs containing pointers
7922    
7923    Prevents trying to free uninitialized pointers if we have to bail out
7924    partway through setup, such as if we receive a corrupted or incomplete
7925    connection setup block from the server.
7926    
7927    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7928
7929commit d38527e25f8b6e2f1174ecc21260c5c5416f972e
7930Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7931Date:   Thu Mar 7 23:46:05 2013 -0800
7932
7933    Remove more unnecessary casts from Xmalloc/calloc calls
7934    
7935    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7936
7937commit b2c86b582c58f50c7b14da01cf7ebd20ef12a6b2
7938Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7939Date:   Sat Mar 2 16:56:16 2013 -0800
7940
7941    Convert more _XEatData callers to _XEatDataWords
7942    
7943    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7944    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7945
7946commit 192bbb9e2fc45df4e17b35b6d14ea0eb418dbd39
7947Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7948Date:   Sat Mar 9 11:04:37 2013 -0800
7949
7950    Make XGetWindowProperty() always initialize returned values
7951    
7952    Avoids memory corruption and other errors when callers access them
7953    without checking to see if XGetWindowProperty() returned an error value.
7954    
7955    Callers are still required to check for errors, this just reduces the
7956    damage when they don't.
7957    
7958    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7959    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7960    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7961
7962commit db1b1c871da29aa0545182bf888df81627f165a5
7963Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7964Date:   Sat Mar 2 15:08:21 2013 -0800
7965
7966    Avoid overflows in XListExtensions() [CVE-2013-1997 15/15]
7967    
7968    Ensure that when breaking the returned list into individual strings,
7969    we don't walk past the end of allocated memory to write the '\0' bytes
7970    
7971    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7972    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7973
7974commit 8d5936594993921acdfec778dd8f41b555e2543a
7975Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7976Date:   Sat Mar 2 15:08:21 2013 -0800
7977
7978    Avoid overflows in XGetFontPath() [CVE-2013-1997 14/15]
7979    
7980    Ensure that when breaking the returned list into individual strings,
7981    we don't walk past the end of allocated memory to write the '\0' bytes
7982    
7983    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7984    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7985
7986commit 0c404db6a92dc2c198328bf586c02d8abbe02013
7987Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7988Date:   Sat Mar 2 15:08:21 2013 -0800
7989
7990    Avoid overflows in XListFonts() [CVE-2013-1997 13/15]
7991    
7992    Ensure that when breaking the returned list into individual strings,
7993    we don't walk past the end of allocated memory to write the '\0' bytes
7994    
7995    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7996    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7997
7998commit 0b0f5d4358c3de7563d6af03f0d2ce454702a06a
7999Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8000Date:   Sat Mar 2 15:08:21 2013 -0800
8001
8002    integer overflow in XGetModifierMapping() [CVE-2013-1981 13/13]
8003    
8004    Ensure that we don't underallocate when the server claims a very large reply
8005    
8006    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8007    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8008
8009commit a351b8103b2ba78882e1c309e85893ca3abe2073
8010Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8011Date:   Sat Mar 2 15:08:21 2013 -0800
8012
8013    integer overflow in XGetPointerMapping() & XGetKeyboardMapping() [CVE-2013-1981 12/13]
8014    
8015    Ensure that we don't underallocate when the server claims a very large reply
8016    
8017    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8018    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8019
8020commit 833f6b70bc789d33607f6dbfee9e0a4178ec4b59
8021Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8022Date:   Sat Mar 2 15:08:21 2013 -0800
8023
8024    integer overflow in XGetImage() [CVE-2013-1981 11/13]
8025    
8026    Ensure that we don't underallocate when the server claims to have sent a
8027    very large reply.
8028    
8029    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8030    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8031
8032commit 79d8dc08eb98842173ce239b9dd60df0e9e9ae72
8033Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8034Date:   Fri Mar 8 22:25:35 2013 -0800
8035
8036    integer overflow in XGetWindowProperty() [CVE-2013-1981 10/13]
8037    
8038    If the reported number of properties is too large, the calculations
8039    to allocate memory for them may overflow, leaving us returning less
8040    memory to the caller than implied by the value written to *nitems.
8041    
8042    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8043    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8044    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8045
8046commit 164bf4dfe839b1cc75cdeee378a243d04a8200e4
8047Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8048Date:   Sat Mar 2 13:18:48 2013 -0800
8049
8050    integer overflows in TransFileName() [CVE-2013-1981 9/13]
8051    
8052    When trying to process file paths the tokens %H, %L, & %S are expanded
8053    to $HOME, the standard compose file path & the xlocaledir path.
8054    If enough of these tokens are repeated and values like $HOME are set to
8055    very large values, the calculation of the total string size required to
8056    hold the expanded path can overflow, resulting in allocating a smaller
8057    string than the amount of data we'll write to it.
8058    
8059    Simply restrict all of these values, and the total path size to PATH_MAX,
8060    because really, that's all you should need for a filename path.
8061    
8062    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8063    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8064    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8065
8066commit 460e8a223b87d4fa0ea1e97823e998a770e0f2a2
8067Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8068Date:   Fri Mar 1 18:37:37 2013 -0800
8069
8070    integer truncation in _XimParseStringFile() [CVE-2013-1981 8/13]
8071    
8072    Called from _XimCreateDefaultTree() which uses getenv("XCOMPOSEFILE")
8073    to specify filename.
8074    
8075    If the size of off_t is larger than the size of unsigned long (as in
8076    32-bit builds with large file flags), a file larger than 4 gigs could
8077    have its size truncated, leading to data from that file being written
8078    past the end of the undersized buffer allocated for it.
8079    
8080    While configure.ac does not use AC_SYS_LARGEFILE to set large file mode,
8081    builders may have added the large file compilation flags to CFLAGS on
8082    their own.
8083    
8084    size is left limited to an int, because if your Xim file is
8085    larger than 2gb, you're doing it wrong.
8086    
8087    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8088    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8089    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8090
8091commit 226622349a4b1e16064649d4444a34fb4be4f464
8092Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8093Date:   Sat Mar 2 12:39:58 2013 -0800
8094
8095    Unbounded recursion in _XimParseStringFile() when parsing include files [CVE-2013-2004 2/2]
8096    
8097    parseline() can call _XimParseStringFile() which can call parseline()
8098    which can call _XimParseStringFile() which can call parseline() ....
8099    eventually causing recursive stack overflow and crash.
8100    
8101    Limit is set to a include depth of 100 files, which should be enough
8102    for all known use cases, but could be adjusted later if necessary.
8103    
8104    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8105    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8106    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8107
8108commit 236b603d235dc264d1c6250dca09c745458a9088
8109Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8110Date:   Sat Mar 2 12:01:39 2013 -0800
8111
8112    Unbounded recursion in GetDatabase() when parsing include files [CVE-2013-2004 1/2]
8113    
8114    GetIncludeFile() can call GetDatabase() which can call GetIncludeFile()
8115    which can call GetDatabase() which can call GetIncludeFile() ....
8116    eventually causing recursive stack overflow and crash.
8117    
8118    Easily reproduced with a resource file that #includes itself.
8119    
8120    Limit is set to a include depth of 100 files, which should be enough
8121    for all known use cases, but could be adjusted later if necessary.
8122    
8123    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8124    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8125    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8126
8127commit 076428918e6c35f66b9b55c3fa097ff06496d155
8128Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8129Date:   Fri Mar 1 18:37:37 2013 -0800
8130
8131    integer overflow in ReadInFile() in Xrm.c [CVE-2013-1981 7/13]
8132    
8133    Called from XrmGetFileDatabase() which gets called from InitDefaults()
8134    which gets the filename from getenv ("XENVIRONMENT")
8135    
8136    If file is exactly 0xffffffff bytes long (or longer and truncates to
8137    0xffffffff, on implementations where off_t is larger than an int),
8138    then size may be set to a value which overflows causing less memory
8139    to be allocated than is written to by the following read() call.
8140    
8141    size is left limited to an int, because if your Xresources file is
8142    larger than 2gb, you're very definitely doing it wrong.
8143    
8144    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8145    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8146    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8147
8148commit 90fd5abac2faca86f9f100353a3c9c7b89f31484
8149Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8150Date:   Sat Mar 2 11:44:19 2013 -0800
8151
8152    Integer overflows in stringSectionSize() cause buffer overflow in ReadColornameDB() [CVE-2013-1981 6/13]
8153    
8154    LoadColornameDB() calls stringSectionSize() to do a first pass over the
8155    file (which may be provided by the user via XCMSDB environment variable)
8156    to determine how much memory needs to be allocated to read in the file,
8157    then allocates the returned sizes and calls ReadColornameDB() to load the
8158    data from the file into that newly allocated memory.
8159    
8160    If stringSectionSize() overflows the signed ints used to calculate the
8161    file size (say if you have an xcmsdb with ~4 billion lines in or a
8162    combined string length of ~4 gig - which while it may have been
8163    inconceivable when Xlib was written, is quite possible today), then
8164    LoadColornameDB() may allocate a memory buffer much smaller than the
8165    amount of data ReadColornameDB() will write to it.
8166    
8167    The total size is left limited to an int, because if your xcmsdb file
8168    is larger than 2gb, you're doing it wrong.
8169    
8170    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8171    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8172    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8173
8174commit b9ba832401734e1cbd30a930c0d11d850293f3f9
8175Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8176Date:   Sat Mar 2 11:25:25 2013 -0800
8177
8178    unvalidated length in _XimXGetReadData() [CVE-2013-1997 12/15]
8179    
8180    Check the provided buffer size against the amount of data we're going to
8181    write into it, not against the reported length from the ClientMessage.
8182    
8183    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8184    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8185    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8186
8187commit de2e6c322c4aca22856b380f67f8e488e7510015
8188Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8189Date:   Sat Mar 2 11:11:08 2013 -0800
8190
8191    unvalidated index/length in _XkbReadGetNamesReply() [CVE-2013-1997 11/15]
8192    
8193    If the X server returns key name indexes outside the range of the number
8194    of keys it told us to allocate, out of bounds memory writes could occur.
8195    
8196    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8197    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8198    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8199
8200commit 2df882eeb3a70256170127a746a9ba26376599a1
8201Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8202Date:   Sat Mar 2 11:01:04 2013 -0800
8203
8204    unvalidated index in _XkbReadVirtualModMap() [CVE-2013-1997 10/15]
8205    
8206    If the X server returns modifier map indexes outside the range of the number
8207    of keys it told us to allocate, out of bounds memory writes could occur.
8208    
8209    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8210    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8211    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8212
8213commit 4d7c422a37eb9617fb22f8e37527c2b34b105665
8214Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8215Date:   Sat Mar 2 11:04:44 2013 -0800
8216
8217    unvalidated index in _XkbReadExplicitComponents() [CVE-2013-1997 9/15]
8218    
8219    If the X server returns key indexes outside the range of the number of
8220    keys it told us to allocate, out of bounds memory writes could occur.
8221    
8222    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8223    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8224    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8225
8226commit e56a2ada719c5cfac5ed61a52a80ade86c0f5957
8227Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8228Date:   Sat Mar 2 10:51:51 2013 -0800
8229
8230    unvalidated index in _XkbReadModifierMap() [CVE-2013-1997 8/15]
8231    
8232    If the X server returns modifier map indexes outside the range of the number
8233    of keys it told us to allocate, out of bounds memory writes could occur.
8234    
8235    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8236    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8237    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8238
8239commit 06c086e8a1d8374ea9a95ff989f053c96bb1bdca
8240Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8241Date:   Sat Mar 2 10:39:21 2013 -0800
8242
8243    unvalidated index in _XkbReadKeyBehaviors() [CVE-2013-1997 7/15]
8244    
8245    If the X server returns key behavior indexes outside the range of the number
8246    of keys it told us to allocate, out of bounds memory writes could occur.
8247    
8248    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8249    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8250    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8251
8252commit 00626c3830b869259098985afa38933d77ccec72
8253Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8254Date:   Sat Mar 2 09:40:22 2013 -0800
8255
8256    unvalidated index in _XkbReadKeyActions() [CVE-2013-1997 6/15]
8257    
8258    If the X server returns key action indexes outside the range of the number
8259    of keys it told us to allocate, out of bounds memory access could occur.
8260    
8261    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8262    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8263    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8264
8265commit fd7d4956bc7a1c4b5c38661b12777ebee4d685d9
8266Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8267Date:   Sat Mar 2 09:28:33 2013 -0800
8268
8269    unvalidated index in _XkbReadKeySyms() [CVE-2013-1997 5/15]
8270    
8271    If the X server returns keymap indexes outside the range of the number of
8272    keys it told us to allocate, out of bounds memory access could occur.
8273    
8274    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8275    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8276    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8277
8278commit 59ae16a00d18588e98af57d26e442af8ea42b7aa
8279Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8280Date:   Sat Mar 2 09:18:26 2013 -0800
8281
8282    unvalidated indexes in _XkbReadGetGeometryReply() [CVE-2013-1997 4/15]
8283    
8284    If the X server returns color indexes outside the range of the number of
8285    colors it told us to allocate, out of bounds memory access could occur.
8286    
8287    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8288    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8289    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8290
8291commit bff938b9fe1629cbacb726509edfa2a3840b7207
8292Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8293Date:   Sat Mar 2 09:12:47 2013 -0800
8294
8295    unvalidated indexes in _XkbReadGeomShapes() [CVE-2013-1997 3/15]
8296    
8297    If the X server returns shape indexes outside the range of the number
8298    of shapes it told us to allocate, out of bounds memory access could occur.
8299    
8300    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8301    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8302    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8303
8304commit f293659d5a4024bda386305bb7ebeb4647c40934
8305Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8306Date:   Fri Mar 1 22:49:01 2013 -0800
8307
8308    unvalidated index in _XkbReadGetDeviceInfoReply() [CVE-2013-1997 2/15]
8309    
8310    If the X server returns more buttons than are allocated in the XKB
8311    device info structures, out of bounds writes could occur.
8312    
8313    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8314    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8315    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8316
8317commit cddc4e7e3cb4b9b7ad25f8591971a86901c249f2
8318Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8319Date:   Fri Mar 1 19:30:09 2013 -0800
8320
8321    unvalidated lengths in XAllocColorCells() [CVE-2013-1997 1/15]
8322    
8323    If a broken server returned larger than requested values for nPixels or
8324    nMasks, XAllocColorCells would happily overflow the buffers provided by
8325    the caller to write the results into.
8326    
8327    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8328    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8329    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8330
8331commit 2cd62b5eb99ffbb2fce99f3c459455e630b35bf7
8332Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8333Date:   Fri Mar 1 22:49:01 2013 -0800
8334
8335    integer overflow in XListHosts() [CVE-2013-1981 5/13]
8336    
8337    If the reported number of host entries is too large, the calculations
8338    to allocate memory for them may overflow, leaving us writing beyond the
8339    bounds of the allocation.
8340    
8341    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8342    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8343    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8344
8345commit 1f6a3dbf699b85c0ea715ef21de7e7095a714e12
8346Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8347Date:   Fri Mar 1 22:49:01 2013 -0800
8348
8349    integer overflow in XGetMotionEvents() [CVE-2013-1981 4/13]
8350    
8351    If the reported number of motion events is too large, the calculations
8352    to allocate memory for them may overflow, leaving us writing beyond the
8353    bounds of the allocation.
8354    
8355    v2: Ensure nEvents is set to 0 when returning NULL events pointer
8356    
8357    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8358    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8359
8360commit 39515b7c3ba8cae9021bf6695e378ae19487082f
8361Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8362Date:   Fri Mar 1 22:49:01 2013 -0800
8363
8364    integer overflow in XListFontsWithInfo() [CVE-2013-1981 3/13]
8365    
8366    If the reported number of remaining fonts is too large, the calculations
8367    to allocate memory for them may overflow, leaving us writing beyond the
8368    bounds of the allocation.
8369    
8370    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
8371    v3: On error cases, also set values to be returned in pointer args to 0/NULL
8372    
8373    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8374    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8375
8376commit 5669a220816b7d58fcaf0c302ead16fbe5c87817
8377Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8378Date:   Fri Mar 1 21:05:27 2013 -0800
8379
8380    integer overflow in _XF86BigfontQueryFont() [CVE-2013-1981 2/13]
8381    
8382    Similar to _XQueryFont, but with more ways to go wrong and overflow.
8383    Only compiled if libX11 is built with XF86BigFont support.
8384    
8385    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
8386    
8387    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8388    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8389
8390commit 6df8a63d34b7514077188e2062a13774f920c085
8391Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8392Date:   Fri Mar 1 21:05:27 2013 -0800
8393
8394    integer overflow in _XQueryFont() on 32-bit platforms [CVE-2013-1981 1/13]
8395    
8396    If the CARD32 reply.nCharInfos * sizeof(XCharStruct) overflows an
8397    unsigned long, then too small of a buffer will be allocated for the
8398    data copied in from the reply.
8399    
8400    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
8401    
8402    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
8403    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8404
8405commit 9f5d83706543696fc944c1835a403938c06f2cc5
8406Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8407Date:   Fri Mar 1 20:54:24 2013 -0800
8408
8409    Add _XEatDataWords to discard a given number of 32-bit words of reply data
8410    
8411    Matches the units of the length field in X protocol replies, and provides
8412    a single implementation of overflow checking to avoid having to replicate
8413    those checks in every caller.
8414    
8415    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8416    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8417
8418commit d7f04c340ade3834e603c23d543132e1ee4e0c63
8419Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8420Date:   Sat Mar 2 13:03:55 2013 -0800
8421
8422    Move repeated #ifdef magic to find PATH_MAX into a common header
8423    
8424    Lets stop duplicating the mess all over
8425    
8426    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8427    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
8428
8429commit f3a553a4e4a55d9d19deda1ea01883e1d5d682b1
8430Author: Pander <pander@users.sourceforge.net>
8431Date:   Tue May 7 18:38:14 2013 -0400
8432
8433    Add compose sequences for J́ and j́.
8434    
8435    The resulting sequences are:
8436    
8437      U+004A LATIN CAPITAL LETTER J  U+0301 COMBINING ACUTE ACCENT
8438      U+006A LATIN SMALL LETTER J    U+0301 COMBINING ACUTE ACCENT
8439    
8440    Used in Dutch, per:
8441    
8442      http://lists.x.org/archives/xorg-devel/2013-February/035514.html
8443      https://nl.wikipedia.org/wiki/Accenttekens_in_de_Nederlandse_spelling
8444    
8445    Signed-off-by: Pander <pander@users.sourceforge.net>
8446    Signed-off-by: James Cloos <cloos@jhcloos.com>
8447
8448commit f49bb2dd6d4ea45c55bd21acc0efe2b764441020
8449Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8450Date:   Sat Mar 16 18:30:56 2013 -0700
8451
8452    Move big request comment in XOpenDisplay to the right place
8453    
8454    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8455
8456commit 3996543c1b2919e97d61a5d70fe1ebd7cd76fc83
8457Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8458Date:   Sat Mar 9 19:16:03 2013 -0800
8459
8460    libX11 1.5.99.901 (1.6 RC1)
8461    
8462    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8463
8464commit f9cd175a471116a616e681fb0ca1a61b3d84a6a0
8465Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8466Date:   Fri Mar 8 22:33:28 2013 -0800
8467
8468    Fix very weird indenting in src/GetFProp.c
8469    
8470    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8471
8472commit e9bd757630368afb374c5d1bcc5d4d85ad3c6c4c
8473Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8474Date:   Fri Mar 8 15:37:33 2013 -0800
8475
8476    XAllocClassHint: Assume calloc sets pointers in allocated memory to NULL
8477    
8478    While the C standard technically allows for the compiler to translate
8479    pointer = 0 or pointer = NULL into something other than filling the
8480    pointer address with 0 bytes, the rest of the Xlib code already assumes
8481    that calloc initializes any pointers in the struct to NULL, and there
8482    are no known systems supported by X.Org where this is not true.
8483    
8484    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8485
8486commit 39547d600a13713e15429f49768e54c3173c828d
8487Author: Karl Tomlinson <xmail@karlt.net>
8488Date:   Mon Feb 18 01:25:34 2013 +0000
8489
8490    MakeBigReq: don't move the last word, already handled by Data32
8491    
8492    MakeBigReq inserts a length field after the first 4 bytes of the request
8493    (after req->length), pushing everything else back by 4 bytes.
8494    
8495    The current memmove moves everything but the first 4 bytes back.
8496    If a request aligns to the end of the buffer pointer when MakeBigReq is
8497    invoked for that request, this runs over the buffer.
8498    Instead, we need to memmove minus the first 4 bytes (which aren't moved),
8499    minus the last 4 bytes (so we still align to the previous tail).
8500    
8501    The 4 bytes that fell out are already handled with Data32, which will
8502    handle the buffermax correctly.
8503    
8504    The case where req->length = 1 was already not functional.
8505    
8506    Reported by Abhishek Arya <inferno@chromium.org>.
8507    
8508    https://bugzilla.mozilla.org/show_bug.cgi?id=803762
8509    
8510    Reviewed-by: Jeff Muizelaar <jmuizelaar@mozilla.com>
8511    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8512    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8513
8514Notes:
8515    CVE-2013-7439
8516
8517commit 3cdb6c3a1646f670afa03d424ec12ac418181d1e
8518Author: Quentin Glidic <sardemff7+git@sardemff7.net>
8519Date:   Tue Jan 15 21:07:17 2013 +0000
8520
8521    nls/Makefile.am: Use LOG_COMPILER
8522    
8523    TESTS_ENVIRONMENT is deprecated
8524    
8525    Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
8526    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8527
8528commit df66d7a98e2bc7f44fb5583b645df87d525f07f1
8529Author: Quentin Glidic <sardemff7+git@sardemff7.net>
8530Date:   Tue Jan 15 21:07:16 2013 +0000
8531
8532    nls/Makefile.am: Remove unneeded $(srcdir)
8533    
8534    Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
8535    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
8536    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8537
8538commit 3facbe5c0df1b5597571b7b00d5f7bdbc92fb278
8539Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8540Date:   Sat Mar 2 12:01:39 2013 -0800
8541
8542    Add <X11/Xresource.h> hint to all Xrm* man pages
8543    
8544    Help users figure out which header file they need to #include
8545    
8546    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8547
8548commit 466404007f2c8f7166e4faddfea1454c5bfe1e9a
8549Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8550Date:   Fri Mar 8 17:13:09 2013 -0800
8551
8552    _xudc_code_to_glyph: check for NULL pointer *before* writing to it, not after
8553    
8554    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8555
8556commit 47d2eff64d167b3245d346d7f38ca703be645e26
8557Author: Ken Moffat <ken@linuxfromscratch.org>
8558Date:   Thu Mar 7 18:27:54 2013 -0500
8559
8560    dead_double_grave and dead_inverted_breve should only have one underscore.
8561    
8562    Correct instances of dead_double_grave and dead_inverted_breve to
8563    dead_doublegrave and dead_invertedbreve.
8564    
8565    Signed-off-by: Ken Moffat <ken@linuxfromscratch.org>
8566    Signed-off-by: James Cloos <cloos@jhcloos.com>
8567
8568commit c23d61d1b84dca3740bf4786978c7908d0065fb9
8569Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8570Date:   Fri Mar 1 18:10:27 2013 -0800
8571
8572    Assume size_t is always available, since it was defined in C89
8573    
8574    Don't provide a fallback definition #ifdef X_NOT_POSIX anymore.
8575    We already use size_t throughout the rest of Xlib, just had this
8576    one instance left in XKBGAlloc.c of a fallback definition.
8577    
8578    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8579
8580commit 9bcfd84aa1410387bc8cf002a5f90f44705aa0d1
8581Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8582Date:   Fri Mar 1 18:09:07 2013 -0800
8583
8584    unifdef XKB_IN_SERVER
8585    
8586    Leftovers from XKB files that were previously shared between the client
8587    and server code, but aren't any more.
8588    
8589    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8590    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8591
8592commit 769a0efa2298040fe8316a89fc9e75fb61e288e5
8593Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8594Date:   Thu Feb 28 20:04:25 2013 -0800
8595
8596    unifdef CRAY & _CRAY
8597    
8598    (mostly performed with unifdef, followed by some manual cleanup of
8599     the remaining code)
8600    
8601    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8602    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8603
8604commit ca106eb03e5f5468df8033300c5caae3d3c6936b
8605Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8606Date:   Thu Feb 28 20:04:25 2013 -0800
8607
8608    unifdef WORD64
8609    
8610    WORD64 seems to have only been defined in <X11/Xmd.h> when building for
8611    CRAY, to handle int being a 64-bit value (ILP64, not LP64) and having
8612    64-bit alignment requirements.
8613    
8614    It hadn't been fully supported even before autotooling, as can be
8615    seen by removed code such as:
8616    
8617     #ifdef WORD64
8618     _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8
8619     #endif
8620    
8621    (mostly performed with unifdef, followed by some manual cleanup of
8622     the remaining code)
8623    
8624    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8625    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8626
8627commit 9399caf2c12cbe1ed56f4f6b368c5811cb5d0458
8628Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8629Date:   Thu Feb 28 20:04:25 2013 -0800
8630
8631    unifdef MUSTCOPY
8632    
8633    MUSTCOPY seems to have only been defined in <X11/Xmd.h> when building for
8634    CRAY, to handle missing some sizes of integer type.
8635    
8636    (mostly performed with unifdef, followed by some manual cleanup of
8637     spacing/indenting in the remaining code)
8638    
8639    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8640    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8641
8642commit b687440c28c7da6ee0ae44514d20248db5161606
8643Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8644Date:   Sat Feb 16 10:42:23 2013 -0800
8645
8646    Convert more sprintf calls to snprintf
8647    
8648    You could analyze most of these and quickly recognize that there was no
8649    chance of buffer overflow already, but why make everyone spend time doing
8650    that when we can just make it obviously safe?
8651    
8652    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8653
8654commit b092dabbd712d7b656abcf572d253b9b206c0237
8655Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8656Date:   Fri Feb 15 23:43:12 2013 -0800
8657
8658    XKeysymToString: move variable declarations to the scope of their usage
8659    
8660    Makes it easier for readers to understand scope of variable usage, and
8661    clears up gcc warning:
8662    
8663    KeysymStr.c: In function 'XKeysymToString':
8664    KeysymStr.c:128:13: warning: declaration of 'i' shadows a previous local [-Wshadow]
8665    KeysymStr.c:73:18: warning: shadowed declaration is here [-Wshadow]
8666    
8667    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8668
8669commit f0b171c8ea7b055ba520272ea9a2604e18841ac7
8670Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8671Date:   Fri Feb 15 23:34:40 2013 -0800
8672
8673    Preserve constness in casting arguments through the Data*() routines
8674    
8675    Casts were annoying gcc by dropping constness when changing types,
8676    when routines simply either copy data into the request buffer or
8677    send it directly to the X server, and never modify the input.
8678    
8679    Fixes gcc warnings including:
8680    ChProp.c: In function 'XChangeProperty':
8681    ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8682    ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8683    ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8684    ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8685    ChProp.c:83:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8686    SetHints.c: In function 'XSetStandardProperties':
8687    SetHints.c:262:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8688    SetPntMap.c: In function 'XSetPointerMapping':
8689    SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8690    SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8691    StBytes.c: In function 'XStoreBuffer':
8692    StBytes.c:97:33: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8693    StName.c: In function 'XStoreName':
8694    StName.c:40:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8695    StName.c: In function 'XSetIconName':
8696    StName.c:51:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8697    
8698    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8699
8700commit 6c558ee357292dd9dfc6d9006f4525f625327c52
8701Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8702Date:   Fri Feb 15 22:58:54 2013 -0800
8703
8704    Fix comment typo & confusing indentation levels in Data() macro definition
8705    
8706    The final } matches the one on the #define line, not one that doesn't
8707    appear after the else statement it was lined up with
8708    
8709    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8710
8711commit afd6593da90e51234d59f8921c411317f91ab48b
8712Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8713Date:   Fri Feb 15 23:25:38 2013 -0800
8714
8715    XStringToKeysym: preserve constness when casting off unsignedness for strcmp
8716    
8717    Fixes gcc warning:
8718    StrKeysym.c:97:17: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8719    
8720    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8721
8722commit 7e3bf4dd83fec22bd568146de75e6d59eff74e21
8723Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8724Date:   Fri Feb 15 23:14:40 2013 -0800
8725
8726    XRebindKeysym: Drop unnecessary const-removing cast
8727    
8728    C89 defines memcpy as taking a const void *, so casting from
8729    const unsigned char * to char * simply angers gcc for no benefit:
8730    
8731    KeyBind.c:1017:24: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8732    
8733    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8734
8735commit 54527eab93d46055cf11eb6c18abb353a03ae544
8736Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8737Date:   Fri Feb 15 22:45:19 2013 -0800
8738
8739    cmsColNm.c: maintain constness of arguments to qsort helper function
8740    
8741    Fixes gcc warning:
8742    
8743    cmsColNm.c: In function 'FirstCmp':
8744    cmsColNm.c:257:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8745    cmsColNm.c:257:45: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8746    
8747    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8748
8749commit deedeada53676ee529d700bf96fde0b29a3a1def
8750Author: Nickolai Zeldovich <nickolai@csail.mit.edu>
8751Date:   Tue Jan 22 10:03:00 2013 -0500
8752
8753    XListFontsWithInfo: avoid accessing realloc'ed memory
8754    
8755    If exactly one of the two reallocs in XListFontsWithInfo() fails, the
8756    subsequent code accesses memory freed by the other realloc.
8757    
8758    Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
8759    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8760    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8761
8762commit f57fd760cba92ad846917f21e94e73e9c846185f
8763Author: Colin Walters <walters@verbum.org>
8764Date:   Wed Jan 4 17:37:06 2012 -0500
8765
8766    autogen.sh: Implement GNOME Build API
8767    
8768    http://people.gnome.org/~walters/docs/build-api.txt
8769    
8770    Signed-off-by: Adam Jackson <ajax@redhat.com>
8771
8772commit 51c102d39e855cf1704d9eb3afba76a2e73c6b81
8773Author: Adam Jackson <ajax@redhat.com>
8774Date:   Tue Jan 15 14:28:48 2013 -0500
8775
8776    configure: Remove AM_MAINTAINER_MODE
8777    
8778    Signed-off-by: Adam Jackson <ajax@redhat.com>
8779
8780commit 3cd974b1d4d1fa6389d3695fa9fcc0c22a51d50c
8781Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8782Date:   Wed Dec 26 22:57:39 2012 -0800
8783
8784    Remove unused DECnet ("DNETCONN") code from Xlib
8785    
8786    Has never been converted to build in modular builds, so has been unusable
8787    since X11R7.0 release in 2005.  DNETCONN support was removed from xtrans
8788    back in 2008.
8789    
8790    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8791
8792commit a6e5b36a3e6d4a7a9fb4bad905ed127e67b1957e
8793Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8794Date:   Wed Dec 26 22:56:38 2012 -0800
8795
8796    Remove unused TLI ("STREAMSCONN") code from Xlib
8797    
8798    Has never been converted to build in modular builds, so has been unusable
8799    since X11R7.0 release in 2005.  All known platforms with TLI/XTI support
8800    that X11R7 & later releases run on also have (and mostly prefer) BSD
8801    socket support for their networking API.
8802    
8803    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8804
8805commit 59c9ee8cd58857c5769b643611cbe526005a9e45
8806Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8807Date:   Sun Dec 16 17:44:42 2012 -0800
8808
8809    Tell clang not to report -Wpadded warnings on public headers we can't fix
8810    
8811    Better to silence the compiler warning than break ABI.
8812    
8813    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8814
8815commit 0b148750027fd0557c5ed93afda861ddf4b92e0f
8816Author: Jon TURNEY <jon.turney@dronecode.org.uk>
8817Date:   Mon Nov 12 17:27:52 2012 +0000
8818
8819    Fix config check for loadable modules
8820    
8821    The config check of the results of testing for dlfcn.h or dl.h just tests the
8822    value of the ac_cv_ variables, which will be 'yes' or 'no', rather than checking
8823    it is 'yes', so loadable module support would always be detected.
8824    
8825    This is necessary for successful compilation for the MinGW target without the
8826    optional dlfcn-win32 library.
8827    
8828    v2: Also, fixed typoed name of ac_cv_header_dlfcn_h, so check still works
8829    correctly when dlfcn.h is available
8830    
8831    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8832    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
8833
8834commit cd25cab4b5b957641183ce72dd1ae0424aff1663
8835Author: Egbert Eich <eich@freedesktop.org>
8836Date:   Fri May 20 18:27:02 2011 +0200
8837
8838    Install Xcms.txt in $(datadir) rather than $(libdir).
8839    
8840    This file is an architecture independent data and should be where
8841    other databases are.
8842    This is the Xlib provided sample file, applications are free to
8843    use a different one specifying its location in the XCMSDB env
8844    variable.
8845    
8846    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8847    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8848
8849commit 51fef7e273b5c6256e4c9604e3e1afe5dc8f6a1a
8850Author: Egbert Eich <eich@freedesktop.org>
8851Date:   Fri May 20 18:25:24 2011 +0200
8852
8853    Don't hard code path to Xcms.txt file.
8854    
8855    The path to this file is configurable at build time. The source
8856    however contains a hard coded path.
8857    
8858    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8859    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8860
8861commit 9833489e6c3829a1e835bc0a11f028fc180809e4
8862Author: Colin Walters <walters@verbum.org>
8863Date:   Fri Dec 7 08:51:21 2012 -0500
8864
8865    Fix build after dropping locales
8866    
8867    They also needed to be removed from configure.ac
8868    
8869    Signed-off-by: Colin Walters <walters@verbum.org>
8870    Signed-off-by: Julien Cristau <jcristau@debian.org>
8871
8872commit 0a740a574aaf0c0eec78859b773a532cff3b74c3
8873Author: Egbert Eich <eich@freedesktop.org>
8874Date:   Fri May 20 13:04:11 2011 +0200
8875
8876    i18n: Uppercased all occurances if 'iso8859' in the full locale name.
8877    
8878    Making all occurances of iso8859 upper case in the full local name
8879    makes the alias entries more consistent and match the entries on
8880    locale.dir.
8881    
8882    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8883    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8884
8885commit 5e7d589697755a70fb22d85c6a1ae82b39843e53
8886Author: Egbert Eich <eich@freedesktop.org>
8887Date:   Fri May 20 17:55:49 2011 +0200
8888
8889    i18n: Remove ja.S90 and ja.U90 locales.
8890    
8891    Both locales carry a copyright notice and a prorietary statement:
8892    
8893    Copyright 1995 by FUJITSU LIMITED
8894    This is source code modified by FUJITSU LIMITED under the Joint
8895    Development Agreement for the CDEnext PST.
8896    This is unpublished proprietary source code of FUJITSU LIMITED
8897    
8898    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8899    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8900
8901commit 99eae3dcb7bf6764e3b243d2a2934a4d1ecce90e
8902Author: Egbert Eich <eich@freedesktop.org>
8903Date:   Fri May 20 16:46:15 2011 +0200
8904
8905    i18n: Bring locale.dir and compose.dir in sync.
8906    
8907    Some entries for locale/encoding combinations were missing from
8908    either file or just misspelled, some entries were wrong or just
8909    aliases.
8910    
8911    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8912    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8913
8914commit d1e6baa4e290b758e430077cb74e7c03ad850771
8915Author: Egbert Eich <eich@freedesktop.org>
8916Date:   Fri May 20 14:02:39 2011 +0200
8917
8918    i18n: Treat 'a3_AZ' as an alias for 'az_AZ'.
8919    
8920    locale.alias contains a comment:
8921    XCOMM a3 is not an ISO 639 language code, but in Cyrillic, "Z" looks like "3".
8922    Thus lets treat 'a3' as an alias for 'az'.
8923    
8924    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8925    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8926    Reviewed-by: Magnus Kessler <Magnus.Kessler@gmx.net>
8927
8928commit fa2aab0bea18efa26a56977d3166277582ab7b07
8929Author: Egbert Eich <eich@freedesktop.org>
8930Date:   Fri May 20 13:09:38 2011 +0200
8931
8932    i18n: Fixed typos in full locale names.
8933    
8934    Fixing those typos those names actually match entries in
8935    locale.dir.
8936    
8937    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8938    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8939
8940commit b88dd95005dce4c40f9b4d5f938f945e7955fd04
8941Author: Egbert Eich <eich@freedesktop.org>
8942Date:   Sat May 21 07:29:38 2011 +0200
8943
8944    i18n: Remove duplicates and aliases to oneself.
8945    
8946    Aliases to itself don't make much sense.
8947    This changes occurances of:
8948    xy_UV.UTF-8:                    xy_UV.UTF-8
8949    to
8950    xy_UV:                  xy_UV.UTF-8
8951    where applicable.
8952    
8953    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8954    Reviewed-by: James Cloos <cloos@jhcloos.com>
8955
8956commit f198c6aa98f88ff285d903175a3c4c0fd33a4575
8957Author: Jens Herden <jens.herden@email.de>
8958Date:   Fri May 20 17:29:03 2011 +0200
8959
8960    i18n: Add support for Khmer locale and compose table.
8961    
8962    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8963    Reviewed-by: James Cloos <cloos@jhcloos.com>
8964
8965commit 40761898692e5063957bfa2518cca3d35b2e354a
8966Author: Kalman Kemenczy <kkemenczy@novell.com>
8967Date:   Fri May 20 17:03:24 2011 +0200
8968
8969    i18n: Add support for Serbian specific compose table entries.
8970    
8971    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8972    Reviewed-by: James Cloos <cloos@jhcloos.com>
8973
8974commit 7c14aacc9f01d7a975f8d9d033b9b13cbd777a61
8975Author: Egbert Eich <eich@freedesktop.org>
8976Date:   Fri May 20 14:00:35 2011 +0200
8977
8978    i18n: Adding and removing comments.
8979    
8980    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8981    Reviewed-by: James Cloos <cloos@jhcloos.com>
8982
8983commit 7754d68976106183751243c2c35a84134be17b34
8984Author: Egbert Eich <eich@freedesktop.org>
8985Date:   Fri May 20 13:45:31 2011 +0200
8986
8987    i18n: Add missing locales which existed in locale.alias.
8988    
8989    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8990    Reviewed-by: James Cloos <cloos@jhcloos.com>
8991
8992commit 952eccd0d25ba66023acfd31873eee2e71c38c42
8993Author: Egbert Eich <eich@freedesktop.org>
8994Date:   Fri May 20 10:57:57 2011 +0200
8995
8996    i18n: Consolidate compose handling for locales with UTF-8 encoding.
8997    
8998    - add an entry to include the default en_US compose file
8999      for the ja_JP, ko_KR, th_TH, zh_CN, zh_HK and zh_TW locales.
9000    - add missing entries for zh_CN. and zh_HK and am_ET.
9001    - change entries for the UTF-8 encoding for ru_RU, ja_JP,
9002      ko_KR, th_TH and zh_TW to point to their native directory
9003      entries.
9004    
9005    Signed-off-by: Egbert Eich <eich@freedesktop.org>
9006    Reviewed-by: James Cloos <cloos@jhcloos.com>
9007
9008commit d14b6a250f004fa405179db7020f6953001d17b9
9009Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
9010Date:   Mon Oct 22 13:54:11 2012 -0500
9011
9012    XIM: remove Private and Public macros
9013    
9014    Private is a struct member name in mingw-w64 <winioctl.h>, causing this
9015    useless define in a private header to break the build.
9016    
9017    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
9018    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9019
9020commit f2a8def423a46d52e834cf7ea49fa0079427663a
9021Author: Adam Jackson <ajax@redhat.com>
9022Date:   Wed Oct 17 14:40:43 2012 -0400
9023
9024    XErrorDB: Add GLXBadProfileARB
9025    
9026    Signed-off-by: Adam Jackson <ajax@redhat.com>
9027
9028commit d45b3fc19fbe95c41afc4e51d768df6d42332010
9029Author: Ross Burton <ross.burton@intel.com>
9030Date:   Wed Sep 12 14:39:40 2012 +0100
9031
9032    Allow overriding location of keysymdef.h
9033    
9034    Currently keysymdef.h is found by using the includedir of xproto.  This doesn't
9035    work when cross-compiling with a sysroot as that ends up being /usr/include/X11,
9036    not a path into the cross-build environment.
9037    
9038    So, add an option to allow explicitly specifying the location of keysymdef.h,
9039    and verify that the specified or found path exists.
9040    
9041    (original patch by Martin Jansa <martin.jansa@gmail.com>, revised by myself)
9042    
9043    Signed-off-by: Ross Burton <ross.burton@intel.com>
9044    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
9045    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
9046
9047commit 44cdc0dc2c68d67654023ec707b807145d3a38c0
9048Author: Eric S. Raymond <esr@thyrsus.com>
9049Date:   Thu Aug 23 19:15:07 2012 -0400
9050
9051    Renove some unnecessary low-level markup.
9052    
9053    Also, SYNTAX -> SYNOPSIS so function prototypes get parsed by doclifter.
9054    This appears to have been somebody's thinko, it's only in a few of the files.
9055    
9056    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9057
9058commit b686600ab5ca93b5750f827786e79c329ab2db4d
9059Author: Eric S. Raymond <esr@thyrsus.com>
9060Date:   Thu Aug 23 10:53:33 2012 -0400
9061
9062    The .NT/.NE macro pair is no longer used.  Remove it.
9063    
9064    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9065
9066commit b83f2898528c5dd683acd2a2143879a760a8dad5
9067Author: Eric S. Raymond <esr@thyrsus.com>
9068Date:   Thu Aug 23 10:42:46 2012 -0400
9069
9070    Remove the one and only use of the .NT/.NE pair.
9071    
9072    The problem with these macros is that they rely on being able to
9073    center the note label.  That doesn't play well with modern HTML,
9074    not anyway without coomplications like CSS.  This use was just a cute
9075    trick, not adding enough value to be kept.
9076
9077commit e9509fa6745d25eee01ec6f1c34edf8a806d66b3
9078Author: Eric S. Raymond <esr@thyrsus.com>
9079Date:   Thu Aug 23 10:23:45 2012 -0400
9080
9081    The .C{ and .C} macros are never used.  Remove them.
9082    
9083    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9084
9085commit 6b2f7ddeea6c87dbec4ddfc19b1fed33f1bb8575
9086Author: Eric S. Raymond <esr@thyrsus.com>
9087Date:   Thu Aug 23 10:08:08 2012 -0400
9088
9089    The .FN macro, paired with .FD, is also never used. Remove it.
9090    
9091    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9092
9093commit cac6572701c976542da562b2b277b07ffb892a6f
9094Author: Eric S. Raymond <esr@thyrsus.com>
9095Date:   Thu Aug 23 10:03:00 2012 -0400
9096
9097    The ".FD" macro is never used.  Remove it.
9098    
9099    It was a temptation to presentation-level klugery and is best gone.
9100    
9101    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9102
9103commit 6e27a828f39f7028bc7f4a8736e7262fca250632
9104Author: Eric S. Raymond <esr@thyrsus.com>
9105Date:   Thu Aug 23 09:54:25 2012 -0400
9106
9107    Clean up, my last commit missed four cases.
9108    
9109    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9110
9111commit a7fb575957ff4d9dd3671994a005ac3be8bb10fe
9112Author: Eric S. Raymond <esr@thyrsus.com>
9113Date:   Thu Aug 23 08:24:17 2012 -0400
9114
9115    Eliminate all uses of tab stops in the libX11 man pages.
9116    
9117    Two steps: First, expand tabs to 8 spaces in code and structure
9118    listings.  Second, make the .Ds used to wrap code listings switch to
9119    constant-width font (CW) rather than numeric font position 1, which
9120    maps to R on most systems.
9121    
9122    It is possible some archaic systems won't know what CW is, but the
9123    only risk is that code listings won't look quite right on troff
9124    devices; the PostScript and DVI drivers definitely grok it, so those
9125    important cases are OK.
9126    
9127    The purpose of these changes is to get rid of presentation-level
9128    markup so these pages will lift clean to DocBook.
9129    
9130    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
9131
9132commit 8042f88ace33573f9d0dfaa21ed54ac7cef266d5
9133Author: Will Thompson <will@willthompson.co.uk>
9134Date:   Mon Jul 9 18:00:27 2012 +0100
9135
9136    Add compose sequences for "therefore" and "because".
9137    
9138    These sequences look sensible to me. I added them to the APL-related
9139    section of Mathematical Operators—they're in that section of Unicode
9140    anyway.
9141    
9142    https://bugs.freedesktop.org/show_bug.cgi?id=51922
9143    
9144    Signed-off-by: Will Thompson <will@willthompson.co.uk>
9145    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
9146
9147commit 65358ea5079236b2508f787ac2fb2024a477e36d
9148Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9149Date:   Fri Jun 29 23:08:04 2012 -0700
9150
9151    Convert XCreate{Pix,Bit}map...Data to use C99 designated initializers
9152    
9153    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9154
9155commit 0dc93f7e43deb102b1f8fb7c4c4844cdce7ffd1e
9156Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9157Date:   Fri Jun 29 22:57:13 2012 -0700
9158
9159    XCreate{Pix,Bit}map...Data: Free pixmap in error path if XCreateGC fails
9160    
9161    Fixes leaks in error paths found by Parfait 1.0.0:
9162    
9163    Error: X Resource Leak
9164       Leaked X Resource pix
9165            at line 62 of CrBFData.c in function 'XCreateBitmapFromData'.
9166              pix initialized at line 60 with XCreatePixmap
9167    Error: X Resource Leak
9168       Leaked X Resource pix
9169            at line 70 of CrPFBData.c in function 'XCreatePixmapFromBitmapData'.
9170              pix initialized at line 66 with XCreatePixmap
9171    
9172    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9173    Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
9174
9175commit dce84b8c39ad5a8908c29bb6de25b6c3004c1ab7
9176Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9177Date:   Wed Jun 6 13:31:16 2012 -0700
9178
9179    libX11 spec: Correct prototype for XConvertSelection
9180    
9181    selection & target parameters were accidentally run together
9182    
9183    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9184
9185commit 6c5cb2a90a6479f56855e5167039c37c234cdfe7
9186Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9187Date:   Fri Jun 1 23:37:09 2012 -0700
9188
9189    libX11 1.5.0
9190    
9191    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9192
9193commit 05c587ec3be880721131a17c1dd4366e458fdd8b
9194Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9195Date:   Sat May 26 14:37:28 2012 -0700
9196
9197    libX11 1.4.99.902 (1.5 RC2)
9198    
9199    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9200
9201commit d817834d9772e3500d5102e2eae868b04a33c31f
9202Author: Pander <pander@users.sourceforge.net>
9203Date:   Sat May 5 19:02:10 2012 +0200
9204
9205    Compose: Reassigned squences with minus and a or o
9206    
9207    Reassigned squences with minus and a or o (vice versa and lower and
9208    upper case) to conform existing series and not resulting in tilde.
9209    Also added noe missing underscore sequence.
9210    
9211    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9212
9213commit 6bfd1dca6e7cb2046ee6bf9dbbddc0af5ef7cc00
9214Author: Pander <pander@users.sourceforge.net>
9215Date:   Sat May 5 16:56:05 2012 +0200
9216
9217    Compose: Removed <slash> <U> and vice versa
9218    
9219    Combination with lower case u suffices
9220    
9221    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9222
9223commit 3a8b1637132d1e36eb5e11f52dfb284081772d14
9224Author: Pander <pander@users.sourceforge.net>
9225Date:   Sat May 5 16:44:31 2012 +0200
9226
9227    Compose: Removed <n> <minus> for n with tilde, also vice versa & for upper case
9228    
9229    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9230
9231commit fd514d7a35119dd4413b51c84f2a536f2ca501a6
9232Author: Pander <pander@users.sourceforge.net>
9233Date:   Sat May 5 16:40:20 2012 +0200
9234
9235    Compose: Removed <S> in combination with a numeral for superscript
9236    
9237    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9238
9239commit 1b5cad3ca54410c4edbca79c23c463e9e088bc0d
9240Author: Pander <pander@users.sourceforge.net>
9241Date:   Sat May 5 16:28:01 2012 +0200
9242
9243    Compose: Reassigned <o> <apostrophe> to oacute, also for upper case.
9244    
9245    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9246
9247commit 0bbc0d5e605e2a4a3eb00a229b651d4546e8aef4
9248Author: James Cloos <cloos@jhcloos.com>
9249Date:   Thu May 10 12:27:28 2012 -0400
9250
9251    Remove duplicate compose sequences from commit e51e37c118ae.
9252    
9253    The compose-check script only handles compiled Compose files, not
9254    the Compose.pre files.  One must remember to use:
9255    
9256        ./autogen.sh; make; make check
9257    
9258    when reviewing patches to the Compose.pre files....
9259    
9260    Signed-off-by: James Cloos <cloos@jhcloos.com>
9261
9262commit e51e37c118ae6cb9ced8244ce1c410677e0279ce
9263Author: Geoff Streeter <geoff@dyalog.com>
9264Date:   Thu Mar 22 15:02:00 2012 +0000
9265
9266    Add APL support to compose
9267    
9268    Signed-off-by: Geoff Streeter <geoff@dyalog.com>
9269    Signed-off-by: James Cloos <cloos@jhcloos.com>
9270
9271commit dac90324cee224df977a428afe80d960dceca769
9272Author: Julien Cristau <jcristau@debian.org>
9273Date:   Sat May 5 16:05:07 2012 +0200
9274
9275    configure: make previous change work with older autoconf
9276    
9277    autoconf 2.63 doesn't seem to like the nested AC_CHECK_DECL/FUNC.  So do
9278    the tests separately.
9279    
9280    Reported-by: Dave Airlie
9281    Signed-off-by: Julien Cristau <jcristau@debian.org>
9282
9283commit f5b50af4324186962e258ffe9be78d5ee4681982
9284Author: Julien Cristau <jcristau@debian.org>
9285Date:   Sun Apr 29 16:43:09 2012 +0200
9286
9287    configure: check if issetugid is declared
9288    
9289    GNU/kFreeBSD has issetugid in libc (for legacy apps?), but doesn't
9290    declare it anywhere, causing gcc to error out with
9291    -Werror=implicit-function-declaration.  Use AC_CHECK_DECL in addition to
9292    AC_CHECK_FUNC so we disable this code instead of failing to build it.
9293    
9294    Debian bug#669670 <http://bugs.debian.org/669670>
9295    
9296    Signed-off-by: Julien Cristau <jcristau@debian.org>
9297    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9298    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9299
9300commit 52e1b5cc3b6de76ccf4285b55652474a522ed9a8
9301Author: Peter Hutterer <peter.hutterer@who-t.net>
9302Date:   Mon Apr 30 16:36:47 2012 +1000
9303
9304    Typo fix
9305    
9306    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9307
9308commit b64969f0e510d5d3300cf968741a3726a6409577
9309Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9310Date:   Tue Apr 17 18:12:02 2012 -0700
9311
9312    Add X11R7 sections to the libX11 & XKBlib credits to cover Docbook conversion
9313    
9314    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9315
9316commit 9ea611696f317ac3b3fb67893f1d6d87d49e3b5e
9317Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9318Date:   Tue Apr 17 18:01:36 2012 -0700
9319
9320    Add olinks from libX11 & localedb specs to ICCCM spec
9321    
9322    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9323
9324commit b3c1b8cdab7d14220426c9b997ac362dc16318fc
9325Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9326Date:   Tue Apr 17 17:49:44 2012 -0700
9327
9328    Add olinks from libX11 spec to ICCCM spec
9329    
9330    Also convert ICCCM title mentions from <emphasis> to <citetitle>
9331    
9332    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9333
9334commit ebebb65e753007ad01966dccc90bd6ca9a826488
9335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9336Date:   Sat Apr 14 23:40:01 2012 -0700
9337
9338    libX11 AppC: Fix section headers that didn't translate from nroff properly
9339    
9340    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9341    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
9342
9343commit d5ab4ae0e74ae1fb30fb72add0751effe2759bf2
9344Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9345Date:   Sat Apr 14 23:13:05 2012 -0700
9346
9347    Add olinks from libX11 spec to x11protocol spec
9348    
9349    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9350    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
9351
9352commit 83878a0e34fffd255597300dd3e6cd43fcd645b0
9353Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9354Date:   Sat Apr 14 22:28:53 2012 -0700
9355
9356    libX11 spec: Remove .br nroff macro left behind in XGetWindowProperty prototype
9357    
9358    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9359
9360commit abc523fce31fcf2687229697a8eb656e343ecb0c
9361Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9362Date:   Thu Mar 15 22:14:45 2012 -0700
9363
9364    libX11 1.4.99.901 (1.5 RC1)
9365    
9366    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9367
9368commit b2cc5905a4a6d519957223e8ba2caef71520040c
9369Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9370Date:   Thu Mar 15 22:03:21 2012 -0700
9371
9372    Remove "register" qualifier that annoys Solaris Studio compiler
9373    
9374    Fixes warning:
9375    "Xrm.c", line 1094: warning: storage class after type is obsolescent
9376    
9377    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9378
9379commit 62d42953893f93a98db0504eaf06d650ceaf5811
9380Author: James Cloos <cloos@jhcloos.com>
9381Date:   Wed Mar 14 17:25:46 2012 -0400
9382
9383    Fix the gtk+ additions
9384    
9385    (Some of) the Dstroke and dstroke entries already were present as U011[01],
9386    even though XK_Dstroke and XK_dstroke are part of the latin2 set in keysymdef.h.
9387    
9388    The addition of <Multi_key> <o> <apostrophe> as a postfix version of
9389    <Multi_key> <apostrophe> <o> blocks the existing entries for ǻ and Ǻ.
9390    That prevents its and <Multi_key> <O> <apostrophe>’s addition.
9391    
9392    Signed-off-by: James Cloos <cloos@jhcloos.com>
9393
9394commit 91bcce48d94792f78333d2aea73961cc2e739d2e
9395Author: Pander <pander@users.sourceforge.net>
9396Date:   Wed Mar 14 12:54:53 2012 -0400
9397
9398    Complete compose key sequences for musical symbols
9399    
9400    Signed-off-by: Pander <pander@users.sourceforge.net>
9401    Signed-off-by: James Cloos <cloos@jhcloos.com>
9402
9403commit cf040016d455bc37f7665d6714337c5eafd8ea94
9404Author: Pander <pander@users.sourceforge.net>
9405Date:   Wed Mar 14 12:46:25 2012 -0400
9406
9407    Add compose sequences from gtk+ to X.Org
9408    
9409    Signed-off-by: Pander <pander@users.sourceforge.net>
9410    Signed-off-by: James Cloos <cloos@jhcloos.com>
9411
9412commit a4c591ced5cac9301b9abfa0e521be2d0b267882
9413Author: Keith Packard <keithp@keithp.com>
9414Date:   Sun Mar 4 02:00:13 2012 -0800
9415
9416    Block for other threads in _XUserLockDisplay
9417    
9418    Wait for all other threads to release the user-level lock when
9419    acquiring it. This ensures that only one thread at a time holds the
9420    user-level lock, necessary as it is a nesting lock and a single
9421    variable is used to determine when the lock is nesting and when it is
9422    contended.
9423    
9424    Signed-off-by: Keith Packard <keithp@keithp.com>
9425    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9426
9427commit ed00b460acb08787b695f27b864e96102dfd4867
9428Author: Jon TURNEY <jon.turney@dronecode.org.uk>
9429Date:   Fri Oct 28 11:09:20 2011 -0500
9430
9431    Don't use caddr_t casts
9432    
9433    (caddr_t) isn't used anywhere else in xcb or libX11.
9434    Cast to (char *) for consistency.
9435    
9436    Removing this cast allows building for MinGW without patching.
9437    
9438    v2: Cast to (char *) rather than just dropping the cast
9439    
9440    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9441    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9442
9443commit 20adca02c2a1d0b7c95ecbe387d68f881fd57754
9444Author: Julien Cristau <jcristau@debian.org>
9445Date:   Sun Feb 19 13:27:38 2012 +0100
9446
9447    Revert "xcb: Add TCP fallback"
9448    
9449    This reverts commit f09c5299a381e2729e800a0ac43f1c0e371f65f6.
9450    
9451    The TCP fallback ended up falling back to UNIX socket connection if
9452    $DISPLAY was set to e.g. some.host:0 and the initial attempt failed.
9453    
9454    Debian bug#659558 <http://bugs.debian.org/659558>
9455    
9456    Signed-off-by: Julien Cristau <jcristau@debian.org>
9457    
9458    Conflicts:
9459    
9460            src/OpenDis.c
9461
9462commit f4378193619baa9bb973c1b5b718721bbcbe92c7
9463Author: James Cloos <cloos@jhcloos.com>
9464Date:   Wed Feb 22 14:13:20 2012 -0500
9465
9466    Make the compose sequence for ẞ work.
9467    
9468    There is no XK_Ssharp symbol for U+1E9E LATIN CAPITAL LETTER SHARP S,
9469    so use the U1e9e symbol in the Compose sequence.
9470    
9471    (Compose sequences do not work when the target symbol is unknown.)
9472    
9473    Signed-off-by: James Cloos <cloos@jhcloos.com>
9474
9475commit 2ca641c3a506dcbee97e279b67990d5387389f36
9476Author: Marko Myllynen <myllynen@redhat.com>
9477Date:   Mon Feb 20 17:04:59 2012 +0200
9478
9479    Use ezh/EZH in compose maps
9480    
9481    Related: https://bugs.freedesktop.org/show_bug.cgi?id=19687
9482    
9483    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
9484    Reviewed-by: Matt Dew <marcoz@osource.org>
9485    Signed-off-by: James Cloos <cloos@jhcloos.com>
9486
9487commit 61725822f20f47684a545c1797183ee7075243ac
9488Author: Marko Myllynen <myllynen@redhat.com>
9489Date:   Sun Feb 20 17:09:43 2011 +0200
9490
9491    Provide translation from XK_permille (ad5) to Unicode (U2030)
9492    
9493    https://bugs.freedesktop.org/show_bug.cgi?id=19687
9494    
9495    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
9496    Reviewed-by: Matt Dew <marcoz@osource.org>
9497    Signed-off-by: James Cloos <cloos@jhcloos.com>
9498
9499commit d2cce0abba0fa0143f49026442c8cab5ed721625
9500Author: Jeremy Huddleston <jeremyhu@apple.com>
9501Date:   Sun Feb 12 19:01:43 2012 -0800
9502
9503    nls: Use LC_CTYPE=C for sed magic
9504    
9505    Stricter versions of sed can trip up if the input does not match
9506    LC_CTYPE
9507    
9508    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9509
9510commit f180e043f55531933bedfa6e0ff36a00a9ec07f3
9511Author: Frédéric Boiteux <fboiteux@calistel.com>
9512Date:   Wed Nov 30 12:47:31 2011 +0000
9513
9514    Compositions with the dead greek symbol
9515    
9516    FreeDesktop Bug 21475 <https://bugs.freedesktop.org/show_bug.cgi?id=21475>
9517    
9518    Signed-off-by: Frédéric Boiteux <fboiteux@calistel.com>
9519    Signed-off-by: James Cloos <cloos@jhcloos.com>
9520
9521commit d58e8f8e27790017fcfdeca0843b7318d541c189
9522Author: Ryan Pavlik <rpavlik@iastate.edu>
9523Date:   Sun Jan 1 21:04:52 2012 +0000
9524
9525    Use pthreads on MinGW also
9526    
9527    Use pthreads (provided by the pthreads-win32 compatability library which implements
9528    them using native Win32 threading) on MinGW
9529    
9530    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9531
9532commit bf15ccb6821664746ec23d769d757edf8059007e
9533Author: Ryan Pavlik <rpavlik@iastate.edu>
9534Date:   Mon Jan 23 14:18:02 2012 +0000
9535
9536    Add XWindows.h include to Xxcbint.h
9537    
9538    This avoids some conflicting type re-definition errors which occur if
9539    we attempt to include Windows headers after Xmd.h
9540    
9541    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
9542
9543commit cadcbd376f0a5d17a71a2fe2f8ced5d93232921a
9544Author: Matt Dew <marcoz@osource.org>
9545Date:   Sat Jan 21 17:59:51 2012 -0700
9546
9547    informaltable & table cleanup
9548    
9549    On certain tables, add top and bottom borders to table header
9550    and a bottom border to the table. This matches what those
9551    tables in the old pdfs looked like.
9552    
9553    the <?dbfo keep-together='always'> prevents tables from
9554    splitting across pages. Useful for tiny tables.
9555    
9556    Converting the colwidth to a floating point, IE, 1* -> 1.0*
9557    cleans up these build errors:
9558    WARNING: table-layout="fixed" and column-width unspecified =>
9559    falling back to proportional-column-width(1)
9560    
9561    Signed-off-by: Matt Dew <marcoz@osource.org>
9562
9563commit bb551f654df8f647c867f79252241964521e689e
9564Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9565Date:   Wed Dec 28 21:22:41 2011 -0800
9566
9567    Add more Xkb man pages to the See Also lists for core keyboard functions
9568    
9569    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9570    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9571
9572commit b7bb23fe7c8b8a17128e5bc98a05f68595190438
9573Author: Gaetan Nadon <nadon@memsize.(none)>
9574Date:   Fri Dec 30 17:08:14 2011 -0500
9575
9576    docbook.am: embed css styles inside the HTML HEAD element
9577    
9578    Rather than referring to the external xorg.css stylesheet, embed the content
9579    of the file in the html output produced. This is accomplished by using
9580    version 1.10 of xorg-xhtml.xsl.
9581    
9582    This makes the whole html docs tree much more relocatable.
9583    In addition, it eliminates xorg.css as a runtime file which makes
9584    xorg-sgml-doctools a build time only package.
9585    
9586    Signed-off-by: Gaetan Nadon <nadon@memsize.(none)>
9587
9588commit 70505468b7c4a7068cc39be42e421dcee34ec595
9589Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9590Date:   Tue Dec 13 19:58:30 2011 -0800
9591
9592    makekeys: move buf declaration from global to main to silence gcc -Wshadow
9593    
9594    The global was only referenced in the main() function, which passes it
9595    as an argument of the same name to the parse_line() function, leading
9596    to gcc -Wshadow warnings:
9597    
9598    makekeys.c: In function ‘parse_line’:
9599    makekeys.c:58:24: warning: declaration of ‘buf’ shadows a global declaration
9600    makekeys.c:54:13: warning: shadowed declaration is here
9601    
9602    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9603
9604commit 87e10a7b9a97c951ab4d477f61177779ac0a6a66
9605Author: Kusanagi Kouichi <slash@ac.auone-net.jp>
9606Date:   Wed Dec 14 02:17:55 2011 -0500
9607
9608    XQueryColors: Split a request into multiple requests if necessary
9609    
9610    https://bugs.freedesktop.org/show_bug.cgi?id=9160
9611    
9612    Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
9613    Signed-off-by: James Cloos <cloos@jhcloos.com>
9614
9615commit d3b3570592e9b9e57f270a0bd86762fd205a2833
9616Author: Andreas Wettstein <wettstein509@solnet.ch>
9617Date:   Tue Nov 8 20:18:09 2011 +0000
9618
9619    XIM: Allow modifier releases in compose sequences (#26705)
9620    
9621    Currently, only non-modifier keys (actually, keysyms) can be part of a compose
9622    sequence, and they are matched against the defined compose sequences at the
9623    time the key is pressed.  The patch allows to use modifier keys an well, but
9624    matches them on key release, and only if no other key has been pressed after
9625    the modifier.
9626    
9627    Releasing a non-matched modifier during an ongoing compose sequence only aborts
9628    the sequence if any modifier release would have matched.  In particular, if no
9629    compose sequences with modifiers are specified, the compose mechanism works
9630    exactly as without this patch.
9631    
9632    Even if modifiers are part of a compose sequence, they are not filtered.  This
9633    is because modifiers affect the keyboard state no matter what we do here and,
9634    therefore, filtering them only could confuse clients.
9635    
9636    The purpose is this extension to the compose mechanism is to allow to make
9637    better use of keys in convenient reach for touch typing.
9638    
9639    Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
9640    Signed-off-by: James Cloos <cloos@jhcloos.com>
9641
9642commit 56448a626fc90bcf75a1fa2f4c294b0eb1f23bd6
9643Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9644Date:   Tue Nov 29 23:56:57 2011 -0800
9645
9646    Reject negative string counts in copy_string_list
9647    
9648    Silences parfait warning of a potential memory leak:
9649       Memory leak of pointer 'dst' allocated with malloc(length)
9650            at line 160 of FSWrap.c in function 'copy_string_list'.
9651              'dst' allocated at line 145 with malloc(length).
9652              dst leaks when count <= 0 at line 154.
9653    
9654    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9655
9656commit bf2d7c8c6d70539c72560b1921e18df2610acf29
9657Author: Peter Hutterer <peter.hutterer@who-t.net>
9658Date:   Fri Nov 11 14:56:36 2011 +1000
9659
9660    libX11 1.4.99.1
9661    
9662    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9663
9664commit 24283d40b1e4314c6647dda49d2a159833341a8b
9665Author: Derek Buitenhuis <derek.buitenhuis@gmail.com>
9666Date:   Thu Sep 22 18:44:13 2011 -0400
9667
9668    makekeys: Fix build/target word size mismatch when cross-compiling
9669    
9670    Since makekeys is built using build environment's gcc and
9671    runs natively, we have to make sure that the size of the
9672    Signature type is the same on both the native environment
9673    and the target, otherwise we get mismatches upon running X,
9674    and some LSB test failures (xts5).
9675    
9676    Use an unsigned 32-bit integer on all platforms. Also,
9677    eliminate the redundant multiple typedefs for the
9678    Signature type.
9679    
9680    Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
9681    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
9682
9683commit 1c41f3b9b86b5eeedfa3bff92e519d45aa097587
9684Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9685Date:   Wed Nov 9 21:00:36 2011 -0800
9686
9687    XlcSL.c: convert old-style function definitions to ANSI C89 style
9688    
9689    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9690
9691commit f6dad6aaa384bf836baa28ccb72b476a85c40eff
9692Author: Peter Hutterer <peter.hutterer@who-t.net>
9693Date:   Thu Oct 27 13:53:22 2011 +1000
9694
9695    Use GetReqSized for GetReq and GetReqExtra
9696    
9697    GetEmptyReq and GetResReq cannot do this due to the final typecast -
9698    typically requests that need either of those do not have their own typedef
9699    in the protocol headers.
9700    
9701    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9702    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9703
9704commit c9c99058b9d98789c0b2d7e78a23443c2b57a047
9705Author: Peter Hutterer <peter.hutterer@who-t.net>
9706Date:   Fri Oct 14 14:51:06 2011 +1000
9707
9708    include: Add GetReqSized() for request buffers of specific size
9709    
9710    Some XI2 requests change in size over different versions and libXi would
9711    need to hack around GetReq and GetReqExtra. Add a new GetReqSized so the
9712    library can explicitly specify the size of the request in 4-byte units.
9713    
9714    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9715    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9716
9717commit ba8a7a19165e30d14bc165f43f67c19b6a115585
9718Author: Peter Hutterer <peter.hutterer@who-t.net>
9719Date:   Thu Oct 27 13:24:10 2011 +1000
9720
9721    Switch GetEmptyReq and GetResReq to call _XGetRequest
9722    
9723    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9724    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9725
9726commit 4a060f993bf676cf21ad9784e010f54134da7b40
9727Author: Peter Hutterer <peter.hutterer@who-t.net>
9728Date:   Mon Oct 17 09:45:15 2011 +1000
9729
9730    Add _XGetRequest as substitute for GetReq/GetReqExtra
9731    
9732    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9733    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9734
9735commit d8956520deb79c1cbb5e974c175bf8493859b22b
9736Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9737Date:   Fri Oct 14 17:53:00 2011 -0700
9738
9739    Fix "nomal" -> "normal" typo in several comments
9740    
9741    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9742
9743commit 002b36e308a26a152504f9b40aa08a0dce9a7991
9744Author: Xue Wei <Wei.Xue@Sun.COM>
9745Date:   Fri Oct 14 17:39:21 2011 -0700
9746
9747    mbtocs should not truncate input
9748    
9749    Fixes pasting more than 1024 bytes into xterm, as described in
9750    https://bugs.freedesktop.org/show_bug.cgi?id=25209
9751    
9752    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9753
9754commit 9b8d8c9e5b27273e8856a3851ba9b68022bed3cd
9755Author: Marko Myllynen <myllynen@redhat.com>
9756Date:   Mon Oct 10 09:41:13 2011 +0300
9757
9758    Add new compose sequences
9759    
9760    This patch adds few new compose sequences to the en_US map, ligatures
9761    common in typesetting and sequences already in use in the fi_FI map.
9762    
9763    https://bugs.freedesktop.org/show_bug.cgi?id=30621
9764    https://bugs.freedesktop.org/show_bug.cgi?id=34523
9765    
9766    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9767
9768commit 738f7b867341c1da87ee667d48815822715c3e75
9769Author: Choe Hwanjin <choe.hwanjin@gmail.com>
9770Date:   Thu Oct 13 07:58:02 2011 +0900
9771
9772    XIM: Make Xim handle NEED_SYNC_REPLY flag
9773    
9774    NEED_SYNC_REPLY flag should be in Xim not in Xic.
9775    Because the focused Xic can be changed before sending sync reply.
9776    After focused Xic changed, the new Xic doesn't have NEED_SYNC_REPLY
9777    flag enabled, so libX11 doesn't send XIM_SYNC_REPLY packet.
9778    
9779    This patch adds sync reply flag to Xim and removes sync reply
9780    from Xic.
9781    
9782    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=7869
9783    
9784    Signed-off-by: Choe Hwanjin <choe.hwanjin@gmail.com>
9785    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9786
9787commit 50e15379331c436851abb194251cde96999e67b3
9788Author: Bodo Graumann <mail@bodograumann.de>
9789Date:   Mon Oct 10 16:40:52 2011 +0200
9790
9791    libX11: Fixing modifier key range in Xutil.h (Bug #21910)
9792    
9793    IsModifierKey, defined in include/X11/Xutil.h, is a macro determining,
9794    which keys are regarded as modifiers. The constants ISO_Level5_Shift,
9795    ISO_Level5_Latch and ISO_Level5_Lock where excluded previously, leaving
9796    some Neo2 modifiers functionless in combination with compose.
9797    This patch adjusts the range to include the correct, full range of
9798    modifier constants.
9799    
9800    Neo2 Bug 277 <http://wiki.neo-layout.org/ticket/277>
9801    
9802    X.Org Bug 21910 <http://bugs.freedesktop.org/show_bug.cgi?id=21910>
9803    
9804    Signed-off-by: Bodo Graumann <mail@bodograumann.de>
9805    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
9806
9807commit b5a108624331fabf393223c0891914cc54d4caf4
9808Author: Yann Droneaud <yann@droneaud.fr>
9809Date:   Tue Oct 11 17:27:59 2011 +0200
9810
9811    Return name instead of False in XSetICValues()
9812    
9813    In case of error, XSetICValues() must return the first argument
9814    that failed to be set.
9815    
9816    But in some error paths, it returns False, which is converted to NULL,
9817    so the function returns OK in case of error.
9818    
9819    Signed-off-by: Yann Droneaud <yann@droneaud.fr>
9820    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9821
9822commit 10992cb8dc5e4d938e7e5a633b68a81b5875f3ba
9823Author: Yann Droneaud <yann@droneaud.fr>
9824Date:   Sun Oct 9 17:56:45 2011 +0200
9825
9826    Return name instead of value in XGetIMValues() and XSetIMValues()
9827    
9828    As stated in man page (XOpenIM) and Xlib documentation (chapter 13.5.3),
9829    XGetIMValues() and XSetImValues() "returns the name of the first argument
9830    that could not be obtained."
9831    
9832    But currently,
9833    
9834      err = XGetIMValues(im, "invalid", &arg, NULL);
9835    
9836    returns &arg instead of "invalid".
9837    
9838    This patch fixes https://bugs.freedesktop.org/show_bug.cgi?id=12897
9839    
9840    Signed-off-by: Yann Droneaud <yann@droneaud.fr>
9841    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9842
9843commit f2651e03f3295a453a2965c3749bc8b6e66f1c09
9844Author: Jeremy Huddleston <jeremyhu@apple.com>
9845Date:   Mon Oct 10 14:09:17 2011 -0700
9846
9847    Mark XKeycodeToKeysym as _X_DEPRECATED
9848    
9849    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9850
9851commit 25d7bb9c705623ebbb3afba3b86d0ded5b6bb4fb
9852Author: Jeremy Huddleston <jeremyhu@apple.com>
9853Date:   Sun Oct 9 02:38:28 2011 -0700
9854
9855    Fix nobreakspace for pt_BR.UTF-8
9856    
9857    https://bugs.freedesktop.org/show_bug.cgi?id=31334
9858    
9859    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9860
9861commit 913603660c17ec59a4d39d3e3b9d60469ee5c060
9862Author: Jeremy Huddleston <jeremyhu@apple.com>
9863Date:   Sun Oct 9 02:26:45 2011 -0700
9864
9865    Fix potential uninitialized variable access in _XimMakeICAttrIDList
9866    
9867    Found by clang static analysis
9868    
9869    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9870
9871commit 566ceaf5a92c721ac7155528e4d0d2e5cbef023f
9872Author: Jeremy Huddleston <jeremyhu@apple.com>
9873Date:   Sun Oct 9 02:25:50 2011 -0700
9874
9875    Remove self-resolving aliases
9876    
9877    https://bugs.freedesktop.org/show_bug.cgi?id=30112
9878    
9879    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9880
9881commit 22ba43d198dcca86c88eb15a56fc7d8fc47c422e
9882Author: Matt Dew <marcoz@osource.org>
9883Date:   Fri Oct 7 22:52:30 2011 -0600
9884
9885    Cleanup IDs and links in doc
9886    
9887    1 - fix the capitalization of the ID attributes to match either the
9888         <title> or <funcdef> string it goes with.
9889    2 - fix any <linkend>'s that were affected by 1.
9890    3 - any <function> in the docs that has an actual funcdef,
9891    will become an olink.
9892    
9893    Signed-off-by: Matt Dew <marcoz@osource.org>
9894
9895commit f858f3326adbc0c5711669b92a64a84a9083a055
9896Author: James Cloos <cloos@jhcloos.com>
9897Date:   Tue Oct 4 17:11:11 2011 -0400
9898
9899    [nls] Fix typo/synco.
9900    
9901    The iso8859-1 Compose table includes an optional section which uses
9902    Ctrl<T> as a substitute for <Multi_key>.  In that section the sequence
9903    to generate an @ (at) either was incorrectly copied from the Multi_key
9904    section or was not kept in sync with the Multi_key section.
9905    
9906    Fixing this eliminates the warning from compose-check.pl:
9907    
9908    ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
9909    ┃ Clash with existing sequence in iso8859-1/Compose.pre
9910    ┃  on line 661: Ctrl<T> <A> <A>
9911    ┃    line #661: Ctrl<T> <A> <A>         : "\305"        Aring
9912    ┃    line #480: Ctrl<T> <A> <A>         : "@"   at
9913    ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
9914    
9915    Signed-off-by: James Cloos <cloos@jhcloos.com>
9916
9917commit 7f35f7efc23234b3f8529fee31a802c21846d9ab
9918Author: Jeremy Huddleston <jeremyhu@apple.com>
9919Date:   Mon Sep 26 11:32:56 2011 -0700
9920
9921    Remove conflicting compose sequences for cent and colon
9922    
9923    Regression from 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4
9924    Found by tinderbox
9925    
9926    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9927
9928commit 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4
9929Author: Jeremy Huddleston <jeremyhu@apple.com>
9930Date:   Sun Sep 25 16:29:17 2011 -0700
9931
9932    Add additional compose sequences for pound sterling, yen, and cent (mixed case)
9933    
9934    https://bugs.freedesktop.org/show_bug.cgi?id=1013
9935    
9936    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9937    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
9938
9939commit 0ca4153a373e2af7a9eb3cef003393ab332bc79e
9940Author: Jeremy Huddleston <jeremyhu@apple.com>
9941Date:   Sat Sep 24 16:44:04 2011 -0700
9942
9943    Use a configure check for seteuid
9944    
9945    HP-UX doesn't have seteuid
9946    
9947    https://bugs.freedesktop.org/show_bug.cgi?id=1497
9948    
9949    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9950
9951commit 79594b4d66344f248eb4314ecb78eb81f632c3ab
9952Author: Gaetan Nadon <memsize@videotron.ca>
9953Date:   Tue Sep 20 16:06:10 2011 -0400
9954
9955    localedb specs: use <copyright> for first holder of multi license
9956    
9957    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9958
9959commit 154430268cf004b06920edde2f1812dc16d71fae
9960Author: Gaetan Nadon <memsize@videotron.ca>
9961Date:   Tue Sep 20 16:01:52 2011 -0400
9962
9963    libX11 specs: use <copyright> for first holder of multi license
9964    
9965    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9966
9967commit 66e621f58dee804acb795ff2c79ce6dd38b34a3f
9968Author: Tollef Fog Heen <tfheen@err.no>
9969Date:   Tue Sep 20 13:07:37 2011 -0400
9970
9971    NLS: Add more vulgar fractions
9972    
9973    Add 1/7, 1/9, 1/10 and 0/3 vulgar fractions.
9974    
9975    Signed-off-by: Tollef Fog Heen <tfheen@err.no>
9976    Signed-off-by: James Cloos <cloos@jhcloos.com>
9977
9978commit e37a6da814b5653be46000a9a76902729660a2e6
9979Author: Alexander Polakov <polachok@gmail.com>
9980Date:   Sat Sep 17 20:54:58 2011 +0400
9981
9982    XGrabKey manual page: change XAllowAccess to XAllowEvents in See Also
9983    
9984    There is no XAllowAccess man page to see.
9985    
9986    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9987    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9988
9989commit a23f3323f269b33a43d16dce01395cd28bc41b45
9990Author: Gaetan Nadon <memsize@videotron.ca>
9991Date:   Mon Sep 19 15:27:44 2011 -0400
9992
9993    XKB: provide adequate quotes for the license text
9994    
9995    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9996
9997commit a9c7a5cad91d968f5536ef8fc735036e921832a3
9998Author: Gaetan Nadon <memsize@videotron.ca>
9999Date:   Mon Sep 19 15:18:53 2011 -0400
10000
10001    XIM: refactor the multi licensing legal text
10002    
10003    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10004
10005commit c46f934ed89e7de79746a0387c9f998d91994ea6
10006Author: Gaetan Nadon <memsize@videotron.ca>
10007Date:   Mon Sep 19 10:33:30 2011 -0400
10008
10009    xim trans: restore Fujitsu copyright legal text
10010    
10011    Somehow lost during docbook conversion. text from x.org ftp R7.5.
10012    
10013    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10014
10015commit 3d75f993382bfdc89d31668d7dfc71c91222e0d7
10016Author: Gaetan Nadon <memsize@videotron.ca>
10017Date:   Mon Sep 19 10:11:46 2011 -0400
10018
10019    xtrans: restore X Consortium original legal text
10020    
10021    Asking X Consortium permission to use The Open Group name makes no sense.
10022    Even more so in 1994 before X Window System was passed on to the Open Group.
10023    
10024    Using original text from xorg-docs/general/License
10025    
10026    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10027
10028commit 136a381585a4a9686c11bad1a6130837978e677f
10029Author: Gaetan Nadon <memsize@videotron.ca>
10030Date:   Mon Sep 19 10:03:21 2011 -0400
10031
10032    Framework: restore X Consortium copyright
10033    
10034    Somewhat dammaged during docbook conversion.
10035    Also restore pasrt of the original license text
10036    
10037    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10038
10039commit 33f346878406762704a0c13cdc018111c666f3c1
10040Author: Gaetan Nadon <memsize@videotron.ca>
10041Date:   Mon Sep 19 09:50:00 2011 -0400
10042
10043    localedb: restore X Consortium original legal text
10044    
10045    Asking X Consortium permission to use The Open Group name makes no sense.
10046    Even more so in 1994 before X Window System was passed on to the Open Group.
10047    
10048    Using original text from xorg-docs/general/License
10049    
10050    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10051
10052commit e99c9338e413efca026210b9d830efb5c74d34e5
10053Author: Gaetan Nadon <memsize@videotron.ca>
10054Date:   Mon Sep 19 08:23:18 2011 -0400
10055
10056    specs: support multi licensed copyright notice and license text
10057    
10058    For documentation having multiple licenses, the copyright and legalnotice
10059    elements sequence cannot instantiated multiple times.
10060    The copyright notice and license text are therefore coded inside a
10061    legalnotice element. The role attribute on the paragraph is used to allow
10062    styling of the copyright notice text which should not be italicized.
10063    
10064    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10065
10066commit b9dedc757e667333ce899115d618f25cdaa4dd5e
10067Author: Gaetan Nadon <memsize@videotron.ca>
10068Date:   Mon Sep 19 08:17:50 2011 -0400
10069
10070    localedb: add release info to spec
10071    
10072    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10073
10074commit 4519c89a87ddce01e599542737a27c26030fe4ff
10075Author: Gaetan Nadon <memsize@videotron.ca>
10076Date:   Mon Sep 19 08:05:32 2011 -0400
10077
10078    specs: fix The Open Group license text
10079    
10080    The warranty referred to the X Consortium
10081    
10082    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10083
10084commit 08ac378423e7be72e340197ba5559a1a36f1783c
10085Author: Gaetan Nadon <memsize@videotron.ca>
10086Date:   Mon Sep 19 08:03:39 2011 -0400
10087
10088    specs: The strandard name is still "X Consortium Standard"
10089    
10090    This spec, and fsproto spec, are the only two docs with a different
10091    standard name.
10092    
10093    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10094
10095commit 8dfbeb1b1b1e8aa17f69d0a373155419a5f6a172
10096Author: Gaetan Nadon <memsize@videotron.ca>
10097Date:   Sun Sep 18 13:22:34 2011 -0400
10098
10099    specs: support multi licensed copyright notice and license text
10100    
10101    For documentation having multiple licenses, the copyright and legalnotice
10102    elements sequence cannot instantiated multiple times.
10103    The copyright notice and license text are therefore coded inside a legalnotice
10104    element. The role attribute on the paragraph is used to allow styling of the
10105    copyright notice text which should not be italicized.
10106    
10107    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10108
10109commit 278ca8947c1dabb2b819527dca0fa6190c034f67
10110Author: Gaetan Nadon <memsize@videotron.ca>
10111Date:   Fri Sep 16 11:45:56 2011 -0400
10112
10113    docs: merge copyright holder under the same copyright notice
10114    
10115    As per the docbook markup dtd.
10116    
10117    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10118
10119commit 7ff012bb433109a622ac122ade71669e842ebdcf
10120Author: Gaetan Nadon <memsize@videotron.ca>
10121Date:   Fri Sep 16 10:09:57 2011 -0400
10122
10123    specs: handle multiple sets of copyright notice/license/warranty
10124    
10125    Docbook groups all the <copyright> elements together and all the
10126    <legalnotice> elements together.
10127    
10128    We cannot have a sequence:
10129    <copyright> <legalnotice> <copyright> <legalnotice> [...]
10130    
10131    A workaround, which was done in some documents, is to put the copyright
10132    notice inside the legalnotice in plain text without the <copyright> element.
10133    A formal paragraph title is added here which makes the copyright notice bold,
10134    and makes it much easier to locate.
10135    
10136    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10137    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10138
10139commit 4a550c71b8221c37b1a9378d5a170da9eaa03405
10140Author: Gaetan Nadon <memsize@videotron.ca>
10141Date:   Wed Sep 14 15:55:42 2011 -0400
10142
10143    specs: remove orphan affiliation.
10144    
10145    Authors affiliation are correct.
10146    
10147    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10148
10149commit 0cc02a6df65214560b7575e89cebee741d6d2469
10150Author: Gaetan Nadon <memsize@videotron.ca>
10151Date:   Tue Sep 13 17:39:02 2011 -0400
10152
10153    specs: use appropriate markup for Copyright statements
10154    
10155    Also move <releaseinfo> to match order of appearance
10156    
10157    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10158
10159commit afe13e19ebf0fc4e3460644164433af016f0add7
10160Author: Gaetan Nadon <memsize@videotron.ca>
10161Date:   Mon Sep 12 16:54:45 2011 -0400
10162
10163    docs: use the &fullrelvers; entity to set X11 release information
10164    
10165    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10166
10167commit 22a2153282db13e9d94881022b8c979539841a2c
10168Author: Gaetan Nadon <memsize@videotron.ca>
10169Date:   Sun Sep 11 19:49:53 2011 -0400
10170
10171    docs: remove <productnumber> which is not used by default
10172    
10173    This element is not rendered by default on the title. A template
10174    customization is required to display it.
10175    X Window System does not have a product number.
10176    
10177    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10178
10179commit 719f16570d9fcfd15247813ee51fa51ac8a6ff4c
10180Author: Gaetan Nadon <memsize@videotron.ca>
10181Date:   Sun Sep 11 17:40:21 2011 -0400
10182
10183    docs: use the &fullrelvers; entity to set X11 release information
10184    
10185    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10186
10187commit a6b2992f50b571d612ea9ade631c432c3099bc12
10188Author: Gaetan Nadon <memsize@videotron.ca>
10189Date:   Sun Sep 11 17:38:42 2011 -0400
10190
10191    docs: remove orphan <affiliation>
10192    
10193    Somehow created during the conversion from roff. Unable to locate
10194    the author to which it belongs.
10195    
10196    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10197
10198commit c7420060b6e47b8467ba50c796ec3c1bab090bc7
10199Author: Gaetan Nadon <memsize@videotron.ca>
10200Date:   Sun Sep 11 17:38:05 2011 -0400
10201
10202    docs: remove <productnumber> which is not used by default
10203    
10204    This element is not rendered by default on the title. A template
10205    customization is required to display it.
10206    X Window System does not have a product number.
10207    
10208    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10209
10210commit 7d5b718c1edbc43c54b7001be548e515a49540ea
10211Author: Gaetan Nadon <memsize@videotron.ca>
10212Date:   Fri Sep 9 14:46:39 2011 -0400
10213
10214    docbook.am: embed css styles inside the HTML HEAD element
10215    
10216    Rather than referring to the external xorg.css stylesheet, embed the content
10217    of the file in the html output produced. This is accomplished by using
10218    version 1.10 of xorg-xhtml.xsl.
10219    
10220    This makes the whole html docs tree much more relocatable.
10221    In addition, it eliminates xorg.css as a runtime file which makes
10222    xorg-sgml-doctools a build time only package.
10223    
10224    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10225
10226commit 24632d280491c465dfe208ef788815890892e326
10227Author: Gaetan Nadon <memsize@videotron.ca>
10228Date:   Mon Sep 5 12:53:12 2011 -0400
10229
10230    compose: upgrade makefile to support olinking on chunked html
10231    
10232    The essential differences over the regular docbook.am are:
10233    Adding root.filename parameter for naming of chapters html files.
10234    Using xhtml xmlto format and xorg-chunk.xsl stylesheet
10235    Set olink.base.uri for pdf but not for chunked html
10236    Olink is not applicable to ps and txt formats.
10237    
10238    Html chapters are added to shelf_DATA as they are also installed.
10239    The xml is generated from a perl script and not distributed.
10240    
10241    Requires version 1.10 of xorg-sgml-doctools.
10242    
10243    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10244
10245commit 93ab0a1c9a4f9094cf0c8e1c686130673e681798
10246Author: Gaetan Nadon <memsize@videotron.ca>
10247Date:   Wed Aug 31 19:46:20 2011 -0400
10248
10249    docbook.am: refactor common flags for xmlto and xsltproc
10250    
10251    Maximize reuse and reduce risk of setting the wrong flag
10252    at the wrong place.
10253    
10254    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10255
10256commit 85b725aa9ef76aeb5e94d42e0cd60f896da7ab97
10257Author: Gaetan Nadon <memsize@videotron.ca>
10258Date:   Wed Aug 31 10:46:59 2011 -0400
10259
10260    docbook.am: add search path for local entities
10261    
10262    Currently, only $(XORG_SGML_PATH)/X11 is searched for xml entities.
10263    A module may want to add entities that apply only to itself and
10264    not to all modules, like the xserver does.
10265    
10266    This feature may or may not be used in this module, but all modules
10267    share a copy of docbook.am.
10268    
10269    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10270
10271commit b5ed2d368d47f244d43d2513329fa2758bc4ef54
10272Author: Gaetan Nadon <memsize@videotron.ca>
10273Date:   Thu Aug 25 14:39:20 2011 -0400
10274
10275    docbook.am: explicitly list xmlto flags for each target
10276    
10277    Normal evolution of make targets make it impractical to factor out
10278    common command flags for xmlto.
10279    
10280    The targets now list each command option as its presence or absence
10281    needs to be justified.
10282    
10283    xorg.ss is only needed by xmlto for html.
10284    masterdb is only needed by xmlto.
10285    img.src.path must not be used by html.
10286    xsltproc need to use customization layer xorg-*.xsl.
10287    txt format is not required to search masterdb.
10288    
10289    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10290
10291commit d5b3605f3b0639452e892e9ff3afb18bb7442069
10292Author: Gaetan Nadon <memsize@videotron.ca>
10293Date:   Thu Aug 25 13:12:49 2011 -0400
10294
10295    docbook.am: do not generate docs if docbook customization layer is missing
10296    
10297    The stylesheets used to be only about style, fonts or colors.
10298    Complex features are now used like olink and chunked html which
10299    may cause a build break when stylesheets are missing or lead to hard
10300    to find problem.
10301    
10302    Some modules may be built when stylesheets are present while others
10303    are built without. There is no requirement to build crippled docs.
10304    
10305    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10306
10307commit 7f23c72c94d9d14122426b0d8c66054f1402769a
10308Author: Gaetan Nadon <memsize@videotron.ca>
10309Date:   Fri Aug 5 14:33:32 2011 -0400
10310
10311    libX11 specs: review doclifter generated tables
10312    
10313    Many tables had a questionnable layout and some had information dropped.
10314    Each table was cross-referenced with a pre-docbook version
10315    to ensure semantic integrity.
10316    
10317    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10318
10319commit 1efdbeb8cd08f443188ac783cbab6ba0654a66d6
10320Author: Gaetan Nadon <memsize@videotron.ca>
10321Date:   Wed Jul 20 15:28:47 2011 -0400
10322
10323    credits.xml: remove toc from Acknowledgments
10324    
10325    There should be no toc for a simple preface with only one
10326    Acknowledgments section.
10327    
10328    Use <simplesect> markup rather than sect1.
10329    
10330    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10331
10332commit 77cc17ed0817c1df3ac5feb9dc83a5ba5d021a4b
10333Author: Gaetan Nadon <memsize@videotron.ca>
10334Date:   Thu Aug 4 19:59:37 2011 -0400
10335
10336    libX11 specs: move </para> above <varaiablelist>
10337    
10338    Many, but not all, function synopsis have a paragraph that
10339    nests the variable list. The code was generated by doclifter
10340    and there is no apprent reasons as to why it was done this way.
10341    Found while investigating a spacing issue.
10342    
10343    Before the patch:
10344    <para>Some blurb about the function
10345    <funcsynopsis>
10346    </funcsynopsis>
10347    <variablelist>
10348    </variablelist>
10349    </para>
10350    
10351    After the patch:
10352    <para>Some blurb about the function</para>
10353    <funcsynopsis>
10354    </funcsynopsis>
10355    <variablelist>
10356    </variablelist>
10357    
10358    There are no noticable differences when reading the doc
10359    other than than the removed few pixels of spacing between
10360    the function sysnopsis and the variable list block.
10361    
10362    In some cases, there are no "blurb about the fucntion"
10363    and the empty paragraph is removed.
10364    
10365    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10366
10367commit 18595f0d8f95ff1c12e7b9f73bd4b30dda5c7168
10368Author: Gaetan Nadon <memsize@videotron.ca>
10369Date:   Thu Jul 28 19:46:14 2011 -0400
10370
10371    compose specs: generate chunked html
10372    
10373    For large 600 page documents such as this one, chunked html
10374    provide faster browser load time and better navigation.
10375    
10376    Simply click on the locale of your choice in the toc and
10377    the browser loads just that one file. Being a DocBook,
10378    it benefits from all of the usual features and can be
10379    integrated with the rest of the documentation.
10380    
10381    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10382    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10383
10384commit 5cc8815b8a117cf0b6af37e7d55c5a420d9fff57
10385Author: Gaetan Nadon <memsize@videotron.ca>
10386Date:   Thu Jun 23 15:16:47 2011 -0400
10387
10388    specs: build compose keys tables in specs/i18n/compose
10389    
10390    Build the tables article together with the rest of the specs.
10391    Make it transparent that the source in generated in nls.
10392    Reuse docbook.am and get all the features such as olink.
10393    
10394    The docbook article file stem and id must be the same. The new name
10395    for the main article is libX11-keys. The new installation location
10396    is $docdir/libX11/i18n/compose.
10397    
10398    The nls dir retains the role of generating the DocBook/XML source
10399    but does not build neither installs output formats or generated source.
10400    
10401    The tables article now has to specify that each included locale
10402    section is a dependency. It did not matter before as they were web links.
10403    
10404    The xorg-sgml-doctools masterdb shall be updated to include this new doc.
10405    Install location moved from doc/libX11/Compose to doc/libX11/i18n/compose.
10406    
10407    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10408    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10409
10410commit adebbe3856b4aee39c0c7af902c1dea847dd6d7c
10411Author: Gaetan Nadon <memsize@videotron.ca>
10412Date:   Thu Jun 23 14:43:31 2011 -0400
10413
10414    nls: restructure charts as a single article with sections
10415    
10416    Looks more like a real article with a toc rather than individual
10417    web pages. Looks nicer in pdf.
10418    
10419    Each locale is a "section" rather than an "article".
10420    Using XInclude to aggregate xml source files gets you the toc for free.
10421    
10422    The single document is over 600 pages while there were 62 separate
10423    documents previously. FOP version 1.0 is required to handle missing
10424    character like capital sharp s.
10425    
10426    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10427    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10428
10429commit fc74dc12b1ff3c43e240e1a713316ce1bf525d61
10430Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10431Date:   Thu Jul 28 19:32:40 2011 -0700
10432
10433    libX11 1.4.4
10434    
10435    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10436
10437commit ab1f30231755d99ed123af6873b981834640f01d
10438Author: Matt Dew <marcoz@osource.org>
10439Date:   Wed Jul 13 12:33:40 2011 -0600
10440
10441    Add id attributes to error codes to allow linking from other docs.
10442    
10443    Signed-off-by: Matt Dew <marcoz@osource.org>
10444    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10445    Acked-by: Gaetan Nadon <memsize@videotron.ca>
10446
10447commit 5c831fef402914ccf2ec14005c25be48852f119b
10448Author: David Coppa <dcoppa@gmail.com>
10449Date:   Tue Jul 12 10:05:47 2011 +0200
10450
10451    Fix libpthread linkage on OpenBSD.
10452    
10453    OpenBSD prefers to use the -pthread to fetch pthread libs when needed.
10454    
10455    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
10456    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10457
10458commit 445b0b3efc04a74fc77cd3e483b25cfb5da82f12
10459Author: Gaetan Nadon <memsize@videotron.ca>
10460Date:   Wed Jun 22 20:46:26 2011 -0400
10461
10462    config: use XORG_WITH_PERL macro to replace custom program check
10463    
10464    New in version util-macros 1.15, the macro does the usual checking
10465    for program path and provides an interface for makefile variables.
10466    
10467    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10468    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10469
10470commit 47d99a324db9232029a22dd523e3bc3ed0936aca
10471Author: Gaetan Nadon <memsize@videotron.ca>
10472Date:   Wed Jun 22 20:41:18 2011 -0400
10473
10474    config: update XORG_WITH_FOP macro usage
10475    
10476    A new feature to test for a minimum version has been added.
10477    
10478    The parameter position changed. Due to limited usage
10479    and requiring 1.15, it turns out to be backward compatible.
10480    
10481    There is no functional change to how docs are build.
10482    
10483    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10484
10485commit e659683ad51ecc26c989fbdefdfe20f83af2c5a5
10486Author: Gaetan Nadon <memsize@videotron.ca>
10487Date:   Wed Jun 22 11:48:49 2011 -0400
10488
10489    Remove unused xmlrules.in in specs, now in /nls
10490    
10491    The specs dir has been converted to docbook.am
10492    
10493    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10494
10495commit 9992092609bd83a0db778e9a9ac1e879d445e637
10496Author: Gaetan Nadon <memsize@videotron.ca>
10497Date:   Thu Mar 31 10:23:22 2011 -0400
10498
10499    nls: move xmlrules.in from specs to nls
10500    
10501    It is no longer used in the specs dir.
10502    Remove html from CLEANFILES as it breaks. Make do not clean directories
10503    Remove SUFFIXES as pattern rules are used rather than suffix rules.
10504    
10505    Other improvements are possible, the first priority is to move it out
10506    of the specs dir.
10507    
10508    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
10509    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10510
10511commit b370eeb0191b0e2a8bde2f03a442a8ca62e35d58
10512Author: Gaetan Nadon <memsize@videotron.ca>
10513Date:   Sun Jun 5 16:27:36 2011 -0400
10514
10515    Install target dbs alongside generated documents
10516    
10517    This matches a change in xorg-sgml-docs whereby the masterdb will look for
10518    the target dbs into the same location as the generated documents.
10519    
10520    The target dbs are now installed alongside the generated documents.
10521    Previously they are installed in $prefix/sgml/X11/dbs alongside masterdb which
10522    has the potential of installing outside the package prefix and cause
10523    distcheck to fail when user does not have write permission in this package.
10524    
10525    Requires XORG_CHECK_SGML_DOCTOOLS(1.8) which was released 2011-06-11
10526
10527commit 4e741654b7e0a283779b8b4af5bb32e05b7fc678
10528Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10529Date:   Tue Jun 7 18:41:30 2011 -0700
10530
10531    specs/libX11: Turn appendix references into links
10532    
10533    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10534
10535commit a3d1152bc4b300f59ea2f08925739aab1ebfbc4e
10536Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10537Date:   Tue Jun 7 18:37:41 2011 -0700
10538
10539    specs/libX11: Turn chapter references into links
10540    
10541    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10542
10543commit d0cc949dfe44a41a71f19e4fe6c7547bb3d9bdbe
10544Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10545Date:   Tue Jun 7 00:52:32 2011 -0700
10546
10547    specs/libX11: Turn many "see section ..." into links
10548    
10549    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10550
10551commit 02d4c08dead2f266809b3f93ec72377783423ac7
10552Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10553Date:   Sun Jun 5 22:50:37 2011 -0700
10554
10555    specs/libX11: Convert some header filenames to filename tags
10556    
10557    perl -i -p -e 's{^&lt;(.*\.h)&gt;\ *}{<filename class="headerfile">&lt;\1&gt;</filename>}' *.xml
10558    
10559    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10560
10561commit c51a011329afa6e5b9d2b285349c132683ecf9fb
10562Author: Matt Dew <marcoz@osource.org>
10563Date:   Tue May 31 20:03:23 2011 -0600
10564
10565    Add id attributes to funcsynopsis to allow other docs to olink to them.
10566    
10567    Signed-off-by: Matt Dew <marcoz@osource.org>
10568    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10569    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10570
10571commit 6841260c8bb15404a0b4805bee3b0bdfec7176b3
10572Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10573Date:   Fri May 27 20:14:36 2011 -0700
10574
10575    Bug 37624 - parameter x is missing from synopsis of XDrawString in libX11.html
10576    
10577    https://bugs.freedesktop.org/show_bug.cgi?id=37624
10578    
10579    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10580
10581commit 205af6a4e557c62a6395feadc1c89f4a9fe1713b
10582Author: Matt Dew <marcoz@osource.org>
10583Date:   Sat May 21 00:23:06 2011 -0600
10584
10585    Add id's to functions, to make clickable links.
10586    
10587    Signed-off-by: Matt Dew <marcoz@osource.org>
10588    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10589
10590commit 5c06bc594473f6ab234724cd90db32e7b57fe811
10591Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10592Date:   Fri May 20 15:38:08 2011 -0700
10593
10594    libX11 spec: fix monospaced column alignment after deligaturization
10595    
10596    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10597
10598commit ef0019c714c273cb9b4ad3eba52d0b905109826b
10599Author: James Cloos <cloos@jhcloos.com>
10600Date:   Fri May 20 15:51:46 2011 -0400
10601
10602    Expand latin ligatures out to their NFKC equivalents.
10603    
10604    Ligatures should be done by the typesetting system.
10605    Manuscripts should not bother with them.
10606    
10607    Signed-off-by: James Cloos <cloos@jhcloos.com>
10608    Reviewed-by: Keith Packard <keithp@keithp.com>
10609
10610commit 761b8aa0c9b3c58c478ac5ea1b3aaafadcfc1325
10611Author: Daniel Stone <daniel@fooishbar.org>
10612Date:   Thu May 12 16:21:50 2011 +0200
10613
10614    XCB: Add more friendly error messages for common asserts
10615    
10616    This patch adds more friendly error messages for three common classes of
10617    assertion:
10618        - missed sequence numbers due to being griefed by another thread
10619        - unknown requests in queue due to being griefed by another thread
10620        - extensions dequeuing too much or too little reply data
10621    
10622    It adds error messages offering advice (e.g. call XInitThreads() first)
10623    on stderr, but still generates actual assertions.  Hopefully this means
10624    it's a little more Googleable and a little less frightening.
10625    
10626    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
10627    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10628
10629commit 159bf292477048b9a2f074735afc516f52c93d80
10630Author: Gaetan Nadon <memsize@videotron.ca>
10631Date:   Thu May 19 20:22:11 2011 -0400
10632
10633    Remove misplaced hyphens in libX11 DocBook/XML #37364
10634    
10635    Reported-by: Christopher Yeleighton <giecrilj@stegny.2a.pl>
10636    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10637    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10638    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10639
10640commit 0ee6d8247d397500ae183ef180b6ff21bceefa1c
10641Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10642Date:   Tue May 17 20:49:59 2011 -0700
10643
10644    Fix man page and comment references to use XFreeModifiermap (lowercase map)
10645    
10646    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10647
10648commit 3275ec4ca8790d571b6a1902367ca5f68a2dc5fd
10649Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10650Date:   Tue May 17 20:30:54 2011 -0700
10651
10652    XKB man pages: Fix coordinantes typo in multiple pages
10653    
10654    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10655
10656commit ca33a80606cd68e4572fa49ed6aaa1d1fc31a47f
10657Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10658Date:   Tue May 17 16:27:32 2011 -0700
10659
10660    DisplayOfCCC.man: Fix typo "ClientWhitePointOfCC" -> "ClientWhitePointOfCCC"
10661    
10662    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10663
10664commit ae39d82b01359b07dd1fe092d867f79ea95a5438
10665Author: Jeremy Huddleston <jeremyhu@apple.com>
10666Date:   Sun May 8 09:08:07 2011 -0700
10667
10668    Silence clang static analysis warnings for SetReqLen
10669    
10670    This provides a simplified version of the SetReqLen macro when using clang for
10671    static analysis.  Prior to this change, we would see many Idempotent operation
10672    warnings inside this macro due to the common case of calling with arg2 and
10673    arg3 being the same variable.  This has no effect on code produced during
10674    compilation, but it silences a number of false positives in static analysis.
10675    
10676    XIPassiveGrab.c:170:5: warning: Assigned value is always the same as the existing value
10677        SetReqLen(req, num_modifiers, num_modifiers);
10678        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
10679    In file included from XIPassiveGrab.c:26:
10680    .../include/X11/Xlibint.h:580:8: note: instantiated from:
10681                n = badlen; \
10682                  ^
10683    
10684    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10685
10686commit 130af2bc02090ee2526adb2f9803fc07019b8f1f
10687Author: Jeremy Huddleston <jeremyhu@apple.com>
10688Date:   Wed May 4 15:34:31 2011 -0700
10689
10690    XKB: XkbComputeSectionBounds: Check correct bounds in default switch-case
10691    
10692    XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds')
10693            _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1);
10694                                   ^~~~~~~
10695    
10696    Found-by: clang static analyzer
10697    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10698    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
10699
10700commit c987ef04bea345d5810e60cf6c53b747ddbe910e
10701Author: Jeremy Huddleston <jeremyhu@apple.com>
10702Date:   Wed May 4 15:31:17 2011 -0700
10703
10704    Revert "XKB: Avoid a possible NULL dereference"
10705    
10706    Sorry for the noise.  I accidentally pushed and didn't mean to. =(
10707    
10708    This reverts commit 4024091678ea07e0d898b798df9b29f3bf68eb08.
10709
10710commit 46d04bc4e7ff6c86385002c929d8eb0310f737a8
10711Author: Jeremy Huddleston <jeremyhu@apple.com>
10712Date:   Wed May 4 11:35:31 2011 -0700
10713
10714    Dead code removal
10715    
10716    XKBGeom.c:118:27: warning: Assigned value is always the same as the existing value
10717        for (key=row->keys,pos=k=0;k<row->num_keys;k++,key++) {
10718                           ~~~^~~~
10719    XKBGeom.c:115:5: warning: Value stored to 'pos' is never read
10720        pos= 0;
10721        ^    ~
10722    
10723    Found-by: clang static analyzer
10724    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10725
10726commit 4024091678ea07e0d898b798df9b29f3bf68eb08
10727Author: Jeremy Huddleston <jeremyhu@apple.com>
10728Date:   Wed May 4 11:34:53 2011 -0700
10729
10730    XKB: Avoid a possible NULL dereference
10731    
10732    XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds')
10733            _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1);
10734                                   ^~~~~~~
10735    
10736    Found-by: clang static analyzer
10737    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10738
10739commit 626e5e34be55b7cf734d745bd40a7ee3359029f8
10740Author: Jeremy Huddleston <jeremyhu@apple.com>
10741Date:   Wed May 4 11:21:41 2011 -0700
10742
10743    Annotate _XIOError as _X_NORETURN
10744    
10745    Found-by: clang static analyzer
10746    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10747
10748commit d3d7896408b435ab42656398839ff4351a37724d
10749Author: Jeremy Huddleston <jeremyhu@apple.com>
10750Date:   Tue May 3 09:32:53 2011 -0700
10751
10752    clang analyzer: Don't warn about Xmalloc(0)
10753    
10754    This will prevent a number of false positives in where clang's
10755    static analysis reports about calls to malloc(0).
10756    
10757    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10758
10759commit 393921cf2188b2b0713cc157effaf17d0abab783
10760Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10761Date:   Thu Apr 21 16:21:04 2011 -0700
10762
10763    Delete special case code to append "/sparcv9" to i18n module path
10764    
10765    Was triggered by defined(__sparcv9) so only built on Solaris SPARC 64-bit.
10766    Inconsistent with all other platforms, and a bit overcomplicated.
10767    
10768    Should anyone need to continue using that path, simply build with
10769    a #define POSTLOCALELIBDIR "sparcv9" to get the same result.
10770    
10771    Fixes Solaris bug 7038737:
10772     sparcv9 Xlib looking in wrong path for i18n loadable modules
10773    
10774    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10775    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
10776
10777commit b9ff9cfd5c56ae6d52c6503f7019756d9de29557
10778Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10779Date:   Thu Apr 14 13:47:12 2011 -0700
10780
10781    specs/libX11: Fix XOpenDisplay() prototype in chapter 2 [bug 36244]
10782    
10783    Was incorrectly showing AllPlanes() instead when describing XOpenDisplay()
10784    
10785    https://bugs.freedesktop.org/show_bug.cgi?id=36244
10786    
10787    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10788
10789commit b850adbdebcf500c659f85285d4d7374e15857f5
10790Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10791Date:   Tue Apr 12 22:30:45 2011 -0700
10792
10793    Convert malloc(strlen()); strcpy() sets to strdup
10794    
10795    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10796    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
10797
10798commit 09194042d3dc44a463add1f7c122a68ffd5ef0bf
10799Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10800Date:   Tue Apr 12 21:27:45 2011 -0700
10801
10802    Replace Xmalloc+bzero pairs with Xcalloc calls
10803    
10804    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10805    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
10806
10807commit ac1e2bff7121987fd768500a11d428d9fb9447c5
10808Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10809Date:   Fri Mar 25 14:47:35 2011 +0200
10810
10811    om: Fix memory leaks on get_font_name error paths.
10812    
10813    While at it, remove unneeded check for NULL before Xfree.
10814    
10815    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10816    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10817
10818commit 7c362e275c93c92b4e68fe862e73ee36665de703
10819Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10820Date:   Wed Mar 30 15:06:10 2011 +0300
10821
10822    xcms: Fix memory leaks on LINEAR_RGB_InitSCCData error path.
10823    
10824    pScreenData is replaced when building per visual intensity tables. If
10825    malloc failed the old value of pScreenData (stored also in
10826    pScreenDefaultData) was being leaked. Also, property_return wasn't
10827    free'd in that case.
10828    
10829    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10830    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10831
10832commit 5c810e2ac233e00f361549bafb9b59e8a9e05eff
10833Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10834Date:   Wed Mar 30 15:19:28 2011 +0300
10835
10836    xcms: Fix error on LINEAR_RGB_InitSCCData error path.
10837    
10838    Due to what looks like a copy & paste error, pScreenData->pBlueTbl would
10839    be accessed after being free'd.
10840    
10841    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10842    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10843
10844commit d749948f9492fd9b61c74655a08e32c595e0e3a5
10845Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10846Date:   Fri Mar 25 16:06:15 2011 +0200
10847
10848    om: Fix potential memory leak in init_om.
10849    
10850    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10851    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10852
10853commit d0749d6abdf0fd4d8b4e59b02dad8ccda3f10995
10854Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10855Date:   Thu Mar 24 20:40:41 2011 +0200
10856
10857    om: Fix memory leak on read_EncodingInfo error path.
10858    
10859    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10860    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10861
10862commit 4b2e8d00f5b6969c14003ee8eb258b9f0e4dd7c3
10863Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10864Date:   Thu Mar 24 19:55:33 2011 +0200
10865
10866    Fix memory leaks on _XimCbDispatch error path.
10867    
10868    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10869    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10870
10871commit 46e6c78b1a89e4774e0f7e0f4d6d0fd060c3897e
10872Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10873Date:   Thu Mar 24 19:52:05 2011 +0200
10874
10875    Fix memory leak on _XimCommitRecv error path.
10876    
10877    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10878    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10879
10880commit 0ace642a2d47265f01450bfa2b2fd48eb6956a83
10881Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10882Date:   Thu Mar 24 19:48:44 2011 +0200
10883
10884    Fix memory leaks on _XimWriteCachedDefaultTree error paths.
10885    
10886    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10887    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10888
10889commit e29be94edbb58b3b8dab545377a710d1f73b61e0
10890Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10891Date:   Thu Mar 24 19:42:36 2011 +0200
10892
10893    Fix memory leaks on _XimGetAttributeID error paths.
10894    
10895    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10896    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10897
10898commit d144a50512466766f55ce61e3884925334b08f0d
10899Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10900Date:   Thu Mar 24 19:36:56 2011 +0200
10901
10902    Fix memory leaks on _XimProtoCreateIC error paths.
10903    
10904    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10905    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10906
10907commit 6a452f7a98499508f753cb8a7c3f08bcbec736b9
10908Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10909Date:   Thu Mar 24 16:24:29 2011 +0200
10910
10911    Fix leaks in _XimEncodingNegotiation error paths.
10912    
10913    name_ptr and detail_ptr weren't free'd in some cases before returning
10914    False.
10915    
10916    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10917    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10918
10919commit 396e5a452a59c1f121220ba72167b720a863b30f
10920Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10921Date:   Thu Mar 24 14:17:44 2011 +0200
10922
10923    FSWrap: fix potential leak in copy_string_list
10924    
10925    If list_count is 0, dst would be allocated and leaked.
10926    
10927    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10928    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10929
10930commit 1a944260182bb552b954d69f6355c2760d4415df
10931Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10932Date:   Wed Mar 23 14:36:22 2011 +0200
10933
10934    Fix memory leak with broken bitmap files in XReadBitmapFileData
10935    
10936    Bitmap file data is read looping through the lines in the input file. If
10937    there is extra data after the bitmap, these lines will be processed and
10938    if this data represents another bitmap it will replace the one read
10939    before causing the memory allocated for bits to leak.
10940    
10941    This changes the code to stop processing the file once a bitmap was
10942    read.
10943    
10944    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10945    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10946
10947commit c1c91e9a221a0e762d96cc12a3f189d6e59f4865
10948Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10949Date:   Tue Apr 12 20:50:42 2011 -0700
10950
10951    Fix "attrinute" typo in comments in ximcp
10952    
10953    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10954
10955commit 7103b83a88edb9c93acfa68f3b556595b087baff
10956Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10957Date:   Fri Apr 8 21:36:11 2011 -0700
10958
10959    Make doc install subdirectories more like the pre-docbook.am paths
10960    
10961    Haven't restored the "specs" path in the install path, but have restored
10962    libX11 & i18n subdirectories in the path for better logical grouping.
10963    
10964    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10965    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10966
10967commit 9a1ae7a90f815a737a942e42d5eb1ffff400977b
10968Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10969Date:   Fri Apr 8 13:24:28 2011 -0700
10970
10971    Install xml versions of specs even if HAVE_XMLTO is false
10972    
10973    Moves HAVE_XMLTO check into docbook.am, more closely matches behaviour
10974    from before docbook.am changes (commit e8c76407d2f6e3)
10975    
10976    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10977    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10978
10979commit a0ad0d5c99023bb9a8ce3944dbc3267f5265721e
10980Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10981Date:   Tue Apr 5 13:29:04 2011 -0700
10982
10983    libX11 1.4.3
10984    
10985    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10986
10987commit e8c76407d2f6e3b11babdb84426e43e780c859c3
10988Author: Gaetan Nadon <memsize@videotron.ca>
10989Date:   Fri Feb 25 09:23:54 2011 -0500
10990
10991    Documentation: add Docbook external references support
10992    
10993    When writing technical documentation, it is often necessary to cross
10994    reference to other information. When that other information is not in the
10995    current document, additional support is needed, namely <olink>.
10996    
10997    A new feature with version 1.7 of xorg-sgml-doctools adds references to
10998    other documents within or outside this package.
10999    
11000    This patch adds technical support for this feature but does not change
11001    the content of the documentation as seen by the end user.
11002    
11003    Each book or article must generate a database containing the href
11004    of sections that can be referred to from another document. This database
11005    is installed in DATAROOTDIR/sgml/X11/dbs. There is a requirement that
11006    the value of DATAROOTDIR for xorg-sgml-doctools and for the package
11007    documentation is the same. This forms a virtual document tree.
11008    
11009    This database is consulted by other documents while they are being generated
11010    in order to fulfill the missing information for linking.
11011    Refer to the xorg-sgml-doctools for further technical information.
11012    
11013    Co-authored-by: Matt Dew <marcoz@osource.org>
11014    
11015    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11016
11017commit 25eb76b3d2726f2f1de759901971ae53b2539dc4
11018Author: Harshula Jayasuriya <harshula@gmail.com>
11019Date:   Mon Mar 21 14:49:37 2011 +0000
11020
11021    Add #define XK_SINHALA
11022    
11023    Add #define XK_SINHALA so that the Sinhala keysyms can be used by
11024    the lk xkb keymap.
11025    
11026    Signed-off-by: Harshula Jayasuriya <harshula@gmail.com>
11027    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
11028
11029commit db8b20b789112717ac0590b40f0b4dc2171797d0
11030Author: Jeremy Huddleston <jeremyhu@apple.com>
11031Date:   Thu Mar 17 16:15:00 2011 -0700
11032
11033    configure.ac: Bump version to 1.4.2
11034    
11035    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
11036
11037commit 83e1ba59c48c79f8b0a7e7aa0b9c9cfd84fa403d
11038Author: Jamey Sharp <jamey@minilop.net>
11039Date:   Tue Mar 15 16:48:07 2011 -0700
11040
11041    Call _XErrorFunction without holding the Display lock.
11042    
11043    Historically, Xlib dropped the Display lock around the upcall to any
11044    user-supplied _XErrorFunction, but somewhere along the way I quit doing
11045    that if you built with XCB. The reasons are lost somewhere in the
11046    pre-git history of Xlib/XCB, and I can't now see any reason to hold the
11047    lock.
11048    
11049    The documentation for XSetErrorHandler still applies though:
11050    
11051        Because this condition is not assumed to be fatal, it is acceptable
11052        for your error handler to return; the returned value is ignored.
11053        However, the error handler should not call any functions (directly
11054        or indirectly) on the display that will generate protocol requests
11055        or that will look for input events.
11056    
11057    So while you are now once again permitted to re-enter Xlib from the
11058    error handler, you're only allowed to call non-protocol functions.
11059    
11060    Signed-off-by: Jamey Sharp <jamey@minilop.net>
11061
11062commit fd85aca7a616c595fc17b2520f84316a11e8906f
11063Author: Jamey Sharp <jamey@minilop.net>
11064Date:   Mon Mar 14 14:45:35 2011 -0700
11065
11066    Ignore user locks after sleeping in _XReply and _XReadEvents.
11067    
11068    This bug appears as a hang in applications that wait for replies from
11069    multiple threads, where one such thread has taken a user lock using
11070    XLockDisplay.
11071    
11072    Prior to this fix, the code could deadlock in this way: If thread 1 goes
11073    to sleep waiting for a reply, and then thread 2 takes a user lock and
11074    waits for a reply, then thread 2 will wait for thread 1 to process its
11075    reply (because responses must be processed in order), but thread 1 will
11076    wait for thread 2 to drop its user lock.
11077    
11078    Fixed by making thread 1 not wait for thread 2 to drop its user lock.
11079    This makes the semantics of user locks hard to define, but they were
11080    already hard to define. The new behavior appears to be consistent with
11081    the way Xlib worked historically, anyway.
11082    
11083    Fixes: http://lists.freedesktop.org/archives/xcb/2011-March/006802.html
11084    
11085    There was a similar potential for deadlock in _XReadEvents, fixed the
11086    same way, with the same caveats about user-lock semantics.
11087    
11088    Signed-off-by: Jamey Sharp <jamey@minilop.net>
11089
11090commit 690f8bffd48a4e7e74298360ddd0431dc95dcd3f
11091Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11092Date:   Tue Jan 18 15:58:20 2011 +0200
11093
11094    xkb: XkbPropertyPtr determined allocation success from wrong variables
11095    
11096    Cannot reach dead statement "return NULL;"
11097    
11098    Check for the NULLness of prop->name and prop->value instead of
11099    name and value, which was checked earlier anyway. Decided against
11100    using strdup due to curious memory allocation functions and the
11101    rest of the xkb not using it either.
11102    
11103    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11104    Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>
11105
11106commit 6a4d027284e7bb5dd458157947bbb1ff580ad071
11107Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11108Date:   Mon Jan 10 16:37:22 2011 +0200
11109
11110    keyBind: Use Xcalloc to initialize allocated _XKeytrans
11111    
11112    Using uninitialized value "p->modifiers"
11113    
11114    Small fix by using Xcalloc instead of Xmalloc
11115    
11116    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11117    Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>
11118
11119commit b993d73bb3214ecc24646f5427c8003b816c6921
11120Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11121Date:   Mon Jan 10 16:22:45 2011 +0200
11122
11123    im/ximcp: release modifiermap before returning
11124    
11125    Variable "map" goes out of scope
11126    
11127    Release modifiermap before returning. Reordered code to call
11128    XGetModifierMapping after the first return from the function.
11129    
11130    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11131    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
11132
11133commit 807a7fc0354f2212dfa5ff1f9e4ede56d8e69ef4
11134Author: Gaetan Nadon <memsize@videotron.ca>
11135Date:   Fri Feb 25 09:23:54 2011 -0500
11136
11137    Docbook: change the book id to match the xml file basename
11138    
11139    This is required for the up-coming external references support.
11140    
11141    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11142
11143commit 40812b53ff5fe548f6eaf43ba4c8781cb43dab43
11144Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
11145Date:   Mon Feb 21 21:54:17 2011 +0100
11146
11147    Make the Local XIM request key releases for braille
11148    
11149    Braille chords management needs key release events. We need to explicitly
11150    request then, else GTK would not pass them throught XFilterEvent and braille
11151    wouldn't work.
11152    
11153    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
11154
11155commit c97c42c49cd5095462abecdf908b416fb0b540b6
11156Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
11157Date:   Mon Feb 21 17:27:38 2011 +0100
11158
11159    Match braille patterns with compose tree
11160    
11161    Braille patterns should also be usable in Compose.  This combines the
11162    implementation of braille chords and compose tree: only emit the braille
11163    pattern if it can not be found in the compose tree, if any.
11164    
11165    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
11166
11167commit 0c6ca565d7c8a47ef3ea823569a9ca5298a5307d
11168Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
11169Date:   Mon Feb 21 15:56:54 2011 +0100
11170
11171    Fix status reporting for braille patterns
11172    
11173    _XimLocalMbLookupString can return a braille keysym even if _Xlcwctomb can't
11174    convert to the current MB charset.
11175    _XimLocalUtf8LookupString needs to set the braille keysym and status too.
11176    
11177    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
11178
11179commit 993abe751f4141f54d8d28b8b73588a1c9085970
11180Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11181Date:   Fri Feb 11 14:49:17 2011 -0800
11182
11183    Clean up memory when first XCreateRegion succeeds and second fails
11184    
11185    Error: Memory leak (CWE 401)
11186       Memory leak of pointer 's' allocated with XCreateRegion()
11187            at line 387 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XShrinkRegion'.
11188              's' allocated at line 387 with XCreateRegion().
11189              s leaks when s != 0 at line 387.
11190    Error: Memory leak (CWE 401)
11191       Memory leak of pointer 'tra' allocated with XCreateRegion()
11192            at line 1452 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XXorRegion'.
11193              'tra' allocated at line 1451 with XCreateRegion().
11194              tra leaks when tra != 0 at line 1451.
11195    
11196    [ This bug was found by the Parfait 0.3.6 bug checking tool.
11197      For more information see http://labs.oracle.com/projects/parfait/ ]
11198    
11199    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11200
11201commit 6ac417cea1136a3617f5e40f4b106aaa3f48d6c2
11202Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11203Date:   Fri Feb 11 14:20:24 2011 -0800
11204
11205    ximcp: Prevent memory leak & double free if multiple %L in string
11206    
11207    In the highly unlikely event that TransFileName was passed a path
11208    containing multiple %L entries, for each entry it would call
11209    _XlcFileName, leaking the previous results, and then for each entry it
11210    would copy from that pointer and free it, resulting in invalid pointers
11211    & possible double frees for each use after the first one freed it.
11212    
11213    Error: Use after free (CWE 416)
11214       Use after free of pointer 'lcCompose'
11215            at line 358 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
11216              Previously freed at line 360 with free.
11217    Error: Use after free (CWE 416)
11218       Use after free of pointer 'lcCompose'
11219            at line 359 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
11220              Previously freed at line 360 with free.
11221    Error: Double free (CWE 415)
11222       Double free of pointer 'lcCompose'
11223            at line 360 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
11224              Previously freed at line 360 with free.
11225    
11226    [ This bug was found by the Parfait 0.3.6 bug checking tool.
11227      For more information see http://labs.oracle.com/projects/parfait/ ]
11228    
11229    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11230
11231commit 502d414118c97d35a44f8e295709682022876331
11232Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11233Date:   Thu Feb 3 17:08:57 2011 +0200
11234
11235    xcms/cmsProp: don't deal with uninitialized values, fail instead
11236    
11237    Properly handle the return value of XGetWindowProperty by considering
11238    if after the loop as well.
11239    
11240    Using freed pointer "prop_ret"
11241    
11242    There were numerous things wrong in how this function interacted with
11243    XGetWindowProperty.
11244    
11245    None of the local variables were initialized and remained that way if
11246    the call to XGetWindowProperty returned 1 (not Succeed). That doesn't
11247    result in after_ret being initialized in which case if it happens to
11248    be 0, the loop was exited. In that case format_ret and nitems_ret were
11249    uninitialized and the function might return with success (but with
11250    uninitialized pointer in prop_ret) or XcmsFailure.
11251    
11252    As the buffer enlarging code was called only when XGetWindowProperty
11253    failed (returned not Success), after_ret would not have been
11254    initialized. It would have been initialized only if the
11255    XGetWindowProperty has returned Success earlier, but in that case the
11256    code fragment would not have been reached.
11257    
11258    This patch alters the function to return XcmsFailure if the call to
11259    XGetWindowProperty fails.
11260    
11261    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11262    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11263    Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
11264    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11265
11266commit c37e278993b9e5b3d7025ef4c434373a011996ec
11267Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11268Date:   Mon Jan 31 14:02:07 2011 +0200
11269
11270    xcms/LRGB: don't double-free property_return
11271    
11272    property_return was free'd before and in the case the conditional is true,
11273    the call to XcmsGetProperty failed which means that property_return wasn't
11274    set so there is no need to free it again.
11275    
11276    Double free of pointer "property_return" in call to "free"
11277    
11278    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11279    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11280    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11281
11282commit 50f4107811249806718a100f9d34f996c58e5e25
11283Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11284Date:   Wed Feb 2 08:58:45 2011 -0800
11285
11286    Xrm.c: ReadInFile: refactor fstat error handling
11287    
11288    We can simplify the fstat failure case now that the GetFileSize macro
11289    has been expanded inline.
11290    
11291    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11292    Reviewed-by: Julien Cristau <jcristau@debian.org>
11293
11294commit 5e9c40fcb5da43c9fdacf12967d090bf202daf2a
11295Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11296Date:   Wed Feb 2 08:56:00 2011 -0800
11297
11298    Expand GetSizeOfFile() macro at the one place it's called
11299    
11300    Removes XrmI.h header that only contained this single macro
11301    
11302    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11303    Reviewed-by: Julien Cristau <jcristau@debian.org>
11304
11305commit 450e17422c0e374d25c643f343ea268cec68da38
11306Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11307Date:   Mon Jan 31 14:01:57 2011 +0200
11308
11309    XlibInt: Use strncpy+zero termination instead of strcpy to enforce buffer size
11310    
11311    Possible overrun of 8192 byte fixed size buffer "buffer" by copying
11312    "ext->name" without length checking
11313    
11314    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11315    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11316    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11317    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11318
11319commit e2566e43b02d2d7b7c1c3bb7db7c5ae81c1245fa
11320Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11321Date:   Mon Jan 31 14:02:13 2011 +0200
11322
11323    lc/def/lcDefConv: Use Xcalloc to avoid use of uninitialized memory
11324    
11325    Fixed by zero'ing conv on allocation with Xcalloc. Then
11326    close_converter works properly.
11327    
11328    Using uninitialized value "conv->state" in call to function "close_converter"
11329    
11330    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11331    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11332    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11333    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11334
11335commit 03f81ad8fb9783986cf9b17661dd31e95c396681
11336Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11337Date:   Mon Jan 31 14:02:16 2011 +0200
11338
11339    xlibi18n/lcFile: Removed superfluous check for NULL target_dir
11340    
11341    The situation is already handled before this code.
11342    
11343    Cannot reach dead expression "0U" inside statement "if (1U + (target_dir ? strl..."
11344    
11345    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11346    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11347    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11348    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11349
11350commit 1346b9ea3b3882201ff8c3ee462ff4b0d4edf639
11351Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11352Date:   Mon Jan 31 14:02:15 2011 +0200
11353
11354    ximcp/imLckup: Handle negative return value from _Xlcwctomb
11355    
11356    Fixed by negative value to memcpy by checking for the negative return
11357    value of _Xlcwctomb and returning 0/XLookupNone in that case.
11358    
11359    a negative value was passed to memcpy
11360    
11361    Unfortunately the other return values for *status don't fit into the
11362    error (which appears to indicate some internal error or running out of
11363    memory). The other valid status codes are XBufferOverflow,
11364    XLookupNone, XLookupChars, XLookupKeySym, and XLookupBoth. Each of
11365    these has a specific meaning attached.
11366    
11367    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11368    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11369    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11370
11371commit 79a5c86e020f08cc108184298a72e2777036de39
11372Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11373Date:   Mon Jan 31 14:02:14 2011 +0200
11374
11375    ximcp/imTrX: Handle failing XGetWindowProperty
11376    
11377    Checked return value of XGetWindowProperty and return false if it fails.
11378    
11379    Return value of "XGetWindowProperty(im->core.display, spec->lib_connect_wid, prop, 0L, (length + bytes_after_ret + 3UL) / 4UL, 1, 0UL, &type_ret, &format_ret, &nitems, &bytes_after_ret, &prop_ret)" is not checked
11380    
11381    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11382    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11383    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11384
11385commit 2ace8d5c89c8f6d9f42b4068f4b508ca28f0ced1
11386Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11387Date:   Mon Jan 31 14:02:12 2011 +0200
11388
11389    XlibInt: info_list->watch_data was reallocated, but result was discarded
11390    
11391    info_list->watch_data was being reallocated, but the return value of
11392    the reallocation was stored only into a local variable. This might
11393    cause some funky behavior and crashes.
11394    
11395    Variable "wd_array" goes out of scope
11396    Value "wd_array" is overwritten in "wd_array = (XPointer*)realloc((char*)info_list->watch_data, (((dpy->watcher_count + 1) * 4U == 0U) ? 1U : ((dpy->watcher_count + 1) * 4U)))"
11397    
11398    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11399    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11400    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11401    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11402
11403commit 2b16a7e683e355c9746290b2cee2fd0dd2bf342a
11404Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11405Date:   Mon Jan 31 14:02:10 2011 +0200
11406
11407    GetProp: Zero-initialized error so its resourceID field is initialized
11408    
11409    Using uninitialized value "error.resourceID" in call to function "_XError"
11410    
11411    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11412    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11413    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11414    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11415
11416commit 85e9f38e016137f0ff2791eb0d092ab027382d2c
11417Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11418Date:   Tue Feb 1 11:07:25 2011 +0200
11419
11420    xcms/LRGB: Add a label for freeing property_return.
11421    
11422    The rest of the code uses goto's to free memory allocated later
11423    and prevent memory leaks, but there were several paths were
11424    property_return was free'd just before a goto.
11425    
11426    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11427    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11428    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11429    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11430
11431commit 3161dc57d4e9b70f852f05e5e474455e121b06ab
11432Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11433Date:   Mon Jan 31 14:02:08 2011 +0200
11434
11435    xcms/LRGB: Fix potential resource leak.
11436    
11437    property_return was not free'd if the allocation of pRedTbl failed.
11438    
11439    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11440    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11441    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11442
11443commit be3e6c205d94dedc1cdebf5d17b987f0f828377a
11444Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11445Date:   Mon Jan 31 14:02:06 2011 +0200
11446
11447    Xrm: Handle the extremely unlikely situation of fstat failing
11448    
11449    Tracked variable "size" was passed to a negative sink.
11450    
11451    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11452    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11453    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11454    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11455
11456commit 3fae16c64d6ef76fd4a25a54c7f7de76596457db
11457Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11458Date:   Mon Jan 31 14:02:05 2011 +0200
11459
11460    Xrm: NEWTABLE had a memory leak after a memory allocation error
11461    
11462    The NEWTABLE macro missed freeing its allocated memory on subsequent
11463    memory allocation errors. Added call to Xfree.
11464    
11465    Variable "table" goes out of scope
11466    
11467    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11468    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11469    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11470    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11471
11472commit 7110ac653349a23c80c365f11f6270dc27c8975a
11473Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11474Date:   Mon Jan 31 14:02:03 2011 +0200
11475
11476    ImUtil: Handle a memory leak in one early return branch
11477    
11478    Fixed memory leak by adding Xfree for image
11479    
11480    Variable "image" goes out of scope
11481    
11482    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11483    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11484    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11485    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11486
11487commit d695f5da9f5b778e54b6987d9177048e32818c4d
11488Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11489Date:   Mon Jan 31 14:02:01 2011 +0200
11490
11491    ximcp/imRm: Handle leaking colormap_ret
11492    
11493    Fixed memory leak by adding Xfree for colormap_ret
11494    
11495    Variable "colormap_ret" goes out of scope
11496    
11497    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11498    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11499    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11500    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11501
11502commit f9eda19d08c1ea0973dfe0bc10a2519d6fd26cc7
11503Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11504Date:   Mon Jan 31 14:01:59 2011 +0200
11505
11506    ximcp/imRmAttr: Handle leaking missing_list
11507    
11508    Fixed memory leak by adding Xfree and initializing missing_list with NULL
11509    
11510    Variable "missing_list" goes out of scope
11511    
11512    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11513    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11514    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11515    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11516
11517commit 3183269e0979c9dbce9d55d9e03937897dc9fb3b
11518Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11519Date:   Mon Jan 31 14:01:58 2011 +0200
11520
11521    ximcp/imRmAttr: Handle leaking colormap_ret
11522    
11523    XFree colormap_ret and initialize it when appropriate.
11524    
11525    Variable "colormap_ret" goes out of scope
11526    
11527    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11528    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11529    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11530
11531commit f66a032a937030f2d9baa81744d36dc585bb085c
11532Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11533Date:   Mon Jan 31 14:01:56 2011 +0200
11534
11535    xlibi18n/lcGeneric: Initialize uninitialized local variable
11536    
11537    Using uninitialized value "new"
11538    
11539    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11540    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11541    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11542    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11543
11544commit d0266e06d38110ec908ca28379014eff743630b7
11545Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11546Date:   Mon Jan 31 14:01:54 2011 +0200
11547
11548    xcmx/cmxColNm: Removed unused assignments to pBuf (in two functions)
11549    
11550    Pointer "pBuf" returned from "fgets(buf, 256, stream)" is never used
11551    
11552    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11553    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11554    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11555    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11556
11557commit cc686655d7bfdeab8b67e01a24bd452a2e9e3fcf
11558Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11559Date:   Tue Feb 1 12:30:56 2011 +0200
11560
11561    XDefaultOMIF.c: Fix memory leaks in get_font_name
11562    
11563    Instead of copying the value returned by get_prop_name and then releasing it,
11564    directly use the return value of get_prop_name, which allocates memory for the
11565    name.
11566    
11567    If get_prop_name returns NULL, continue on to XFreeFont to release the font
11568    before returning the NULL via the normal function return.
11569    
11570    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11571    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11572
11573commit 95796351435d81eaef8166d49ba3a6e7b633d70a
11574Author: Gaetan Nadon <memsize@videotron.ca>
11575Date:   Mon Jan 31 14:50:19 2011 -0500
11576
11577    config: comment, minor upgrade, quote and layout configure.ac
11578    
11579    Group statements per section as per Autoconf standard layout
11580    Quote statements where appropriate.
11581    Autoconf recommends not using dnl instead of # for comments
11582    
11583    Use AC_CONFIG_FILES to replace the deprecated AC_OUTPUT with parameters.
11584    Remove redundant AC_CANONICAL_HOST included in XORG_DEFAULT_OPTIONS
11585    
11586    This helps automated maintenance and release activities.
11587    Details can be found in http://wiki.x.org/wiki/NewModuleGuidelines
11588    
11589    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11590
11591commit e994ab227ed28655565c3193e63733630105e7f9
11592Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11593Date:   Tue Jan 18 12:49:48 2011 +0200
11594
11595    Comparing array against NULL is not useful "&xkb->server->vmods != NULL"
11596    
11597    Removed superfluous comparison.
11598    
11599    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
11600    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11601
11602commit 4115f051e62f9b098efce691e070d44f09f30f1c
11603Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11604Date:   Wed Jan 12 15:51:11 2011 +0200
11605
11606    Variable "entry" tracked as NULL was dereferenced.
11607    
11608    Check entry for non-nullness before dereferencing it
11609    
11610    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
11611    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11612    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11613
11614commit 0edb76d1d5fdca5a2543332699be2e72386dab24
11615Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11616Date:   Thu Jan 27 09:54:00 2011 +0200
11617
11618    Dereferencing possibly NULL "str" in call to function "memcpy" (Deref assumed on the basis of 'nonnull' parameter attribute.)
11619    
11620    If _XkbGetReadBufferPtr returns NULL, goto BAILOUT
11621    
11622    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
11623    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11624    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11625
11626commit 59da8a211ef723909d0530c0331d541db8e63378
11627Author: Erkki Seppälä <erkki.seppala@vincit.fi>
11628Date:   Mon Jan 10 16:17:47 2011 +0200
11629
11630    Using freed pointer "e"
11631    
11632    Reordered code to first to do the comparison and then to release data
11633    
11634    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11635    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
11636    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
11637
11638commit 0f11c229f7099f7c5aeed4691b358dca151dac7d
11639Author: Matt Dew <marcoz@osource.org>
11640Date:   Thu Jan 27 00:31:23 2011 -0700
11641
11642    Remove <literal> tags from compose-chart.pl so pdf building doesn't die on soft-hyphens.
11643    
11644    Signed-off-by: Matt Dew <marcoz@osource.org>
11645    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11646
11647commit ca00bb202f3afcdbbcb8c4cb50308c5dd03f2322
11648Author: Gaetan Nadon <memsize@videotron.ca>
11649Date:   Fri Jan 28 16:07:07 2011 -0500
11650
11651    config: replace deprecated AC_HELP_STRING with AS_HELP_STRING
11652    
11653    This silences an Automake warning.
11654    
11655    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11656
11657commit 8dac08c7515cb6f8a8f4d0a721713697eae755d8
11658Author: Gaetan Nadon <memsize@videotron.ca>
11659Date:   Fri Jan 28 14:59:04 2011 -0500
11660
11661    config: remove unrequired AC_HEADER_STDC
11662    
11663    Autoconf says:
11664    "This macro is obsolescent, as current systems have conforming
11665    header files. New programs need not use this macro".
11666
11667commit d4483375e37a72631d5821413cb75a423e990ffe
11668Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11669Date:   Tue Jan 18 19:37:02 2011 -0800
11670
11671    config: Use correct AC_CONFIG_HEADERS macro
11672    
11673    Replaces obsolete AM_CONFIG_HEADER and undocumented AC_CONFIG_HEADER
11674    
11675    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11676    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11677    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
11678
11679commit c6405fe931f313a0d159bdad00244fc8aae12468
11680Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11681Date:   Sun Jan 16 11:55:23 2011 -0800
11682
11683    config: remove AC_PROG_CC as it overrides AC_PROG_C_C99
11684    
11685    XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls
11686    AC_PROG_C_C99. This sets gcc with -std=gnu99.
11687    If AC_PROG_CC macro is called afterwards, it resets CC to gcc.
11688    
11689    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11690    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11691
11692commit 17aa1f2305da0952ad594c7329d08aba21f13ac8
11693Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11694Date:   Sun Jan 16 11:40:59 2011 -0800
11695
11696    Update to xf86bigfontproto >= 1.2.0 header name
11697    
11698    Clears compile-time warning of:
11699    "X11/extensions/xf86bigfstr.h", line 1: #warning: "xf86bigfstr.h is obsolete and may be removed in the future."
11700    "X11/extensions/xf86bigfstr.h", line 2: #warning: "include <X11/extensions/xf86bigfproto.h> for the protocol defines."
11701    
11702    Requires xf86bigfontproto >= 1.2.0 if --disable-xf86bigfont is not
11703    passed to configure.
11704    
11705    Also removes unnecessary AC_SUBST of BIGFONT_CFLAGS & BIGFONT_LIBS
11706    that PKG_CHECK_MODULES does automatically
11707    
11708    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11709    Reviewed-by: Julien Cristau <jcristau@debian.org>
11710
11711commit a9228fcc676aacf9a760dd94891c89f9bc82b20d
11712Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11713Date:   Tue Jan 11 17:55:22 2011 -0800
11714
11715    libX11 1.4.1
11716    
11717    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11718
11719commit 92fa96451af37a7a0f2592de07643fdaed7a9efd
11720Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
11721Date:   Mon Jan 3 12:25:28 2011 -0500
11722
11723    Initialize event type
11724    
11725    If we receive unsupported event closing connection triggers valgrind
11726    error.
11727    
11728    ==12017== Conditional jump or move depends on uninitialised value(s)
11729    ==12017==    at 0x487D454: _XFreeDisplayStructure (OpenDis.c:607)
11730    ==12017==    by 0x486857B: XCloseDisplay (ClDisplay.c:72)
11731    *snip*
11732    ==12017==  Uninitialised value was created by a heap allocation
11733    ==12017==    at 0x4834C48: malloc (vg_replace_malloc.c:236)
11734    ==12017==    by 0x4894147: _XEnq (XlibInt.c:877)
11735    ==12017==    by 0x4891BF3: handle_response (xcb_io.c:335)
11736    ==12017==    by 0x4892263: _XReply (xcb_io.c:626)
11737    *snip*
11738    
11739    Problem is that XFreeDisplaySturture is checking for qelt->event.type ==
11740    GenericEvent while _XUnknownWireEvent doesn't store the type.
11741    
11742    Reviewed-by: Adam Jackson <ajax@redhat.com>
11743    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11744    Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
11745
11746commit fa9747b9d4443eef9b1687b8b1208801f5c585da
11747Author: Fernando Carrijo <fcarrijo.lists@gmail.com>
11748Date:   Tue Dec 28 16:18:57 2010 -0500
11749
11750    specs: convert images from gif to svg format.
11751    
11752    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11753
11754commit 0012e2a4eaddc0840617b4758931ad976ca2eb7c
11755Author: Philipp Reh <sefi@s-e-f-i.de>
11756Date:   Wed Dec 22 15:14:05 2010 +0000
11757
11758    Events: Store event cookie when dequeuing event
11759    
11760    When we dequeue an event in XCheckTypedEvent or XCheckTypedWindowEvent,
11761    make sure to store the corresponding cookie too.
11762    
11763    Signed-off-by: Philipp Reh <sefi@s-e-f-i.de>
11764    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
11765
11766commit d2714d65e85b44abedf5f82e1a31506dba397ef2
11767Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11768Date:   Tue Dec 21 18:47:16 2010 -0800
11769
11770    Revert "Mark atom names argument to XInternAtoms as const"
11771    
11772    This reverts commit c8701115462b482d99ecff24d9de0f2806084ba5.
11773    
11774    The constification of a pointer to a pointer caused unexpected issues,
11775    and xorg-devel was unable to come up with a clean, safe, reasonable way
11776    to handle them, so we're chalking this up for now as yet another mistake
11777    in the Xlib API definition we'll be living with.
11778    
11779    See https://bugs.freedesktop.org/show_bug.cgi?id=32098 for details.
11780    
11781    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11782
11783commit 6459f9a48a82de3bc0235bd36acab2df11e436d5
11784Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11785Date:   Mon Dec 20 20:54:07 2010 -0800
11786
11787    specs/XKB: make acknowlegement section <preface> instead of <chapter>
11788    
11789    Fixes numbering of actual chapters to match their filenames and to
11790    make the table/figure numbering match the references in the text.
11791    
11792    Reported-by: jelmd on freenode irc
11793    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11794
11795commit 42c653d5413fc5870486a9cf51dd28d1e88f7c69
11796Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11797Date:   Sun Dec 19 23:53:02 2010 -0800
11798
11799    Add XKBlib spec images to dist_spec_DATA for distribution & installation
11800    
11801    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11802    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11803
11804commit 8f8a3c8c42af82976d261ac2447cfbb0f90fa77d
11805Author: Matt Dew <matt@osource.org>
11806Date:   Tue Nov 30 10:52:07 2010 -0500
11807
11808    specs: convert XKBlib spec from Framemaker to DocBook.xml
11809    
11810    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11811
11812commit c8701115462b482d99ecff24d9de0f2806084ba5
11813Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11814Date:   Sun Nov 21 20:49:05 2010 -0800
11815
11816    Mark atom names argument to XInternAtoms as const
11817    
11818    Updates code & docs for XInternAtoms.
11819    
11820    The single atom name argument to XInternAtom was already const char *
11821    in the code, but not the docs, so updated it in the docs too.
11822    
11823    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11824    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11825
11826commit 24db5b5ff0d51b0ed9d9701a792c5f14f23b638d
11827Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11828Date:   Sun Nov 21 14:04:32 2010 -0800
11829
11830    libX11 1.4.0
11831    
11832    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11833
11834commit 18d3c3e9de19e936281b111d5a82acfb3a3f552e
11835Author: Dan Nicholson <dbn.lists@gmail.com>
11836Date:   Sat Nov 6 21:58:10 2010 +0000
11837
11838    config: Check host string when deciding architecture to build for
11839    
11840    When checking for the OS/2 platform, $target_os is used. However, unless
11841    building a cross compiler, the $host* strings contain the platform
11842    details for the build system. See:
11843    
11844    http://www.gnu.org/software/automake/manual/automake.html#Cross_002dCompilation
11845    
11846    $host_os is already being used to determine the transport and threading
11847    options.
11848    
11849    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11850    Reviewed-by: Julien Cristau <jcristau@debian.org>
11851    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11852
11853commit affc2488a7f2660a74dc8354fc3e0bff2c4f879c
11854Author: Dan Nicholson <dbn.lists@gmail.com>
11855Date:   Sat Nov 6 21:58:09 2010 +0000
11856
11857    config: Fix architecture check for OS/2 to skip nios2 cpu
11858    
11859    The OS/2 platform requires some utility functions as well as having a
11860    non-32 bit wchar_t. Fix the configure check so that it doesn't also
11861    affect the nios2 cpu, which wouldn't influence these operating system
11862    issues.
11863    
11864    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11865    Tested-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
11866    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11867    Reviewed-by: Julien Cristau <jcristau@debian.org>
11868    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11869
11870commit 8cbca8a10761d1ea75a75bafa647632d6c0dac71
11871Author: Gaetan Nadon <memsize@videotron.ca>
11872Date:   Tue Nov 9 13:04:44 2010 -0500
11873
11874    config: HTML file generation: use the installed copy of xorg.css
11875    
11876    Currenlty the xorg.css file is copied in each location
11877    where a DocBook/XML file resides. This produces about
11878    70 copies in the $(docdir) install tree.
11879    
11880    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11881
11882commit 3e8907305e1818369aef5a5c0da61f09e20de4f8
11883Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11884Date:   Fri Oct 29 22:02:10 2010 -0700
11885
11886    libX11 1.3.99.903 (1.4.0 RC3)
11887    
11888    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11889
11890commit 5d245d8013289b13e0c42100951b26166c7fada4
11891Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11892Date:   Fri Oct 29 22:01:39 2010 -0700
11893
11894    Require xorg-macros 1.11 now that it is released
11895    
11896    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11897
11898commit 1ec89689fc771f116a6165226b9e076f54254a40
11899Author: Dan Nicholson <dbn.lists@gmail.com>
11900Date:   Fri Oct 22 19:09:07 2010 -0700
11901
11902    docs: Disable fop documentation by default
11903    
11904    fop is used to generate the pdf and ps formats of the documentation.
11905    This can significantly slow down the build, especially when creating all
11906    the compose key charts. Since few people probably want the full set of
11907    doc formats, set the default to 'no'.
11908    
11909    The default parameter for XORG_WITH_FOP is only available in recent
11910    macros. Users generating configure from older macros will just get
11911    'auto' as the default.
11912    
11913    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11914    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11915    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11916    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11917
11918commit 03877bda911b72ac60a7144d6eced7d5be4b4d72
11919Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11920Date:   Wed Apr 14 13:38:18 2010 +0100
11921
11922    Add C.UTF-8 and C.ASCII locale aliases
11923    
11924    Add C.UTF-8 locale as an alias for en_US.UTF-8
11925    Add C.ASCII locale as an alias for C
11926    
11927    (C.UTF-8 is the default locale for cygwin.  It also exists in debian,
11928    although I don't think it's intended for use by humans.)
11929    
11930    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11931    Acked-by: Julien Cristau <jcristau@debian.org>
11932    Reviewed-by: James Cloos <cloos@jhcloos.com>
11933
11934commit 81d1d9a6bdca779a44f931730b866eea75777c00
11935Author: Dan Nicholson <dbn.lists@gmail.com>
11936Date:   Sat Oct 9 16:00:47 2010 -0700
11937
11938    docs: Remove directory prerequisites from make rules
11939    
11940    Make expects prerequisites to be files with valid timestamps, and
11941    directories are treated as always being out of date. Thus, any targets
11942    depending on directories will always be rebuilt.
11943    
11944    Instead, the doc rules are changed to always create the target's leading
11945    directory. This should prevent the documentation from being rebuilt when
11946    "make install" is run.
11947    
11948    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11949    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
11950    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11951
11952commit d0cbf388919364fe6b5b9127f36426eb744090a0
11953Author: Jeremy Huddleston <jeremyhu@apple.com>
11954Date:   Sun Sep 26 21:21:16 2010 -0700
11955
11956    Add an "X11_" string to header guards to avoid possible collision
11957    
11958    This addresses a build failure which can result from <X11/Xlocale.h> and
11959    <xlocale.h> being included in the same code since they both used the same
11960    _XLOCALE_H_ protection.
11961    
11962    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
11963
11964commit b51ac675f44f5ce1a28c5734bf3c26983f8b4192
11965Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11966Date:   Tue Sep 21 19:53:03 2010 -0700
11967
11968    libX11 1.3.99.902 (1.4.0 RC2)
11969    
11970    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11971
11972commit 986bb6d1d54368fe91e3ea24f518d43ce6179782
11973Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11974Date:   Tue Sep 14 00:10:31 2010 -0700
11975
11976    Bug 19379 - Provide docs with overview of all compose key combinations
11977    
11978    Adds compose-chart.pl to generate DocBook/XML documents listing compose
11979    keys, and Makefile rules to generate HTML & PDF output from them if xmlto
11980    is present.
11981    
11982    https://bugs.freedesktop.org/show_bug.cgi?id=19379
11983    
11984    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11985    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
11986    Reviewed-by: James Cloos <cloos@jhcloos.com>
11987    Tested-by: Gaetan Nadon <memsize@videotron.ca>
11988
11989commit 3eb064071695ebf0f371163ed818a428dfeba8e6
11990Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11991Date:   Sat Sep 11 00:49:21 2010 -0700
11992
11993    Make locale data build non-recursive / parallelizable
11994    
11995    On a 4 core CPU with gmake -j 16 the nls subdir builds in half the time,
11996    plus this simplifies the next set of changes.
11997    
11998    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11999    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
12000    Reviewed-by: James Cloos <cloos@jhcloos.com>
12001    Tested-by: Gaetan Nadon <memsize@videotron.ca>
12002
12003commit a3fc78ac352e5a70a958996ef6aec50f653974d1
12004Author: Gaetan Nadon <memsize@videotron.ca>
12005Date:   Fri Sep 17 17:38:11 2010 -0400
12006
12007    config: remove man page suffix from bottom summary
12008    
12009    The man page suffix is the same for all libraries on a given
12010    platform and is not configurable. It should have been removed
12011    in commit 09edc6de6.
12012    
12013    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12014
12015commit a52b0068f36c739eb7e426214cd72a8d9bb776ca
12016Author: Gaetan Nadon <memsize@videotron.ca>
12017Date:   Fri Sep 17 17:29:30 2010 -0400
12018
12019    config: remove unhelpful comment in .gitignore
12020    
12021    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12022
12023commit 10dd881e7818be9074da4ab3a68577adcdaf71d2
12024Author: Adam Jackson <ajax@redhat.com>
12025Date:   Wed Sep 8 10:44:23 2010 -0400
12026
12027    Zero buffer data in BufAlloc()
12028    
12029    Inspired by a pattern in NoMachine's NX.  Consistently zeroed buffers
12030    compress better with ssh and friends.  Note that you'll need to rebuild
12031    all your protocol libraries to take advantage of this.
12032    
12033    Signed-off-by: Adam Jackson <ajax@redhat.com>
12034    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
12035    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12036
12037commit da099f0d2015831ea5f4a5d0740ed962aa4d5d8b
12038Author: James Cloos <cloos@jhcloos.com>
12039Date:   Sun Sep 12 17:00:54 2010 -0400
12040
12041    Revert “Dolt-ify"
12042    
12043    Libtool’s is now sufficiently fast that DOLT is no longer
12044    worth the bother, even on those few systems where is works.
12045    
12046    This reverts commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60
12047    and part of commit d31e644c65c52828ea3e7abd94a8cf9aee12265c.
12048    
12049    It conflicted with commit f6a4fd0c7615684d08e848245849dea4017a5214
12050    which moved dolt from configure.ac to m4/dolt.m4.
12051    
12052    And it addresses: http://bugs.freedesktop.org/show_bug.cgi?id=28188
12053    
12054    Signed-off-by: James Cloos <cloos@jhcloos.com>
12055
12056commit 2661fbe6b809c937a60c2c1ca5ddb8280e399bd8
12057Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12058Date:   Thu Sep 9 19:24:29 2010 -0700
12059
12060    XIM spec title page minor formatting cleanup
12061    
12062    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12063
12064commit f99f1380bbf7bb2a0c491acad4a3d8db44bfd752
12065Author: Jens Petersen <petersen@redhat.com>
12066Date:   Mon Aug 23 18:08:10 2010 -0700
12067
12068    Bug 29773: aliases for nb_NO.utf8 and nn_NO.utf8
12069    
12070    <https://bugs.freedesktop.org/show_bug.cgi?id=29773>
12071    
12072    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12073
12074commit b15b8a558ec64c834cc8f6a52d7f3c4f530c8c4b
12075Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12076Date:   Fri Sep 3 23:11:53 2010 -0700
12077
12078    Sun's copyrights are now owned by Oracle
12079    
12080    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12081
12082commit 7742bf62b1fa652da4270587e280249945367cae
12083Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12084Date:   Fri Sep 3 18:31:44 2010 -0700
12085
12086    libX11 1.3.99.901 (1.4.0 RC1)
12087    
12088    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12089
12090commit 5a02d6d74b5451d48d8f55709dfd4ecd3c068586
12091Author: Jeroen Hoek <mail@jeroenhoek.nl>
12092Date:   Sun Aug 29 09:47:06 2010 -0400
12093
12094    libX11/nls compose tables, add LATIN CAPITAL LETTER SHARP S
12095    
12096    Add LATIN CAPITAL LETTER SHARP S
12097    (See https://bugs.freedesktop.org/show_bug.cgi?id=29448 for rationale)
12098    
12099    Signed-off-by: James Cloos <cloos@jhcloos.com>
12100
12101commit 09edc6de6619a2eabda3b808ebff4165550664f1
12102Author: Gaetan Nadon <memsize@videotron.ca>
12103Date:   Fri Aug 13 14:21:08 2010 -0400
12104
12105    config: remove man-pages configuration option
12106    
12107    This option was added in commit 6e752ea120 with no explanation.
12108    The section number is provoded by XORG_MANPAGE_SECTIONS
12109    There is no case where libX11 should be different than other libs
12110    The option was also used to disable building of the man pages,
12111    which build in 14 secs. No indication this is required.
12112    
12113    If there is a requirement from system builders to disable building
12114    of man pages, it could be done consistently for all modules.
12115    
12116    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12117
12118commit f92e754297ec5fdb81068b56a4435026666224fa
12119Author: Aaron Plattner <aplattner@nvidia.com>
12120Date:   Sun Aug 15 21:51:38 2010 -0700
12121
12122    XOpenDisplay: save the correct display_name value
12123    
12124    The X Test Suite's XDisplayString test checks the invariant
12125    XDisplayString(XOpenDisplay(str)) == str.  The Xlib XOpenDisplay violates this
12126    invariant by expanding str to the canonical form "host:display.scrn" (unless
12127    HAVE_LAUNCHD is set and it starts with "/tmp/launch").  E.g., this expands ":1"
12128    to ":1.0":
12129    
12130      400|26 1 1 19:26:41|IC Start
12131      200|26 1 19:26:41|TP Start
12132      520|26 1 00032625 1 1|VSW5TESTSUITE PURPOSE 1
12133      520|26 1 00032625 1 2|Assertion XDisplayString-1.(A)
12134      520|26 1 00032625 1 3|A call to XDisplayString returns the string that was used
12135      520|26 1 00032625 1 4|as the argument to the XOpenDisplay call that returned the
12136      520|26 1 00032625 1 5|value used as the display argument.
12137      520|26 1 00032625 1 6|METH: Open a connection using XOpenDisplay.
12138      520|26 1 00032625 1 7|METH: Obtain the display string using XDisplayString.
12139      520|26 1 00032625 1 8|METH: Verify that the value of the string is the parameter used in XOpenDisplay.
12140      520|26 1 00032625 1 9|METH: Close the display using XCloseDisplay.
12141      520|26 1 00032625 1 10|REPORT: XDisplayString() returned ":1.0" instead of ":1".
12142      220|26 1 1 19:26:41|FAIL
12143      410|26 1 1 19:26:41|IC End
12144    
12145    Fix this by deleting all of the code to construct the canonical path and just
12146    stashing a copy of the original display_name in dpy->display_name.
12147    
12148    Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
12149    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
12150
12151commit 1dc401f678469d0235e6d0b28eb4356f03327c9f
12152Author: Gaetan Nadon <memsize@videotron.ca>
12153Date:   Fri Aug 13 13:06:51 2010 -0400
12154
12155    man: xkb: remove unused variable LIB_MAN_DIR_SUFFIX
12156    
12157    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12158
12159commit cbe9eebd11e5ebab4280d25b57b4cd4001241619
12160Author: Gaetan Nadon <memsize@videotron.ca>
12161Date:   Fri Aug 13 11:42:29 2010 -0400
12162
12163    man: simplify building of shadow man pages
12164    
12165    Store the shadow files in git as any other man page.
12166    Move man pages to man dir and use the common makefile
12167    
12168    Local fix in CVS for bug 5628 is not required
12169    as the problem has been fixed in
12170    util-macros d9062e4077ebfd0985baf8418f3d0f111b9ddbba
12171    
12172    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12173
12174commit bfd899c9af2318ecbe84cb69007dba4223fdb502
12175Author: Gaetan Nadon <memsize@videotron.ca>
12176Date:   Mon Aug 9 09:39:26 2010 -0400
12177
12178    config: reinstate XORG_PROG_RAWCPP erroneoulsy removed
12179    
12180    in commit 76e07ef6911734eac418e399b114f1b544512736.
12181    
12182    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12183
12184commit 76e07ef6911734eac418e399b114f1b544512736
12185Author: Gaetan Nadon <memsize@videotron.ca>
12186Date:   Fri Aug 6 14:01:51 2010 -0400
12187
12188     man: using the C preprocessor is not required for man pages.
12189    
12190    There were no special symbols needing cpp.
12191    Everything can be handled by the default MAN_SUBSTS in util-macros.
12192    
12193    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12194
12195commit 4b8ff7db39f2fe7ef12968d462aaf3f9054b6c18
12196Author: Jamey Sharp <jamey@minilop.net>
12197Date:   Fri Aug 6 15:51:56 2010 -0700
12198
12199    Fix use-after-free in _XReply on X errors.
12200    
12201    _XReply would always call dequeue_pending_request on errors.  When it
12202    got an error for the current request, it would call dequeue, then break
12203    out of the loop; then, if it had an error in the event queue, it would
12204    compare it with the sequence number of the now-freed pending request.
12205    _XReply already stored that sequence number in dpy->last_request_read
12206    before freeing it, so look at that instead.
12207    
12208    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29412
12209    
12210    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12211    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12212
12213commit 9fa146b30046396b70d64986e50d6617b3a8ac48
12214Author: Gaetan Nadon <memsize@videotron.ca>
12215Date:   Sat Jul 31 16:15:35 2010 -0400
12216
12217    specs: xsl stylesheet requires abs path to images
12218    
12219    Using abs_builddir requires automake 1.10 or later.
12220    
12221    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12222
12223commit a8727d4bc39015e303b4128d0ad8aaf6d5fc9f0a
12224Author: Gaetan Nadon <memsize@videotron.ca>
12225Date:   Sun Aug 1 14:13:07 2010 -0400
12226
12227    specs: update .gitignore now that all groff generated files are gone
12228    
12229    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12230
12231commit f70527f05677aaf6a493ba42d2d95f738aaebafa
12232Author: Adam Jackson <ajax@redhat.com>
12233Date:   Thu Jul 29 11:11:21 2010 -0400
12234
12235    nls: Switch one of the interrobang sequences to gnaborretni
12236    
12237    Since gnaborretni is primarily used in LTR locales, the ? part of the
12238    sequence reflects the position of the bulge in the ?.  When scanning an
12239    inverted interrobang left-to-right, you see the bulge first, so the
12240    compose sequence ?! is used; upright interrobang shows the bulge
12241    last, so it's !?.
12242    
12243    This is a change in behaviour, but I was unable to come up with anything
12244    better.  ??!! or !!?? would match the patterns for ¿ and ¡, but they
12245    would delay evaluation of them; "¿Qué?" wouldn't show the ¿ until you hit
12246    Q.  Likewise ?!?! and !?!? would delay showing the interrobang itself.
12247    ~!? and ~?! were considered but are arguably less intuitive.
12248    
12249    Reviewed-by: James Cloos <cloos@jhcloos.com>
12250    Signed-off-by: Adam Jackson <ajax@redhat.com>
12251
12252commit 511c4f6d29b2da4f71093feabcbb3913cb5d12a7
12253Author: Gaetan Nadon <memsize@videotron.ca>
12254Date:   Wed Jul 28 10:27:39 2010 -0400
12255
12256    specs: move indexterm from glossdef to glossterm
12257    
12258    This move fixes a Java class cast exception in the glossary.
12259    The problem was introduced in commit
12260    26f4f0d50840fe5ba4c46aae0a8e68db0059434b
12261    
12262    It may not happen on all versions of the doc toolchain.
12263    There is no reason why indexterm cannot appear in glossdef,
12264    this is a workaround to an implementation problem found by
12265    trial and error.
12266    
12267    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12268    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12269    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12270
12271commit 87a108f3aa565aa803dacb3a53b811f45025b609
12272Author: Matt Dew <matt@osource.org>
12273Date:   Sat Jul 24 14:46:57 2010 -0400
12274
12275    specs: replace troff source with docbook-xml source
12276    
12277    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12278
12279commit 8e603413586ff42b1351688f6f99318cbe243bca
12280Author: Julien Cristau <jcristau@debian.org>
12281Date:   Sat Jul 24 22:13:07 2010 +0200
12282
12283    Compose.man: default user compose file is .XCompose, not .Xcompose
12284    
12285    The path was correct in the DESCRIPTION section, but not in FILES.
12286    
12287    Signed-off-by: Julien Cristau <jcristau@debian.org>
12288
12289commit 554da76ece85d0fc0cada45a86860e69c2107e9a
12290Author: Daniel Stone <daniel@fooishbar.org>
12291Date:   Tue Jul 20 12:34:48 2010 +0100
12292
12293    NLS: Add \o/ Compose sequence
12294    
12295    Unicode is even more pointlessly awesome! Add a binding for Compose-\o/
12296    to U+1F64C PERSON RAISING BOTH HANDS IN CELEBRATION.
12297    
12298    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12299
12300commit 2e7a18b6a617b9b4bfcea2d36f2bd2d7e0c4a3dd
12301Author: Daniel Stone <daniel@fooishbar.org>
12302Date:   Fri Jul 9 18:13:13 2010 +0100
12303
12304    XStringToKeysym: Cope with 0x1234cafe-style input
12305    
12306    If we get input in the style of 0xdeadbeef, just return that exact
12307    keysym.  Introduces a dependency on strtoul, which I'm told is OK on all
12308    the systems we care about.
12309    
12310    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12311
12312commit 3df45ed0c29b98ff468a0ff0ba24830bb664fd5a
12313Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12314Date:   Mon Jul 19 17:58:27 2010 -0700
12315
12316    specs/libX11: Fix column count of Gravity Attributes table
12317    
12318    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12319
12320commit 53bcba0d1d2dc3fd5e0de4bae3da30a1aa31a0c4
12321Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12322Date:   Mon Jul 19 14:43:38 2010 -0700
12323
12324    specs/libX11: Convert \- to &minus; and \^ to either &hairsp; or removed
12325    
12326    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12327
12328commit 5a0b45275638281e1bb2ae7d3e16b98e6470dae9
12329Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12330Date:   Mon Jul 19 11:28:50 2010 -0700
12331
12332    specs/libX11: Manual cleanup pass over Ch. 1
12333    
12334    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12335
12336commit 21567992958a6f12b3d0186f2f27a5b2a41b8448
12337Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12338Date:   Mon Jul 19 14:51:10 2010 -0700
12339
12340    specs/libX11: make sure all files have DOCTYPEs so standard entities work
12341    
12342    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12343
12344commit c944a8521f5760b485192658b921145159cdb439
12345Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12346Date:   Sun Jul 18 13:12:20 2010 -0700
12347
12348    specs/libX11: Fix up list of header files in Ch.1 & add index entries
12349    
12350    Combination of manual editing and automatic substitution via:
12351    
12352    perl -i -p -00 -e 's{<varlistentry>(\s+)<term>&lt;X11/([^&]+)&gt;</term>(\s+)<listitem>(\s+)<para>}{<varlistentry id="Standard_Header_Files:$2">$1<term><filename class="headerfile">&lt;X11/$2&gt;</filename></term>$3<listitem>$4<indexterm type="file"><primary><filename class="headerfile">X11/$2</filename></primary></indexterm>$4<indexterm><primary>Files</primary><secondary>&lt;X11/$2&gt;</secondary></indexterm>$4<indexterm><primary>Headers</primary><secondary>&lt;X11/$2&gt;</secondary></indexterm>$4<para>}g' CH01.xml
12353    
12354    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12355
12356commit bb66e8f1ce330043278d3e8cbc7d5d3bc56ee030
12357Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12358Date:   Sun Jul 18 12:47:27 2010 -0700
12359
12360    specs/libX11: Add indexterms for headerfiles that .hN used to provide
12361    
12362    Automatic substitution performed via:
12363    perl -i -p -e 's{^&lt;<filename class="headerfile">([^<]+)</filename>&gt;(.*)$}{<filename class="headerfile">&lt;$1&gt;</filename>$2\n<indexterm type="file"><primary><filename class="headerfile">$1</filename></primary></indexterm>\n<indexterm><primary>Files</primary><secondary><filename class="headerfile">&lt;$1&gt;</filename></secondary></indexterm>\n<indexterm><primary>Headers</primary><secondary><filename class="headerfile">&lt;$1&gt;</filename></secondary></indexterm>}' *.xml
12364    
12365    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12366
12367commit d0b0d215cb85da48b323b1392149ce181e7ef5ec
12368Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12369Date:   Fri Jul 16 21:15:13 2010 -0700
12370
12371    specs/libX11: convert multicolumn lists from tables to <simplelist>
12372    
12373    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12374
12375commit e679f80c6210c0aefe3669f8f536353619097a4b
12376Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12377Date:   Fri Jul 16 20:51:52 2010 -0700
12378
12379    specs/libX11: Tag WM_* as <property>
12380    
12381    Mass substitution done by:
12382     perl -i -p -e 'if ($_ !~ m{^\<}) { $_ =~ s{(WM_\w+)}{<property>$1</property>}g; }' *.xml
12383    
12384    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12385
12386commit c7b7e59b3b22221d0be6286c540001c360308f69
12387Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12388Date:   Fri Jul 16 20:41:33 2010 -0700
12389
12390    specs/libX11: Convert simpler eqn markup to docbook tags
12391    
12392    Mostly "sup" to <superscript>
12393    
12394    There's several more complicated equations that will probably need
12395    MathML or SVG to solve.
12396    
12397    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12398
12399commit 897486c54c6a54771867d667441aaf9a4b9c35ca
12400Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12401Date:   Fri Jul 16 00:50:39 2010 -0700
12402
12403    specs/libX11: App. D: convert literallayouts to synopsis or programlisting
12404    
12405    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12406
12407commit 692906c3b3d8aa8b8927fbc230c3050d633785d2
12408Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12409Date:   Fri Jul 16 00:36:43 2010 -0700
12410
12411    specs/libX11: Fix section headers in Appendix D
12412    
12413    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12414
12415commit bcc41baa02e137884d847aec0f3ff8ca5c85c32f
12416Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12417Date:   Wed Jul 14 07:22:26 2010 -0700
12418
12419    specs/libX11: Explicitly tag document as English
12420    
12421    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12422
12423commit 6783544706ff370e900c137f951e90230586d6dc
12424Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12425Date:   Mon Jul 12 22:03:18 2010 -0700
12426
12427    specs/libX11: Appendix C: convert literallayouts to synopsis or programlisting
12428    
12429    Fixes display of a bunch of function prototypes and sample code
12430    
12431    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12432
12433commit 8834cd9285e6aaf437aee56292c508d99a090fa9
12434Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12435Date:   Mon Jul 12 18:37:16 2010 -0700
12436
12437    xmlrules.in: Add chunked-html rule to manually generate smaller files
12438    
12439    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12440
12441commit 53f78680c34ebc94ce26b5e14c2a0003435cf10d
12442Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12443Date:   Mon Jul 12 18:35:59 2010 -0700
12444
12445    specs/libX11: Convert Appendix A tables to real tables
12446    
12447    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12448
12449commit 705a1257a8a47f4a1f94979aee09a837500ede7d
12450Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12451Date:   Mon Jul 12 18:38:09 2010 -0700
12452
12453    specs/libX11: Fix out-of-place text in Chapter 1
12454    
12455    DTD violation error introduced in 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4
12456    
12457    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12458
12459commit a151346a2b7810e988f1de4b1e00b37672dc587a
12460Author: Jeremy Huddleston <jeremyhu@apple.com>
12461Date:   Mon Jul 12 16:52:12 2010 -0700
12462
12463    launchd: Explicitly search /sbin
12464    
12465    Previously, launchd wasn't found if /sbin wasn't in the user's PATH.
12466    https://bugs.freedesktop.org/show_bug.cgi?id=29028
12467    
12468    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12469
12470commit 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4
12471Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12472Date:   Mon Jul 12 13:38:04 2010 -0700
12473
12474    Replace untranslated .hN macros with <filename> tags to show header names
12475    
12476    Translation performed by:
12477     perl -i -p -e 's{\<\!-- .hN (\S+) (\S+)?\s*-->}
12478                     {&lt;<filename class="headerfile">$1</filename>&gt;$2}' *.xml
12479    
12480    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12481
12482commit 75480440946603b8efdbbf78b88d59c641b6d2c8
12483Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12484Date:   Sat Jul 10 00:06:13 2010 -0700
12485
12486    specs/libX11: Mass substitution of <function> tags with more specific tags
12487    
12488    Matched names from X headers & "nm libX11.so" output to names in spec to
12489    map to more specific tags.   Tags used:
12490    
12491                        <code>   code fragments
12492                    <constant>   enum values
12493                   <errorname>   X protocol errors
12494                    <filename>   filenames
12495                    <function>   functions
12496                    <function>   function-like macros (#define foo(a,b)...)
12497                 <returnvalue>   function return codes
12498                 <structfield>   struct members
12499                  <structname>   struct names (even when typedefed)
12500                      <symbol>   simple value #defines (#define NAME value)
12501                  <systemitem>   X protocol requests
12502    <systemitem class="event">   X protocol events
12503                        <type>   non-struct typedefs
12504                     <varname>   global variables
12505    
12506    (Also fixed a couple typos detected by failures of this matching, such as
12507     XESSetPrintErrorValues -> XESetPrintErrorValues.)
12508    
12509    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12510
12511commit 26f4f0d50840fe5ba4c46aae0a8e68db0059434b
12512Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12513Date:   Fri Jul 9 00:37:23 2010 -0700
12514
12515    specs/libX11: Glossary cross-reference links
12516    
12517    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12518
12519commit 930b52d84031de0e15e0a11cf4ecfd3b2f59e073
12520Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12521Date:   Thu Jul 8 21:19:08 2010 -0700
12522
12523    specs/libX11: Make whitespace around <function> tags more uniform
12524    
12525    Simplifies regular expressions for further mass substitutions.
12526    
12527    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12528
12529commit 730ce6b8aca77b77385aa894d234cbde5fcbce37
12530Author: Gaetan Nadon <memsize@videotron.ca>
12531Date:   Fri Jul 9 20:10:37 2010 -0400
12532
12533    specs: use pattern rules rather than suffix rules
12534    
12535    This allows target to rebuild when included .xml files are changed.
12536    
12537    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12538
12539commit 5527b4bc8ce7de60123eb28789ddef54aa48a378
12540Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12541Date:   Thu Jul 8 20:56:41 2010 -0700
12542
12543    specs/libX11: Move punctuation outside of the <function> tags
12544    
12545    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12546
12547commit d66d2134dfc38cd866c7c2d9ea45ed4b4dad23f0
12548Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12549Date:   Thu Jul 8 20:42:50 2010 -0700
12550
12551    specs/libX11: Mark a number of <acronym>s
12552    
12553    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12554
12555commit d5bbb12f55ac9f691eab6242cedc53207275131b
12556Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12557Date:   Thu Jul 8 19:31:36 2010 -0700
12558
12559    specs/libX11: Glossary terms should not be marked as functions
12560    
12561    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12562
12563commit 5decf7bc5dae0ae0c45c47eaaa9c4fdf9515ca1c
12564Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12565Date:   Thu Jul 8 19:28:18 2010 -0700
12566
12567    specs/libX11: Clean up author list & acknowledgements a little
12568    
12569    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12570
12571commit 0cd29bdb332c333e5123dce65b25ad1d97fbdae3
12572Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12573Date:   Thu Jul 8 17:30:34 2010 -0700
12574
12575    specs/libX11: Convert troff .IN macros to docbook <indexterm> tags
12576    
12577    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12578
12579commit a67d99ccbe22c8ce5f7a12f13b8991d9e0cf4491
12580Author: Gaetan Nadon <memsize@videotron.ca>
12581Date:   Thu Jul 8 17:29:21 2010 -0400
12582
12583    specs: specdir is required to install xml files
12584    
12585    The source is installed as well as the targets.
12586    This failed when configuring with --without-xmlto
12587    
12588    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12589
12590commit e14273f44c1501ab51db4adcb83b18a1073787d8
12591Author: Matt Dew <matt@osource.org>
12592Date:   Thu Jul 8 14:42:32 2010 -0400
12593
12594    specs: replace troff source with docbook-xml source
12595    
12596    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12597
12598commit bea0873caf50e9ed1b89255775d9ab912cbecd45
12599Author: Daniel Stone <daniel@fooishbar.org>
12600Date:   Thu Jul 8 16:49:51 2010 +0100
12601
12602    XStringToKeysym: Check strdup() return value
12603    
12604    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12605    Reviewed-by: Keith Packard <keithp@keithp.com>
12606
12607commit cffa71b4a5743e2b6675b9a917d15aef4177d513
12608Author: Marko Myllynen <myllynen@redhat.com>
12609Date:   Mon Jun 28 15:08:05 2010 +0300
12610
12611    Fix two typos in SFS 5966 Annex 3
12612    
12613    X.Org bug#28792 <https://bugs.freedesktop.org/show_bug.cgi?id=28792>
12614    
12615    Signed-off-by: Julien Cristau <jcristau@debian.org>
12616
12617commit eb023c0f8919e809b8b609e1467b14d20a290aa7
12618Author: Daniel Stone <daniel@fooishbar.org>
12619Date:   Tue Jun 15 18:49:43 2010 +0100
12620
12621    Delete now-redundant XKeysymDB
12622    
12623    Since XStringToKeysym now supports all the vendor keysyms, just delete
12624    our XKeysymDB, which was incomplete at best, misleading at worst, and
12625    always an annoyance.
12626    
12627    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12628    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12629
12630commit 00175397480b76d32bf82b0c7c94c91a2a95954e
12631Author: Daniel Stone <daniel@fooishbar.org>
12632Date:   Tue Jun 15 18:48:48 2010 +0100
12633
12634    makekeys: Scan vendor keysyms as well as core
12635    
12636    Since we can't really live without vendor keysyms, scan them all in to
12637    generate ks_tables.h, rather than only doing the core ones, and leaving
12638    the vendor syms to be manually synchronised with XKeysymDB.
12639    
12640    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12641    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12642
12643commit ebd6ef0a4db0ddef0ae17ad14571518ccdeea5ba
12644Author: Daniel Stone <daniel@fooishbar.org>
12645Date:   Tue Jun 15 18:47:37 2010 +0100
12646
12647    XStringToKeysym: Special case for XF86 keysyms
12648    
12649    Some XFree86 keysyms were in XKeysymDB as XF86_foo, despite really being
12650    XF86foo.  So, if we get to the bottom of XStringToKeysym and haven't
12651    found our XF86_foo, try it again as XF86foo.
12652    
12653    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12654    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12655
12656commit 8c2ffce9e5c6eef8a04f7e2732db46b3b62e13bb
12657Author: Daniel Stone <daniel@fooishbar.org>
12658Date:   Tue Jun 15 17:20:48 2010 +0100
12659
12660    configure.ac: Change from deprecated AC_DEFINE_DIR to AX_
12661    
12662    AC_DEFINE_DIR is deprecated as it's squatting on the autoconf-builtin
12663    namespace, so start using the more proper AX_DEFINE_DIR instead.
12664    
12665    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
12666    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
12667    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12668
12669commit 6de368c9aa7ccd2fcd62fca5a2b278913db4d03d
12670Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
12671Date:   Thu Jul 1 06:50:47 2010 -0300
12672
12673    Purge macros NEED_EVENTS and NEED_REPLIES
12674    
12675    Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
12676    Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
12677    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12678
12679commit 0b724231be17f19538ee115a03b82b221b1f83c0
12680Author: Adam Jackson <ajax@redhat.com>
12681Date:   Wed Jun 30 16:30:49 2010 -0400
12682
12683    Use -version-number consistently instead of -version-info
12684    
12685    The latter is libtool braindamage.
12686    
12687    Signed-off-by: Adam Jackson <ajax@redhat.com>
12688
12689commit 241a990afcf9c967c587ad6fd245df21b68b1ad9
12690Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
12691Date:   Tue Jun 29 18:32:42 2010 +0200
12692
12693    Make Compose-comma map to ogonek for I and U in UTF-8 locales.
12694    
12695    With the preceding patch, it makes the UTF-8 compose map consistent with
12696    the ISO 8859-4 and -13 maps.
12697
12698commit af55e582f485a668c2bf43129be972bc65f03c60
12699Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
12700Date:   Tue Jun 29 18:20:18 2010 +0200
12701
12702    Make Compose-comma map to Ogonek for A and E in UTF-8 locales.
12703    
12704    This makes the UTF-8 compose file consistent with the ISO 8859-2 compose file
12705    as far as Polish is concerned.  We only sacrifice one pair of characters,
12706    e-cedilla, which is not used in any language.
12707
12708commit d6ba13009fc9ef876a104b907ffef73c6e405d4b
12709Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12710Date:   Thu Jun 24 13:13:11 2010 -0700
12711
12712    Define FILE_MAN_DIR_SUFFIX so XCompose shadow page has correct path
12713    
12714    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12715
12716commit 978c7238789037de917b9423ea9adfb006da1260
12717Author: Jamey Sharp <jamey@minilop.net>
12718Date:   Sat Jun 19 10:44:55 2010 -0700
12719
12720    poll_for_response: Really handle xcb_poll_for_reply getting a reply.
12721    
12722    Don't lose async replies. That's bad.
12723    
12724    `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the
12725    _XReply path worked; that path waited for replies, rather than polling.
12726    
12727    However, XRecordProcessReplies, which does nothing but call XPending,
12728    relied on the event-handling path to process async replies, and that was
12729    busted.
12730    
12731    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595
12732    
12733    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12734
12735commit ef67486c5a8aada24fd95b2991a3c4979f53805f
12736Author: Jamey Sharp <jamey@minilop.net>
12737Date:   Mon Jun 21 13:34:15 2010 -0700
12738
12739    Revert "xcb_io.c: poll_for_response doesn't guarantee there's a pending request."
12740    
12741    This reverts commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e, because as
12742    a matter of fact, if poll_for_response returns NULL when we know
12743    dpy->xcb->next_event is non-NULL, there *is* guaranteed to be a pending
12744    request.
12745
12746commit 301ec5b41e9d253a446db656e1789ac7345dc433
12747Author: Jamey Sharp <jamey@minilop.net>
12748Date:   Mon Jun 21 13:24:24 2010 -0700
12749
12750    Revert "poll_for_response: Really handle xcb_poll_for_reply getting a reply."
12751    
12752    This reverts commit c115095d7f2bc4f5a4fb26380e3698fefdad7611. We *do*
12753    need to check poll_for_event every time through the loop in
12754    poll_for_response, so the commit did too much.
12755
12756commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e
12757Author: Jamey Sharp <jamey@minilop.net>
12758Date:   Mon Jun 21 10:54:02 2010 -0700
12759
12760    xcb_io.c: poll_for_response doesn't guarantee there's a pending request.
12761    
12762    Fixes the second bug reported in:
12763    https://bugs.freedesktop.org/show_bug.cgi?id=28595
12764    
12765    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12766
12767commit c115095d7f2bc4f5a4fb26380e3698fefdad7611
12768Author: Jamey Sharp <jamey@minilop.net>
12769Date:   Sat Jun 19 10:44:55 2010 -0700
12770
12771    poll_for_response: Really handle xcb_poll_for_reply getting a reply.
12772    
12773    Don't lose async replies. That's bad.
12774    
12775    `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the
12776    _XReply path worked; that path waited for replies, rather than polling.
12777    
12778    However, XRecordProcessReplies, which does nothing but call XPending,
12779    relied on the event-handling path to process async replies, and that was
12780    busted.
12781    
12782    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595
12783    
12784    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12785    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12786
12787commit a25ae169862ab9b76daf259613b37c6b07bc2ef2
12788Author: Marko Myllynen <myllynen@redhat.com>
12789Date:   Sun Jun 13 19:23:05 2010 +0300
12790
12791    Implement SFS 5966 Annex 4 for Finland
12792    
12793    This patch adds Annex 4 of SFS 5966 for Finland and fixes two typos in
12794    Annex 3 (which were actually copied verbatim from the standard text).
12795    
12796    The implementation of the standard is now complete.
12797    
12798    X.Org bug#28498 <https://bugs.freedesktop.org/show_bug.cgi?id=28498>
12799    
12800    Signed-off-by: Julien Cristau <jcristau@debian.org>
12801
12802commit 47b04195d8a31c8f9e6dd804196162c6cfca3ac6
12803Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12804Date:   Fri Jun 4 13:20:17 2010 -0700
12805
12806    LINEAR_RGB_InitSCCData: When malloc fails, don't try to free unallocated bits
12807    
12808    One of the malloc failure checks had a goto to the wrong spot in the
12809    list of cleanup free() calls to unwind at the end, and was freeing
12810    bits that hadn't been initialized/allocated yet, since they would be
12811    stored in the struct that just failed to be allocated.
12812    
12813    Error: Null pointer dereference (CWE 476)
12814       Read from pointer that could be constant 'NULL'
12815            at line 805 of /export/alanc/X.Org/sx86/lib/libX11/src/xcms/LRGB.c in function 'LINEAR_RGB_InitSCCData'.
12816              Pointer checked against constant 'NULL' at line 754 but does not protect the dereference.
12817    
12818    [ This bug was found by the Parfait bug checking tool.
12819      For more information see http://research.sun.com/projects/parfait ]
12820    
12821    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12822
12823commit aebbf3623888119b43893b253195d93b613a6e1e
12824Author: Josh Triplett <josh@joshtriplett.org>
12825Date:   Sun Jun 6 19:11:55 2010 -0700
12826
12827    Mark the rest of Xprivate.h as _X_HIDDEN.
12828    
12829    None of the functions in Xprivate.h should have any callers outside of
12830    Xlib, by definition.
12831    
12832    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12833    Suggested-by: Jamey Sharp <jamey@minilop.net>
12834    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12835
12836commit abcd1b67c8135ad652833e5a60d818837c39fce1
12837Author: Julien Cristau <jcristau@debian.org>
12838Date:   Fri Jun 4 13:57:45 2010 +0200
12839
12840    Hide _XSeqSyncFunction
12841    
12842    Commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43 made _XSeqSyncFunction
12843    non-static, but we don't need to export it.
12844    
12845    Signed-off-by: Julien Cristau <jcristau@debian.org>
12846    Reviewed-by: Adam Jackson <ajax@redhat.com>
12847    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12848
12849commit 15e5eaf62897b3179d1fbe457cb19f886f0449f8
12850Author: Josh Triplett <josh@joshtriplett.org>
12851Date:   Thu Jun 3 09:41:01 2010 -0700
12852
12853    Remove support for building without XCB
12854    
12855    And there was much rejoicing.
12856    
12857    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12858    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12859    
12860    Consensus on #xorg-devel agrees with removing --without-xcb; in
12861    particular, acks from Adam Jackson, Daniel Stone, Kristian Høgsberg,
12862    Julien Cristau, and Rémi Cardona.
12863
12864commit fb19eb767a32fd4ff74021c51bc6f60eb8bdff74
12865Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12866Date:   Thu Jun 3 15:21:40 2010 -0700
12867
12868    libX11 1.3.4
12869    
12870    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12871
12872commit 8f3db40ca5108a919244f3fff7466d01a14b3ce2
12873Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12874Date:   Thu Jun 3 13:08:17 2010 -0700
12875
12876    Workaround bug in groff flag processing that breaks distcheck
12877    
12878    At least with the groff 1.19.2 package I have installed, groff passes
12879    on the -I flags for the include path to grohtml, which if they come
12880    after the -P-I... flag we pass to grohtml to specify the image file
12881    name pattern cause it to override that flag and put the images in
12882    the wrong place, breaking "make distcheck" - changing the flag order
12883    works around this.
12884    
12885    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12886
12887commit bdb31a1fb707cebccc9efbbf68cb55c10fa4ea3e
12888Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12889Date:   Thu Jun 3 11:25:33 2010 -0700
12890
12891    Fix typo that made configure always report "none" for man page suffix
12892    
12893    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12894
12895commit 6bb7d2d21d6b58019fb37b724b59c9744f30b9ce
12896Author: Paul Bender <pebender@gmail.com>
12897Date:   Thu Jun 3 11:20:26 2010 -0700
12898
12899    Bug 22591 - configure does not obey the --enable-*-transport options
12900    
12901    https://bugs.freedesktop.org/show_bug.cgi?id=22591
12902    
12903    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12904
12905commit d7f35946d673bb00f2768e025b6e1017422b2174
12906Author: Paul Bender <pebender@gmail.com>
12907Date:   Thu Jun 3 11:17:45 2010 -0700
12908
12909    Bug 22590 - libX11 1.2.1 has broken abstract namespace support
12910    
12911    https://bugs.freedesktop.org/show_bug.cgi?id=22590
12912    
12913    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12914
12915commit b2487d07f7b355f230a56e32c763acd08862735c
12916Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12917Date:   Thu Jun 3 11:06:15 2010 -0700
12918
12919    Bug 22584 - libX11 does not cross compile
12920    
12921    Adaptation of patch submitted by Paul Bender in attachment 27301 to
12922    https://bugs.freedesktop.org/show_bug.cgi?id=22584
12923    
12924    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12925    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
12926    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
12927
12928commit 4378219ff8c1418418c70086085358b69d574e74
12929Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12930Date:   Tue Jun 1 18:55:48 2010 -0700
12931
12932    Clarify requirements in XRestackWindows man page
12933    
12934    The required common parent window is not specified in the arguments,
12935    just implied as the parent of the first window in the list.
12936    
12937    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12938
12939commit f09c5299a381e2729e800a0ac43f1c0e371f65f6
12940Author: Jeremy Huddleston <jeremyhu@apple.com>
12941Date:   Wed May 12 16:42:18 2010 -0700
12942
12943    xcb: Add TCP fallback
12944    
12945    If we fail to connect to a UNIX socket and the transport isn't specified,
12946    fallback on TCP.  This matches behavior with the xtrans codepath and the
12947    Xlib spec.
12948    
12949    http://lists.x.org/archives/xorg-devel/2010-April/007915.html
12950    
12951    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12952    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12953
12954commit fd82552d5c0ce1931f29006a0c36f5e03cf8577e
12955Merge: aae2a4a7 933aee1d
12956Author: Jamey Sharp <jamey@minilop.net>
12957Date:   Mon May 10 16:51:24 2010 -0700
12958
12959    Merge branch 'xlib-xcb-thread-fixes'
12960
12961commit aae2a4a7aab26de3fa715d6ecd0a0e0926b37fc9
12962Author: Jeremy Huddleston <jeremyhu@apple.com>
12963Date:   Fri Apr 23 21:50:29 2010 -0700
12964
12965    Don't append the screen number when using a launchd socket
12966    
12967    ssh gets confused by this.  XQuartz is the only DDX using this
12968    functionality, and it doesn't support different screens, so
12969    let's just not include this until most ssh know how to handle
12970    this.
12971    
12972    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12973
12974commit adcd0ec209abf832a769d52db660fb37eaad6e0c
12975Author: Jeremy Huddleston <jeremyhu@apple.com>
12976Date:   Fri Apr 23 16:33:44 2010 -0700
12977
12978    Remove launchd logic from _XConnectXCB as it's handled in XCB
12979    
12980    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12981
12982commit d232b259c36fdde1f4179822809fec1480867dc5
12983Author: Jeremy Huddleston <jeremyhu@apple.com>
12984Date:   Tue Feb 2 17:01:28 2010 -0800
12985
12986    Fix various build warnings
12987    
12988    imLcIm.c: In function '_XimCachedFileName':
12989    imLcIm.c:361: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int'
12990    imLcIm.c:364: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int'
12991    
12992    imRm.c: In function '_XimDefaultArea':
12993    imRm.c:597: warning: cast from pointer to integer of different size
12994    imRm.c: In function '_XimDefaultColormap':
12995    imRm.c:626: warning: cast from pointer to integer of different size
12996    
12997    lcFile.c:224: warning: no previous prototype for 'xlocaledir'
12998    
12999    lcUTF8.c: In function 'iconv_cstombs':
13000    lcUTF8.c:1841: warning: assignment discards qualifiers from pointer target type
13001    lcUTF8.c:1869: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
13002    lcUTF8.c:1873: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
13003    lcUTF8.c: In function 'iconv_mbstocs':
13004    lcUTF8.c:1935: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
13005    lcUTF8.c: In function 'iconv_mbtocs':
13006    lcUTF8.c:2031: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
13007    lcUTF8.c: In function 'iconv_mbstostr':
13008    lcUTF8.c:2121: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
13009    lcUTF8.c: In function 'iconv_strtombs':
13010    lcUTF8.c:2180: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
13011    lcUTF8.c: In function '_XlcAddGB18030LocaleConverters':
13012    lcUTF8.c:2367: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13013    lcUTF8.c:2368: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13014    lcUTF8.c:2373: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13015    lcUTF8.c:2374: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13016    lcUTF8.c:2375: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13017    lcUTF8.c:2376: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13018    lcUTF8.c:2377: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
13019    
13020    XlibInt.c: In function '_XGetHostname':
13021    XlibInt.c:3441: warning: implicit declaration of function 'gethostname'
13022    XlibInt.c:3441: warning: nested extern declaration of 'gethostname'
13023    
13024    ConnDis.c: In function '_XDisconnectDisplay':
13025    ConnDis.c:540: warning: old-style function definition
13026    ConnDis.c: In function '_XSendClientPrefix':
13027    ConnDis.c:554: warning: old-style function definition
13028    ConnDis.c: In function 'XSetAuthorization':
13029    ConnDis.c:677: warning: old-style function definition
13030    
13031    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
13032
13033commit 3e11c73187acb2e2be9a812840bbbea947527ccb
13034Author: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
13035Date:   Fri Apr 23 00:01:54 2010 +0200
13036
13037    Allow X11 users to compose anarchism
13038    
13039    I can easily compose the hammer and sickle (the symbol representative
13040    of communism) by pressing the key sequence:
13041    
13042      <Compose> <C> <C> <C> <P> -> ☭
13043    
13044    But i can't easily make the circled A (the symbol representative of
13045    anarchism).
13046    
13047    I'd like to be able to use <Compose> <O> <A> (this is a
13048    currently unused key sequence) to generate the symbol:  Ⓐ
13049    
13050    Debian bug#555938 <http://bugs.debian.org/555938>
13051    
13052    Signed-off-by: Julien Cristau <jcristau@debian.org>
13053
13054commit 933aee1d5c53b0cc7d608011a29188b594c8d70b
13055Author: Jamey Sharp <jamey@minilop.net>
13056Date:   Fri Apr 16 20:18:28 2010 -0700
13057
13058    Fix Xlib/XCB for multi-threaded applications (with caveats).
13059    
13060    Rather than trying to group all response processing in one monolithic
13061    process_responses function, let _XEventsQueued, _XReadEvents, and
13062    _XReply each do their own thing with a minimum of code that can all be
13063    reasoned about independently.
13064    
13065    Tested with `ico -threads 20`, which seems to be able to make many
13066    icosahedrons dance at once quite nicely now.
13067    
13068    Caveats:
13069    
13070    - Anything that was not thread-safe in Xlib before XCB probably still
13071      isn't. XListFontsWithInfo, for instance.
13072    
13073    - If one thread is waiting for events and another thread tries to read a
13074      reply, both will hang until an event arrives. Previously, if this
13075      happened it might work sometimes, but otherwise would trigger either
13076      an assertion failure or a permanent hang.
13077    
13078    - Versions of libxcb up to and including 1.6 have a bug that can cause
13079      xcb_wait_for_event or xcb_wait_for_reply to hang if they run
13080      concurrently with xcb_writev or other writers. So you'll want that fix
13081      as well.
13082    
13083    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13084    Reviewed-by: Josh Triplett <josh@freedesktop.org>
13085
13086commit aab43278ae619eb57d2dd9c7396f460f078588fc
13087Author: Jamey Sharp <jamey@minilop.net>
13088Date:   Fri Apr 16 22:24:16 2010 -0700
13089
13090    Use InternalLockDisplay on code paths called from LockDisplay.
13091    
13092    It's easier to reason about the code when we can't re-enter the
13093    Xlib-private sync-handlers while they're already running.
13094    
13095    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13096    Reviewed-by: Josh Triplett <josh@freedesktop.org>
13097
13098commit 660b7d05f4ca4ab4661c9fe7ce655a4909b4e556
13099Author: Jamey Sharp <jamey@minilop.net>
13100Date:   Fri Apr 16 19:45:11 2010 -0700
13101
13102    Fix _XSend to enqueue the right range of pending requests.
13103    
13104    _XSend was off-by-one on both ends. It should not re-enqueue the last
13105    request that was already flushed, but it should enqueue the last request
13106    currently being flushed.
13107    
13108    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13109    Reviewed-by: Josh Triplett <josh@freedesktop.org>
13110
13111commit f2735889908d6e5a7f8dbee42f00c54a52665191
13112Author: Jamey Sharp <jamey@minilop.net>
13113Date:   Fri Apr 16 19:45:11 2010 -0700
13114
13115    Pending requests are always added in-order.
13116    
13117    Replace insert_pending_request, which did an in-order search of the
13118    queue to find the right insertion point, with a simpler
13119    append_pending_request, and use that in _XSend as well.
13120    
13121    Includes assertions to check that the list of pending requests is in
13122    order by sequence number and does not have duplicates.
13123    
13124    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13125    Reviewed-by: Josh Triplett <josh@freedesktop.org>
13126
13127commit d9cf5885b0f97942fbbd2a7cc50118132ece50f6
13128Author: Markus Duft <markus.duft@salomon.at>
13129Date:   Fri Apr 16 08:39:34 2010 -0700
13130
13131    Bug 26839: Fix build problem on Interix (POSIX layer on Windows)
13132    
13133    https://bugs.freedesktop.org/show_bug.cgi?id=26839
13134    
13135    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13136
13137commit a3f5f1b90936d23e9894e3261b2d77fb7b32a51a
13138Author: Josh Triplett <josh@freedesktop.org>
13139Date:   Thu Apr 15 14:24:21 2010 -0700
13140
13141    Stop returning an int from _XIDHandler and _XSeqSyncFunction
13142    
13143    _XIDHandler and _XSeqSyncFunction originally ran from dpy->synchandler, and
13144    thus had to return an int.  Now, they only run from _XPrivSyncHandler or
13145    LockDisplay, neither of which needs to check their return value since they
13146    always returned 0.  Make them both void.
13147    
13148    Signed-off-by: Josh Triplett <josh@freedesktop.org>
13149    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13150
13151commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43
13152Author: Jamey Sharp <jamey@minilop.net>
13153Date:   Thu Apr 15 13:05:08 2010 -0700
13154
13155    Move XID and sync handling from SyncHandle to LockDisplay to fix races.
13156    
13157    XID and sync handling happened via _XPrivSyncHandler, assigned to
13158    dpy->synchandler and called from SyncHandle.  _XPrivSyncHandler thus ran
13159    without the Display lock, so manipulating the Display caused races, and
13160    these races led to assertions in multithreaded code (demonstrated via
13161    ico).
13162    
13163    In the XTHREADS case, after you've called XInitThreads, we can hook
13164    LockDisplay and UnlockDisplay.  Use that to run _XIDHandler and
13165    _XSeqSyncHandler from LockDisplay rather than SyncHandle; we then know
13166    that we hold the lock, and thus we can avoid races.  We think it makes
13167    sense to do these both from LockDisplay rather than UnlockDisplay, so
13168    that you know you have valid sync and a valid XID before you start
13169    setting up the request you locked to prepare.
13170    
13171    In the !XTHREADS case, or if you haven't called XInitThreads, you don't
13172    get to use Xlib from multiple threads, so we can use the logic we have
13173    now (with synchandler and savedsynchandler) without any concern about
13174    races.
13175    
13176    This approach gets a bit exciting when the XID and sequence sync
13177    handlers drop and re-acquire the Display lock. Reacquisition will re-run
13178    the handlers, but they return immediately unless they have work to do,
13179    so they can't recurse more than once.  In the worst case, if both of
13180    them have work to do, we can nest the Display lock three deep.  In the
13181    case of the _XIDHandler, we drop the lock to call xcb_generate_id, which
13182    takes the socket back if it needs to request more XIDs, and taking the
13183    socket back will reacquire the lock; we take care to avoid letting
13184    _XIDHandler run again and re-enter XCB from the return_socket callback
13185    (which causes Very Bad Things, and is Not Allowed).
13186    
13187    Tested with ico (with 1 and 20 threads), and with several test programs
13188    for XID and sequence sync.  Tested with and without XInitThreads(), and
13189    with and without XCB.
13190    
13191    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=23192
13192    
13193    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13194    Signed-off-by: Josh Triplett <josh@freedesktop.org>
13195
13196commit b089b53b697c2851db2985d32af3b29f1da5e31e
13197Author: Jamey Sharp <jamey@minilop.net>
13198Date:   Wed Apr 14 12:59:11 2010 -0700
13199
13200    Honest. Extensions get to filter async errors too.
13201    
13202    Under some circumstances, errors are picked up by calling
13203    xcb_poll_for_reply, rather than xcb_poll_for_event, because Xlib issued
13204    the requests with the XCB_REQUEST_CHECKED flag. That happens when either
13205    an async handler is queued at the time the requests are flushed, or when
13206    XSetEventQueueOwner has been used to prevent Xlib from processing XCB's
13207    event queue.
13208    
13209    This bugfix extends 405132dab64bf2375f8e57d02b1b53da2311933a to cover
13210    those cases.
13211    
13212    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545
13213    
13214    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13215
13216commit a15c31274650e391bc6de5d0951eb4464c228139
13217Author: Jamey Sharp <jamey@minilop.net>
13218Date:   Wed Apr 14 12:10:34 2010 -0700
13219
13220    _XError already runs async handlers; only call them directly for replies.
13221    
13222    The previous behavior probably would have triggered bug reports someday.
13223    
13224    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13225
13226commit 405132dab64bf2375f8e57d02b1b53da2311933a
13227Author: Jamey Sharp <jamey@minilop.net>
13228Date:   Tue Apr 13 12:12:36 2010 -0700
13229
13230    Extensions get to filter async errors too.
13231    
13232    Apparently I misread XlibInt.c:_XReply and thought that handlers set
13233    with XESetError should be consulted only for the sequence number that
13234    _XReply is currently looking for. In fact, the error handlers are also
13235    consulted when an error arrives for a request that was not expected to
13236    have a reply.
13237    
13238    However, in an odd twist, the error handlers are *not* consulted outside
13239    of _XReply--that is, when looking for events, or waiting to be able to
13240    flush the output queue. So this patch takes some small pains to preserve
13241    that behavior, though it seems likely to have been unintentional.
13242    
13243    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545
13244    
13245    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13246
13247commit d3eab4a38f3e53ef21eb3b6fa66ead7afecf6227
13248Author: Jamey Sharp <jamey@minilop.net>
13249Date:   Tue Apr 13 12:49:59 2010 -0700
13250
13251    Prefer the xcb_generic_error_t we already have over casting to xEvent.
13252    
13253    Just a minor cleanup.
13254    
13255    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13256
13257commit 75ea8c37935ccc911557d16a303ba595b8ab106b
13258Author: Jamey Sharp <jamey@minilop.net>
13259Date:   Mon Apr 12 11:30:20 2010 -0700
13260
13261    Run the user's synchandler as well as any internal synchandlers.
13262    
13263    Fixes https://bugs.freedesktop.org/show_bug.cgi?id=27595
13264    
13265    Signed-off-by: Jamey Sharp <jamey@minilop.net>
13266
13267commit c3f3e4a9e531d010312c97e753d6e543e607094d
13268Author: Kusanagi Kouichi <slash@ac.auone-net.jp>
13269Date:   Fri Feb 12 17:16:10 2010 +0900
13270
13271    Fix manual typos.
13272    
13273    Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
13274    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13275
13276commit 75c6c5595408caba9ef8233839d77021d02f86f2
13277Author: Dirk Wallenstein <dirkwallenstein@t-online.de>
13278Date:   Fri Feb 5 20:42:46 2010 +0100
13279
13280    man: Correct the XkbAllAccessXEventsMask mask name
13281    
13282    This mask has probably been renamed but not been updated in the manuals.
13283    
13284    Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de>
13285    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13286
13287commit 9262aceaff24c954cab8c252690032bc2ec25e7d
13288Author: Dirk Wallenstein <dirkwallenstein@t-online.de>
13289Date:   Fri Feb 5 20:42:45 2010 +0100
13290
13291    man: Add missing geometry component flag
13292    
13293    Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de>
13294    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13295
13296commit 50cf134465f6fd263f37acecac94518468ef1c6d
13297Author: Dirk Wallenstein <halsmit@t-online.de>
13298Date:   Tue Feb 2 21:24:32 2010 +0100
13299
13300    man: Return value of XkbGetState is Status and not Bool
13301    
13302    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
13303    Reviewed-by: Julien Cristau <jcristau@debian.org>
13304    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13305
13306commit 7a93ae2bb5292a97fdd27ed818bc81248b37f641
13307Author: Dirk Wallenstein <halsmit@t-online.de>
13308Date:   Tue Feb 2 21:24:30 2010 +0100
13309
13310    man: Fix return value specification of XkbKeyActionEntry
13311    
13312    The XkbKeyActionEntry macro expands to a pointer.
13313    
13314    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
13315    Reviewed-by: Julien Cristau <jcristau@debian.org>
13316    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13317
13318commit e9884d4a05a5661ec343ea8a2aa0562b6419e086
13319Author: Dirk Wallenstein <halsmit@t-online.de>
13320Date:   Tue Feb 2 21:24:29 2010 +0100
13321
13322    man: Redirect users from XKeycodeToKeysym to XkbKeycodeToKeysym #25732
13323    
13324    XKeycodeToKeysym keeps compatibility with pre-XKB and thus only sees 2
13325    groups with 2 levels each. It wraps the index into the next group.
13326    This behavior confuses the unaware user, and therefore this will add a
13327    reference to XkbKeycodeToKeysym in the corresponding man paragraph.
13328    
13329    Another bug had that issue, too. #5349
13330    
13331    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
13332    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13333
13334commit 6474cdf97cd648afaeaa872155f3fa21cfb43f80
13335Author: Alan Coopersmith <alan.coopersmith@oracle.com>
13336Date:   Thu Apr 8 19:59:55 2010 -0700
13337
13338    Fix typo in new fi_FI.UTF-8 that was reported by "make check"
13339    
13340    Unrecognized pattern in Compose on line #154:
13341      <dead_diaeresis> <space>                " "¨"
13342    
13343    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
13344
13345commit 91bd5e7af2d3a9526aefb73dc10f417896f0e4dd
13346Author: Marko Myllynen <myllynen@redhat.com>
13347Date:   Thu Apr 8 19:52:42 2010 -0700
13348
13349    Bug 27465 - Rewritten fi_FI.UTF-8 Compose file
13350    
13351    I've rewritten the fi_FI.UTF-8 Compose file so that it
13352    
13353    - includes en_US.UTF-8/Compose for base compose definitions (and thus gets any
13354    possible additions to en_US.UTF-8/Compose automatically included)
13355    - overrides any en_US.UTF-8/Compose definitions with fi_FI.UTF-8/Compose
13356    definitions (thus making it safe to include en_US.UTF-8/Compose)
13357    - lists all the definitions specified in the SFS 5966 standard (thus making it
13358    easy to verify that the implementation is valid and complete)
13359    - adds a header about the file itself as required by the standard
13360    
13361    The rewritten version is ~320 lines compared to over 5000 lines of the previous
13362    version.
13363    
13364    https://bugs.freedesktop.org/show_bug.cgi?id=27465
13365
13366commit f6a4fd0c7615684d08e848245849dea4017a5214
13367Author: Gaetan Nadon <memsize@videotron.ca>
13368Date:   Thu Apr 1 21:46:12 2010 -0400
13369
13370    config: update and relocate AC_DEFINE_DIR macro
13371    
13372    Remove deprecated acinclude.m4 macro container file
13373    Use separate macro files as per autoconf recommendation
13374    Use the latest macro from GNU (ax) which replaces
13375    the non-gnu version (ac)
13376    This preserves the Autoconf macro AC namespace.
13377    
13378    Also moved out of acinclude.m4 is the DOLT macro
13379    
13380    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13381
13382commit 61d5231db418cec51bd718633c3bba39b18689d6
13383Author: Gaetan Nadon <memsize@videotron.ca>
13384Date:   Mon Mar 29 14:53:48 2010 -0400
13385
13386    config: remove the pkgconfig pc.in file from EXTRA_DIST
13387    
13388    Automake always includes it in the tarball.
13389    
13390    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13391
13392commit 69fc76a0321c6b25cc4286173c53435c69ab8e0e
13393Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
13394Date:   Wed Mar 17 23:53:16 2010 -0300
13395
13396    No need for req->firstKeyCode to be set twice.
13397    
13398    There is no reason to set req->firstKeyCode twice when a client, wishful
13399    for changing keyboard mappings, calls XChangeKeyboardMapping. This patch
13400    fixes the mistake by making no functional changes to libX11 whatsoever.
13401    
13402    Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
13403    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13404
13405commit b5797597f83ee62228abfb88f026ef92163e07a5
13406Author: Gaetan Nadon <memsize@videotron.ca>
13407Date:   Tue Feb 16 10:37:21 2010 -0500
13408
13409    config: move CWARNFLAGS from configure.ac to Makefile.am
13410    
13411    Compiler warning flags should be explicitly set in the makefile
13412    rather than being merged with other packages compiler flags.
13413    
13414    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
13415    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13416
13417commit 052c4d7995b0e52a038933beb85cd544571c6b7e
13418Author: Gaetan Nadon <memsize@videotron.ca>
13419Date:   Tue Feb 16 14:28:21 2010 -0500
13420
13421    specs: change install cmd due to automake 1.11
13422    
13423    specData_INSTALL is defined in 1.9 and 1.10 but not 1.11
13424    
13425    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13426
13427commit 7277a18cda9f5a6807d11cf4ceb9449b1c302c18
13428Author: Gaetan Nadon <memsize@videotron.ca>
13429Date:   Tue Feb 9 17:27:08 2010 -0500
13430
13431    doc: use $(mkdir_p) rather than $(MKDIR_P) due to automake 1.9.6
13432    
13433    $(MKDIR_P) is not defined in automake 1.9.
13434    
13435    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13436
13437commit e66eda95b520727969d721416d4ea49b54ef2b03
13438Author: Gaetan Nadon <memsize@videotron.ca>
13439Date:   Tue Feb 2 16:29:26 2010 -0500
13440
13441    specs: install html images in $docdir with html files
13442    
13443    The images required by the html files have been omitted.
13444    
13445    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13446
13447commit 137d1ff6be99d040e77c8c4c52029c5bdf1f4808
13448Author: Gaetan Nadon <memsize@videotron.ca>
13449Date:   Sun Jan 31 14:16:20 2010 -0500
13450
13451    doc: use new macros to control doc generation
13452    
13453    Namely XORG_WITH_GROFF for the groff generation tool
13454    XORG_WITH_PS2PDF for the conversion of PS docs to PDF
13455    XORG_ENABLE_SPECS for the generation of specs
13456    
13457    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13458
13459commit 9ca583c5d6fcf9cd8151ee7b019630b141d32fab
13460Author: Gaetan Nadon <memsize@videotron.ca>
13461Date:   Tue Jan 26 19:41:08 2010 -0500
13462
13463    doc: clean-up generated html images
13464    
13465    Generate images in /images as is the convention
13466    Provide a base file name for images rather than process ID
13467    Remove images directory when running make clean
13468    
13469    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
13470
13471commit eb289b34bfc9978eef724e0f268975938334c06c
13472Author: Kristian Høgsberg <krh@bitplanet.net>
13473Date:   Thu Jan 21 13:39:17 2010 -0500
13474
13475    XErrorDB: Add new DRI2 request names
13476
13477commit 6babf2123d05adb9349394c49cc2d81d1f66cf9e
13478Author: Julien Cristau <jcristau@debian.org>
13479Date:   Sat Jan 16 22:47:03 2010 +0000
13480
13481    man: Fix typo in Makefile
13482    
13483    Missing line continuation was preventing the XCompose alias from being
13484    generated.
13485    
13486    Signed-off-by: Julien Cristau <jcristau@debian.org>
13487    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13488
13489commit 68530599e61cf9fef3fcc91650e7cef21837fce7
13490Author: Alan Coopersmith <alan.coopersmith@sun.com>
13491Date:   Sat Jan 16 11:41:32 2010 -0800
13492
13493    libX11 1.3.3
13494    
13495    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13496
13497commit 1e5af224b9fbdb154620f898558704f2bc4bf4e9
13498Author: Alan Coopersmith <alan.coopersmith@sun.com>
13499Date:   Thu Jan 14 18:33:39 2010 -0800
13500
13501    Update COPYING file with actual licenses from libX11 code & docs
13502    
13503    COPYING was previously a years out-of-date copy of the LICENSES
13504    doc summarizing all licenses for all XFree86 components, many of
13505    which did not apply to libX11.
13506    
13507    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13508
13509commit ddb1786720d6136b1b28be599c75c10ad1b76a2a
13510Author: Alan Coopersmith <alan.coopersmith@sun.com>
13511Date:   Thu Jan 14 17:38:26 2010 -0800
13512
13513    Purge CVS/RCS id tags
13514    
13515    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13516
13517commit c1db9ddcfa7091f173478dca45bf720badedfca4
13518Author: Alan Coopersmith <alan.coopersmith@sun.com>
13519Date:   Thu Jan 14 15:39:14 2010 -0800
13520
13521    Update Sun license notices to current X.Org standard form
13522    
13523    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13524
13525commit 1aaa7c0358c5c035b99625f5715cb722bc2d2a2f
13526Author: Alan Coopersmith <alan.coopersmith@sun.com>
13527Date:   Thu Jan 14 08:28:22 2010 -0800
13528
13529    Add %S substitutions to Compose man page
13530    
13531    Commit 9df349a7894725f9469b106af645f57f7f3f9af3 added support for %S
13532    in Compose file include statements - this documents it in the Compose
13533    file man page.
13534    
13535    Also changes the existing substitution documentation to list format
13536    instead of a paragraph format to allow easier adding of %S.
13537    
13538    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13539
13540commit 94b45db93058cffa25598fe27dd3e385ace9d305
13541Author: Richard Purdie <rpurdie@linux.intel.com>
13542Date:   Thu Jan 14 08:04:28 2010 -0800
13543
13544    Fix configure.ac PKG_ macro calls
13545    
13546    If the first call to PKG_CHECK_MODULES() may not be called during execution
13547    which is the case here since its inside a case statement, call
13548    PKG_PROG_PKG_CONFIG() to ensure things worked correctly.
13549    
13550    Fixes a configure failure introduced by commit
13551    87529c039050ce3336ff9ce00f1b5a21d15690da when xcb is enabled.
13552    
13553    http://bugs.freedesktop.org/show_bug.cgi?id=26041
13554    
13555    Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
13556    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13557
13558commit df559d8bc2238a2c369152fcb06e5b5ab132a5dc
13559Author: Alan Coopersmith <alan.coopersmith@sun.com>
13560Date:   Tue Jan 12 09:18:15 2010 -0800
13561
13562    Raise xorg-macros requirement to 1.4
13563    
13564    Needed since the changes in 464390f16d7ed4aa5bf80f89863ba92273075ec2
13565    depend on the INSTALL file delivered in xorg-macros 1.4
13566    
13567    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13568
13569commit 9df349a7894725f9469b106af645f57f7f3f9af3
13570Author: James Cloos <cloos@jhcloos.com>
13571Date:   Wed Jan 13 11:04:38 2010 -0500
13572
13573    Allow inclusion of system level compose files.
13574    
13575    With the release of XFree86 4.4 an inclusion system of compose files was
13576    added to Xlib to allow inclusion of the default compose file (with %L),
13577    any compose files from user’s home directory (with %H), or a compose
13578    file with a hard coded path¹. However, even today including system level
13579    compose files is not possible in a platform independent manner although
13580    the machinery for including compose files and overriding previously
13581    defined compositions is already in place.
13582    
13583    With the ability to include system level compose files one could greatly
13584    reduce the need for compose file rule duplication and the work needed to
13585    propagate changes in one compose file to others. For example, currently
13586    the Finnish compose file fi_FI.UTF-8 weights over 5000 lines² but it
13587    is almost identical with en_US.UTF-8 except for perhaps half a dozen
13588    compositions.
13589    
13590    This commit allows one to include system level compose files with the
13591    following kind of syntax:
13592    
13593    include "%S/en_US.UTF-8/Compose"
13594    
13595    1] http://www.xfree86.org/4.4.0/RELNOTES5.html#42
13596    2] http://cgit.freedesktop.org/xorg/lib/libX11/tree/nls/fi_FI.UTF-8/Compose.pre
13597    
13598    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
13599    Signed-off-by: James Cloos <cloos@jhcloos.com>
13600
13601commit ccf21a8877f40136e25f2f62d0668b0c76b3cfbd
13602Author: Thien-Thi Nguyen <ttn@gnuvola.org>
13603Date:   Tue Dec 1 10:31:47 2009 +0100
13604
13605    libX11: Fix comment: Invert polarity (direction) of reformat description.
13606    
13607    Signed-off-by: Thien-Thi Nguyen <ttn@gnuvola.org>
13608    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13609
13610commit 4d4676c0d6324d2ff9d00d15ab485fa5d848a369
13611Author: Osamu Sayama <osamu.sayama@sun.com>
13612Date:   Tue Jan 5 17:26:40 2010 -0800
13613
13614    set_fontset_extents crash after get_rotate_fontname fix in 2bef065b70f7
13615    
13616    - In set_fontset_extents, check font_data is not NULL before running
13617      loop that may increment it to a non-NULL invalid pointer.
13618    - Make sure get_rotate_fontname counts the final field
13619    
13620    Fixes OpenSolaris bug 6913809: X application dumps core in ja_JP.UTF-8 locale
13621    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6913809>
13622    
13623    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13624    Reviewed-by: Adam Jackson <ajax@redhat.com>
13625
13626commit 87529c039050ce3336ff9ce00f1b5a21d15690da
13627Author: Alan Coopersmith <alan.coopersmith@sun.com>
13628Date:   Tue Jan 5 18:02:37 2010 -0800
13629
13630    Merge X11, XKBPROTO, & XPROTO pkg-config lists
13631    
13632    Since the XPROTO_CFLAGS & XKBPROTO_CFLAGS are just merged into X11_CFLAGS
13633    in configure.ac anyway, might as well combine the lists passed to
13634    PKG_CHECK_MODULES to reduce duplication in the flags.
13635    
13636    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13637    Reviewed-by: Rémi Cardona <remi@gentoo.org>
13638    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
13639
13640commit 2dd053667b44c55e2bf601aec7b75fc6fee1cf44
13641Author: Alan Coopersmith <alan.coopersmith@sun.com>
13642Date:   Tue Jan 5 17:56:58 2010 -0800
13643
13644    Add XCompose man page shadow for Compose man page
13645    
13646    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13647
13648commit 02543bba816e065c02cd36e0c9e21519a9ee9c96
13649Author: Alan Coopersmith <alan.coopersmith@sun.com>
13650Date:   Tue Jan 5 17:47:44 2010 -0800
13651
13652    Add $(AM_V_GEN) to silence lint rules too
13653    
13654    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13655
13656commit db7c6fdeeaef9475458498e4cf09d6b1329e9aa3
13657Author: Alan Coopersmith <alan.coopersmith@sun.com>
13658Date:   Tue Jan 5 17:42:42 2010 -0800
13659
13660    Remove GCC_WARNINGS now that XORG_CWARNFLAGS sets them for us
13661    
13662    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13663
13664commit 6b4e526e7ba105622f6b143e8e2639b44cb6cc19
13665Author: Gaetan Nadon <memsize@videotron.ca>
13666Date:   Sun Nov 29 10:44:24 2009 -0500
13667
13668    configure.ac: restore correct order for XORG_DEFAULT_OPTIONS
13669    
13670    It must be following AC_USE_SYSTEM_EXTENSION
13671
13672commit 3dbaa11d1c8705e1b309c6686f354766b651ad20
13673Author: Gaetan Nadon <memsize@videotron.ca>
13674Date:   Sun Nov 29 10:01:44 2009 -0500
13675
13676    Add .gitignore in /specs for generated files
13677    
13678    Being in /specs, it will apply for all 3 subdirs
13679
13680commit f45d39d37aab04742e44cd8c3c993aad7587e40c
13681Author: Gaetan Nadon <memsize@videotron.ca>
13682Date:   Fri Nov 27 20:56:03 2009 -0500
13683
13684    Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES
13685    
13686    Now that the INSTALL file is generated.
13687    Allows running make maintainer-clean.
13688
13689commit 464390f16d7ed4aa5bf80f89863ba92273075ec2
13690Author: Gaetan Nadon <memsize@videotron.ca>
13691Date:   Wed Oct 28 14:09:10 2009 -0400
13692
13693    INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206
13694    
13695    Add missing INSTALL file. Use standard GNU file on building tarball
13696    README may have been updated
13697    Remove AUTHORS file as it is empty and no content available yet.
13698    Remove NEWS file as it is empty and no content available yet.
13699
13700commit 6360e7f0ce6a75da2bed33ede7fea783a1fb80e1
13701Author: Gaetan Nadon <memsize@videotron.ca>
13702Date:   Tue Oct 27 15:07:25 2009 -0400
13703
13704    Deploy the new XORG_DEFAULT_OPTIONS #24242
13705    
13706    This macro aggregate a number of existing macros that sets commmon
13707    X.Org components configuration options. It shields the configuration file from
13708    future changes.
13709
13710commit d02f943c98fde2f14319bc57fd9ad77eb9a2a572
13711Author: Gaetan Nadon <memsize@videotron.ca>
13712Date:   Mon Oct 26 22:08:42 2009 -0400
13713
13714    Makefile.am: ChangeLog not required: EXTRA_DIST or *CLEANFILES #24432
13715    
13716    ChangeLog filename is known to Automake and requires no further
13717    coding in the makefile.
13718
13719commit f77c89c751a6a63c8ef11ecdddca2aed11ff6e29
13720Author: Gaetan Nadon <memsize@videotron.ca>
13721Date:   Thu Oct 22 12:34:19 2009 -0400
13722
13723    .gitignore: use common defaults with custom section # 24239
13724    
13725    Using common defaults will reduce errors and maintenance.
13726    Only the very small or inexistent custom section need periodic maintenance
13727    when the structure of the component changes. Do not edit defaults.
13728
13729commit 46c7b0e9d0d85cf5ccc0d40d19821fcc3335503b
13730Author: Alan Coopersmith <alan.coopersmith@sun.com>
13731Date:   Thu Nov 12 23:05:40 2009 -0800
13732
13733    Add compose-check.pl to EXTRA_DIST
13734    
13735    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13736
13737commit 54c64267cc8bc98641cc39a22cb7bd71673e89e0
13738Author: Julien Cristau <jcristau@debian.org>
13739Date:   Thu Oct 29 17:17:24 2009 +0100
13740
13741    man: fix XCopyGC argument order
13742    
13743    Ubuntu bug#408337
13744
13745commit aad10032651cdc2a53b359035954454a28d6db67
13746Author: Alan Coopersmith <alan.coopersmith@sun.com>
13747Date:   Fri Oct 23 13:55:14 2009 -0700
13748
13749    libX11 1.3.2
13750    
13751    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13752
13753commit 6303ada89cb228c0f9656036f798703afb42fc29
13754Author: Peter Hutterer <peter.hutterer@who-t.net>
13755Date:   Wed Oct 21 12:42:07 2009 +1000
13756
13757    Add smiley faces to compose sequences.
13758    
13759    I wonder how we could have lasted that long without them.
13760    
13761    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13762    Acked-by: Daniel Stone <daniel@fooishbar.org>
13763    Acked-By: James Cloos <cloos@jhcloos.com>
13764
13765commit 0e104ebd8628803c27e36b16922ad1edd891325a
13766Author: Alan Coopersmith <alan.coopersmith@sun.com>
13767Date:   Thu Oct 22 23:12:30 2009 -0700
13768
13769    Add man page for Compose file format
13770    
13771    Based on grammar description in modules/im/ximcp/imLcPrs.c and
13772    note on XFree86 changes formerly found in xorg-docs RELNOTES.sgml
13773    
13774    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13775
13776commit 9c95f2af7c442b3a59b1a30cf804f1ef4e7fc5b5
13777Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
13778Date:   Tue Oct 20 12:46:03 2009 -0700
13779
13780    Add extra configuration and sanity checks for groff and ps2pdf
13781    
13782    1) Add AC_ARG_VAR for GROFF and PS2PDF to inform users of these
13783       environment variables.
13784    2) Check that groff -ms works
13785    
13786    Some distributions ship the ms macros as a separate package which may
13787    not be installed together with groff, so we need to make sure that groff
13788    works and the required macros are actually installed before attempting
13789    to build the specs.
13790    
13791    Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
13792    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
13793    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13794
13795commit d3f801fd2f9198eaad6797414dba652f9c006c6d
13796Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13797Date:   Sun Oct 18 17:34:53 2009 -0500
13798
13799    Fix VPATH build of libX11 specs
13800    
13801    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13802
13803commit ad15e1a89d30ccc11d80745897b83def1448e8c0
13804Author: Alan Coopersmith <alan.coopersmith@sun.com>
13805Date:   Sat Oct 17 16:14:34 2009 -0700
13806
13807    libX11 1.3.1
13808    
13809    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13810
13811commit 0cbf98c17a7484bb58b2464b98d63bb3b4ea2594
13812Author: Alan Coopersmith <alan.coopersmith@sun.com>
13813Date:   Sat Oct 17 16:28:25 2009 -0700
13814
13815    Fix make distcheck
13816    
13817    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13818
13819commit 082e62ad268ef16a3bebc5a3c9fa008dbdc483ed
13820Author: Alan Coopersmith <alan.coopersmith@sun.com>
13821Date:   Fri Oct 16 19:25:35 2009 -0700
13822
13823    Use $(AM_V_GEN) to silence echo commands for generating shadow man pages
13824    
13825    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13826
13827commit 4e66da0783b2e5e3b288aaecd3c89396ed425c20
13828Author: Alan Coopersmith <alan.coopersmith@sun.com>
13829Date:   Wed Oct 14 16:18:24 2009 -0700
13830
13831    Move libX11 & XIM/locale specs from xorg-docs
13832    
13833    If groff is found, and --disable-specs is not passed to configure,
13834    specs will be converted to text, html and ps (or pdf if ps2pdf is
13835    found) and installed to $(docdir)
13836    
13837    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13838
13839commit 5d3d817a42ddcc8d0c6efd33efd1442fe14f5c6b
13840Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13841Date:   Tue Oct 13 19:15:51 2009 -0500
13842
13843    Provide _Xsetlocale compat wrappers on Cygwin
13844    
13845    Previous versions of Cygwin did not have proper locale support, so Cygwin/X
13846    defined X_LOCALE, using _Xsetlocale instead.  Cygwin 1.7 has added locale
13847    support, but we can't remove the _Xsetlocale entry point without breaking
13848    ABI.
13849    
13850    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13851
13852commit 2c8b3a877a713bb66a6316a7051b43c46af6e1a0
13853Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13854Date:   Tue Oct 13 19:15:49 2009 -0500
13855
13856    dolt: add Cygwin to supported platforms
13857    
13858    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13859
13860commit a2c8e3e34b15b57ff881a52101fc961a602f35e4
13861Author: Alan Coopersmith <alan.coopersmith@sun.com>
13862Date:   Wed Oct 14 13:23:30 2009 -0700
13863
13864    Recognize XSUNBUFFERSIZE alias for XLIBBUFFERSIZE on Solaris
13865    
13866    Also fix indenting of the XLIBBUFFERSIZE code to match surrounding code
13867    
13868    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13869
13870commit 34ddfca7b7d58240b0fe74bec6b2e0652d15c051
13871Author: Jon TURNEY <jon.turney@dronecode.org.uk>
13872Date:   Tue Feb 10 17:47:25 2009 +0000
13873
13874    Include sys/select.h for select() and struct timeval, if it exists
13875    
13876    This is a cygwin build fix
13877
13878commit 383165916ddac91740d4c780174d4c0d07cdb994
13879Author: Xake <xake@rymdraket.net>
13880Date:   Sun Sep 27 11:16:36 2009 +0200
13881
13882    Use AM_V_GEN instead of customized macros for AM_SILENT_RULES
13883    
13884    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13885
13886commit 854269d55cdda8caf425515bfed2855e211a5ada
13887Author: Julien Cristau <jcristau@debian.org>
13888Date:   Tue Oct 6 16:11:24 2009 +0200
13889
13890    configure: quote argument to m4_pattern_forbid
13891    
13892    Without this, configure spits out
13893    ../configure: line 12364: ac_fn_c_check_member: command not found
13894    ../configure: line 12378: ac_fn_c_check_type: command not found
13895    
13896    Also anchor the pattern to make it stricter.
13897    
13898    Signed-off-by: Julien Cristau <jcristau@debian.org>
13899
13900commit 3bb020587ce74e0737ec7aceb20041f1e77d3b87
13901Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
13902Date:   Thu Oct 1 22:20:38 2009 -0700
13903
13904    Split CFLAGS into CPPFLAGS and CFLAGS
13905    
13906    On some build systems, CPPFLAGS is set to "-I/some/prefix/include".  If older
13907    X11 headers are in /some/prefix/include, they will be preferred over the
13908    shipped headers.  This corrects that problem.
13909
13910commit d54caf1c9c55af8247621b7ba6afb20b23699839
13911Author: Peter Hutterer <peter.hutterer@who-t.net>
13912Date:   Fri Oct 2 10:59:08 2009 +1000
13913
13914    libX11 1.3
13915    
13916    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13917
13918commit dd201bcf9e7f8863b7ef32273a5ef021678133c1
13919Author: Peter Hutterer <peter.hutterer@who-t.net>
13920Date:   Fri Oct 2 12:10:29 2009 +1000
13921
13922    nls: remove duplicate Compose sequences from pt_BR.UTF-8
13923    
13924    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13925
13926commit a293ae9e83739067fc92db1a39c262857bc2283e
13927Author: Peter Hutterer <peter.hutterer@who-t.net>
13928Date:   Fri Sep 25 11:19:41 2009 +1000
13929
13930    Add XF86TouchpadToggle to XKeysymDB
13931    
13932    Lenovo laptops provide a key to enable or disable the touchpad and the
13933    trackstick. This key is usually located on Fn + F8.
13934    
13935    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13936    Acked-by: Adam Jackson <ajax@redhat.com>
13937
13938commit 69839f8903a24eab08f17a781b3797fb64dce9cf
13939Author: Alan Coopersmith <alan.coopersmith@sun.com>
13940Date:   Sun Sep 27 10:34:16 2009 -0700
13941
13942    Bug 24173: libX11 from git fails to build with automake older then 1.11
13943    
13944    AM_CONDITIONAL must come *before* the AC_OUTPUT that creates the
13945    Makefiles, instead of after.
13946    <http://bugs.freedesktop.org/show_bug.cgi?id=24173>
13947    
13948    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13949
13950commit f5effd041f58ef07703cca2b4f396758811e1eec
13951Author: Alan Coopersmith <alan.coopersmith@sun.com>
13952Date:   Wed Apr 15 10:59:23 2009 -0700
13953
13954    Resolve conflicting Compose sequences in iso8859-2, el_GR.UTF-8 & pt_BR.UTF-8
13955    
13956    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13957
13958commit 3843778358d3a0cd6a2d07dba5dd061248053ac9
13959Author: Alan Coopersmith <alan.coopersmith@sun.com>
13960Date:   Wed Apr 15 10:56:09 2009 -0700
13961
13962    Add perl script to check for duplicate or conflicting compose file entries
13963    
13964    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13965
13966commit 19cc5e1fa17a285045662820a8b4de2a0f9a194d
13967Author: Alan Coopersmith <alan.coopersmith@sun.com>
13968Date:   Fri Sep 18 17:10:04 2009 -0700
13969
13970    Use make rules instead of shell for loops to generate shadow man pages
13971    
13972    Allows parallel make and simpler build logs/error reporting
13973    
13974    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13975
13976commit 7dabcac973d0b6692a3cd62bd6d8e0467b24200b
13977Author: Alan Coopersmith <alan.coopersmith@sun.com>
13978Date:   Fri Sep 18 16:58:53 2009 -0700
13979
13980    Add AM_SILENT_RULES support for cpp rules for man & nls files
13981    
13982    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13983
13984commit bfa19cddd8546b4930a773f3bbd81137c362d82b
13985Author: Alan Coopersmith <alan.coopersmith@sun.com>
13986Date:   Fri Sep 18 16:58:16 2009 -0700
13987
13988    Update to using xorg-macros 1.3 & XORG_DEFAULT_OPTIONS
13989    
13990    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13991
13992commit 51396066c8003a96a6399c9a4bed2a81e512b582
13993Author: Mikko Niskanen <mikko.niskanen@iki.fi>
13994Date:   Fri Sep 4 10:11:15 2009 +1000
13995
13996    Fix wrong typedef on HP-UX (#18998)
13997    
13998    shl_dt doesn't exist, the type is shl_t.
13999    
14000    X.Org Bug 18998 <http://bugs.freedesktop.org/show_bug.cgi?id=18998>
14001    
14002    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14003
14004commit 615220a312b9430a580fe6dcf51703c6ef244f66
14005Author: Paul Bender <pebender@gmail.com>
14006Date:   Fri Sep 4 09:57:10 2009 +1000
14007
14008    Don't require xdmcp in configure.ac (#22583)
14009    
14010    X.Org Bug 22583 <http://bugs.freedesktop.org/show_bug.cgi?id=22583>
14011    
14012    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14013
14014commit 20f9ecd86ad2a9ac6966f7eee32234cb5ef77c29
14015Author: Peter Hutterer <peter.hutterer@who-t.net>
14016Date:   Thu Sep 3 14:02:44 2009 +1000
14017
14018    man: fix parameters to XkbAllocGeomOverlay{Rows|Keys} (#23499)
14019    
14020    X.Org Bug 23499 <http://bugs.freedesktop.org/show_bug.cgi?id=23499>
14021    
14022    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14023
14024commit c2814a614dc4d9409bfa0f44c18bfd69ad7f7b85
14025Author: Peter Hutterer <peter.hutterer@who-t.net>
14026Date:   Thu Sep 3 12:05:25 2009 +1000
14027
14028    man: XQueryTree may return BadWindow. (#23416)
14029    
14030    X.Org Bug 23416 <http://bugs.freedesktop.org/show_bug.cgi?id=XXX>
14031    
14032    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14033
14034commit dbe98d456ccc6eeca9fa3e241a3db0a4d83c5a65
14035Author: Alan Coopersmith <alan.coopersmith@sun.com>
14036Date:   Fri Aug 28 23:07:58 2009 +0800
14037
14038    Fix version tag in .TH line of several XKB man pages
14039    
14040    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14041
14042commit bf24400936c10af6f5aa0c75cfe2207ab9b680b4
14043Author: Alan Coopersmith <alan.coopersmith@sun.com>
14044Date:   Fri Aug 28 23:04:38 2009 +0800
14045
14046    XkbSetDeviceButtonActions.man: remove non-existent actions argument
14047    
14048    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14049
14050commit 53affa9335116f2d367f041e6502a411d4619e47
14051Author: Alan Coopersmith <alan.coopersmith@sun.com>
14052Date:   Fri Aug 28 23:00:17 2009 +0800
14053
14054    XkbQueryExtension.man: Arguments should be pointers
14055    
14056    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14057
14058commit 6233948885acc5873a5abddfff235afec555f3c2
14059Author: Alan Coopersmith <alan.coopersmith@sun.com>
14060Date:   Fri Aug 28 22:53:03 2009 +0800
14061
14062    XkbSAActionSetCtrls.man: Fix typo in formatting macro
14063    
14064    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14065
14066commit 28a9ca57cdec4aa9ca92322f963b01f0c2daf47a
14067Author: Alan Coopersmith <alan.coopersmith@sun.com>
14068Date:   Fri Aug 28 22:49:31 2009 +0800
14069
14070    Convert Xkb API man pages to ANSI prototypes
14071    
14072    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14073
14074commit fa2eecca85baab9616f7143cc1a853a594b4a90c
14075Author: James Cloos <cloos@jhcloos.com>
14076Date:   Mon Aug 24 06:35:17 2009 -0400
14077
14078    Add some (Serbian) Cyrillic NFD sequences.
14079    
14080    A number of characters in use in the various countries which use the
14081    Cyrillic script do not appear as pre-composed characters in The UCS
14082    or Unicode; they are only available as combining-character sequences.
14083    
14084    This commit adds support for using (prefix) dead keys and Multi_key-
14085    initiated sequences to enter a number of these combining-character
14086    sequences.  This ensures that users can enter these scripts even
14087    when using the current Cyrillic keymaps, which lack support for
14088    the combining characters.
14089    
14090    Please see the discussions on the xkb mailing list.
14091    
14092    Signed-off-by: James Cloos <cloos@jhcloos.com>
14093
14094commit d1bdc909f9246119696c8b0d9afb7bd8afb71b60
14095Author: Julien Cristau <jcristau@debian.org>
14096Date:   Wed Aug 5 18:14:23 2009 +0200
14097
14098    man/xkb: delete spurious newline in .TH headers
14099
14100commit ee723b83b24682db833a2f0abd96cd319b8a62af
14101Author: Julien Cristau <jcristau@debian.org>
14102Date:   Wed Aug 5 16:45:19 2009 +0200
14103
14104    man: use __libmansuffix__ instead of 3X11 for references to other pages
14105
14106commit 595e204feb82c798a92eea41fea03be6476ac181
14107Author: Julien Cristau <jcristau@debian.org>
14108Date:   Wed Aug 5 16:43:36 2009 +0200
14109
14110    man/xkb: use __libmansuffix__ instead of hardcoding 3Xkb for manpage sections
14111
14112commit 9da7e230d5320e1556ad2084fcd06ee7994385ea
14113Author: Peter Hutterer <peter.hutterer@who-t.net>
14114Date:   Wed Aug 5 14:15:02 2009 +1000
14115
14116    Bump to 1.2.99.901 (1.3 RC1)
14117    
14118    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14119
14120commit 8f78c7b4e3570cd46c5a220982963c17fe2157b8
14121Author: Filippo Giunchedi <filippo@debian.org>
14122Date:   Sat Jun 6 16:56:54 2009 +0200
14123
14124    nls: add {left,right}wards arrow to compose table
14125    
14126    Debian bug#532117 <http://bugs.debian.org/532117>
14127    
14128    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14129
14130commit 7949bfa00390241d994f32463e50d4bd78920568
14131Author: Julien Cristau <jcristau@debian.org>
14132Date:   Fri Jul 31 13:33:52 2009 +0200
14133
14134    Update library version for new symbols
14135    
14136    Commit 554f755e5545f63d3c8f299297927238da155773 added generic event
14137    cookie handling.  Bump libX11 version number accordingly.
14138    
14139    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14140
14141commit 640fec5f4feacd01a00eea3dcd4edb220907d3dc
14142Author: Julien Cristau <jcristau@debian.org>
14143Date:   Sun Aug 2 17:18:31 2009 +0200
14144
14145    Add _XFUNCPROTOBEGIN/END to Xlib-xcb.h
14146    
14147    X.Org bug#22252 <https://bugs.freedesktop.org/show_bug.cgi?id=22252>
14148    
14149    Reported-by: Riku Salminen <rsalmin2@cc.hut.fi>
14150    Signed-off-by: Julien Cristau <jcristau@debian.org>
14151
14152commit bc06d49e9dac1836d6824769ddb2ac5ba9f14df7
14153Author: Peter Hutterer <peter.hutterer@who-t.net>
14154Date:   Wed Jul 29 08:44:09 2009 +1000
14155
14156    Fix compiler warning 'unused variable qelt'
14157    
14158    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14159
14160commit 03f4907e14f5755e72309f08742977b871e81e33
14161Author: Peter Hutterer <peter.hutterer@who-t.net>
14162Date:   Wed Jul 29 08:34:57 2009 +1000
14163
14164    Add utlist.h to the Makefile.am
14165    
14166    utlist.h contains the linked list macros, it was added with the recent
14167    addition of event cookies but utlist.h wasn't added to the Makefile.am. As a
14168    result, make dist failed.
14169    
14170    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14171
14172commit 554f755e5545f63d3c8f299297927238da155773
14173Author: Peter Hutterer <peter.hutterer@who-t.net>
14174Date:   Fri Jun 26 11:27:43 2009 +1000
14175
14176    Add generic event cookie handling to libX11.
14177    
14178    Generic events require more bytes than Xlib provides in the standard XEvent.
14179    Memory allocated by the extension and stored as pointers inside the event is
14180    prone to leak by simple 'while (1) { XNextEvent(...); }' loops.
14181    
14182    This patch adds cookie handling for generic events. Extensions may register
14183    a cookie handler in addition to the normal event vectors. If an extension
14184    has registered a cookie handler, _all_ generic events for this extensions
14185    must be handled through cookies. Otherwise, the default event handler is
14186    used.
14187    
14188    The cookie handler must return an XGenericEventCookie with a pointer to the
14189    data.The rest of the event (type, serialNumber, etc.) are to be filled as
14190    normal. When a client retrieves such a cookie event, the data is stored in
14191    an internal queue (the 'cookiejar'). This data is freed on the next call to
14192    XNextEvent().
14193    
14194    New extension interfaces:
14195        XESetWireToEventCookie(display, extension_number, cookie_handler)
14196    
14197    Where cookie_handler must set cookie->data. The data pointer is of arbitray
14198    size and type but must be a single memory block. This memory block
14199    represents the actual extension's event.
14200    
14201    New client interfaces:
14202        XGetEventData(display, *cookie);
14203        XFreeEventData(display, *cookie);
14204    
14205    If the client needs the actual event data, it must call XGetEventData() with
14206    the cookie. This returns the data pointer (and removes it from the cookie
14207    jar) and the client is then responsible for freeing the event with
14208    XFreeEventData(). It is safe to call either function with a non-cookie
14209    event. Events unclaimed or not handled by the XGetEventData() are cleaned up
14210    automatically.
14211    
14212    Example client code:
14213        XEvent event;
14214        XGenericEventCookie *cookie = &ev;
14215    
14216        XNextEvent(display, &event);
14217        if (XGetEventData(display, cookie)) {
14218            XIEvent *xievent = cookie->data;
14219            ...
14220        } else if (cookie->type == GenericEvent) {
14221            /* handle generic event */
14222        } else {
14223            /* handle extension/core event */
14224        }
14225        XFreeEventData(display, cookie);
14226    
14227    Cookies are not multi-threading safe. Clients that use XGetEventData() must
14228    lock between XNextEvent and XGetEventData to avoid other threads freeing
14229    cookies.
14230    
14231    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14232
14233commit d7675cb8fa7155e7aff1459636a117a97aa1bf28
14234Author: Peter Hutterer <peter.hutterer@who-t.net>
14235Date:   Mon Jul 6 13:17:35 2009 +1000
14236
14237    Bump to 1.2.99.1
14238    
14239    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14240
14241commit 75fe48e7a42a685d7098e8d7108b9b956c471563
14242Author: Peter Hutterer <peter.hutterer@who-t.net>
14243Date:   Fri Jul 10 14:07:34 2009 +1000
14244
14245    Bump to 1.2.2
14246    
14247    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14248
14249commit 5d0fe0e0e92759af5667c5dca2eacb1b6f2d66ea
14250Author: Peter Hutterer <peter.hutterer@who-t.net>
14251Date:   Thu Jul 2 09:10:25 2009 +1000
14252
14253    XMaskEvent/XCheckMaskedEvents must not check for GenericEvents.
14254    
14255    GenericEvent cannot be selected for in the core event masks and they must
14256    thus be treated like extension events.
14257    
14258    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14259
14260commit c1bf65b89f4e361f6178a73dd5334c8f2bd95732
14261Author: Peter Hutterer <peter.hutterer@who-t.net>
14262Date:   Thu Jul 2 09:06:05 2009 +1000
14263
14264    XWindowEvent/XCheckWindowEvent must not return GenericEvents.
14265    
14266    GenericEvents have no fixed position for the window, so they must be treated
14267    like extension events.
14268    
14269    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
14270
14271commit 38f9054554d63525d2dd51aafb5eb57821158ab9
14272Author: Alan Coopersmith <alan.coopersmith@sun.com>
14273Date:   Mon Jun 15 19:00:43 2009 -0700
14274
14275    Drop ancient USG SysV #ifdefs
14276    
14277    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14278
14279commit 7bfe1323f16a1a69cc474659f7ac0c2570b1cf42
14280Author: Adam Jackson <ajax@redhat.com>
14281Date:   Fri Jun 12 12:44:01 2009 -0400
14282
14283    Remove X_NOT_STDC_ENV usage. (#6527)
14284
14285commit 6dd74d7fb414ca1e99bae5c13e333961f396eb36
14286Author: Julien Cristau <jcristau@debian.org>
14287Date:   Fri May 29 11:18:11 2009 +0200
14288
14289    nls: remove more duplicated aliases
14290    
14291    Signed-off-by: Julien Cristau <jcristau@debian.org>
14292
14293commit e4b0899f516da224010e68bd2d953d5293d94993
14294Author: parag <parag@rawhideTM.pnq.redhat.com>
14295Date:   Thu May 28 11:29:35 2009 +0530
14296
14297    nls: Add pa_PK locale information and make pa_IN as default for pa.
14298    
14299    X.Org bug#21954 <http://bugs.freedesktop.org/show_bug.cgi?id=21954>
14300    
14301    [jcristau: removed the pa_PK.UTF-8 alias to itself]
14302    
14303    Signed-off-by: parag <pnemade@redhat.com>
14304    Signed-off-by: Julien Cristau <jcristau@debian.org>
14305
14306commit f0ea1f6d51145592f8617854f9320ec5dbff3299
14307Author: Julien Cristau <jcristau@debian.org>
14308Date:   Fri May 29 10:58:20 2009 +0200
14309
14310    nls: remove broken sd_IN.UTF-8 alias
14311    
14312    Signed-off-by: Julien Cristau <jcristau@debian.org>
14313
14314commit e29e010dabdb17d6498f2ef1786f69b8830c18ca
14315Author: Julien Cristau <jcristau@debian.org>
14316Date:   Fri May 29 10:57:43 2009 +0200
14317
14318    nls: remove duplicated en_US* aliases
14319    
14320    Signed-off-by: Julien Cristau <jcristau@debian.org>
14321
14322commit a89a300d87852c84389ad97db66dcb8930cb45dd
14323Author: Caolan McNamara <caolanm@redhat.com>
14324Date:   Thu May 21 18:41:05 2009 +0200
14325
14326    man: missing space in XAllocColor man page
14327    
14328    X.Org bug#21854 <http://bugs.freedesktop.org/show_bug.cgi?id=21854>
14329
14330commit c1c001e36504fd304f76f69bf6af3643225c49ea
14331Author: James Cloos <cloos@jhcloos.com>
14332Date:   Wed May 13 13:03:54 2009 -0400
14333
14334    [nls] Replace remaining UCS Combining Characters in Compose sequences.
14335    
14336    The replaces the instances of keysyms which match <U03[0-6][0-9A-Fa-f]>,
14337    where the keysym is used as a dead_key, with an actual dead_key symbol.
14338    
14339    The only remaining instances of UCS combining characters in the
14340    compose sequences are of U0338 COMBINING LONG SOLIDUS OVERLAY
14341    used as a suffix in Multi_key-initiated sequences to create
14342    mathematics characters such as ∉ U+2209 NOT AN ELEMENT OF
14343    from ∈ U+2208 ELEMENT OF.
14344
14345commit e2b0bad3d3b9e9ca781fc264eb7584afbe2a1a4f
14346Author: James Cloos <cloos@jhcloos.com>
14347Date:   Wed May 13 09:58:59 2009 -0400
14348
14349    [nls] Remove extraneous instances of UCS Combining Characters in Compose sequences.
14350    
14351    This removes those instances of keysyms which match <U03[0-6][0-9A-Fa-f]>,
14352    where the matching keysym is used as a dead_key, and for which alternative
14353    compose sequences exist.
14354
14355commit 79f47e6dff2f0a0b673bbfecc47528edca814baa
14356Author: James Cloos <cloos@jhcloos.com>
14357Date:   Fri May 8 20:11:54 2009 -0400
14358
14359    [nls] Remove combining_ keysyms from the Compose files
14360    
14361    Some of the UTF-8 Compose tables included combining_ keysyms in
14362    the compose sequences as though they were dead symbols.  This
14363    is contrary to how combining characters are used in the UCS.
14364    Therefore, those lines have been removed from the Compose tables.
14365    
14366    There were also some combining_ keysyms as targets.  As those
14367    are not included in x11proto’s keysymdef.h, and as those do
14368    exist there as Uxxxx keysyms, they are replaced with the Uxxxx
14369    keysym names.
14370    
14371    This addresses http://bugzilla.freedesktop.org/show_bug.cgi?id=5107
14372    and is based on attachment 25644 by samuel.thibault@ens-lyon.org.
14373    
14374    Signed-off-by: James Cloos <cloos@jhcloos.com>
14375
14376commit 4a08a3dfbda497b2be46e3e5fe6b777815ea27f9
14377Author: parag <pnemade@redhat.com>
14378Date:   Tue May 5 16:50:47 2009 +0530
14379
14380    libX11: Add new Indic language information to nls directory files. #21560
14381    
14382    Signed-off-by: parag <pnemade@redhat.com>
14383
14384commit aaf81096eb44b4c2812108721ba02738391884da
14385Author: Alan Coopersmith <alan.coopersmith@sun.com>
14386Date:   Sat May 2 01:38:14 2009 -0700
14387
14388    Correct return type in XkbGetKeyboard man page
14389    
14390    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14391
14392commit 2bef065b70f70af520b5de8fb23529254d15f003
14393Author: Christoph Pfister <christophpfister@gmail.com>
14394Date:   Mon Apr 27 22:32:57 2009 -0700
14395
14396    X.Org Bug #21117: crash in get_rotate_fontname (omGeneric.c)
14397    
14398    http://bugs.freedesktop.org/show_bug.cgi?id=21117
14399    
14400    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14401
14402commit 5c1cde105db10df24d3c532f032cbc59050c7313
14403Author: Alan Coopersmith <alan.coopersmith@sun.com>
14404Date:   Fri Apr 17 22:14:47 2009 -0700
14405
14406    Use AC_USE_SYSTEM_EXTENSIONS instead of hand-rolled check for _GNU_SOURCE
14407    
14408    Raises minimum autoconf version required to 2.60
14409    
14410    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14411
14412commit 84b7a91ef84f345384e4b0e13907385ca3ca3255
14413Author: Julien Cristau <jcristau@debian.org>
14414Date:   Tue Apr 14 15:59:57 2009 +0100
14415
14416    Fix fi_FI.UTF-8, again
14417    
14418    Commit 97fc6babd4ccaf300e25708868aa2a738893dc30 "NLS: Add UTF-8 compose
14419    file for Finnish" made fi_FI.UTF-8 use a broken empty XLC_LOCALE file.
14420    This reverts it back to using the en_US.UTF-8 one.
14421
14422commit 128daff4422f973ea40dd1e31b2db230e643549e
14423Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
14424Date:   Thu Apr 9 12:01:07 2009 +0700
14425
14426    Thai XIM not retrieve MB surrounding on UTF-8 LC
14427    
14428    On th_TH.UTF-8 locale, Thai XIM rejects all combining characters for GTK+ apps
14429    that use X Input Method.
14430    
14431    This is because GTK+ imxim immodule passes surrounding text in locale encoding,
14432    which is UTF-8 for UTF-8 locales. But current Thai XIM in Xlib assumes the
14433    multi-byte StringConversionText response for the StringConversionCallback to
14434    always be TIS-620, by retrieving a single byte and using it as-is.
14435    
14436    If the Thai XIM tries to convert the multi-byte text based on locale codeset
14437    before using it, it will work again.
14438    
14439    X.Org But 12759 <http://bugs.freedesktop.org/show_bug.cgi?id=12759>
14440    
14441    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
14442    Signed-off-by: Julien Cristau <jcristau@debian.org>
14443
14444commit e09f0d227fbf95b6252759af9d426efd57686f9f
14445Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
14446Date:   Thu Apr 9 11:47:55 2009 +0700
14447
14448    Thai XIM not filters when NumLock or CapsLock is on
14449    
14450    The Thai XIM component in libx11 activated on 'th*' locales normally filters
14451    input sequence according to orthographic rules. However, when NumLock/CapsLock
14452    is on, this stops working. All sequences are passed through.
14453    
14454    This is caused by missing masks in _XimThaiFilter(), which normally screens out
14455    certain special keys from entering orthographic rules. Unfortunately, this
14456    included events with NumLock/CapsLock on. Negating the masks from the check
14457    allows the events to be tested by the rules.
14458    
14459    X.Org Bug 12517 <http://bugs.freedesktop.org/show_bug.cgi?id=12517>
14460    
14461    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
14462    Signed-off-by: Julien Cristau <jcristau@debian.org>
14463
14464commit d108d3c706af3502820b5202564488ea19908b77
14465Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
14466Date:   Thu Apr 9 11:25:25 2009 +0700
14467
14468    CharSet-to-CompoundText Conversion Failed for Thai Locales
14469    
14470    SCIM fails to commit Thai input characters on Thai locales, because it commits
14471    string in compound text form, which was converted via
14472    XwcTextListToTextProperty(). But the XLC_LOCALE for th_TH and th_TH.UTF-8
14473    declares cs1's ct_encoding as TIS620-0:GR, which was commented out in
14474    src/xlibi18n/lcCT.c default_ct_data, in favor of ISO8859-11 ESC sequence.
14475    So, declaring cs1 as ISO8859-11:GR instead makes it work.
14476    
14477    Besides, for th_TH.UTF-8, adding cs2 class with ISO10646-1 encoding also adds
14478    support for UTF-8 input.
14479    
14480    And, along discussion in the bug, a similar problem was found for fontset, too,
14481    by causing delays on X apps startups on systems without tis620-0 fonts. This
14482    is normally the case, as mkfontdir and mkfontscale generate iso8859-11 entries
14483    by default for Thai X fonts. So, Thai fontset charset is also patched.
14484    
14485    X.Org Bug 16475 <http://bugs.freedesktop.org/show_bug.cgi?id=16475>
14486    
14487    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
14488    Signed-off-by: Julien Cristau <jcristau@debian.org>
14489
14490commit fd62d3318c846cd43d66a505946e94704d7d83dc
14491Author: Alan Coopersmith <alan.coopersmith@sun.com>
14492Date:   Wed Apr 8 19:42:25 2009 -0700
14493
14494    Revert "Change masculine to ordmasculine in Compose file comments"
14495    
14496    This reverts commit 892b401d5acc055803a20e349ede0d64490f2230.
14497    
14498    As Julien Cristau correctly points out, I misread the Compose file grammar
14499    in modules/im/ximcp/imLcPrs.c, and those are keysyms, not comments, and
14500    the keysym is named XK_masculine in keysymdef.h.   This change is thus a
14501    bug in the Solaris compose tables to be fixed, not an improvement to bring
14502    upstream.
14503
14504commit eac57c77afdf44f50692225b8b0345a7c927bc84
14505Author: Alan Coopersmith <alan.coopersmith@sun.com>
14506Date:   Tue Apr 7 15:38:45 2009 -0700
14507
14508    Version bump: 1.2.1
14509    
14510    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14511
14512commit bfbec08baec33c5024510b0bcbbee6e4a8473e79
14513Author: Yaakov Selkowitz (Cygwin Ports maintainer) <yselkowitz@users.sourceforge.net>
14514Date:   Tue Apr 7 13:46:57 2009 -0700
14515
14516    Bug 20773: Xcursor dynamic loading on Cygwin
14517    
14518    X.Org Bug #20773 <http://bugs.freedesktop.org/show_bug.cgi?id=20773>
14519    Patch #24096 <http://bugs.freedesktop.org/attachment.cgi?id=24096>
14520    
14521    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14522
14523commit c8c41614911be4fa222fa22478677d263b41c751
14524Author: Alan Coopersmith <alan.coopersmith@sun.com>
14525Date:   Mon Apr 6 16:52:46 2009 -0700
14526
14527    Fix a several sparse warnings: Using plain integer as NULL pointer
14528    
14529    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14530
14531commit b336c3d0cc2aefc8926500cff5f76b5a3e803886
14532Author: Alan Coopersmith <alan.coopersmith@sun.com>
14533Date:   Mon Apr 6 16:32:05 2009 -0700
14534
14535    Further ansify prototypes & reduce #ifdefs in locking.c
14536    
14537    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14538
14539commit 892b401d5acc055803a20e349ede0d64490f2230
14540Author: Alan Coopersmith <alan.coopersmith@sun.com>
14541Date:   Mon Apr 6 10:50:09 2009 -0700
14542
14543    Change masculine to ordmasculine in Compose file comments
14544    
14545    Matches the ordfeminine name used for the matching character,
14546    and the ordmasculine name used in many font descriptions of the glyph.
14547    
14548    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14549
14550commit e3198b55dc16ec57346cc28aa8d34165ce8cde96
14551Author: Julien Cristau <jcristau@debian.org>
14552Date:   Fri Mar 13 13:52:33 2009 +0100
14553
14554    Add a ru_RU.UTF-8 locale
14555    
14556    Based on patch by Eugene Konev <ejka@imfi.kspu.ru> for X.Org 6.9.0.
14557    
14558    Debian bug#330144 <http://bugs.debian.org/330144>
14559    X.Org bug#15887 <http://bugs.freedesktop.org/show_bug.cgi?id=15887>
14560
14561commit d239de9452691d6f875e6e5ace3d499ec3bf14d9
14562Author: Alan Coopersmith <alan.coopersmith@sun.com>
14563Date:   Wed Mar 25 17:59:09 2009 -0700
14564
14565    Delete some unused "#ifdef notdef" static functions
14566    
14567    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14568
14569commit 4281892e31058ef3aecc96a5767824b34d88d415
14570Author: Alan Coopersmith <alan.coopersmith@sun.com>
14571Date:   Wed Mar 25 17:52:48 2009 -0700
14572
14573    Remove _XP_PRINT_SERVER_ #ifdefs from Xrm.c
14574    
14575    This copy of Xrm.c is never compiled into the Xprint server any more, so
14576    this old code-sharing #ifdef from the monolith tree isn't needed.
14577    
14578    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14579
14580commit 27a2e16e8ea11c1604e28645fae4d6ba4371d513
14581Author: Alan Coopersmith <alan.coopersmith@sun.com>
14582Date:   Tue Mar 17 18:38:58 2009 -0700
14583
14584    makekeys: combine malloc(strlen)+strcpy into strdup
14585    
14586    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14587
14588commit 78a894231ae8ec4959007b6d9b8d2a15d6333a1e
14589Author: Alan Coopersmith <alan.coopersmith@sun.com>
14590Date:   Tue Mar 17 15:42:19 2009 -0700
14591
14592    When makekeys fails to find a good hash, print error instead of divide-by-zero
14593    
14594    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14595
14596commit c9f84af591d15fbc3fa890bcd955d94f1ff82a0b
14597Author: Alan Coopersmith <alan.coopersmith@sun.com>
14598Date:   Tue Mar 17 14:59:16 2009 -0700
14599
14600    Remove ifdef checks for macII left over from ancient A/UX 3.0 support
14601    
14602    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14603
14604commit 4b90da0ffb32b791d915ecae11996cc2f2cac9a7
14605Author: Alan Coopersmith <alan.coopersmith@sun.com>
14606Date:   Mon Mar 16 18:37:49 2009 -0700
14607
14608    XErrorDB additions for DRI2 requests
14609    
14610    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14611
14612commit 273f45af1441be9d2135e4cac8c46ceb33470236
14613Author: Alan Coopersmith <alan.coopersmith@sun.com>
14614Date:   Mon Mar 16 18:35:44 2009 -0700
14615
14616    XErrorDB additions for XInput errors and new XInput 1.5 requests
14617    
14618    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14619
14620commit 90ef36f6336068183bf9d13ca972202db35b1202
14621Author: Alan Coopersmith <alan.coopersmith@sun.com>
14622Date:   Mon Mar 16 18:23:05 2009 -0700
14623
14624    XErrorDB additions for RANDR 1.3
14625    
14626    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14627
14628commit 124a4efaccf329f1a44f6b824e70278bdd1cad83
14629Author: Lubos Lunak <l.lunak@suse.cz>
14630Date:   Mon Mar 16 18:03:36 2009 -0700
14631
14632    XErrorDB updates for XTEST, RANDR, DAMAGE extensions
14633    
14634    From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html
14635    
14636    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14637
14638commit 80811846e37b805fddb37c71589fd5f6f6037b3f
14639Author: Lubos Lunak <l.lunak@suse.cz>
14640Date:   Mon Mar 16 17:57:52 2009 -0700
14641
14642    XGetErrorText() fails for extension error codes equal to the error base
14643    
14644    From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html
14645    
14646    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14647
14648commit da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd
14649Author: Milos Komarcevic <miloskomarcevic@netscape.net>
14650Date:   Mon Mar 16 17:43:26 2009 -0700
14651
14652    Bug 11456: Serbian locale updates (sr_RS and sr_ME)
14653    
14654    X.Org Bug #11456 <http://bugs.freedesktop.org/show_bug.cgi?id=11456>
14655    Patch #23937 <http://bugs.freedesktop.org/attachment.cgi?id=23937>
14656    
14657    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14658
14659commit 934a6c0519a9e32505beee55b004f36c2a710217
14660Author: Alan Coopersmith <alan.coopersmith@sun.com>
14661Date:   Mon Mar 16 14:55:22 2009 -0700
14662
14663    Bug 10082: Compose entries for some standard mathematical operators
14664    
14665    X.Org Bug #10082 <http://bugs.freedesktop.org/show_bug.cgi?id=10082>
14666    
14667    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14668
14669commit abf4da1ed0f735ca7ce471dc13a0ec3677391486
14670Author: Alan Coopersmith <alan.coopersmith@sun.com>
14671Date:   Mon Mar 16 14:27:46 2009 -0700
14672
14673    Bug 14651: We need to add new locale specification for Belarusian Latin locale
14674    
14675    X.Org Bug #14651 <http://bugs.freedesktop.org/show_bug.cgi?id=14651>
14676    
14677    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14678
14679commit 837703c8651e1321a50147a8311c56e4758ce08a
14680Author: Caolan McNamara <caolanm@redhat.com>
14681Date:   Mon Mar 16 14:15:50 2009 -0700
14682
14683    Bug 20575: man page for XCreatePixmapFromBitmapData doesn't match signature
14684    
14685    X.Org Bug #20575 <http://bugs.freedesktop.org/show_bug.cgi?id=20575>
14686    Patch #23717 <http://bugs.freedesktop.org/attachment.cgi?id=23717>
14687    
14688    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14689
14690commit 22199018879055d8653e59d8236bef57164fac66
14691Author: Alan Coopersmith <alan.coopersmith@sun.com>
14692Date:   Mon Mar 16 13:28:18 2009 -0700
14693
14694    Correct locale alias for sh_BA.ISO8859-2@bosnia (should be sr, not nr)
14695    
14696    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14697
14698commit 4c63c27eab2b88f5556dbf72c36321f50f6de35e
14699Author: Alan Coopersmith <alan.coopersmith@sun.com>
14700Date:   Thu Mar 12 18:57:20 2009 -0700
14701
14702    Bug 9953: Please provide locale alias hu_HU.utf8
14703    
14704    X.Org Bug #9953 <http://bugs.freedesktop.org/show_bug.cgi?id=9953>
14705    Debian Bug #407573 <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=407573>
14706    
14707    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14708
14709commit 501f4e0ada1690783ada05ad412e4b191ad55336
14710Author: Alan Coopersmith <alan.coopersmith@sun.com>
14711Date:   Thu Mar 12 17:38:21 2009 -0700
14712
14713    Bug 6820: Xlib shouldn't handle EAGAIN as a fatal IO error
14714    
14715    X.Org Bug #6820 <http://bugs.freedesktop.org/show_bug.cgi?id=6820>
14716    Patch #17637 <http://bugs.freedesktop.org/attachment.cgi?id=17637>
14717    
14718    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14719
14720commit 7b9c543e7210c7da204871c31a160e79d3a949b6
14721Author: Paul Bender <pebender@gmail.com>
14722Date:   Thu Mar 12 17:11:42 2009 -0700
14723
14724    Bug 15664: xau & xdmcp not needed in x11.pc dependencies when built with xcb
14725    
14726    X.Org bug #15664 <https://bugs.freedesktop.org/show_bug.cgi?id=15664>
14727    Patch #16128 <https://bugs.freedesktop.org/attachment.cgi?id=16128>
14728    
14729    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14730
14731commit fd2cf1ef66c2aff3dc758956c9e9e567b9892c06
14732Author: Xue Wei <Wei.Xue@Sun.COM>
14733Date:   Wed Mar 4 19:32:29 2009 -0800
14734
14735    Add UTF-8 locale entries for es_US, kk_KZ, mt_MT, and sh_BA
14736    
14737    Sun bug 6809309 Add new utf8 locales supported by Xlib
14738    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6809309>
14739    
14740    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14741
14742commit cb70c9bc43267577859a3674ca9de9be396ba69e
14743Author: Alan Coopersmith <alan.coopersmith@sun.com>
14744Date:   Mon Feb 23 19:29:15 2009 -0800
14745
14746    Add --with-locale-lib-dir configure option to set locale lib install dir
14747    
14748    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14749
14750commit ccd3584f0330db8dac90b9313c33ab8b5b2ec6af
14751Author: Alan Coopersmith <alan.coopersmith@sun.com>
14752Date:   Mon Feb 23 18:33:51 2009 -0800
14753
14754    Incorporate more locale names/aliases from Solaris libX11
14755    
14756    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14757
14758commit 83ce4daefdf544f801c7d666c89162690a36ce41
14759Author: Alan Coopersmith <alan.coopersmith@sun.com>
14760Date:   Mon Feb 23 18:32:34 2009 -0800
14761
14762    Incorporate char range comments from Solaris version of ksc5601.h
14763    
14764    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14765
14766commit ee279c84e34f1ebb8a6ff17d54ee146d11e29764
14767Author: Chris Ball <cjb@laptop.org>
14768Date:   Sat Feb 21 14:48:42 2009 -0500
14769
14770    Fix fi_FI locale install directory.
14771    
14772    fi_FI was setting "x11thislocaledir" to en_US, with the result that its
14773    locale data was written in that locale dir.
14774    
14775    Signed-off-by: Chris Ball <cjb@laptop.org>
14776
14777commit da6bbca07c796c69172a649405474f03bee66754
14778Author: Emilio Jesús Gallego Arias <egallego@babel.ls.fi.upm.es>
14779Date:   Sat Feb 21 20:17:23 2009 +0100
14780
14781    xcb_io: Avoid datatype overflow on AMD64 and friends.
14782
14783commit 4ef6491afa69e8441caee7bbebc583e6e796275e
14784Author: Chris Ball <cjb@laptop.org>
14785Date:   Sat Feb 21 12:51:03 2009 -0500
14786
14787    Build fix for fi_FI.
14788    
14789    Commit 642c4e928e770e0.. instructs make to enter nls/fi_FI, but no
14790    Makefile is written there by configure.
14791    
14792    Signed-off-by: Chris Ball <cjb@laptop.org>
14793
14794commit 642c4e928e770e012379539a6ce09e11c02f09a6
14795Author: Julien Cristau <jcristau@debian.org>
14796Date:   Sat Feb 21 03:12:05 2009 +0100
14797
14798    nls: actually use the fi_FI.UTF-8 files
14799    
14800    The subdir wasn't added to nls/Makefile.am
14801
14802commit 9bad8309ef289bb943651abf6967b24fa2252aac
14803Author: Alan Coopersmith <alan.coopersmith@sun.com>
14804Date:   Fri Feb 20 14:45:54 2009 -0800
14805
14806    flags member of Display structure needs to be marked volatile
14807    
14808    Since the Xlib multithreaded code checks the flags variable in _XFlushInt
14809    to see if the other threads are done yet, it has to be marked volatile so
14810    the compiler doesn't optimize out re-loading it on each trip through the
14811    while loop and end up in an impossible-to-exit infinite loop of CPU chewing.
14812    
14813    Part of fix for Sun bug 6409332: infinite loop in XFlushInt() on x86/32-bit
14814    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6409332>
14815    
14816    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14817
14818commit ef5c446395aa30d7b1096a112e241a81c5b358e7
14819Author: Xue Wei <Wei.Xue@Sun.COM>
14820Date:   Fri Feb 20 15:12:35 2009 -0800
14821
14822    Add nn_NO.UTF-8 to compose.dir.pre for Norwegian Nynorsk
14823    
14824    Sun bug 6691236: Swing applications dump core when locale is nn_NO.UTF-8
14825    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691236>
14826    
14827    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14828
14829commit 3822f2654e9630167f0c6cae317b472c09771672
14830Author: Xue Wei <Wei.Xue@Sun.COM>
14831Date:   Fri Feb 20 15:03:51 2009 -0800
14832
14833    Add locale aliases for no_NO & sh_BA locale variants
14834    
14835    Fixes Sun bug id 6691219: xterm refuses to start in some locales
14836    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691219>
14837    
14838    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14839
14840commit 63b6b5b5f522c0a606b32163c643edb64ca91d54
14841Author: Xue Wei <Wei.Xue@Sun.COM>
14842Date:   Fri Feb 20 14:23:11 2009 -0800
14843
14844    Add kk_KZ.UTF-8 to locale.dir.pre for Kazakhstan
14845    
14846    Fixes Sun bug id 6737254 ("kk_KZ.UTF-8 locale: In Java applications
14847     changing keyboard layout with gimlet does not work")
14848    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6737254>
14849    
14850    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14851
14852commit d497301707962f3b94542e999a36abffcfbd303d
14853Author: Julien Cristau <jcristau@debian.org>
14854Date:   Tue Feb 17 16:09:41 2009 +0100
14855
14856    Bump to 1.2
14857
14858commit 990e71361d1d7b79bf07b1dc93e4e905d4f1bdaf
14859Author: Julien Cristau <jcristau@debian.org>
14860Date:   Tue Feb 17 15:23:40 2009 +0100
14861
14862    Check Xmalloc return value in _XConnectXCB
14863    
14864    X.Org bug#19137 <http://bugs.freedesktop.org/show_bug.cgi?id=19137>
14865    
14866    Signed-off-by: Julien Cristau <jcristau@debian.org>
14867
14868commit b4b5893f69419ff577bbaa4d18f78e4ffd729a0c
14869Author: James Cloos <cloos@jhcloos.com>
14870Date:   Sat Feb 14 12:35:56 2009 -0500
14871
14872    dolt: allow older versions of bash to compile the library
14873    
14874    Cf xserver commit 7be6520d and bugzilla #19031.
14875
14876commit 20982d6866e24453642b0b592fa0f13a88aa747c
14877Author: Will Thompson <will@willthompson.co.uk>
14878Date:   Thu Feb 5 02:53:06 2009 +1100
14879
14880    NLS: Compose: Non-aliasing CCCP
14881    
14882    Oops, cccp aliased cc for question mark.  Upper-case it to avoid fail.
14883    
14884    Signed-off-by: Will Thompson <will@willthompson.co.uk>
14885    Signed-off-by: Daniel Stone <daniel@fooishbar.org> (sorry)
14886
14887commit f052665394f3f0319e93a98f1d5d4ea287e1dd07
14888Author: Will Thompson <will@willthompson.co.uk>
14889Date:   Wed Feb 4 14:51:11 2009 +0000
14890
14891    Add two essential compose sequences
14892    
14893    Signed-off-by: Will Thompson <will@willthompson.co.uk>
14894    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14895
14896commit d7bea6fa909bf34c43efe0ca8239ab0f9f3a415f
14897Author: Alan Coopersmith <alan.coopersmith@sun.com>
14898Date:   Mon Feb 2 20:34:31 2009 -0800
14899
14900    Add README with pointers to mailing list, bugzilla & git repos
14901    
14902    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14903
14904commit f682c27e93512773122887d2cbabb1657af45d2e
14905Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14906Date:   Mon Feb 2 16:36:39 2009 -0200
14907
14908    Check if a function argument is NULL.
14909    
14910      This was an addition to patch (also by me)
14911    https://bugs.freedesktop.org/attachment.cgi?id=14660
14912    that was not added when rediscovering/correcting the problem.
14913
14914commit 427e9d45d424b84efd9fc499aebf8d72392844c5
14915Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14916Date:   Thu Feb 28 15:58:12 2008 -0300
14917
14918    Allow multiple inclusions of cursorfont.h, cosmetic patch.
14919
14920commit b91524a53e691f6a5d278fd8972b48a14ebeedeb
14921Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14922Date:   Thu Feb 28 15:54:43 2008 -0300
14923
14924    Don't add prototypes for functions that don't exist.
14925    
14926    Note that a full review was not done, only for functions that receive
14927    char/short arguments, or one of it's parameters is a function pointer
14928    that requires char/short arguments.
14929
14930commit 537eb52fe266ac439c4b383bb04a70017b709911
14931Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14932Date:   Thu Feb 28 15:50:27 2008 -0300
14933
14934    WORD64 compile fix. This bug catched on a overview of the code.
14935    
14936    The code is wrong since the first git revision, so it seens that it has
14937    not been compiled with WORD64 for quite some time, there is also another
14938    interesting code in xkb/XKBRdBuf.c:
14939      <hash>ifdef WORD64
14940      _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8
14941      <hash>endif
14942    and possibly there are other similar problems.
14943
14944commit ffd0300fb74c6183208ae599133f2ded09e08d97
14945Author: Brian Rogers <brian@xyzw.org>
14946Date:   Sat Jan 31 10:37:51 2009 -0800
14947
14948    Initialize event_notify after allocating the memory for it.
14949    
14950    An uninitialized or otherwise invalid condition variable can apparently
14951    cause a hang in pthread_cond_broadcast. Ekiga, openoffice, and xine
14952    at least are freezing as a result of event_notify never being initialized.
14953    
14954    Signed-off-by: Brian Rogers <brian@xyzw.org>
14955    Signed-off-by: Bart Massey <bart@cs.pdx.edu>
14956
14957commit 97fc6babd4ccaf300e25708868aa2a738893dc30
14958Author: James Cloos <cloos@jhcloos.com>
14959Date:   Thu Jan 29 20:10:41 2009 -0500
14960
14961    NLS: Add UTF-8 compose file for Finnish
14962    
14963    From bug report:
14964    
14965      https://bugs.freedesktop.org/show_bug.cgi?id=18747
14966
14967commit 1bd2966ed88f83479a066c6ca7da23a515979550
14968Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14969Date:   Thu Jan 29 20:25:15 2009 -0200
14970
14971    patches to avoid gcc warnings for libX11 (#4)
14972    
14973    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14974    Bug #17946, attachment #19443
14975    
14976    This patch avoids the gcc warning
14977            ../../../../libX11-1.1.5/modules/im/ximcp/imDefLkup.c:223: warning: passing arg 1 of `_XimProcSyncReply' from incompatible pointer type
14978    (same as already done at other places)
14979    
14980    BTW: what is the difference between XIM (the type of ic->core.im)
14981    and Xim ?
14982
14983commit f16dd6af3eb17a25b8ee03d6617a7acc6e919fb0
14984Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14985Date:   Thu Jan 29 20:22:21 2009 -0200
14986
14987    patches to avoid gcc warnings for libX11 (#3)
14988    
14989    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14990    Bug #17946, attachment #19441
14991    
14992    This patch avoids the two gcc warnings
14993            ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:413: warning: assignment discards qualifiers from pointer target type
14994            ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:450: warning: assignment discards qualifiers from pointer target type
14995    
14996    Note, that this as a rather crude fix of the problem (and it is really a
14997    shame to cast name_table to non-const).
14998    
14999    The right solution would be to declare XIMValuesList.supported_values
15000    (in include/X11/Xlib.h) as 'const char **' (or '_Xconst char **').
15001    This will, however, require extensive modifications in various places.
15002
15003commit cce75c5dce73fe1f8626ed9e6798138ada09a860
15004Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
15005Date:   Thu Jan 29 20:20:18 2009 -0200
15006
15007    patches to avoid gcc warnings for libX11 (#2)
15008    
15009    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
15010    Bug #17946, attachment #19440
15011    
15012    Avoid a preprocessor message
15013            <stdin>:194: warning: no newline at end of file
15014    
15015    Two more such warnings (in XkbSAGroup.man and XkbSASetGroup.man)
15016    seem to be caused by a truncated (or otherwise incomplete)
15017    manpage.
15018
15019commit 692baebcc50f1e952800bfe4e2e6bc42f54e62fe
15020Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
15021Date:   Thu Jan 29 20:12:24 2009 -0200
15022
15023    patches to avoid gcc warnings for libX11 (#1)
15024    
15025    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
15026    Bug #17946, attachment #19439
15027    
15028    Define as 1 (one) as done by autoconf and the command line
15029    option, e.g. -DX11_t, not as empty.
15030    
15031    This avoids the gcc (3.4.6) warnings:
15032            ../../libX11-1.1.5/src/x11_trans.c:27:1: warning: "X11_t" redefined
15033            <command line>:7:1: warning: this is the location of the previous definition
15034            ../../libX11-1.1.5/src/x11_trans.c:28:1: warning: "TRANS_CLIENT" redefined
15035            <command line>:8:1: warning: this is the location of the previous definition
15036    
15037    Similarly, follow the autoconf convention to define XTHREADS
15038    and XUSE_MTSAFE_API as one.
15039    
15040    This avoids analogous warnings when compiling libXcomposite,
15041    libXcursor, and libXdamage.
15042    
15043    No reason to AC_SUBST XTHREADS and XUSE_MTSAFE_API (unused).
15044
15045commit a1977883c9f5ef0e515569d6e2ebccb07411f98c
15046Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
15047Date:   Thu Jan 29 15:01:06 2009 -0200
15048
15049    Janitor: Correct some gcc/sparse warnings.
15050    
15051      Most remaining warnings are about XIM/Xim to/from conversion
15052    and discarding const from pointers.
15053
15054commit 8ba0ca32a63c532f128bdca7f1bf982cab8e12be
15055Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
15056Date:   Wed Jan 28 20:31:42 2009 -0200
15057
15058    Janitor: ansification, make distcheck, compiler warnings.
15059    
15060      Only convert to use "ansi prototypes" the functions warned from
15061    compilation with "./autogen.sh --prefix=/usr", on a Linux computer.
15062    
15063      Also, only address "trivial" compiler warning fixes in this commit.
15064    
15065      The new .gitignore is the output of a command like:
15066    % find . -name .gitignore -exec cat {} \; | sort | uniq
15067    and only the toplevel .gitignore file was kept.
15068
15069commit 091c1624fd2f9d933329d6152e4ecd865aa7903a
15070Author: Peter Hutterer <peter.hutterer@who-t.net>
15071Date:   Tue Jan 13 12:05:54 2009 +1000
15072
15073    Fix wrong implies symbol.
15074    
15075    Quote Simos Xenitellis:
15076    > I checked the gtk+ commit logs that go back to 2000, and I see that still it
15077    > was
15078    >
15079    >   { 0x08ce, 0x21d2 }, /*                     implies ⇒ RIGHTWARDS
15080    > DOUBLE ARROW */
15081    >
15082    > In XFree86, it appears there was an error when they converted the
15083    > original table to
15084    > http://cvsweb.xfree86.org/cvsweb/xc/lib/X11/imKStoUCS.c?rev=1.1&content-type=text/vnd.viewcvs-markup
15085    > and the problem still shows up there (November 2000).
15086    
15087    http://lists.freedesktop.org/archives/xorg/2008-October/039743.html
15088    
15089    Reported by Erik Streb del Toro.
15090
15091commit e32521f19e0b07649b7e3a03d56a2bd556b138fb
15092Author: James Cloos <cloos@jhcloos.com>
15093Date:   Sun Dec 7 04:13:34 2008 -0500
15094
15095    [i18n] s/U00DC/Udiaeresis/g
15096    
15097    The xkeyboard-config keyboards generate the symbol Udiaeresis, not
15098    U00DC.  Make sure the relevant Compose sequences expect the symbol
15099    which the keyboards actually send.
15100
15101commit b7502abfe22f8dc009b21cda1172af221d8f9f32
15102Author: James Cloos <cloos@jhcloos.com>
15103Date:   Sun Dec 7 04:09:46 2008 -0500
15104
15105    Revert "For nls/*.pre, allow people to comment lines by starting them with '##'."
15106    
15107    As of commit c9d20e3 the initial double-hashes are replaced with proper C comments.
15108    
15109    This reverts commit a225a0be48770beb689d5ac5da97073634f7deab.
15110
15111commit c9d20e3f697c9cfae5511412023362c1db7449b1
15112Author: James Cloos <cloos@jhcloos.com>
15113Date:   Sun Dec 7 04:08:23 2008 -0500
15114
15115    Use C comments rather than initial doubled hashes to exclude lines from .pre files
15116
15117commit c34ce54d9eac2d8052dc5f205a2ab09866ef5d25
15118Author: vehemens <vehemens@verizon.net>
15119Date:   Sun Dec 7 01:18:26 2008 -0500
15120
15121    [i18n] Distribute new headers which were added for gb18030 support.
15122    
15123    big5hkscs.h and gbk.h, added in 67e34d7a, need to be in SOURCES to
15124    make it into the tar.
15125    
15126    Completes 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51.
15127    
15128    Signed-off-by: James Cloos <cloos@jhcloos.com>
15129
15130commit 418819558d2c60e58b4e3022ce0fadf2143488ac
15131Author: Stefan Dirsch <sndirsch@suse.de>
15132Date:   Sat Nov 22 22:01:07 2008 +0100
15133
15134    Fixed strange font mixups, when fontsets are still used (#2676, Novell #74299).
15135
15136commit 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51a
15137Author: Stefan Dirsch <sndirsch@suse.de>
15138Date:   Sat Nov 22 19:40:54 2008 +0100
15139
15140    Added remaining xlib patch required for gb18030 support (#1573).
15141
15142commit 55782a0a1fe1560f1a9c0ed78bc7f2575c15abcf
15143Author: Stefan Dirsch <sndirsch@suse.de>
15144Date:   Sat Nov 22 17:53:06 2008 +0100
15145
15146    Added remaining hunk of Egbert's patch to prevent XIM deadlocks (#1182).
15147
15148commit c859446c500c883a67f7a86cab1a44844e24dade
15149Author: Ken Thomases <ken@codeweavers.com>
15150Date:   Fri Nov 21 13:58:10 2008 -0500
15151
15152    [i18n] Provide translation from XK_partialderivative (8ef) to Unicode (U2202)
15153    
15154    Signed-off-by: James Cloos <cloos@jhcloos.com>
15155
15156commit 5e68e94d852c730ef9264fc0d8ca61a2ffe98b53
15157Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
15158Date:   Mon Nov 17 20:47:26 2008 -0800
15159
15160    Force local transport when using the launchd socket.
15161    
15162    Fixes a regression due to
15163    bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
15164
15165commit 7aca689ce14d314b5c8c72c8df76f53f76ab467c
15166Author: Peter Hutterer <peter.hutterer@redhat.com>
15167Date:   Fri Nov 14 10:32:50 2008 +1000
15168
15169    Add XF86Suspend, XF86Hibernate to KeysymDB.
15170
15171commit d16b11f25f8265e651def8d80bcd430c0448e664
15172Author: James Cloos <cloos@jhcloos.com>
15173Date:   Tue Nov 11 19:43:39 2008 -0500
15174
15175    [nls] Annotate the Bépo compose sequences
15176    
15177    Add comments with the UCS names.
15178    Add utf-8 strings for each result.
15179    Format for easy reading.
15180
15181commit 730298464240be6f65b32416b3f9b20062c61825
15182Author: James Cloos <cloos@jhcloos.com>
15183Date:   Tue Nov 11 16:41:34 2008 -0500
15184
15185    [nls]  Add some UTF-8 Compose sequences
15186    
15187    As requested for the Bépo keyboard layout (http://clavier-dvorak.org/wiki/).
15188    
15189    Cf. bugs:
15190    
15191      https://bugs.freedesktop.org/show_bug.cgi?id=17821
15192      https://bugs.freedesktop.org/show_bug.cgi?id=17822
15193
15194commit ad6008a0c92733826983eb93f063d3d6276007d8
15195Author: Colin Harrison <colin.harrison@virgin.net>
15196Date:   Thu Nov 6 17:48:21 2008 -0500
15197
15198    Fix copy/paste typo in imLcPrs
15199    
15200    Signed-off-by: James Cloos <cloos@jhcloos.com>
15201
15202commit 0f0168ad18f8a280fc5a689eb02cfaa62d022ea6
15203Author: Adam Jackson <ajax@redhat.com>
15204Date:   Thu Nov 6 14:54:13 2008 -0500
15205
15206    Fix leak in _XimXGetReadData
15207    
15208    Spotted by Denis Dzyubenko
15209
15210commit ae23c25b9349ab1d7ff81f3075c000cf35fc442b
15211Author: Julien Cristau <jcristau@debian.org>
15212Date:   Wed Nov 5 22:09:37 2008 +0100
15213
15214    Bump to 1.1.99.2
15215
15216commit d31e644c65c52828ea3e7abd94a8cf9aee12265c
15217Author: Julien Cristau <jcristau@debian.org>
15218Date:   Wed Nov 5 21:33:13 2008 +0100
15219
15220    Fix distcheck
15221
15222commit e6a7b70cdb2ae8b713012839a0a0bbb93817b8ef
15223Author: Jamey Sharp <jamey@minilop.net>
15224Date:   Wed Oct 29 14:00:33 2008 -0700
15225
15226    Support multiple independent internal sync handlers
15227    
15228    Xlib has several independent tasks that need to be performed with the
15229    display unlocked. It does this by replacing the existing sync handler with
15230    one of a variety of internal sync handlers. However, if multiple internal
15231    sync handlers need to run, then the last one registering wins and
15232    previously registered internal sync handlers are never invoked. This
15233    manifested as a bug with DRI applications on Xlib/XCB as that requires
15234    both an XID handler after every XID allocation, and the periodic sequence
15235    number handler. The XID handler would win, and the sequence number handler
15236    would never be invoked.
15237    
15238    Fix this by unifying the internal sync handler mechanism into a single
15239    function that calls all of the known internal sync handlers. They all need
15240    to deal with being called when not strictly necessary now.
15241    
15242    Signed-off-by: Keith Packard <keithp@keithp.com>
15243    Signed-off-by: Jamey Sharp <jamey@minilop.net>
15244    Signed-off-by: Josh Triplett <josh@freedesktop.org>
15245
15246commit 2dbaaab9c4e3894b33dcae850551dee5473431d5
15247Author: Keith Packard <keithp@keithp.com>
15248Date:   Sat Oct 11 21:44:21 2008 -0700
15249
15250    Ensure that _XReadEvents always leaves an event in the queue on return
15251    
15252    XNextEvent assumes that the event queue will be non-empty on return from
15253    _XReadEvents, but with multiple event readers running, the previous change
15254    could leave the queue empty on return from process_responses. Re-invoke
15255    process_responses until the queue is non-empty.
15256    
15257    Signed-off-by: Keith Packard <keithp@keithp.com>
15258
15259commit bedfe68259037c5564fe52758c92b9c97729640a
15260Author: Keith Packard <keithp@keithp.com>
15261Date:   Sat Oct 11 21:10:23 2008 -0700
15262
15263    Permit only one Xlib thread to block waiting for events
15264    
15265    As Xlib queues events internally, we must prevent multiple Xlib threads from
15266    entering XCB to wait for an event in case the queued event is to be
15267    delivered to the thread which didn't manage to read it. In other words, let
15268    only one Xlib thread into xcb_wait_for_event at a time.
15269    
15270    Jamey Sharp looked over my shoulder while making this fix and, while hating
15271    my whitespace conventions, appears happy enough with the actual code.
15272    
15273    Signed-off-by: Keith Packard <keithp@keithp.com>
15274
15275commit cc19618d2eb3ed92a0b574aee26a7da8b4aed5d2
15276Author: Jamey Sharp <jamey@minilop.net>
15277Date:   Sun Mar 23 16:33:50 2008 -0700
15278
15279    Fix XAllocID race: hold the user display lock until we have a new XID.
15280    
15281    Xlib built --without-xcb is also vulnerable to this race, and a similar
15282    fix might work there too.
15283    
15284    Also, use an XID that's truly invalid while waiting for the next XID to be
15285    requested.
15286
15287commit 54e5c0941b0ded1628d559a9f0a3451ea96c299b
15288Author: Josh Triplett <josh@freedesktop.org>
15289Date:   Sat Mar 15 17:22:23 2008 -0700
15290
15291    Use XCB's new socket handoff mechanism rather than the old XCB Xlib lock.
15292    
15293    Previously, Xlib/XCB used XCB's Xlib lock to prevent XCB from sending
15294    requests between calls to Xlib's LockDisplay and UnlockDisplay macros.
15295    Xlib/XCB then sent all of its requests using XCB's xcb_send_request, and
15296    had to flush its requests when unlocking the display.
15297    
15298    XCB 1.2 adds a new socket handoff mechanism, xcb_take_socket.  Replace
15299    much of the existing Xlib/XCB implementation with the use of
15300    xcb_take_socket to take ownership of the write side of the X connection
15301    socket, and a return_socket callback which writes any outstanding requests
15302    with xcb_writev.  This approach allows Xlib/XCB to use the same buffering
15303    as traditional Xlib did.  In particular, programs which use Xlib/XCB and
15304    never make XCB calls will never need to hand the socket back to XCB, and
15305    vice versa.
15306    
15307    This allows us to discard large quantities of synchronization code from
15308    Xlib/XCB, together with the synchronization bugs present in that code.
15309    Several test cases which previously failed now work perfectly, including
15310    multi-threaded ico.  In addition, the infamous locking correctness
15311    assertions, triggered when double-locking or when unlocking without a
15312    previous lock, no longer exist, because Xlib/XCB no longer has any reason
15313    to care more about application locking than traditional Xlib does.
15314    
15315    Furthermore, the handoff approach provides great improvements to
15316    performance.  Results from x11perf's XNoOp test, which represented the
15317    worst case for the lock-based Xlib/XCB:
15318    
15319    Traditional Xlib:       average 19100000/sec
15320    Lock-based Xlib/XCB:    average  3350000/sec
15321    Handoff-based Xlib/XCB: average 17400000/sec
15322    
15323    Thus, for no-ops, the handoff mechanism provides more than a 4x speedup to
15324    Xlib/XCB, bringing Xlib/XCB within 9% of traditional Xlib no-op
15325    performance.  Of course, real-world workloads do not use no-op, so your
15326    mileage may vary.  In particular, since no-ops represent the worst case,
15327    we expect real workloads to more closely match the performance of
15328    traditional Xlib.
15329    
15330    While removing synchronization code, we changed _XReply to not drop any
15331    locks when calling xcb_wait_for_reply; previously, we had to carefully
15332    avoid a deadlock between the Display lock and the XCB Xlib lock. Holding
15333    the locks reduces implementation complexity and should not impact
15334    applications.
15335    
15336    Commit by Jamey Sharp and Josh Triplett.
15337    XCB's handoff mechanism inspired by Keith Packard.
15338
15339commit 5a19ac473f7a8046b0421fbd5d53da160c22ed75
15340Author: Chris Ball <cjb@laptop.org>
15341Date:   Mon Nov 3 22:57:29 2008 -0500
15342
15343    Remove configure check for xcb-xlib.
15344    
15345    xcb-xlib has been intentionally removed from libxcb; stop checking for
15346    it at configure-time.
15347
15348commit 34b35dda0bb7f3cf0ad9ab95ad7953d35d24f71b
15349Author: Josh Triplett <josh@freedesktop.org>
15350Date:   Wed Oct 29 14:37:44 2008 -0700
15351
15352    .gitignore: Add dolt files
15353
15354commit 1290cccf2d90083eba852f5f413f7e3dff48ccd2
15355Author: Peter Hutterer <peter.hutterer@redhat.com>
15356Date:   Tue Oct 28 11:56:55 2008 +1030
15357
15358    man: fix formatting error in XkbGetIndicatorState man page.
15359
15360commit b1022fa6d7e97640049e93ffa108083fc8d71b05
15361Author: James Cloos <cloos@jhcloos.com>
15362Date:   Sat Oct 25 09:13:08 2008 -0400
15363
15364    Increase size of working arrays in the makekeys utility program.
15365    
15366    Makekeys is used to create an optimal hash of the keysyms defined
15367    in x11proto’s keysymdef.h.
15368    
15369    The recent addition of new keysyms there has triggered a bug in
15370    makekeys where it tries to use a zero on the rhs of the % (mod)
15371    operator (resulting in a divide by zero error) whenever it fails
15372    to find a solution within its constraints.
15373    
15374    Increasing the size of the arrays allows it to find a solution for
15375    the current set of keysyms.
15376    
15377    Makekeys is only run durring the build process, so this has no impact
15378    on users of libX11, only on the amount of VM needed to build it.
15379    
15380    It still needs a more complete fix, but this allows compiles to
15381    progress until that is completed.
15382
15383commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60
15384Author: James Cloos <cloos@jhcloos.com>
15385Date:   Sat Oct 11 01:03:14 2008 -0400
15386
15387    Dolt-ify
15388    
15389    Add dolt to acinclude.m4 and call it it configure.ac to speed compiles.
15390
15391commit 39c0b266cac8cbc15bf501d7869186862f01d823
15392Author: Peter Hutterer <peter.hutterer@redhat.com>
15393Date:   Wed Oct 15 14:30:20 2008 +1030
15394
15395    Add more keysyms for PS3 BD remotes, Ericsson Phones #16519
15396    
15397    X.Org Bug 16519 <https://bugs.freedesktop.org/show_bug.cgi?id=16519>
15398
15399commit d23aad31338e7d869d878d5aa1b6b91d20287005
15400Author: Peter Hutterer <peter.hutterer@redhat.com>
15401Date:   Mon Oct 13 09:41:59 2008 +1030
15402
15403    Add XF86Battery, XF86Bluetooth, XF86WLAN, XF86UWB to keysymdb.
15404
15405commit 214ea6f5fd6aeaa7303ea4a69f9aedabf219ec4c
15406Author: Peter Hutterer <peter.hutterer@who-t.net>
15407Date:   Thu Jul 24 15:44:26 2008 +0930
15408
15409    xkb: fix out-by-1 error in _XkbWriteKeyExplicit.
15410    
15411    Thanks to Michael Meeks, Novell Bug 369263.
15412    https://bugzilla.novell.com/show_bug.cgi?id=369263
15413
15414commit e7ece39afc8e0adc3b6b1e70b337b98376754462
15415Author: Alan Coopersmith <alan.coopersmith@sun.com>
15416Date:   Tue Oct 7 15:41:38 2008 -0700
15417
15418    Sun bug #6739431: double free in _X11TransConnectDisplay()
15419    
15420    Double free() introduced in bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
15421    After copying original_hostname to phostname, set original_hostname
15422    to NULL, so we don't free the same pointer twice when we free both
15423    original_hostname and phostname.
15424    
15425    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6739431>
15426
15427commit 0877bc916afbd1ff8f1833edc930b765ea783576
15428Author: Daniel Stone <daniel@fooishbar.org>
15429Date:   Tue Sep 23 19:02:02 2008 +0300
15430
15431    configure.ac: Fix CC_FOR_BUILD logic error
15432    
15433    Turns out we were accidentally smashing it so that you couldn't set it
15434    externally at all.  Oops.
15435
15436commit 58bf3aa746908f01c37be7045699e43a4e270944
15437Author: John Tapsell <johnflux@gmail.com>
15438Date:   Tue Sep 23 17:30:13 2008 +0300
15439
15440    Build: Use native compiler for makekeys
15441    
15442    makekeys needs to be run during the build process, as opposed to on the
15443    target, so build it with either of gcc or cc to fix cross-compiling.
15444    This can be overridden by setting $CC_FOR_BUILD.
15445
15446commit 340422a5c7a413faef18666cada27cee14615250
15447Author: Adam Jackson <ajax@redhat.com>
15448Date:   Wed Sep 17 12:54:34 2008 -0400
15449
15450    Fix the previous patch for the BadFont case.
15451
15452commit 2335eafe4b53c27f6f9ee1bab3e1f5842f896428
15453Author: Matthias Clasen <mclasen@redhat.com>
15454Date:   Wed Sep 17 10:43:52 2008 -0400
15455
15456    Bug #17616: Fix an XCB leak when the client has a non-fatal error handler.
15457
15458commit db0b85db29699be6bf7e78dede655d59ba926dfc
15459Author: Rafael Ávila de Espíndola <rafael.espindola@gmail.com>
15460Date:   Sun Sep 14 19:15:26 2008 -0400
15461
15462    Fix problem with <dead_acute> <c> in pt_BR.UTF-8
15463    
15464    The <dead_acute> <C> and <dead_acute> <c> lines in the pt_BR UTF-8
15465    Compose file show "Ç" and "ç" (c with cedilla accent) (akin to the
15466    ISO 8859 pt_BR Compose file) as the string but specify the keysym
15467    and comment for Ć and ć (c with acute accent).
15468    
15469    This commit normalizes those two lines to match the specified string.
15470    
15471    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=4671
15472    
15473    Signed-off-by: James Cloos <cloos@jhcloos.com>
15474
15475commit b065c011baa69b69e3ea77c30d5e153c0d103e2d
15476Author: Michael Verret <michael.verret@gmail.com>
15477Date:   Mon Sep 8 16:33:35 2008 -0400
15478
15479    Fix documentation typo
15480    
15481    Signed-off-by: James Cloos <cloos@jhcloos.com>
15482
15483commit 4213ea95185377bdd1b51e82933f331fc0f52e5b
15484Author: James Cloos <cloos@jhcloos.com>
15485Date:   Sat Sep 6 04:19:19 2008 -0400
15486
15487    Remove extraneous <angle brackets> from the Ethiopic Compose file.
15488    
15489    The am_ET.UTF-8 Compose file submitted in:
15490    
15491    https://bugs.freedesktop.org/show_bug.cgi?id=11307
15492    
15493    for the OLCP project used incorrect syntax.  (It has angle brackets around the
15494    Uxxxx symbols on the right hand side rather than only on the left hand side).
15495    
15496    This bug is noted in OLPC’s ticket:
15497    
15498    http://dev.laptop.org/ticket/7474
15499    http://dev.laptop.org/attachment/ticket/7474/olpc_7474_dead_vowels_libX11.patch
15500
15501commit 9df84b513dd2b6e65e6d528cfac6d4cc3ea46918
15502Author: James Cloos <cloos@jhcloos.com>
15503Date:   Mon Sep 1 17:49:33 2008 -0400
15504
15505    Complete the set of vulgar fractions
15506    
15507    Unicode 1.1 added thirds, fifths, sixths and eights;
15508    we might as well catch up.
15509    
15510    (Unicode and ISO 10646 have 1/7 (U2150), 1/9 (U2151), 1/10 (U2152)
15511    and 0/3 (U2189) in their pipelines, but those four can be added
15512    here after they are published.)
15513
15514commit a788792e9de95f8db0639557859722a35087481d
15515Author: James Cloos <cloos@jhcloos.com>
15516Date:   Wed Aug 20 15:28:07 2008 -0400
15517
15518    nls (en_US) Re-remove long compositions that override shorter
15519    
15520    As reported in <https://bugs.freedesktop.org/show_bug.cgi?id=17228>:
15521    
15522      Commit a6f4bbf7
15523        nls (en_US): remove long compositions that override shorter [...]
15524        removed some longer compose sequences because there are shorter
15525        ones which take preference over the longer. For example the
15526        sequences:
15527    
15528          <Multi_key> <apostrophe> <comma> <c>   : U1E09 # ḉ
15529          <Multi_key> <apostrophe> <comma> <C>   : U1E08 # Ḉ
15530    
15531        were removed becase there already was:
15532    
15533          <Multi_key> <apostrophe> <comma>       : U201A # ‚
15534    
15535      Then commit 4ba09125
15536        Work on making the en_US and pt_BR UTF-8 Compose as similar as
15537        possible added exactly the same key sequences again. Obviusly
15538        they won't work.
15539
15540commit 55248e5c84c3fd8c349a3bb4cb15a1ec86989d74
15541Author: James Cloos <cloos@jhcloos.com>
15542Date:   Thu Jul 17 21:01:42 2008 -0400
15543
15544    Add more <Multi_key> <cedilla> Compose tuples
15545    
15546    The last commit missed the el_GR UTF-8 Compose.pre as well as
15547    the various ISO 8859 locales which have compose sequences
15548    generating ‘WITH CEDILLA’ characters.
15549    
15550    (Interestingly, some of the 8859 locales already supported
15551    <Multi_key> <cedilla> for some CEDILLA characters, but not
15552    for Ç or ç.)
15553    
15554    This is further work on bug 10397.
15555
15556commit 4ba091255bb953d53078ba5619d6751052c739f7
15557Author: James Cloos <cloos@jhcloos.com>
15558Date:   Thu Jul 17 17:16:50 2008 -0400
15559
15560    Work on making the en_US and pt_BR UTF-8 Compose as similar as possible.
15561    
15562    The eventual goal here is to have a single primary UTF-8 Compose
15563    file which the locale-specific UTF-8 Compose.pre files can #include.
15564
15565commit 254522d3c24e0590732fc03cdd61ff4564819d94
15566Author: James Cloos <cloos@jhcloos.com>
15567Date:   Thu Jul 17 17:13:36 2008 -0400
15568
15569    Add <Multi_key> <cedilla> Compose tuples
15570    
15571    The en_US and pt_BR UTF-8 Compose tables had support for using <comma>
15572    with <Multi_key> to enter CEDILLA characters.  Bug 10397 requests
15573    support for using <cedilla> instead of <comma> in said sequences.
15574    
15575    This commit makes both styles work.
15576
15577commit 7dc907f6032e1d5cbe4da0e414bdf2c569c04b44
15578Author: James Cloos <cloos@jhcloos.com>
15579Date:   Sat Jun 28 15:25:23 2008 -0400
15580
15581    Fix commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01
15582    
15583    The new block was added twice to the en_US.UTF-8 Compose.pre;
15584    delete the duplicate.
15585
15586commit 596e081b7457dcd1c4ad555ac140e6999239bc0d
15587Author: Peter Hutterer <peter@cs.unisa.edu.au>
15588Date:   Sat Jun 28 20:14:05 2008 +0930
15589
15590    Fix unbalanced parenthesis in XKBlib.h # 16551
15591    
15592    X.Org Bug 16551 <http://bugs.freedesktop.org/show_bug.cgi?id=16551>
15593
15594commit f6af6dd2f76c12b56ec166bb771457b9f08fe246
15595Author: Adam Jackson <ajax@redhat.com>
15596Date:   Tue Jun 24 13:16:53 2008 -0400
15597
15598    Bug #14898: Don't abuse the sprintf() implementation.
15599    
15600    The thing you're printing into should not itself appear in the list of
15601    things to print from, that's bad juju.  Just use strcat().
15602
15603commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01
15604Author: Khaled Hosny <khaledhosny@eglug.org>
15605Date:   Thu Jun 19 18:26:11 2008 -0400
15606
15607    NLS: Add Arabic Lam-Alef ligature compose sequences (bug #16426)
15608    
15609    Add some Arabic digraphs to utf-8 locales with a Compose.pre
15610    
15611    Signed-off-by: James Cloos <cloos@jhcloos.com>
15612
15613commit bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
15614Author: Alan Coopersmith <alan.coopersmith@sun.com>
15615Date:   Wed Jun 18 20:00:25 2008 -0700
15616
15617    Rework code to choose local connection types and fallback to others
15618    
15619    Adds --with-local-transport-order configure flag if you don't like the
15620    default ordering (which is platform dependent)
15621    
15622    Includes fixes for these Sun/Solaris bug ids:
15623    6678250 X Commands returning incorrect display value unix:0.0 not <system>:0.0
15624    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6678250>
15625    6716481 libX11 should prefer Unix domain sockets over named pipes on Solaris
15626    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6716481>
15627
15628commit cf49e537014c3cb5aaee07e57400933e0bb72b6b
15629Author: Alan Coopersmith <alan.coopersmith@sun.com>
15630Date:   Tue Jun 17 14:41:17 2008 -0700
15631
15632    Strip whitespace from end of lines in source files
15633
15634commit f76fd81dfbbd5cfae75c87ce0511e88e08529cf3
15635Author: Jeff Smith <whydoubt@yahoo.com>
15636Date:   Sun Jun 15 23:52:20 2008 -0500
15637
15638    Fix memory leak in XOpenDisplay
15639    
15640    Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
15641
15642commit fca0b0ba3f72b7284601d4690bba99fc80a92614
15643Author: Jens Herden <jens@khmeros.info>
15644Date:   Tue Jun 10 20:07:30 2008 +0300
15645
15646    NLS: Add Khmer compose sequences (bug #5706)
15647    
15648    Add some Khmer digraphs to all locales with a Compose.pre.
15649
15650commit e54cffb649b1622c17457e470cfab8cc56d38c97
15651Merge: 19802ccd 721b574d
15652Author: Daniel Stone <daniel@fooishbar.org>
15653Date:   Tue Jun 10 20:04:30 2008 +0300
15654
15655    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
15656
15657commit 721b574d36f1884c3f1bf7bd933646e2ed6680b5
15658Author: Peter Hutterer <peter@cs.unisa.edu.au>
15659Date:   Thu May 29 10:57:21 2008 +0930
15660
15661    Bump to 1.1.99.1
15662
15663commit 631d32d13247d1cf52c0833d438c5b38b01b17a4
15664Author: Peter Hutterer <peter@cs.unisa.edu.au>
15665Date:   Wed May 28 17:31:59 2008 +0930
15666
15667    Require xproto 7.0.13 and libxcb 1.1.90 (for GenericEvents)
15668
15669commit e9195db7257e418f83707233baeeb84b225caf4f
15670Merge: c34f76f4 a7f85567
15671Author: Peter Hutterer <peter@cs.unisa.edu.au>
15672Date:   Thu May 22 12:14:28 2008 +0930
15673
15674    Merge branch 'master' into xge
15675
15676commit 19802ccd3909145e3ba2f6c073271cb5f3701685
15677Author: Daniel Stone <daniel@fooishbar.org>
15678Date:   Mon May 19 19:22:31 2008 +0300
15679
15680    gitignore: Update with loads more bits from server
15681
15682commit a7f85567a3e850fba0c44571453d2852ab1a09be
15683Author: Adam Jackson <ajax@redhat.com>
15684Date:   Tue May 13 10:28:39 2008 -0400
15685
15686    Bug #15884: Remove useless sleep()'s from the connection code.
15687    
15688    For network transports, there's enough delay in the network layer
15689    already without adding more.  For local transports, just hurry up
15690    and fail if the server isn't there.
15691
15692commit c34f76f475bc632490122e67b5a82575d69d5569
15693Author: Peter Hutterer <peter@cs.unisa.edu.au>
15694Date:   Mon May 12 21:46:24 2008 +0930
15695
15696    Pull down extra bytes when reading a GenericEvent (non-xcb).
15697    
15698    I refuse to take any responsibily for this code. It works, I guess.
15699    But - all the flushing is done somewhere before that, so we might need to
15700    flush here. Under some circumstances anyway. Don't ask me, I'm an optical
15701    illusion.
15702    
15703    Build with xcb as transport layer highly recommended.
15704
15705commit c9b2ff1e6a607463993afa4a8d085857d97cc2f3
15706Merge: 17d7dcbf 9129057b
15707Author: Peter Hutterer <peter@cs.unisa.edu.au>
15708Date:   Mon May 12 17:58:37 2008 +0930
15709
15710    Merge branch 'master' into xge
15711
15712commit 9129057bdbff0ec9cd8bb780cf7f85f134a291eb
15713Author: Teemu Likonen <tlikonen@iki.fi>
15714Date:   Wed May 7 21:44:22 2008 +0300
15715
15716    Change <dead_belowdot> to <dead_belowring> for U+1E00 and U+U1E01
15717    
15718    Commit 6b6caeea830a977bdb54688cfb648d879821e752 added <dead_belowdot>
15719    <A> and <dead_belowdot> <a> compose sequences for letters U+1E00 and
15720    U+U1E01 (LATIN CAPITAL/SMALL LETTER A WITH RING BELOW). This caused
15721    duplicate compose sequences since these have already been defined. Also,
15722    using <dead_belowring> is more logical since the diacritic is indeed
15723    a "RING BELOW".
15724
15725commit 01a9cb58888d290cc3d319feec4ee4a0297a844c
15726Author: Daniel Stone <daniel@fooishbar.org>
15727Date:   Wed May 7 20:04:44 2008 +0300
15728
15729    NLS: Make UTF-8 the default for Russian
15730    
15731    No-one uses 8859-5 anymore, so make the default for Russian UTF-8; the
15732    only other possible answer would be KOI8-R.
15733    
15734    Signed-off-by: Sergey V. Udaltsov <sergey.udaltsov@gmail.com>
15735
15736commit 407b81bfbbabf6feb565d6da22f9ef9a69016ab8
15737Author: Ross Burton <ross@burtonini.com>
15738Date:   Tue Apr 29 13:38:10 2008 +0300
15739
15740    NLS: Add interrobang to UTF-8 compose tables (bug #15653)
15741    
15742    It is what it says on the box.
15743
15744commit 0b6682303e9c61fefc3818acfda616b1e3691abf
15745Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
15746Date:   Mon Apr 28 11:51:25 2008 +0300
15747
15748    IM: Respect XMODIFIERS for Thai locale (bug #15719)
15749    
15750    When looking at Thai input methods, make sure XMODIFIERS is checked
15751    before jumping straight into built-in Thai processing, so external XIM
15752    servers such as SCIM can be used with Thai.
15753
15754commit c13aded1b2f830ba5004abb0ec5518f9ea16087e
15755Author: Colin Harrison <colin.harrison-at-virgin.net>
15756Date:   Sat Apr 26 18:56:05 2008 +0100
15757
15758    Fix missing error condition
15759
15760commit f5c5ffc175cb383c92ea0fa8c08cfb087c5f3083
15761Author: Colin Harrison <colin.harrison-at-virgin.net>
15762Date:   Mon Apr 21 17:24:33 2008 +0100
15763
15764    Xlib warning fixes
15765
15766commit 6b6caeea830a977bdb54688cfb648d879821e752
15767Author: James Cloos <cloos@jhcloos.com>
15768Date:   Fri Apr 18 02:50:55 2008 -0400
15769
15770    Add some dead_key sequences to en_US.UTF-8 Compose table
15771    
15772    Make use of the new dead key symbols added to x11proto’s
15773    commit 44e24a27bca023cf7b799f191fe6d52e12efbe5f (which
15774    was in responce to bug #15446).
15775
15776commit 8f9b039580deaf658e464b7d6254064fcf183df6
15777Author: Alan Coopersmith <alan.coopersmith@sun.com>
15778Date:   Mon Apr 14 19:09:42 2008 -0700
15779
15780    Update ac_define_dir macro in acinclude.m4 to 2008-04-12 version
15781
15782commit 9f5e96eb91ab55dd441c3e94b75caf48c588778f
15783Author: Alan Coopersmith <alan.coopersmith@sun.com>
15784Date:   Mon Apr 14 18:21:14 2008 -0700
15785
15786    Fix mismatched brace indenting
15787
15788commit a19f9c65ee9e5e5d783feaa84998c36439b0288b
15789Author: Bart Massey <bart@cs.pdx.edu>
15790Date:   Fri Apr 4 18:58:45 2008 -0700
15791
15792    added error check in Xcms color file parser; closes bug #15305
15793
15794commit 12e8d0d01dd72ce98e7683ddb1bde181b7ed246f
15795Author: Christian Weisgerber <naddy@mips.inka.de>
15796Date:   Tue Mar 18 07:30:05 2008 +0100
15797
15798    ConnDis: properly cast 'addr' before accessing it as a byte array.
15799    
15800    If you use XDM-AUTHORIZATION-1 authorization keys for remote X11
15801    clients over IPv6, the clients are liable to segfaults.
15802
15803commit 64325f38bab082a8e0e9ce779a8e582de5c8588e
15804Author: Josh Triplett <josh@freedesktop.org>
15805Date:   Sat Mar 15 12:29:33 2008 -0700
15806
15807    Fix fd.o bug 15023: make Xlib sync correctly given many void requests
15808    
15809    If given many requests without replies, Xlib may not sync until it flushes
15810    the output buffer.  Thus, if Xlib can fit enough requests in the buffer to
15811    pass by the number of requests it would normally sync after (65536 -
15812    BUFSIZE/sizeof(xReq)), it will sync too late.  The test case in bug 15023
15813    demonstrated this by issuing a request with a reply (ListExtensions) at
15814    just the right time to get confused with the GetInputFocus reply issued in
15815    response to the sync 65,536 requests later; the test case used an async
15816    handler to watch the replies, since otherwise it could not issue a request
15817    without waiting for the response.  When the test case failed, Xlib's sync
15818    handler would eat the ListExtensions reply, and the test case's async
15819    handler would see the GetInputFocus reply.
15820    
15821    Fix this by replacing SEQLIMIT with a function sync_hazard() that uses the
15822    buffer size to figure out when the sequence numbers could potentially wrap
15823    before the next flush.
15824    
15825    With this commit, the test case consistently passed, and the async reply
15826    handler always saw the ListExtensions reply.
15827    
15828    Commit by Jamey Sharp and Josh Triplett.
15829
15830commit a5395563bbee15fabe1e8fd7aa86f9f314d8d30e
15831Author: Colin Harrison <colin.harrison@virgin.net>
15832Date:   Sat Mar 15 13:39:13 2008 -0400
15833
15834    Fix typo
15835    
15836    Signed-off-by: James Cloos <cloos@jhcloos.com>
15837
15838commit f07585ca27a8487bc66dfe41486c823f0fdcea7d
15839Author: Daniel Stone <daniel@fooishbar.org>
15840Date:   Sat Mar 15 17:32:57 2008 +0200
15841
15842    configure.ac: Don't search for legacy X11 headers
15843    
15844    This can actually break cross-compiles, so don't do it anymore.
15845
15846commit bf69541238c7df6606340c0f389e5c47149b29c7
15847Author: Matthieu Herrb <matthieu.herrb@laas.fr>
15848Date:   Sun Mar 9 09:08:07 2008 +0100
15849
15850    nuke RCS Ids
15851
15852commit 5e98aed13e529638df744e45893c471d5f2014fb
15853Author: Adam Jackson <ajax@redhat.com>
15854Date:   Thu Mar 6 16:10:33 2008 -0500
15855
15856    libX11 1.1.4
15857
15858commit 8e085971dc661da9f80ff6b67747459c0fb15c08
15859Author: Alan Coopersmith <alan.coopersmith@sun.com>
15860Date:   Thu Feb 28 20:17:41 2008 -0800
15861
15862    Man page typo fixes
15863
15864commit 1a1a42a3ca1dfaf42f1094936b71c140fc030fcb
15865Author: Søren Sandmann Pedersen <sandmann@redhat.com>
15866Date:   Sun Feb 24 20:03:35 2008 -0500
15867
15868    XIM: Fix a hand when switching input context.
15869    
15870    Red Hat bug #201284.
15871
15872commit e02e4ccafcaf3eb8993152dfcbfbee0240ea2db2
15873Author: Adam Jackson <ajax@redhat.com>
15874Date:   Sun Feb 24 20:00:43 2008 -0500
15875
15876    Bug #14029: Don't LockDisplay() recursively.
15877    
15878    See also Red Hat bugzilla #326461.
15879
15880commit e5892467ae3308c8651be76e06db322dcbc08522
15881Author: Alan Coopersmith <alan.coopersmith@sun.com>
15882Date:   Fri Feb 15 17:27:53 2008 -0800
15883
15884    Add support for building lint library with --enable-lint-library
15885
15886commit e3eb83ec6a9bffa63cdffd94f077c12f85ad7240
15887Author: Alan Coopersmith <alan.coopersmith@sun.com>
15888Date:   Mon Feb 11 20:11:43 2008 -0800
15889
15890    Spell out number in XkbGetKeyVirtualModMap man page to avoid cpp errors from #
15891
15892commit d5ceed7a73a6b61758ddb6ff4e194955fbd5c185
15893Author: Alan Coopersmith <alan.coopersmith@sun.com>
15894Date:   Fri Feb 8 16:46:46 2008 -0800
15895
15896    Add WM_LOCALE_NAME to list of properties set in XSetWMProperties comment
15897
15898commit 416a812200f24d19149dcc497e5c51a0608120f6
15899Author: Alan Coopersmith <alan.coopersmith@sun.com>
15900Date:   Fri Feb 8 15:31:31 2008 -0800
15901
15902    XErrorDB updates for Render 0.9 & XFixes 4.0
15903
15904commit 16a76091cd632e5a3708e235ff864b58f3e4613e
15905Author: Kim Woelders <kim@woelders.dk>
15906Date:   Sat Dec 22 21:45:23 2007 +0100
15907
15908    Fix bs_BA entries in locale.dir.pre
15909    
15910    X.Org bug#13786 <http://bugs.freedesktop.org/show_bug.cgi?id=13786>
15911
15912commit 32115c563b87d2f37e3f9de70fbd0f4d9e424aea
15913Merge: 8f0bd3f4 e8d4cefa
15914Author: James Cloos <cloos@jhcloos.com>
15915Date:   Fri Dec 14 22:43:47 2007 -0500
15916
15917    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
15918
15919commit 8f0bd3f445cbdcc67650f6e8e3baf2ed89cb3695
15920Author: James Cloos <cloos@jhcloos.com>
15921Date:   Fri Dec 14 22:42:59 2007 -0500
15922
15923    add a comment to en_US.UTF-8/Compose
15924
15925commit 4d6c45e60ed13d3b0fea10413873d6a74f9d6a3b
15926Author: James Cloos <cloos@jhcloos.com>
15927Date:   Fri Dec 14 22:40:13 2007 -0500
15928
15929    Add <dead_stroke> compose sequences
15930    
15931    The added sequences match the existing <Multi_key> <slash> sequences.
15932    
15933    This is related to bug #12765¹.
15934    
15935    1] https://bugs.freedesktop.org/show_bug.cgi?id=12765
15936
15937commit e8d4cefa0837afa149a10e981528b368485a9e38
15938Author: Jeremy Huddleston <jeremy@yuffie.local>
15939Date:   Mon Dec 10 23:00:44 2007 -0800
15940
15941    Added launchd support.
15942
15943commit 17d7dcbfced4a9417b33507bd3fd9b7dd8268242
15944Merge: 5dfefd38 13ac8046
15945Author: Peter Hutterer <peter@cs.unisa.edu.au>
15946Date:   Thu Dec 6 14:08:29 2007 +1030
15947
15948    Merge branch 'master' into xge
15949
15950commit 5dfefd3829d3ba7e41d5db0ad28e9dfee92fadd5
15951Merge: a68a1cd7 eff33ae5
15952Author: Peter Hutterer <peter@cs.unisa.edu.au>
15953Date:   Thu Dec 6 13:57:09 2007 +1030
15954
15955    Merge branch 'master' into xge
15956    
15957    Conflicts:
15958    
15959            src/xcb_io.c
15960
15961commit 13ac80469f6958cabac596834e203bd9cb6d4c94
15962Author: James Cloos <cloos@jhcloos.com>
15963Date:   Wed Dec 5 20:14:03 2007 -0500
15964
15965    Update the currency symbols block of en_US.UTF-8/Compose.pre
15966    Add XCOMM lines a la the existing NEW SHEQEL SIGN entry for
15967    the KIP, TUGRIK, DRACHMA. GERMAN PENNY, PESO, GUARANI,
15968    AUSTRAL, HRYVNIA and CEDI SIGNs.
15969
15970commit b0a8f2ec4ba698841683f8ce389f9d72e6bce53e
15971Author: Anton Zinoviev <anton@lml.bas.bg>
15972Date:   Wed Dec 5 19:56:03 2007 -0500
15973
15974    Additions to the Compose file for UTF-8
15975    From bug #5371¹
15976    
15977    Commit 5cf5bc76642bfece7cb5b76faf414bf445f14489 left out this change
15978    from those in attachment #4122².  The post³ on xorg resulted in only
15979    a positive reply⁴ from Daniel, so this block is now also commited.
15980    
15981    1] https://bugs.freedesktop.org/show_bug.cgi?id=5371
15982    2] https://bugs.freedesktop.org/attachment.cgi?id=4122
15983    3] http://article.gmane.org/gmane.comp.freedesktop.xorg/20628
15984    4] http://article.gmane.org/gmane.comp.freedesktop.xorg/23966
15985    
15986    Signed-off-by: James Cloos <cloos@jhcloos.com>
15987    Acked-by:  Daniel Stone <daniel@fooishbar.org>
15988
15989commit 438d02ebc08ee171cf1d3936f4c81050d428ab92
15990Author: James Cloos <cloos@jhcloos.com>
15991Date:   Tue Dec 4 17:25:39 2007 -0500
15992
15993    Fix the <U\x+> keysyms in the en_US.UTF-8 Compose file
15994    
15995    Based on src/KeysymStr.c and src/StrKeysym.c and comments in
15996    bugs #11930¹ and #5129² it is clear that <U100XXXXX> is invalid;
15997    those should be in the form U plus the hex of the UCS Code Point.
15998    
15999    The 0x01000000 is ORed in by the code.
16000    
16001    This update fixes all of those.
16002    
16003    1] https://bugs.freedesktop.org/show_bug.cgi?id=11930
16004    2] https://bugs.freedesktop.org/show_bug.cgi?id=5129
16005
16006commit 02e04059c89e175f51647e3b031344f743286b34
16007Merge: 1254c57d b57129ef
16008Author: James Cloos <cloos@jhcloos.com>
16009Date:   Tue Dec 4 06:55:04 2007 -0500
16010
16011    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
16012
16013commit 1254c57dd3a8d6ea87041b2f63024f99094f290f
16014Author: James Cloos <cloos@jhcloos.com>
16015Date:   Tue Dec 4 06:53:55 2007 -0500
16016
16017    Use the new dead_psili and dead_dasia keysyms added to proto/x11proto 7.0.11
16018    Inspired by bug 11930¹:
16019    
16020    Commit 40ed4eef92e31fcf7ea0a436e1a00cdf49484c1b to x11proto added dead_psili
16021    and dead_dasia keysyms.  Make use of them in the en_US.UTF-8 and el_GR.UTF-8
16022    Compose files.
16023    
16024    This was done with a pair of perl scripts based on the one quoted in the
16025    log for commit c76d30253f1483ac8200ad5c032a818907e65030.
16026    
16027    1] https://bugs.freedesktop.org/show_bug.cgi?id=11930
16028
16029commit b57129ef324c73ee91c2a796b800c4b45f4d4855
16030Author: Jeremy Huddleston <jeremy@yuffie.local>
16031Date:   Mon Dec 3 20:04:19 2007 -0800
16032
16033    Use __APPLE__ instead of __DARWIN__
16034
16035commit b9c032e1d5ed08510863dfb91b8bda588e6c8c9e
16036Author: Jeremy Huddleston <jeremy@yuffie.local>
16037Date:   Wed Nov 28 16:43:49 2007 -0800
16038
16039    Define __DARWIN__ when host_os == darwin* as needed in SetLocale.c
16040
16041commit 4b91ed099554626f1ec17d5bdf7bd77ce1a70037
16042Author: Alan Coopersmith <alan.coopersmith@sun.com>
16043Date:   Wed Nov 21 16:50:04 2007 -0800
16044
16045    X.Org Bug #4312: incorrect comment asterisk in XAnyEvent.3x man page
16046    
16047    X.Org Bugzilla #4312 <http://bugs.freedesktop.org/show_bug.cgi?id=4312>
16048    
16049    Protect /* sequences from cpp pre-processor removal without transforming
16050    to Unicode mathematical asterisk character
16051
16052commit 2af660c2fcd15c86c66459bfc074c190ea1462e6
16053Author: Jamey Sharp <jamey@minilop.net>
16054Date:   Mon Oct 29 10:46:20 2007 -0700
16055
16056    Two threads can request sequence sync and XID fetch simultaneously.
16057    
16058    So don't assert that they can't.
16059    
16060    This makes the Xlib/XCB implementation of _XAllocID more closely
16061    resemble the traditional Xlib version.
16062
16063commit 6e5485e0a5e3ab738becad12193e760c5fee83a4
16064Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
16065Date:   Sun Oct 28 04:44:00 2007 -0800
16066
16067    X.Org Bug #12983: Typos in ./man/XChangeKeyboardControl.man
16068    
16069    <http://bugs.freedesktop.org/show_bug.cgi?id=12983>
16070
16071commit 24527c92fd1f433ea135e85ec876a94a529fe500
16072Author: Yann Droneaud <ydroneaud@mandriva.com>
16073Date:   Wed Oct 24 19:26:07 2007 +0300
16074
16075    XIM: Properly initialise client event
16076    
16077    Make sure all ClientMessage fields are initialised to 0 before we send it.
16078
16079commit e41477f37b0d9b74a056d22dbf0073a94eecb9d6
16080Author: Daniel Stone <daniel@fooishbar.org>
16081Date:   Wed Oct 24 19:22:22 2007 +0300
16082
16083    Colours: Fix --disable-xcms
16084    
16085    Disable large tracts of colour management code when passing
16086    --disable-xcms.
16087
16088commit 11ea09745efa8de7dc82fe30ebd2393f08390957
16089Author: Eric Anholt <eric@anholt.net>
16090Date:   Fri Aug 31 17:30:33 2007 -0700
16091
16092    Bug #2081: Note the range limitation of XSetScreenSaver arguments.
16093
16094commit 31540f1438ec63faf37044f2fd654b335ddf80f0
16095Author: Kristian Høgsberg <krh@redhat.com>
16096Date:   Wed Aug 29 19:50:57 2007 -0400
16097
16098    Add GLX 1.4 requests and errors.
16099
16100commit 7c996f78914c77fe17e9f4feede980d895d9df51
16101Author: Eric Anholt <eric@anholt.net>
16102Date:   Tue Aug 28 15:15:11 2007 -0700
16103
16104    Add XF86 keyboard/monitor brightness keysyms to the keysymbdb.
16105
16106commit fa4effe82759f864a22a2dc6c920fa72ddb175a8
16107Author: Eric S. Raymond <esr@thyrsus.com>
16108Date:   Sun Jan 14 10:48:00 2007 -0800
16109
16110    Bug #9658: Bad markup on XIfEvent.3x
16111    
16112    X.Org Bugzilla #9658 <https://bugs.freedesktop.org/show_bug.cgi?id=9658>
16113
16114commit 74cba78daa738ef4d92096107d95c1a585933666
16115Author: Eric S. Raymond <esr@thyrsus.com>
16116Date:   Sun Jan 14 10:43:00 2007 -0800
16117
16118    Bug #9655: Bad markup in XrmUniqueQuark.3x
16119    
16120    X.Org Bugzilla #9655 <https://bugs.freedesktop.org/show_bug.cgi?id=9655>
16121
16122commit 4341d1a34b2a2e460b58131b6fd81935f3355bbc
16123Author: Eric S. Raymond <esr@thyrsus.com>
16124Date:   Sun Jan 14 10:43:00 2007 -0800
16125
16126    Bug #9654: Bad markup in XrmGetFileDatabase.3x
16127    
16128    X.Org Bugzilla #9654 <https://bugs.freedesktop.org/show_bug.cgi?id=9654>
16129
16130commit 2db713252090cae08f0200fecad4fc25fb64c8b1
16131Author: Eric S. Raymond <esr@thyrsus.com>
16132Date:   Sun Jan 14 10:40:00 2007 -0800
16133
16134    Bug #9653: Bad markup in XQueryColor.3x
16135    
16136    X.Org Bugzilla #9653 <https://bugs.freedesktop.org/show_bug.cgi?id=9653>
16137
16138commit 9d3ceea4b902e0471824c1e07ad64342b9a1114e
16139Author: Eric S. Raymond <esr@thyrsus.com>
16140Date:   Sun Jan 14 10:40:00 2007 -0800
16141
16142    Bug #9652: Bad markup in XDrawArc.3x
16143    
16144    X.Org Bugzilla #9652 <https://bugs.freedesktop.org/show_bug.cgi?id=9652>
16145
16146commit 2e7e0748d353d7f53bbd65ec6bf0df8758528ddc
16147Author: Eric S. Raymond <esr@thyrsus.com>
16148Date:   Sun Jan 14 10:39:00 2007 -0800
16149
16150    Bug #9651: Bad markup in XcmsColor.3x
16151    
16152    X.Org Bugzilla #9651 <https://bugs.freedesktop.org/show_bug.cgi?id=9651>
16153
16154commit c316aaf0aab06951db9dc5c9c1148bfea835d885
16155Author: Eric S. Raymond <esr@thyrsus.com>
16156Date:   Sun Jan 14 10:38:00 2007 -0800
16157
16158    Bug #9650: Bad markup in XLoadFont.3x manual page
16159    
16160    X.Org Bugzilla #9650 <https://bugs.freedesktop.org/show_bug.cgi?id=9650>
16161
16162commit ca5d9a625ea0965853fa9e74a448b8c29c78ec95
16163Author: Eric S. Raymond <esr@thyrsus.com>
16164Date:   Tue Jan 2 10:18:00 2007 -0800
16165
16166    Bug 9523: Markup problems in XQueryExtension.3x
16167    
16168    X.Org Bugzilla #9523 <https://bugs.freedesktop.org/show_bug.cgi?id=9523>
16169
16170commit fe713c616e29ba19c179b43c18eca1035079ce18
16171Author: Ian Romanick <idr@us.ibm.com>
16172Date:   Tue Aug 21 14:56:33 2007 -0700
16173
16174    Make sure nls/am_ET.UTF-8/Makefile is created by configure.
16175
16176commit e3430616f26b68e1439143cbe10732f3fc329d20
16177Author: Eric S. Raymond <esr@thyrsus.com>
16178Date:   Tue Jan 2 08:40:00 2007 -0800
16179
16180    Bug #9516: Markup error in XAllocWMHints.3x
16181    
16182    X.Org Bugzilla #9516 <https://bugs.freedesktop.org/show_bug.cgi?id=9516>
16183
16184commit d8fe979fc929833e8c754aed32641786d5a0622b
16185Merge: 21ca9533 4ec1723f
16186Author: James Cloos <cloos@jhcloos.com>
16187Date:   Mon Aug 20 15:34:50 2007 -0400
16188
16189    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
16190
16191commit 21ca953337fb221b85345bf35ce1a98a0dcb2bf2
16192Author: James Cloos <cloos@jhcloos.com>
16193Date:   Mon Aug 20 15:34:03 2007 -0400
16194
16195    Fix typo
16196    The code <U1000000D> was used where <U10000DC> was obviously intended.
16197    
16198    It is possible that <Udiaeresis> should be used instead, if that will
16199    not break anyone’s setup.
16200
16201commit 4ec1723fff729440cd3349c1f95d87d2a6ba89cf
16202Author: James Cloos <cloos@jhcloos.com>
16203Date:   Mon Aug 20 15:25:48 2007 -0400
16204
16205    Add compose file for Ethiopic to match new keyboard in xkeyboard-config
16206    From bug report:
16207    
16208      https://bugs.freedesktop.org/show_bug.cgi?id=11307
16209
16210commit eff33ae525337ce2026be135a26464c7b1237113
16211Author: James Cloos <cloos@jhcloos.com>
16212Date:   Sat Aug 18 17:58:23 2007 -0400
16213
16214    Patch for Catalan locales
16215    From bugzilla bug 10943¹:
16216    
16217      There are several Catalan locale codes which presently can
16218      be used in X11 systems; especially after they were accepted
16219      in belocs-locale-data².
16220    
16221      In the following patches, I³ add ca_AD, ca_FR and ca_IT Catalan
16222      locale codes.  For instance, without this, using ca_AD (actually
16223      a quite used locale⁴) some applications (eg. Emacs or Skype)
16224      cannot display Catalan diacritic marks as you type them.
16225    
16226    1] https://bugs.freedesktop.org/show_bug.cgi?id=10943
16227    2] http://lists.debian.org/debian-devel-changes/2005/07/msg01429.html
16228    3] Toni Hermoso Pulido <toniher@softcatala.org>
16229    4] https://launchpad.net/~ubuntu.cat/+members
16230
16231commit 1f980cb7d022f53d0aee9e793b08203fb888e86e
16232Author: James Cloos <cloos@jhcloos.com>
16233Date:   Sat Aug 18 17:47:04 2007 -0400
16234
16235    Add additional Euro signs to compose
16236    Inspired by bug 7419¹ make all of:
16237    
16238        C=, =C, c=, =c, E=, =E, e=, E=
16239    
16240    after <Multi_key> generate € U+20AC EURO SIGN.
16241    
16242    1] https://bugs.freedesktop.org/show_bug.cgi?id=7419
16243
16244commit 4b0a14521449dfce8b4347bd17243efd1d3eae2d
16245Author: James Cloos <cloos@jhcloos.com>
16246Date:   Sat Aug 18 17:29:08 2007 -0400
16247
16248    Compose fix for Latin-1 (from Debian)
16249    The description from bugzilla bug 7417¹ is:
16250    
16251      We've been shipping this patch for some time in Debian now. The
16252      problem description from the patch header is reproduced below. You
16253      may want to note the licensing issue mentioned below, but we've been
16254      shipping it because the method by which this particular patch was
16255      generated and updated was also given below.
16256    
16257      This patch by Denis Barbier.
16258    
16259      The X11 protocol states that Unicode keysyms are in the range
16260      0x01000100 - 0x0110FFFF.  If the result of composing characters is a
16261      Unicode codepoint, X returns the corresponding Unicode keysym, which
16262      is its Unicode codepoint augmented by 0x01000000.  Latin-1
16263      characters must not appear with their Unicode codepoints in compose
16264      files, otherwise the returned composed character lies in the range
16265      0x01000000 - 0x010000FF which is not valid.
16266    
16267      There are two solutions: either fix composing routines to return
16268      0xZZ instead of 0x010000ZZ (where Z is an hexadecimal digit), or
16269      replace U00ZZ by their corresponding keysyms in compose files.  The
16270      latter is more logical and less error prone, so compose files will
16271      be patched.  Many applications accept these invalid Unicode keysyms,
16272      but few of them don't, most notably xemacs.  Only UTF-8 locales are
16273      affected.
16274    
16275      This has been fixed very recently in XFree86 CVS (but not xorg), but
16276      for licensing reasons, this patch is not grabbed.  Instead automatic
16277      conversion is performed by:
16278    
16279        sed -e '/XK_LATIN1/,/XK_LATIN1/!d' /usr/include/X11/keysymdef.h \
16280        | grep -v deprecated | grep 0x0 \
16281        | sed -e 's/0x0/U0/' -e 's/XK_//' \
16282        | awk '{ printf "s/\\b%s\\b/%s/ig\n", $3, $2; }' > sedfile
16283        for f in nls/*.UTF-8/Compose.pre
16284        do
16285          sed -f sedfile $f > $f.tmp && mv $f.tmp $f
16286        done
16287    
16288    [I edited the quoted script to update it for the current location of
16289    the installed keysymdef.h and the current layout of the libX11
16290    repo. -JimC]
16291    
16292    I applied the script, not the patch attached to the bugreport.
16293    
16294    1] https://bugs.freedesktop.org/show_bug.cgi?id=7417
16295
16296commit 5cf5bc76642bfece7cb5b76faf414bf445f14489
16297Author: James Cloos <cloos@jhcloos.com>
16298Date:   Sat Aug 18 17:13:41 2007 -0400
16299
16300    Add some compose sequences
16301    
16302    Add some compose sequences from the patch in bug 5371 (attachment 4122).
16303    
16304    Cf:
16305    
16306    https://bugs.freedesktop.org/show_bug.cgi?id=5371
16307    https://bugs.freedesktop.org/attachment.cgi?id=4122
16308
16309commit d4002e389dd69780dfc7c2f7bd3cb0c57f05d4f8
16310Author: James Cloos <cloos@jhcloos.com>
16311Date:   Sat Aug 18 13:57:31 2007 -0400
16312
16313    Fix SMP Compose targets
16314    The compose targets from the SMP (plane 1) were incorrect.
16315    
16316    At some point the 0x10000 bit had been lost.
16317
16318commit f1ed3da9a30a1f0264fdc7d1c6466f27fe2a3d7d
16319Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
16320Date:   Thu Aug 16 17:37:22 2007 -0500
16321
16322    Fix a mutex reference-counting bug.
16323    
16324    I was told that some systems have a much more permissive libpthread.
16325    I was asked to commit this. This is from NetBSD's X source.
16326    (I didn't receive any feedback on xorg list for over two weeks
16327    about this.)
16328
16329commit ac00a44b4875de70382da5a40dd87f976e5b9327
16330Author: Tilman Sauerbeck <tilman@code-monkey.de>
16331Date:   Wed Aug 1 20:36:03 2007 +0200
16332
16333    Bumped version to 1.1.3.
16334
16335commit 76fae9cba1e9bdf7f0eb2ff2b90153d622136cf0
16336Author: Joerg Sonnenberger <joerg@netbsd.org>
16337Date:   Fri Jul 27 11:15:47 2007 -0700
16338
16339    Fail properly on errors in recursive make.
16340
16341commit 1a18319b3bde08dd9ef69c7cd735a76000cf3177
16342Author: Brice Goglin <Brice.Goglin@ens-lyon.org>
16343Date:   Thu Jul 26 23:31:15 2007 +0200
16344
16345    Add missing override parameter in XrmCombineDatabase prototype in the manpage
16346    
16347    Reported by Arnaud Giersch <arnaud.giersch@iut-bm.univ-fcomte.fr>
16348    in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=393434
16349    and https://bugs.freedesktop.org/show_bug.cgi?id=9948
16350    
16351    Also add the type of the second argument in XrmMergeDatabases.
16352
16353commit 6f0764d4b56f64786b4980839ca262f10a51af6f
16354Author: Brice Goglin <Brice.Goglin@ens-lyon.org>
16355Date:   Thu Jul 26 22:53:52 2007 +0200
16356
16357    Clarify return value of XGetCommand in case of error in the manpage
16358    
16359    Reported by Sean Perry <shalehperry@attbi.com>
16360    in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=133348
16361    and https://bugs.freedesktop.org/show_bug.cgi?id=9828
16362
16363commit b8bef57342632cc2d25580bb7daa3839bae04d89
16364Author: Alan Coopersmith <alan.coopersmith@sun.com>
16365Date:   Wed Jul 25 17:44:06 2007 -0700
16366
16367    Include comment/copyright/license for AC_DEFINE_DIR in acinclude.m4
16368
16369commit 590cde811a79375231c59ed8583e02b111ed567c
16370Author: Kean Johnston <kean@armory.com>
16371Date:   Thu Jul 19 16:44:20 2007 -0700
16372
16373    Reset hostname when falling back from :0 to hostname/localhost:0 as well
16374
16375commit d334665e619e9db657a2ea2764a8b852401d4a3f
16376Author: Kean Johnston <kean@armory.com>
16377Date:   Tue Jul 17 16:34:30 2007 -0700
16378
16379    LOCALCONN fallback changes DisplayString() output, breaks KDE
16380    
16381    See <http://lists.freedesktop.org/archives/xorg/2007-July/026443.html>
16382
16383commit 8e76bcf3cafda85058ce5f35f1f81929f4772f72
16384Author: Olivier Blin <blino@mandriva.org>
16385Date:   Mon Jun 11 18:32:09 2007 +1000
16386
16387    fix XGetMotionEvents arguments order - Fixes bug 11222
16388
16389commit c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8
16390Author: Jamey Sharp <jamey@minilop.net>
16391Date:   Sun Jun 10 16:19:59 2007 -0700
16392
16393    Fix locking in _XimGetWindowEventmask.
16394    
16395    Now that XFilterEvent drops the Display lock before invoking callback
16396    functions, _XimGetWindowEventmask is called without the lock held. So
16397    when it called _XGetWindowAttributes, a variant of XGetWindowAttributes
16398    that does not lock the Display, Xlib/XCB would assert:
16399    
16400    xcb_xlib.c:50: xcb_xlib_unlock: Assertion `c->xlib.lock' failed.
16401    
16402    Should fix Gentoo #156367, Ubuntu #87947, Debian #427296. And others?
16403
16404commit 65f0ab5d46d80d55fc04d4eb14fa05f130eb8b9c
16405Author: Jamey Sharp <jamey@minilop.net>
16406Date:   Sun Jun 3 21:41:47 2007 -0700
16407
16408    Move security fixes to the top of NEWS, and fix spacing.
16409    
16410    Commit by Josh Triplett and Jamey Sharp.
16411
16412commit a549a258b8fcb1ba9d0c1b01b72967e385f67cab
16413Author: Jamey Sharp <jamey@minilop.net>
16414Date:   Sun Jun 3 21:29:40 2007 -0700
16415
16416    Add NEWS item for bugfix in commit e2c1d788d1fe7bd2d34756493951552441e59b8c.
16417    
16418    Commit by Josh Triplett and Jamey Sharp
16419
16420commit 5123b77a3d32d3ad479462f319762c328278aed9
16421Author: Jamey Sharp <jamey@minilop.net>
16422Date:   Sun Jun 3 21:24:54 2007 -0700
16423
16424    Xlib/XCB: Inline and simplify handle_event, since only process_responses calls it now.
16425    
16426    Commit by Josh Triplett and Jamey Sharp.
16427
16428commit 582ca690ea4f3ffd2b94826c4db97229bd3c7238
16429Author: Jamey Sharp <jamey@minilop.net>
16430Date:   Sun Jun 3 20:59:12 2007 -0700
16431
16432    Xlib/XCB: Fix _XReadEvents to always enqueue a new event, even if an error occurs
16433    
16434    Commit c337f9de7cfd89f983f83956b7457a274dd412f3 broke the invariant that
16435    _XReadEvents always enqueues at least one event even if an error occurred,
16436    because the one call to xcb_wait_for_event would then return an error, not an
16437    event, and nothing else ensured that process_responses would obtain an event.
16438    Fix this by reverting most of c337f9de7cfd89f983f83956b7457a274dd412f3 and
16439    f417570735aac865eb6b576d1ea76b5bfcd8573b and implementing the correct fix.  In
16440    process_responses, wait_for_first_event now serves as a flag, cleared when
16441    actually handling an event.
16442    
16443    Commit by Josh Triplett and Jamey Sharp.
16444
16445commit e2c1d788d1fe7bd2d34756493951552441e59b8c
16446Author: Jamey Sharp <jamey@minilop.net>
16447Date:   Sun Jun 3 17:33:23 2007 -0700
16448
16449    Xlib/XCB: Only remove pending_requests when there are provably no more responses.
16450
16451commit 7a6dbd4b07ca0a49c30ca7a1d2437eafb2e15eab
16452Author: Josh Triplett <josh@freedesktop.org>
16453Date:   Sun Jun 3 15:39:39 2007 -0700
16454
16455    Bump version number to 1.1.2, and add NEWS entry for 1.1.2
16456    
16457    Signed-off-by: Josh Triplett <josh@freedesktop.org>
16458
16459commit 416f38f2e67ee1979b3d2feac6f06b3670238804
16460Author: Josh Triplett <josh@freedesktop.org>
16461Date:   Sun Jun 3 12:13:44 2007 -0700
16462
16463    Revert "Revert "include: don't distribute XlibConf.h""
16464    
16465    This reverts commit 79fa3d8070d95b960ba486f2439225872471dadd.
16466    
16467    Re-revert the XlibConf.h change, which prevented distribution, not
16468    installation.
16469
16470commit 79fa3d8070d95b960ba486f2439225872471dadd
16471Author: Josh Triplett <josh@freedesktop.org>
16472Date:   Sat Jun 2 22:05:16 2007 -0700
16473
16474    Revert "include: don't distribute XlibConf.h"
16475    
16476    This reverts commit c9e28e05ae01ce8a29bea09df759b6271865b44c.
16477    
16478    The installed XlibInt.h includes XlibConf.h , so libX11 should ship
16479    XlibConf.h.  (Commit c9e28e05ae01ce8a29bea09df759b6271865b44c didn't actually
16480    prevent automake from shipping XlibConf.h, because it used
16481    nodist_x11include_HEADERS rather than nodist_HEADERS.)
16482
16483commit f417570735aac865eb6b576d1ea76b5bfcd8573b
16484Author: Jamey Sharp <jamey@minilop.net>
16485Date:   Sat Jun 2 17:59:15 2007 -0700
16486
16487    Xlib/XCB: inline wait_or_poll_for_event, which now had only one caller.
16488    
16489    Commit by Josh Triplett and Jamey Sharp.
16490
16491commit c337f9de7cfd89f983f83956b7457a274dd412f3
16492Author: Jamey Sharp <jamey@minilop.net>
16493Date:   Sat Jun 2 17:46:41 2007 -0700
16494
16495    Xlib/XCB: Ensure _XReadEvents reads at least one new event and blocks for exactly one event.
16496    
16497    Commit by Jamey Sharp and Josh Triplett.
16498
16499commit 7f66c897f04806b75e574b55b48921b48045e3f9
16500Author: Jamey Sharp <jamey@minilop.net>
16501Date:   Sat Jun 2 16:43:39 2007 -0700
16502
16503    Update _XReply's copy of _XCBUnlockDisplay's guts.
16504    
16505    We introduced this bug in 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5.
16506    
16507    Also add a comment in _XCBUnlockDisplay to discourage this problem from
16508    respawning.
16509    
16510    Commit by Josh Triplett and Jamey Sharp.
16511
16512commit 740ead23512f8d2eaafaa69e514f1ebafad475b9
16513Author: Jamey Sharp <jamey@minilop.net>
16514Date:   Sat Jun 2 16:01:01 2007 -0700
16515
16516    Xlib/XCB: Avoid re-crashing after _XIOError.
16517    
16518    Commit by Josh Triplett and Jamey Sharp.
16519
16520commit 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5
16521Author: Jamey Sharp <jamey@minilop.net>
16522Date:   Sat Jun 2 12:30:30 2007 -0700
16523
16524    Hold XCB's Xlib lock even when only the user lock (XLockDisplay) is held.
16525    
16526    An Xlib client can query Display state, such as with NextRequest, while
16527    it holds only the Xlib user lock (between XLockDisplay and
16528    XUnlockDisplay), so XCB requests in other threads should be blocked when
16529    the Xlib user lock is held.
16530    
16531    We acquire the lock even when XInitThreads was not called, so that pure
16532    XCB code can use multiple threads even in an otherwise single-threaded
16533    Xlib application.
16534    
16535    Commit by Josh Triplett and Jamey Sharp.
16536
16537commit 95523387d619af5b400748898d722e080b5ce1a6
16538Author: Jamey Sharp <jamey@minilop.net>
16539Date:   Sat Jun 2 11:57:39 2007 -0700
16540
16541    Allow re-entrant Xlib calls from _XIOError.
16542    
16543    Some libraries try to clean up X resources from atexit handlers, _fini,
16544    or C++ destructors. To make these work, the Display lock should be
16545    downgraded to a user lock (as in XLockDisplay) before calling exit(3).
16546    This blocks Xlib calls from threads other than the one calling exit(3)
16547    while still allowing the exit handlers to call Xlib.
16548    
16549    This assumes that the thread calling exit will call any atexit handlers.
16550    If this does not hold, then an alternate solution would involve
16551    registering an atexit handler to take over the lock, which would only
16552    assume that the same thread calls all the atexit handlers.
16553    
16554    Commit by Josh Triplett and Jamey Sharp.
16555
16556commit 91b02b8064f4e0bcc56019f0722914850008a597
16557Author: Tilman Sauerbeck <tilman@code-monkey.de>
16558Date:   Thu May 17 19:07:34 2007 +0200
16559
16560    More constification.
16561
16562commit a4f3841940158351f9424c3f59b305cce877177d
16563Author: Tilman Sauerbeck <tilman@code-monkey.de>
16564Date:   Thu May 17 00:29:43 2007 +0200
16565
16566    Constified composite text charset table.
16567
16568commit 0581c0aa6039e6b2abb9f7b0a4f9904d8e01f00e
16569Author: Tilman Sauerbeck <tilman@code-monkey.de>
16570Date:   Wed May 16 23:19:22 2007 +0200
16571
16572    Have the compiler fill in hexTable so we don't have to do it at runtime.
16573
16574commit 0e8d9ca47dab0d069e305d5784d05f2ade04f0a8
16575Author: Tilman Sauerbeck <tilman@code-monkey.de>
16576Date:   Wed May 16 19:36:23 2007 +0200
16577
16578    More constification.
16579
16580commit 6d2bed8f04942b4de086a519ac693e729c9fdeea
16581Author: Tilman Sauerbeck <tilman@code-monkey.de>
16582Date:   Wed May 16 19:18:20 2007 +0200
16583
16584    Constify and clean up token table.
16585
16586commit 6c508eab5df5d517f7e4cbe6087308cd53a564b2
16587Author: Tilman Sauerbeck <tilman@code-monkey.de>
16588Date:   Wed May 16 18:24:42 2007 +0200
16589
16590    Constified more tables.
16591
16592commit e699c4231c205ef00d687b6412308d031b99806b
16593Author: Tilman Sauerbeck <tilman@code-monkey.de>
16594Date:   Wed May 16 18:01:23 2007 +0200
16595
16596    Constified error list.
16597
16598commit a68a1cd7cb990ba276fbc36a7591044d78b3d3c1
16599Author: Peter Hutterer <peter@cs.unisa.edu.au>
16600Date:   Tue May 15 16:54:01 2007 +0930
16601
16602    Add XGenericEvent definition and handling for long events.
16603
16604commit c76d30253f1483ac8200ad5c032a818907e65030
16605Author: Jan Willem Stumpel <jstumpel@planet.nl>
16606Date:   Fri May 4 12:00:49 2007 -0700
16607
16608    Update el_GR.UTF-8/Compose.pre to match changes in xkeyboard-config cvs
16609    
16610    Cf:
16611    
16612      https://bugs.freedesktop.org/show_bug.cgi?id=10851
16613      https://bugs.freedesktop.org/show_bug.cgi?id=10824
16614      http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=386385
16615    
16616    The greek keyboard definition was changed to replace dead_horn and
16617    dead_ogonek with U0313 COMBINING COMMA ABOVE (aka Psili) and U0314
16618    COMBINING REVERSED COMMA ABOVE (aka Dasia).
16619    
16620    This patch modifies the Greek Compose.pre to match.
16621    
16622    It is generated by the script Jan Willem Stumpel <jstumpel@planet.nl>
16623    posted to 386385@bugs.debian.org:
16624    
16625    #!/usr/bin/perl
16626    while (<>) {
16627      print $_;
16628      if (/dead_horn/) {
16629        s/dead_horn/U0313/;
16630        print $_;
16631      }
16632      elsif (/dead_ogonek/) {
16633        s/dead_ogonek/U0314/;
16634        print $_;
16635      }
16636    }
16637
16638commit a48386ce6b5f8fd2d9dc11a966c9bf5da59f3831
16639Author: Magnus Kessler <Magnus.Kessler@gmx.net>
16640Date:   Tue May 1 15:20:08 2007 +0200
16641
16642    Switched function definitions from K&R to ANSI style.
16643
16644commit 605d357074d556a05a3fba2e85cbea36a3204248
16645Author: Tilman Sauerbeck <tilman@code-monkey.de>
16646Date:   Tue May 1 14:47:03 2007 +0200
16647
16648    Tweaked configure output about the man pages suffix.
16649
16650commit 9824b40d2af4ca2376512c1be7743da0d5065900
16651Author: Alan Coopersmith <alan.coopersmith@sun.com>
16652Date:   Sat Apr 28 00:42:18 2007 -0700
16653
16654    Fix typo in nroff macro in XkbAddGeomOverlayKey.man
16655
16656commit f93849dcc68bd5042ea0884e5190dc7c35b31d68
16657Author: Alan Coopersmith <alan.coopersmith@sun.com>
16658Date:   Sat Apr 28 00:30:55 2007 -0700
16659
16660    Protect C comments and #defines in XKB man pages from being mangled by cpp
16661
16662commit f2f27d4763c7665e422fab10b96b4cf5ad6c0a6f
16663Author: Alan Coopersmith <alan.coopersmith@sun.com>
16664Date:   Sat Apr 28 00:14:50 2007 -0700
16665
16666    Add Makefile to process/install XKB man pages
16667
16668commit d9954c6f6f3a8c406b946acd0d034ff83c656156
16669Author: Dennis Arellano <Dennis.Arellano@Sun.COM>
16670Date:   Thu Aug 19 00:00:00 1999 -0700
16671
16672    Add man pages for XKB API's
16673    
16674    Man pages originally written for X11R6.4 integration to Solaris 7 11/99
16675    Sun bug id 4258344: Add new XKB API manpages for 6.4 upgrade
16676    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4258344>
16677    
16678    (Volunteer needed to convert prototypes in man pages to ANSI C style...)
16679
16680commit f640a49b5e2ebf29f9d655df544c63bf826f619a
16681Author: Tilman Sauerbeck <tilman@code-monkey.de>
16682Date:   Fri Apr 20 18:39:59 2007 +0200
16683
16684    Markup tweak for XOpenIM.
16685
16686commit e972b0bb255af4f3258217852542faf5afa60b28
16687Author: Julien Cristau <jcristau@debian.org>
16688Date:   Fri Apr 20 18:35:09 2007 +0200
16689
16690    Bug #9695: Fixed a few argument types in the XOpenIM manpage.
16691
16692commit b4e2276f329fa42397cb8609cfcd34ebafd3d96b
16693Author: Julien Cristau <jcristau@debian.org>
16694Date:   Fri Apr 20 18:28:52 2007 +0200
16695
16696    Bug #9697: Fixed documentation of XVisualInfo struct.
16697    
16698    The "depth" member was said to be unsigned int, but it's signed.
16699
16700commit 4068f3dae01c630f825002673b1d3a047ad61863
16701Author: Julien Cristau <jcristau@debian.org>
16702Date:   Fri Apr 20 16:41:21 2007 +0200
16703
16704    Bug #9696: refer to XDefineCursor() instead of XDefineCusor().
16705
16706commit 603c2f88d4e57ce1a3c16e8b6246866e6edd8fa8
16707Author: Tilman Sauerbeck <tilman@code-monkey.de>
16708Date:   Sat Apr 7 14:42:55 2007 +0200
16709
16710    Use unistd.h to get getresuid() and friends.
16711    
16712    This works since we now have _GNU_SOURCE defined.
16713
16714commit 0300f295bbd3a0c7c46baac8e0a27aeaf53c9d9b
16715Author: Tilman Sauerbeck <tilman@code-monkey.de>
16716Date:   Sat Apr 7 14:27:12 2007 +0200
16717
16718    Bug #10562: Define _GNU_SOURCE on glibc systems.
16719
16720commit a225a0be48770beb689d5ac5da97073634f7deab
16721Author: Tilman Sauerbeck <tilman@code-monkey.de>
16722Date:   Fri Apr 6 12:46:25 2007 +0200
16723
16724    For nls/*.pre, allow people to comment lines by starting them with '##'.
16725    
16726    This fixes a bunch of cpp warnings from nls/en_US.UTF-8/Compose.pre.
16727    Fixing that file would have been a larger diff, and using ## may be
16728    nicer to use anyway.
16729
16730commit 1c75a9479011e5f1ad01c950628d0ef5a302d8b6
16731Author: Tilman Sauerbeck <tilman@code-monkey.de>
16732Date:   Fri Apr 6 11:59:31 2007 +0200
16733
16734    Bug #10475: Fixed lots of char*/const char* mixups.
16735    
16736    I didn't fix all of them, as that would require touching
16737    public headers.
16738
16739commit 680dd50193b5b3fcabdd3f1fcbd6a889d5a95c54
16740Author: Matthias Hopf <mhopf@suse.de>
16741Date:   Thu Apr 5 11:53:18 2007 +0200
16742
16743    Fix 64bit issues with reallocation.
16744
16745commit 4d38aeaca42d0bdfe34a833a142ee4d895de03bf
16746Author: Tilman Sauerbeck <tilman@code-monkey.de>
16747Date:   Wed Apr 4 18:41:18 2007 +0200
16748
16749    Fixed a few warnings.
16750
16751commit 7dc7ef398b6ad90ccd1680ed9cd1cfdd47312f5a
16752Author: Matthieu Herrb <matthieu@roadrock.(none)>
16753Date:   Tue Apr 3 15:39:52 2007 +0200
16754
16755    Multiple integer overflows in the XGetPixel() and XInitImage functions
16756    
16757    CVE-2007-1667
16758
16759commit 0284b144340a455a4b5b5011d81ac5a610372291
16760Author: David Baron <dbaron@dbaron.org>
16761Date:   Fri Mar 30 17:07:46 2007 +0200
16762
16763    Bug #7703: Fixed XSetSizeHints() et al wrt use of uninitialized data.
16764    
16765    Now only those fields of the respective hint struct are set that
16766    are actually valid in the input data.
16767    The changed functions are:
16768    XSetSizeHints(), XSetWMHints() and XSetWMSizeHints().
16769
16770commit 0994faa0c76c45b106442db461b8a30a3e1c9395
16771Author: Tilman Sauerbeck <tilman@code-monkey.de>
16772Date:   Thu Mar 29 17:31:25 2007 +0200
16773
16774    Fixed the change from the previous SendEvent commit.
16775    
16776    Testing a different patch than the one you commit is bad, right?
16777
16778commit 398d75528a84f4b8414eb0e363cf53b1b16f6fdf
16779Author: Tilman Sauerbeck <tilman@code-monkey.de>
16780Date:   Wed Mar 28 22:23:44 2007 +0200
16781
16782    Bug #10292: Fixed a memory leak related to XOpenDisplay() in the XCB code.
16783
16784commit ab0bcd07957cecc8e7c0e75d5160a625e91264fe
16785Author: David Baron <dbaron@dbaron.org>
16786Date:   Wed Mar 28 22:21:40 2007 +0200
16787
16788    Bug #7713: Initialize all of the event's fields before sending it.
16789
16790commit bc80f9fe3ccce40ee41246b97470c4f0519756ad
16791Author: Julien Cristau <jcristau@debian.org>
16792Date:   Sun Mar 18 13:14:48 2007 +0100
16793
16794    Bug #9279: Fixed a file descriptor leak.
16795
16796commit c9e28e05ae01ce8a29bea09df759b6271865b44c
16797Author: Daniel Stone <daniel@fooishbar.org>
16798Date:   Sat Dec 16 00:45:19 2006 +0200
16799
16800    include: don't distribute XlibConf.h
16801    
16802    Since XlibConf.h is built by configure, don't distribute it.
16803
16804commit dd1705ced2cac6b4b6b21e79272fcf9bed4bf376
16805Merge: 129bbb9f 769b9854
16806Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
16807Date:   Thu Dec 14 14:23:20 2006 -0600
16808
16809    Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/lib/libX11
16810
16811commit 129bbb9f9114a571556fa3a24f15ba58a5cdb2de
16812Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
16813Date:   Thu Dec 14 14:21:19 2006 -0600
16814
16815    For NetBSD, define the XTHREADLIB and XTHREAD_CFLAGS.
16816
16817commit 769b9854f7eb1d6d20dd0b4a1c1215ad8e1b77b6
16818Author: Daniel Stone <daniel@fooishbar.org>
16819Date:   Wed Dec 6 18:53:00 2006 +0200
16820
16821    Makefile.am: make ChangeLog hook safer
16822    
16823    Make ChangeLog hook as safe as possible.
16824    (cherry picked from f5d6a3d24095c7ffed86705995d0874c885e7676 commit)
16825
16826commit 8a8185a649e93b90ffa820387ffdca831227f5a9
16827Author: Josh Triplett <josh@freedesktop.org>
16828Date:   Thu Nov 30 18:33:13 2006 -0800
16829
16830    Add autogen.sh to EXTRA_DIST.
16831
16832commit efe817f95ef8d05e863c83147e903140bc860de4
16833Author: Jamey Sharp <jamey@minilop.net>
16834Date:   Thu Nov 30 17:58:35 2006 -0800
16835
16836    Release libX11 1.1.1.
16837
16838commit f637a5b03164263a3af2e644cf655e52b015f1bb
16839Author: Ross Combs <rocombs@cs.nmsu.edu>
16840Date:   Sat Nov 25 14:45:17 2006 -0800
16841
16842    Debian bug #354315: Clarify return value in XGetWindowAttributes man page
16843    
16844    This man page does not discuss the actual return values of the
16845    function, but says they are of type "Status".  One might assume
16846    that this means you could compare it with the "Success" macro.
16847    One would be wrong.
16848    
16849    The X functions seem to have two three types representing status.
16850    If it is an "int" there are a number of error codes or "Success"
16851    which can be compared against.  If it is a bool, the result can be
16852    compared with "True" or "False".  If the return type is "Status" it
16853    appears that the return type is either 0 or 1.  Unfortunately the
16854    value for Success is zero, so it is important to distinguish
16855    between the first two types of return values and the third;
16856    otherwise the conditional will be inverted.
16857    
16858    XGetWindowAttributes() is one of the functions which returns zero
16859    for failure.  The man page should make this clear.
16860
16861commit c6a0b0f18ed1242eeb908f5cf767ab8381edd456
16862Author: Jamey Sharp <jamey@minilop.net>
16863Date:   Sat Nov 25 14:23:45 2006 -0800
16864
16865    Bug #9154: Always process an event for _XReadEvents, even if an error occurs
16866    
16867    Previously, process_responses (in the wait_for_first_event case called
16868    from _XReadEvents) considered any return from xcb_wait_for_event
16869    sufficient to think it had processed an event. If xcb_wait_for_event
16870    returned an error, and no more events occurred before process_responses
16871    called xcb_poll_for_event, process_responses would try to return with
16872    dpy->head NULL, and would fail an assertion for the _XReadEvents
16873    postcondition. Now, process_responses continues using xcb_wait_for_event
16874    until it gets an event.
16875
16876commit d56e78acce9b2aa1dd1bf172afedaa3bccd5e1c8
16877Author: Tilman Sauerbeck <tilman@code-monkey.de>
16878Date:   Sat Nov 25 05:29:31 2006 -0800
16879
16880    Bug #9153: Fix access to freed memory.
16881    
16882    The fix for bug #8622 introduced a smaller bug where _XReply would
16883    read memory shortly after freeing it. This patch caches the needed
16884    value in a stack-allocated variable before the heap-allocated memory
16885    is freed.
16886
16887commit 934ca763bbc0dd7ae460469bfc000ba101602bcc
16888Author: Josh Triplett <josh@freedesktop.org>
16889Date:   Fri Nov 24 19:57:58 2006 -0800
16890
16891    libx11 doesn't use inputproto in public headers; don't require it in x11.pc
16892    
16893    Based on a Debian patch.
16894
16895commit 4255997ef2d92740d51f6e63e9eabcfa089683f0
16896Author: Josh Triplett <josh@freedesktop.org>
16897Date:   Thu Nov 23 07:19:32 2006 -0800
16898
16899    Release libX11 1.1
16900
16901commit a1168e11ec9377307c51a7271faec3bf88a63a66
16902Author: Jamey Sharp <jamey@minilop.net>
16903Date:   Tue Nov 21 17:52:34 2006 -0800
16904
16905    Add note in man-page that XListFontsWithInfo is not thread-safe.
16906    
16907    _XReply drops the Display lock, so the value of dpy->request may change
16908    before _XReply is called again.
16909    
16910    I discovered this by inspection a year or two ago. I'm pretty confident
16911    in the claim, and nobody has come up with an argument for why it's safe
16912    despite appearances.
16913
16914commit 67abe024268c6b1fdee516e5d3a046ccffd7e80a
16915Author: Jamey Sharp <jamey@minilop.net>
16916Date:   Sat Nov 18 15:39:26 2006 -0800
16917
16918    Bug #8622: Fix response processing order for threaded apps
16919    
16920    Previously, process_responses (the common code for _XReply,
16921    _XReadEvents, and _XEventsQueued) took the current request sequence
16922    number as an argument, and did some highly complicated processing to
16923    attempt to process responses in order across threads.  This didn't
16924    actually work.
16925    
16926    Now, process_responses handles responses in order, by adding condition
16927    variables to the list of outstanding requests in
16928    dpy->xcb->pending_requests, and blocking on them when those requests
16929    should get processed to allow _XReply to process them; if actually
16930    called from _XReply, it returns when _XReply's request should get
16931    processed.  _XReply broadcasts on the condition variable after it has
16932    read its reply and re-acquired the display lock.
16933    
16934    Another great commit brought to you by Jamey Sharp, Josh Triplett, the
16935    Gobby collaborative text editor, conspy, and ridiculous amounts of SSH
16936    forwarding.
16937    
16938    Signed-off-by: Josh Triplett <josh@freedesktop.org>
16939    Signed-off-by: Jamey Sharp <jamey@minilop.net>
16940
16941commit 941f02ede63baa46f93ed8abccebe76fb29c0789
16942Author: Lars Knoll <lars@trolltech.com>
16943Date:   Wed Nov 8 12:17:41 2006 -0800
16944
16945    Don't hold the display lock around callbacks to the application.
16946    
16947    This fixes an XCB locking assertion failure, particularly with emacs.
16948
16949commit e494ecaac1ec8a22bd9a85f800fca74d02e9d358
16950Author: Diego 'Flameeyes' Pettenò <flameeyes@gentoo.org>
16951Date:   Tue Nov 7 09:32:00 2006 -0800
16952
16953    Add xcb-xlib dependency to x11.pc when built against XCB.
16954
16955commit 2302008a3793eb4df8ede777d54fe06505c47eaf
16956Author: Eric Anholt <eric@anholt.net>
16957Date:   Mon Nov 6 17:11:42 2006 -0800
16958
16959    XCB: Allocate the right amount of memory for dpy->lock_fns.
16960    
16961    Fixes a crash I was experiencing on startup of anything using gdk.
16962
16963commit 5f860655be88108b03ccd97470a0814819254bf0
16964Author: Jamey Sharp <jamey@minilop.net>
16965Date:   Thu Nov 2 17:55:31 2006 -0800
16966
16967    Release libX11 1.1 RC2 (1.0.99.2).
16968
16969commit a6f4bbf7b1d725b0f04bd660f57b861a76b19831
16970Author: Benno Schulenberg <bensberg@justemail.net>
16971Date:   Sun Oct 29 03:10:30 2006 +0300
16972
16973    nls (en_US): remove long compositions that override shorter (bug #2286)
16974    Remove long compositions that override (or get overriden by) later shorter
16975    compositions, e.g. a four-key compose sequence that gets overriden by a
16976    later three-key compose sequence.
16977
16978commit 0280bf11ef88673a9b5bba3a91a599260f1f0949
16979Author: Benno Schulenberg <bensberg@justemail.net>
16980Date:   Sun Oct 29 03:08:36 2006 +0300
16981
16982    nls: remove shadowed compose entries (bug #2286)
16983    Remove compose entries shadowed by others later on.
16984
16985commit d118f2b1ef10997194b281524177dea7396da7dd
16986Author: Benno Schulenberg <bensberg@justemail.net>
16987Date:   Sun Oct 29 03:07:15 2006 +0300
16988
16989    nls: remove duplicate compose entries (bug #2286)
16990    Remove a bunch of duplicate entries from various Compose files.
16991
16992commit 5e1cc2fe20e5904ca1e05a4cb7be13d450a593bb
16993Author: Caolan McNamara <caolanm@redhat.com>
16994Date:   Sun Oct 29 02:46:15 2006 +0300
16995
16996    XKB geometry: fix leaks in GetKeyboardByName (bug #8099)
16997    Don't leak the name and value of every property we parse, as well as the
16998    name of every colour.
16999
17000commit 686bb8b35acf6cecae80fe89b2b5853f5816ce19
17001Author: Matthias Hopf <mhopf@suse.de>
17002Date:   Wed Oct 18 14:25:04 2006 +0200
17003
17004    Fix double open of compose file.
17005    
17006    Issue found by Kees Cook <kees@canonical.com>.
17007
17008commit d3e65cb8cddf08913d83c9df2bb9b1517f2ad3a8
17009Author: Jamey Sharp <jamey@minilop.net>
17010Date:   Sat Oct 14 21:25:10 2006 -0700
17011
17012    XCB: check for and handle I/O errors in _XGetXCBBuffer.
17013
17014commit 256eba6b40c5f811a03b04abf5f85f728ee3ab5d
17015Author: Jamey Sharp <jamey@minilop.net>
17016Date:   Wed Oct 11 00:06:50 2006 -0700
17017
17018    XKB bugfix: SyncHandle must be called after UnlockDisplay, not before.
17019
17020commit 1eedf1bd033e496843cfde42ae4ae5a119298605
17021Author: Jamey Sharp <jamey@minilop.net>
17022Date:   Tue Oct 10 23:03:28 2006 -0700
17023
17024    Add correct Display locking to XKB functions.
17025    
17026    Some XKB functions didn't correctly call LockDisplay or UnlockDisplay.
17027    This patch fixes at least some instances of that problem.
17028    
17029    Thanks to Magnus Kessler <Magnus.Kessler@gmx.net> for finding these bugs
17030    and proposing a fix, which this patch is based on.
17031
17032commit e17c2cbe9fbaa1600d4b9463ec800a874b0d87cd
17033Author: David Nusinow <dnusinow@debian.org>
17034Date:   Tue Oct 10 22:11:05 2006 -0400
17035
17036    Dynamically generate internal manpage section using __libmanpagesuffix__ so that it actually matches the section if you don't use 3X11
17037
17038commit e53557da969b706dbc843f6fde3db31ffe382e0f
17039Author: Jamey Sharp <jamey@minilop.net>
17040Date:   Sat Oct 7 21:00:36 2006 -0700
17041
17042    Release libX11 1.1 RC1 (1.0.99.1).
17043
17044commit bf237409c5fce32c557d298f62f44d456c2b5bc8
17045Merge: ba477191 2d426d1f
17046Author: Jamey Sharp <jamey@minilop.net>
17047Date:   Sat Oct 7 21:07:16 2006 -0700
17048
17049    As XCB support is about to be released in libX11, stable is now subsumed by master.
17050
17051commit ba477191c67ce93e61423cc1abe35275704cce50
17052Author: Jamey Sharp <jamey@minilop.net>
17053Date:   Sat Oct 7 03:48:13 2006 -0700
17054
17055    XCB: Don't rely on having the definition of struct xcb_setup_t available.
17056
17057commit 117b55cbd0b0ce51362df88363ed83d44a493ac7
17058Author: Jamey Sharp <jamey@minilop.net>
17059Date:   Fri Oct 6 16:27:31 2006 -0700
17060
17061    xcb_poll_for_event no longer takes an 'int *error' out-parameter.
17062
17063commit caaa8e8a55e837b3585c1dee7bef194fc4c79d16
17064Author: Josh Triplett <josh@freedesktop.org>
17065Date:   Fri Oct 6 16:26:11 2006 -0700
17066
17067    Actually ship Xxcbint.h
17068
17069commit 7b027e53b5e393082f4f515c8ba18077eb97163f
17070Author: Josh Triplett <josh@freedesktop.org>
17071Date:   Fri Oct 6 16:25:50 2006 -0700
17072
17073    Clean ChangeLog only in "make maintainer-clean", not "make clean"
17074    
17075    ChangeLog requires a git repo to generate; make clean and make distclean
17076    shouldn't get rid of it.
17077
17078commit cab22e02e78b3e5b8a73d1cd55cf6686426b47e0
17079Author: Josh Triplett <josh@freedesktop.org>
17080Date:   Fri Oct 6 16:24:58 2006 -0700
17081
17082    Add ChangeLog and "make dist"-generated files to .gitignore
17083
17084commit e4c7cfdee4a40e466c0c6b370cabd432e9e855a0
17085Author: Josh Triplett <josh@freedesktop.org>
17086Date:   Fri Oct 6 15:53:27 2006 -0700
17087
17088    Add manual pages for XGetXCBConnection and XSetEventQueueOwner
17089
17090commit 688224cea95e453f94c5a602dc6fce84bc93dfc0
17091Author: Josh Triplett <josh@freedesktop.org>
17092Date:   Fri Oct 6 13:41:16 2006 -0700
17093
17094    Remove unnecessary prototype for _XFreeDisplayStructure in xcb_disp.c
17095
17096commit ab728ca372288d0db1b486c265e34c1376f29104
17097Author: Josh Triplett <josh@freedesktop.org>
17098Date:   Fri Oct 6 12:36:39 2006 -0700
17099
17100    XCL is dead; long live Xlib/XCB
17101    
17102    Rename all instances of "XCL" to Xlib/XCB-derived names.
17103    
17104    The only user-visible change: rename the include file <X11/xcl.h> to
17105    <X11/Xlib-xcb.h>; programs will need to change their #include lines to match.
17106    
17107    Remove the XCL cast inlines from Xlib-xcb.h.
17108
17109commit 5b73093203039d307eb7ab3845c3ced207e9e26c
17110Author: Josh Triplett <josh@freedesktop.org>
17111Date:   Fri Oct 6 11:33:49 2006 -0700
17112
17113    Add XCB developers to AUTHORS
17114
17115commit 1cb71ff139276a0a58c60ea5f261f64b94706b9b
17116Author: Josh Triplett <josh@freedesktop.org>
17117Date:   Fri Oct 6 11:21:28 2006 -0700
17118
17119    Fix email addresses in README
17120
17121commit 12f038669278019594ca0ed53dadcf4e84092422
17122Author: Jamey Sharp <jamey@minilop.net>
17123Date:   Fri Oct 6 02:13:05 2006 -0700
17124
17125    XCB: Handle all responses in order of monotonically increasing sequence number.
17126
17127commit f392680273278b43079302206897f794e60f3c70
17128Author: Josh Triplett <josh@freedesktop.org>
17129Date:   Fri Oct 6 01:11:08 2006 -0700
17130
17131    Actually install x11-xcb.pc, and ship x11-xcb.pc.in
17132
17133commit f1fcad2e3fd17aaf1294f1d8e9f406fd5b32a863
17134Author: Jamey Sharp <jamey@minilop.net>
17135Date:   Thu Oct 5 18:32:29 2006 -0700
17136
17137    XCB: correctly handle failure to connect to X server.
17138
17139commit e754b3b078d556c7861da56aad47d244e9199e06
17140Author: Josh Triplett <josh@freedesktop.org>
17141Date:   Thu Oct 5 17:44:22 2006 -0700
17142
17143    Split public Xlib/XCB functions into libX11-xcb
17144    
17145    We can never change the libX11 soname, and we don't want to commit to never
17146    changing the public Xlib/XCB functions, so split them into a separate library
17147    libX11-xcb.  This also means that a program linked solely against libX11
17148    should work with either Xlib or Xlib/XCB, which will make life easier for
17149    package maintainers.
17150    
17151    Signed-off-by: Josh Triplett <josh@freedesktop.org>
17152    Acked-by: Jamey Sharp <jamey@minilop.net>
17153
17154commit ffd367f708b295abaedf3a23a1bfd4710d171d6f
17155Author: Jamey Sharp <jamey@minilop.net>
17156Date:   Wed Oct 4 17:16:46 2006 -0700
17157
17158    No longer #include Xmd from xcl.h: we do not need it.
17159
17160commit 3aff149d42ba8ed620091971b3766bdf96c62aaf
17161Author: Jamey Sharp <jamey@minilop.net>
17162Date:   Wed Oct 4 16:58:32 2006 -0700
17163
17164    XCB: Revert locking to simple wrapper around libX11's normal locks.
17165    
17166    No more recursive mutexes, no more banging XCB's I/O lock in-place, and
17167    reduces the differences between the previous stable release and an
17168    XCB-enabled one. Sadly, Xlib's pluggable thread functions work again
17169    too, now. I apologize to the world.
17170
17171commit 8ff122fb529bdb1c2b9a86b12d06b6da1b35d708
17172Author: Jamey Sharp <jamey@minilop.net>
17173Date:   Mon Sep 25 04:54:52 2006 -0700
17174
17175    Link explicitly against XCB's Xlib compatibility functions.
17176
17177commit bde3cd123d65a2f36ee0c417f5f231b7e01d0671
17178Author: Jamey Sharp <jamey@minilop.net>
17179Date:   Mon Sep 25 04:13:20 2006 -0700
17180
17181    libxcb now installs header files in <xcb>, not <X11/XCB>.
17182
17183commit 87d00207f5a1f25a45a153618739cd6481814f89
17184Author: Ian Osgood <iano@quirkster.com>
17185Date:   Sun Sep 24 23:39:01 2006 -0700
17186
17187    Track XCB's "Great Renaming".
17188
17189commit 85a5e98dff9b7752fae157fad9b8c9825cb0efab
17190Author: Jamey Sharp <jamey@minilop.net>
17191Date:   Tue Sep 12 23:02:42 2006 -0700
17192
17193    Quit using XCBGetQueuedRequestRead.
17194
17195commit a61936fc4e9bd93b108764bbacd5b8f786e51915
17196Author: Tollef Fog Heen <tfheen@err.no>
17197Date:   Wed Aug 30 00:05:54 2006 +0300
17198
17199    nls: use _XlcUtf8Loader for en_US (bug #7982)
17200    Use _XlcUtf8Loader instead of _XlcUnicodeLoade,r bringing it into line with
17201    every other locale.
17202
17203commit abcc7e1865cdfbd591f6520cfe4257f0b0b1c03e
17204Author: Alan Coopersmith <alan.coopersmith@sun.com>
17205Date:   Wed Aug 23 18:49:30 2006 -0700
17206
17207    When opening display, if LOCALCONN fails, fall back to UNIXCONN, then TCPCONN
17208    
17209    Port to X11R7 of Sun bug fix 4061225 by Alex Chen for X11R6 - when failing to
17210    connect on a named pipe, try a Unix socket first, to better support people who
17211    replace their X servers with ones that don't support named pipe transport.
17212    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4061225>
17213
17214commit 1ddc44c1cada7e926bd4787406444ce7c36b61e1
17215Author: Mark Brown <mark.brown@sun.com>
17216Date:   Thu Jul 27 19:17:10 2006 -0700
17217
17218    Sun bug 1149809: Document event delivery when grab is terminated.
17219
17220commit 171107b03ac89d94f9006c7cda242aeefb9ecd16
17221Author: Dennis Arellano <Dennis.Arellano@Sun.COM>
17222Date:   Thu Jul 27 18:47:06 2006 -0700
17223
17224    Sun bug 4091271: XGetWindowProperty is missing a crucial prop_return description
17225    
17226    Document that 32-bit format properties are always returned in arrays of type
17227    long, even on systems where long is 64-bits.
17228    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4091271>
17229
17230commit 8309efe6550877cd0bf22979904b3f6bd3e6cffa
17231Author: Alan Coopersmith <alan.coopersmith@sun.com>
17232Date:   Mon Jul 24 15:52:00 2006 -0700
17233
17234    Add support for "make lint" to check code with lint/sparse/etc.
17235
17236commit 931e02fbd1acd09aae2b0954c34342c86c72dff6
17237Author: Alan Coopersmith <alan.coopersmith@sun.com>
17238Date:   Mon Jul 24 15:50:52 2006 -0700
17239
17240    ANSIfy some static function definitions
17241
17242commit 20b7abcaac324d90454de63f32f4a2b398e69e63
17243Author: Alan Coopersmith <alan.coopersmith@sun.com>
17244Date:   Mon Jul 24 15:01:40 2006 -0700
17245
17246    Fix sparse warning: Using plain integer as NULL pointer
17247
17248commit d158ab29930513c4097f5b67e7bea08ed2bfd62c
17249Author: Alan Coopersmith <alan.coopersmith@sun.com>
17250Date:   Mon Jul 24 14:00:24 2006 -0700
17251
17252    Remove unused variable
17253
17254commit 30377000375bdb958042dcb1f38503c94ef21eaf
17255Author: Eric Anholt <anholt@FreeBSD.org>
17256Date:   Fri Jul 21 18:55:36 2006 -0400
17257
17258    Bug #7188: Fix the documentation of XUrgencyHint (not UrgencyHint).
17259
17260commit 4eba45879aea4e415ab550ee56b900d060099110
17261Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17262Date:   Sun Jul 16 10:55:39 2006 +0200
17263
17264    set GIT_DIR=${srcdir}/.git for git-log
17265
17266commit 2d426d1f2608fedb77bd7d010dabece76b8d4a60
17267Author: Aaron Plattner <aplattner@nvidia.com>
17268Date:   Tue Jul 11 13:27:49 2006 -0700
17269
17270    Add a .PHONY to ensure the ChangeLog isn't stale.
17271    
17272    Setting the ChangeLog rule as phony forces it to be re-run even when the
17273    ChangeLog file already exists. Research indicates .PHONY is portable to BSD and
17274    Solaris make.
17275    (cherry picked from b8a98809ed81e1226775e6447ef219ffc01334b5 commit)
17276
17277commit b8a98809ed81e1226775e6447ef219ffc01334b5
17278Author: Aaron Plattner <aplattner@nvidia.com>
17279Date:   Tue Jul 11 13:27:49 2006 -0700
17280
17281    Add a .PHONY to ensure the ChangeLog isn't stale.
17282    
17283    Setting the ChangeLog rule as phony forces it to be re-run even when the
17284    ChangeLog file already exists. Research indicates .PHONY is portable to BSD and
17285    Solaris make.
17286
17287commit 8f2be66089b88e4ed6acb0211ea107f4bb910bd3
17288Author: Mayank Jain <majain@redhat.com>
17289Date:   Tue Jul 11 19:42:47 2006 +0100
17290
17291    add Indian language locales
17292    Add as, kn, ml, or, ur, and te locales.
17293
17294commit a92eb6785699bbc9c6c3813f6af3bb9431a3f6b0
17295Author: Matthias Hopf <mhopf@suse.de>
17296Date:   Tue Jul 4 12:16:30 2006 +0200
17297
17298    Fix for autoconf 2.60 issue.
17299    
17300    Updated AC_DEFINE_DIR.
17301    Reverted datarootdir change.
17302
17303commit 644f4828b15bce42b597eb123ba0bbc372c46c03
17304Author: Keith Packard <keithp@neko.keithp.com>
17305Date:   Sat Jul 1 21:31:23 2006 -0700
17306
17307    Work around recent autoconf (2.59?) changes in directory expansion.
17308    
17309    Recent autoconf versions have changed how directory names are managed in the
17310    configure.ac script; automatic 'eval' invocations now occur as a part of the
17311    AC_DEFINE_DIR macro which make it imperative that AC_DEFINE_DIR be executed
17312    before the variables are used in further macro definitions. Also, ${datadir}
17313    is apparantly an old name for ${datarootdir} as ${datadir} doesn't get
17314    expanded correctly by AC_DEFINE_DIR. This looks like an autoconf bug, but it
17315    is easy to work around by just using ${datarootdir} instead of ${datadir}.
17316
17317commit be70a31229aa106aff0a09d78c00812682cd3475
17318Author: Keith Packard <keithp@neko.keithp.com>
17319Date:   Sat Jul 1 01:56:05 2006 -0700
17320
17321    Xlib/XCB: handle 32-bit sequence wrap.
17322    
17323    Replace broken sequence compares with XCB_SEQUENCE_COMPARE (copied from
17324    XCB).
17325    Account for XCB sequence 0 handling.
17326
17327commit 99c711707ad08e1396e123b1c7df687c560a489a
17328Author: Donnie Berkholz <spyderous@gentoo.org>
17329Date:   Thu Jun 29 19:43:20 2006 -0700
17330
17331    Bump version to 1.0.3.
17332
17333commit cde3c0dd72af2b490e80cffca962e3487dd31be4
17334Author: Donnie Berkholz <spyderous@gentoo.org>
17335Date:   Thu Jun 29 19:39:36 2006 -0700
17336
17337    Bug #7349: Missed one of the setuid fixes.
17338    (cherry picked from e9614c963b532f46a7932c2305a4b177a996a222 commit)
17339
17340commit df3fef8983d96c59d481c4cdaf1f271d54a116d2
17341Author: Matthias Hopf <mhopf@suse.de>
17342Date:   Thu Jun 29 18:59:57 2006 +0200
17343
17344    Update to final Compose cache directory location.
17345    (cherry picked from abda4d223e9cce9ac6e7b5d82a5680d9a502e52a commit)
17346
17347commit 912ef198292d3053daa810f842510e5d62ded0f0
17348Author: Matthias Hopf <mhopf@suse.de>
17349Date:   Thu Jun 29 17:41:41 2006 +0200
17350
17351    Fix alignment of trees and wide chars in the cache.
17352    (cherry picked from 40a64c61f8bc33d497e1224e02c41dea2d424d97 commit)
17353
17354commit 2ece832118b3ee5d8ed19f1ee9b1c822b70ec6e9
17355Author: Matthias Hopf <mhopf@suse.de>
17356Date:   Wed Jun 28 19:17:03 2006 +0200
17357
17358    First (dummy) entry of compose caches was not initialized and thus contained varying garbage.
17359    (cherry picked from f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3 commit)
17360
17361commit e9614c963b532f46a7932c2305a4b177a996a222
17362Author: Donnie Berkholz <spyderous@gentoo.org>
17363Date:   Thu Jun 29 19:39:36 2006 -0700
17364
17365    Bug #7349: Missed one of the setuid fixes.
17366
17367commit abda4d223e9cce9ac6e7b5d82a5680d9a502e52a
17368Author: Matthias Hopf <mhopf@suse.de>
17369Date:   Thu Jun 29 18:59:57 2006 +0200
17370
17371    Update to final Compose cache directory location.
17372
17373commit 40a64c61f8bc33d497e1224e02c41dea2d424d97
17374Author: Matthias Hopf <mhopf@suse.de>
17375Date:   Thu Jun 29 17:41:41 2006 +0200
17376
17377    Fix alignment of trees and wide chars in the cache.
17378
17379commit f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3
17380Author: Matthias Hopf <mhopf@suse.de>
17381Date:   Wed Jun 28 19:17:03 2006 +0200
17382
17383    First (dummy) entry of compose caches was not initialized and thus contained varying garbage.
17384
17385commit bdbe464d774e01d317f67c63ebbda2fd0edbbdd1
17386Author: Donnie Berkholz <spyderous@gentoo.org>
17387Date:   Thu Jun 22 23:59:03 2006 -0700
17388
17389    Delete ChangeLog, and add a rule to autogenerate it for distribution. Also
17390    add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a
17391    ChangeLog.
17392    (cherry picked from b0edfb8df16ab8c9959b83a4c966d55a59c4e295 commit)
17393
17394commit b0edfb8df16ab8c9959b83a4c966d55a59c4e295
17395Author: Donnie Berkholz <donnie@comet.(none)>
17396Date:   Thu Jun 22 23:47:38 2006 -0700
17397
17398    Delete ChangeLog, and add a rule to autogenerate it for distribution. Also
17399    add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a
17400    ChangeLog.
17401
17402commit dd54981aa76e9dfdc4c3302d6105b4b229447c84
17403Author: Donnie Berkholz <donnie@comet.(none)>
17404Date:   Thu Jun 22 15:47:38 2006 -0700
17405
17406    Bump version to 1.0.2.
17407
17408commit c93539d974a67f596a5eb5b65042d26602546c72
17409Author: Matthieu Herrb <matthieu@deville.herrb.com>
17410Date:   Tue Jun 20 21:04:03 2006 +0200
17411
17412    Check setuid() return value.
17413    (cherry picked from 5169d0e08ff6acb350a6ea768623f5ff0b85b05f commit)
17414
17415commit 605533f814ab7892991578706a6458f61a89ca4d
17416Author: Derek Wang <derek.wang@sun.com>
17417Date:   Mon Jun 19 11:05:37 2006 -0700
17418
17419    Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded
17420    (cherry picked from eff50c94a07194297e705da53d9fbb3a40fb9ad4 commit)
17421
17422commit 5bbd0822c5a926de0ed293437fb9f2b75cf3c4f4
17423Author: Scott Revelt <scott.revelt@sun.com>
17424Date:   Fri Jun 16 19:11:13 2006 -0700
17425
17426    Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators
17427    based on locale that doesn't match those used in the Xcms.txt
17428    (cherry picked from 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa commit)
17429
17430commit 0b05cd4da6134df527fb010384a9fd569bd5d6a3
17431Author: Alan Coopersmith <alanc@alf.(none)>
17432Date:   Fri Jun 16 18:53:33 2006 -0700
17433
17434    Add *~ to ignore emacs droppings
17435    (cherry picked from c33d7b8282ac196b36229be26442296768c16f3e commit)
17436
17437commit 214658b76b56768f69c3959a11525aae7813f448
17438Author: Matthias Hopf <mhopf@suse.de>
17439Date:   Fri Jun 16 15:36:40 2006 +0200
17440
17441    Bug #3104: Compose table cache for faster X11 application starts.
17442    (cherry picked from 1f4c9893ade08bad30c9bd12a36bee57d30b001e commit)
17443
17444commit 13968a23aaea838ba4b69e42e8900f803499e091
17445Author: Matthias Hopf <mhopf@suse.de>
17446Date:   Tue Jun 13 20:23:46 2006 +0200
17447
17448    Bug #3104: Compose table cache for faster X11 application starts.
17449    (cherry picked from 4fe22647e6010a2886c2f3a7093adeaeb6678faa commit)
17450
17451commit e7f8bca08f5476d0ca262097639ac7d424bb4d10
17452Author: Lubos Lunak <llunak@suse.de>
17453Date:   Mon Jun 12 18:48:08 2006 +0200
17454
17455    Bug #3104: Compose table cache for faster X11 application starts.
17456    (cherry picked from 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be commit)
17457
17458commit f506aaf8ac9aae1ee8daaef6cde34ee85aecd641
17459Author: Matthias Hopf <mhopf@suse.de>
17460Date:   Fri Jun 9 18:24:02 2006 +0200
17461
17462    Bug #3104: Compose table cache for faster X11 application starts. Part 1: Pointerless compose data structure, using indices instead of pointers, needed for mmap()ing data structure.
17463    (cherry picked from 9354351fcb8baeaab85250d14409cfb4fa50f3e9 commit)
17464
17465commit 0e6d5e979aacb0c295ce79369ecc5f22ffa7922c
17466Author: Jamey Sharp <jamey@minilop.net>
17467Date:   Wed Jun 7 20:29:05 2006 -0700
17468
17469    Fix bug #7035: unnecessary memmove in XOpenDisplay.
17470    Using memmove on the connection setup data causes a problem for XCB, but making
17471    Xlib stop doing that should be harmless for non-XCB as well.
17472    (cherry picked from b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0 commit)
17473
17474commit cd7328c46ae72903ed02832828891b2dab4d5ee0
17475Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17476Date:   Sat Jun 3 13:51:51 2006 +0300
17477
17478    Fix threading support on GNU/kFreeBSD systems.  (Robert Millan)
17479    (cherry picked from e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb commit)
17480
17481commit c336eb6b80a6f91da1d0b3d28634a2cfde324670
17482Merge: 33556ca8 ad9ebbd2
17483Author: Donnie Berkholz <donnie@comet.(none)>
17484Date:   Thu Jun 22 14:25:35 2006 -0700
17485
17486    Merge branch 'stable' of http://people.freedesktop.org/~jamey/libX11 into stable
17487
17488commit 213dacad21740466e7ab31b01d3fc513fe4b3e74
17489Author: Daniel Stone <daniel@fooishbar.org>
17490Date:   Thu Jun 22 17:20:59 2006 +0300
17491
17492    Bump to 1.0.99.0 to avoid confusion.
17493
17494commit efedfd68e31bcee2d21ac340be8dc9e1825ec890
17495Merge: e3acee88 4b8eb5d4
17496Author: Daniel Stone <daniel@fooishbar.org>
17497Date:   Thu Jun 22 16:53:45 2006 +0300
17498
17499    Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11
17500
17501commit 4b8eb5d4a1da73a94b5a6ab12e34784aae4c79c5
17502Merge: 5169d0e0 eff50c94
17503Author: Matthieu Herrb <matthieu@deville.herrb.com>
17504Date:   Tue Jun 20 21:05:15 2006 +0200
17505
17506    Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11
17507
17508commit 5169d0e08ff6acb350a6ea768623f5ff0b85b05f
17509Author: Matthieu Herrb <matthieu@deville.herrb.com>
17510Date:   Tue Jun 20 21:04:03 2006 +0200
17511
17512    Check setuid() return value.
17513
17514commit eff50c94a07194297e705da53d9fbb3a40fb9ad4
17515Author: Derek Wang <derek.wang@sun.com>
17516Date:   Mon Jun 19 11:05:37 2006 -0700
17517
17518    Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded
17519
17520commit 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa
17521Author: Scott Revelt <scott.revelt@sun.com>
17522Date:   Fri Jun 16 19:11:13 2006 -0700
17523
17524    Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators
17525    based on locale that doesn't match those used in the Xcms.txt
17526
17527commit c33d7b8282ac196b36229be26442296768c16f3e
17528Author: Alan Coopersmith <alanc@alf.(none)>
17529Date:   Fri Jun 16 18:53:33 2006 -0700
17530
17531    Add *~ to ignore emacs droppings
17532
17533commit 1f4c9893ade08bad30c9bd12a36bee57d30b001e
17534Author: Matthias Hopf <mhopf@suse.de>
17535Date:   Fri Jun 16 15:36:40 2006 +0200
17536
17537    Bug #3104: Compose table cache for faster X11 application starts.
17538
17539commit 4fe22647e6010a2886c2f3a7093adeaeb6678faa
17540Author: Matthias Hopf <mhopf@suse.de>
17541Date:   Tue Jun 13 20:23:46 2006 +0200
17542
17543    Bug #3104: Compose table cache for faster X11 application starts.
17544
17545commit 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be
17546Author: Lubos Lunak <llunak@suse.de>
17547Date:   Mon Jun 12 18:48:08 2006 +0200
17548
17549    Bug #3104: Compose table cache for faster X11 application starts.
17550
17551commit 9354351fcb8baeaab85250d14409cfb4fa50f3e9
17552Author: Matthias Hopf <mhopf@suse.de>
17553Date:   Fri Jun 9 18:24:02 2006 +0200
17554
17555    Bug #3104: Compose table cache for faster X11 application starts. Part 1: Pointerless compose data structure, using indices instead of pointers, needed for mmap()ing data structure.
17556
17557commit b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0
17558Author: Jamey Sharp <jamey@minilop.net>
17559Date:   Wed Jun 7 20:29:05 2006 -0700
17560
17561    Fix bug #7035: unnecessary memmove in XOpenDisplay.
17562    Using memmove on the connection setup data causes a problem for XCB, but making
17563    Xlib stop doing that should be harmless for non-XCB as well.
17564
17565commit ad9ebbd2424bc2699944ffdf4e19e13f9dd8ab84
17566Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17567Date:   Sat Jun 3 12:57:55 2006 +0300
17568
17569    Bug #2186: Add cs_CZ.iso8859-2 alias.
17570
17571commit 9e7765e0b1cbaae6643072d91066ba1201b36227
17572Author: Daniel Stone <daniel@fooishbar.org>
17573Date:   Fri Jun 2 02:46:29 2006 +0300
17574
17575    nls: Serbian (sr_CS) update (#5575)
17576    
17577    Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a
17578    corresponding locale change.  Update compose.dir.pre, locale.alias.pre,
17579    and locale.dir.pre.  (Milos Komarcevic)
17580
17581commit a4ac2242b588da23044a20aa999ae84d4de7b2d8
17582Author: Daniel Stone <daniel@fooishbar.org>
17583Date:   Fri Jun 2 02:24:25 2006 +0300
17584
17585    im: add Braille input method (#6296)
17586    
17587    Bug #6296: Add a Braille input method.  (Samuel Thibault)
17588
17589commit 90de1e2e141ec591048a76cb695579ef809a28d3
17590Author: Daniel Stone <daniel@fooishbar.org>
17591Date:   Fri Jun 2 02:22:17 2006 +0300
17592
17593    xkb support: small typo
17594
17595commit 4c3e34bece7402f08139d34d1ef5834e3cf533c7
17596Author: Daniel Stone <daniel@fooishbar.org>
17597Date:   Fri Jun 2 01:50:24 2006 +0300
17598
17599    en_US.UTF-8 Compose.pre: updates from Simos (#5129)
17600    
17601    Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms,
17602    adding Unicode character names, removing duplicate entries, et al.
17603
17604commit 0c6473dd329c7334ae511884bcb6e73e632c784f
17605Author: Daniel Stone <daniel@fooishbar.org>
17606Date:   Fri Jun 2 01:44:53 2006 +0300
17607
17608    nls: fix use of non-keysym dead_space (#5107)
17609    
17610    Bug #5107: Change users of dead_space to space.
17611
17612commit 6f99f6349de5120f1cb7e02fbc97849341bc48e8
17613Author: Daniel Stone <daniel@fooishbar.org>
17614Date:   Fri Jun 2 01:41:18 2006 +0300
17615
17616    optional XKB support fix
17617    
17618    Fix compilation with --disable-xkb.
17619
17620commit 217d43ed44ced901122093af3ef1294e1736bb77
17621Author: Daniel Stone <daniel@fooishbar.org>
17622Date:   Fri Jun 2 01:39:12 2006 +0300
17623
17624    i18n: separate data and lib directories
17625    
17626    Break out locale data into separate data and library directories, under
17627    $(datadir) and $(libdir), respectively, by default.
17628
17629commit 92fa7fcde8df22830fca7c0275ab201033f7909c
17630Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17631Date:   Thu May 11 14:04:48 2006 -0400
17632
17633    libXcursor.so.1, not libXcursor.so
17634
17635commit 135b4df13ed7c35dbae8975f302fc1fb8412d7c0
17636Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17637Date:   Wed May 10 18:06:03 2006 +0300
17638
17639    XKBMisc.c: use Xfree, not xfree
17640    
17641    Use Xfree() instead of xfree() when freeing interps.
17642
17643commit 3518d772b08e3433bc28b4d8d293fa53ca25f0ee
17644Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17645Date:   Wed May 10 14:51:37 2006 +0300
17646
17647    locale.alias.pre: bg_BG typo fix
17648    
17649    Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre.
17650
17651commit cc533db60cb64dc163c66451933a9bf77c519062
17652Author: Daniel Stone <daniels@endtroducing.localdomain>
17653Date:   Sun Apr 9 22:22:03 2006 +0300
17654
17655    Coverity #203, #204: Fix potential NULL dereferences.
17656
17657commit b83adf7dfd6157694fe4f232012fef36cd9666da
17658Author: Daniel Stone <daniels@endtroducing.localdomain>
17659Date:   Sun Apr 9 22:20:25 2006 +0300
17660
17661    Coverity #205: Fix potential NULL dereference.
17662
17663commit 2d0cd10ad907864d0136739eaac459779c9a5332
17664Author: Daniel Stone <daniels@endtroducing.localdomain>
17665Date:   Sun Apr 9 22:18:20 2006 +0300
17666
17667    Coverity #209: Fix potential NULL dereference. (Alan Coopersmith)
17668
17669commit dc2f3966068d66a564aa452cab9f0c26657fa1df
17670Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17671Date:   Fri Apr 7 18:11:52 2006 +0300
17672
17673    Coverity #826: Fix potential memory leak.
17674
17675commit 23df609ec451a01c77e8f31ecc85c5af7c62efed
17676Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17677Date:   Fri Apr 7 17:49:41 2006 +0300
17678
17679    Bug #1625: Include keysym.h from Xutil.h.
17680
17681commit 5262a1945c543a3419ed626e1deb09ef5b4584c1
17682Author: Eric Anholt <anholt@leguin.anholt.net>
17683Date:   Wed Apr 5 17:12:15 2006 -0700
17684
17685    Check if visualList == NULL, not nVisualsMatched == 0. NULL happens in more
17686    
17687    cases (allocation failure) than nVisualsMatched == 0.  Noticed from inspection
17688    of Coverity #599, #600.
17689
17690commit 8b42635f577468bb143ca593cdd9fb3450ad712c
17691Author: Eric Anholt <anholt@leguin.anholt.net>
17692Date:   Wed Apr 5 16:42:26 2006 -0700
17693
17694    Coverity #558: Free newly-allocated Database in error path.
17695
17696commit 6d06e41d1f431b3f1a1fcf69161e0af411325e9f
17697Author: Eric Anholt <anholt@leguin.anholt.net>
17698Date:   Wed Apr 5 16:38:52 2006 -0700
17699
17700    Coverity #582: Free newly-allocated region in error path.
17701
17702commit 5fd8f79ad3e38df74d9a6cb573617542c101df1a
17703Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17704Date:   Fri Mar 17 15:58:39 2006 +0200
17705
17706    Properly clip bounds when only one point is defining an outline.
17707
17708commit 1e1572eb7f8394ce152e99d96f711ccf9083baf9
17709Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17710Date:   Sat Mar 11 15:39:15 2006 +0100
17711
17712    Fix prototype of XConfigureWindow(). Bugzilla #6023.
17713
17714commit 1da8bd904f1fc79d63d368473531b438d08bfe37
17715Author: Jeremy C. Reed <reed@reedmedia.net>
17716Date:   Tue Feb 21 14:10:22 2006 -0800
17717
17718    Set XTHREADLIB correctly for dragonfly platforms.
17719
17720commit c9768133e3f65ac4bb64e3941e2d6ae36897ec9c
17721Author: Jamey Sharp <jamey@minilop.net>
17722Date:   Sun Feb 19 12:29:27 2006 -0800
17723
17724    Update .gitignores for *.o and nls/locale.dir*.
17725
17726commit efcbde6ba0b770bb0d4c7818e307712161011f10
17727Author: Jamey Sharp <jamey@minilop.net>
17728Date:   Sun Feb 19 12:28:41 2006 -0800
17729
17730    Move .cvsignore to .gitignore.
17731
17732commit e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb
17733Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17734Date:   Sat Jun 3 13:51:51 2006 +0300
17735
17736    Fix threading support on GNU/kFreeBSD systems.  (Robert Millan)
17737
17738commit f1bd3152359ddfadd0d043006036c239f3e2907b
17739Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17740Date:   Sat Jun 3 12:57:55 2006 +0300
17741
17742    Bug #2186: Add cs_CZ.iso8859-2 alias.
17743
17744commit 2b1b79d90db1d7f0472718b8c010c36275994195
17745Author: Daniel Stone <daniel@fooishbar.org>
17746Date:   Fri Jun 2 02:46:29 2006 +0300
17747
17748    nls: Serbian (sr_CS) update (#5575)
17749    Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a
17750    corresponding locale change.  Update compose.dir.pre, locale.alias.pre,
17751    and locale.dir.pre.  (Milos Komarcevic)
17752
17753commit d6fba1f44d404362d3be1b421f57d7ccc3c8cdac
17754Author: Daniel Stone <daniel@fooishbar.org>
17755Date:   Fri Jun 2 02:24:25 2006 +0300
17756
17757    im: add Braille input method (#6296)
17758    Bug #6296: Add a Braille input method.  (Samuel Thibault)
17759
17760commit 0fed7d3185addd610e917dcdaa0676f0256c0ec5
17761Author: Daniel Stone <daniel@fooishbar.org>
17762Date:   Fri Jun 2 02:22:17 2006 +0300
17763
17764    xkb support: small typo
17765
17766commit cf7d9f9e46f3ce01ac04a95978918d5c0f3f3cf9
17767Author: Daniel Stone <daniel@fooishbar.org>
17768Date:   Fri Jun 2 01:50:24 2006 +0300
17769
17770    en_US.UTF-8 Compose.pre: updates from Simos (#5129)
17771    Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms,
17772    adding Unicode character names, removing duplicate entries, et al.
17773
17774commit 332d45fce9fdbf59168d90a133af1f580a589e54
17775Author: Daniel Stone <daniel@fooishbar.org>
17776Date:   Fri Jun 2 01:44:53 2006 +0300
17777
17778    nls: fix use of non-keysym dead_space (#5107)
17779    Bug #5107: Change users of dead_space to space.
17780
17781commit 34f59ce3d1e2eb2971b732d11871c6fff8a1c75b
17782Author: Daniel Stone <daniel@fooishbar.org>
17783Date:   Fri Jun 2 01:41:18 2006 +0300
17784
17785    optional XKB support fix
17786    Fix compilation with --disable-xkb.
17787
17788commit c5940a0b85edec4003f91a59fc3c44f538accfe1
17789Author: Daniel Stone <daniel@fooishbar.org>
17790Date:   Fri Jun 2 01:39:12 2006 +0300
17791
17792    i18n: separate data and lib directories
17793    Break out locale data into separate data and library directories, under
17794    $(datadir) and $(libdir), respectively, by default.
17795
17796commit 33556ca81db2419b9d2a37664c4cea2069414b37
17797Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17798Date:   Fri May 12 14:49:17 2006 -0400
17799
17800    Bump to 1.0.1
17801
17802commit 80d88557626fae9debc404de33d7fb5d69a6049d
17803Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17804Date:   Thu May 11 14:06:28 2006 -0400
17805
17806    libXcursor.so.1, not libXcursor.so
17807
17808commit 5384f27dfe3c94c462c137ab3540eaa5646ca4c6
17809Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17810Date:   Thu May 11 14:04:48 2006 -0400
17811
17812    libXcursor.so.1, not libXcursor.so
17813
17814commit 01f4d433eed6b70c6e9636157acac022054fdeb6
17815Author: Jamey Sharp <jamey@minilop.net>
17816Date:   Wed May 10 17:02:52 2006 -0700
17817
17818    Count any partial request towards the current Xlib sequence number.
17819
17820commit 770cfbd1fcc80a83a9be0c4f68727b8af2c8f4a4
17821Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17822Date:   Wed May 10 18:06:03 2006 +0300
17823
17824    XKBMisc.c: use Xfree, not xfree
17825    Use Xfree() instead of xfree() when freeing interps.
17826
17827commit 22a5255b80b80772612279bc840a953edd0e3442
17828Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17829Date:   Wed May 10 14:51:37 2006 +0300
17830
17831    locale.alias.pre: bg_BG typo fix
17832    Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre.
17833
17834commit 9cac8c9824874ca7d835f001a4efa910b7fdd822
17835Merge: 19b8840a e514bc87
17836Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17837Date:   Wed May 10 14:50:37 2006 +0300
17838
17839    Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11
17840
17841commit e514bc875f27f4bf197b06b8315eeca526195915
17842Author: Jamey Sharp <jamey@minilop.net>
17843Date:   Tue May 9 12:41:59 2006 -0700
17844
17845    Assert that dpy->request does not go backwards. Catches #5839 earlier.
17846
17847commit fc1159137365a599bf611ee001f439416952c4e0
17848Author: Jamey Sharp <jamey@minilop.net>
17849Date:   Sun May 7 17:40:01 2006 -0700
17850
17851    In _XPutXCBBuffer, set aside any trailing partial request until the last byte is available.
17852
17853commit c394480a4247213239822808e3f6e7c6cd6decd9
17854Author: Jamey Sharp <jamey@minilop.net>
17855Date:   Sun May 7 16:58:13 2006 -0700
17856
17857    Update for XCB ConnSetupSuccessRep name change.
17858
17859commit 7672bf93bc1200905461aeb0a2dc2c2696410b93
17860Author: Daniel Stone <daniels@endtroducing.localdomain>
17861Date:   Sun Apr 9 22:22:03 2006 +0300
17862
17863    Coverity #203, #204: Fix potential NULL dereferences.
17864
17865commit cfcafbe48d22d9a0cd50eb9454ce0ff88f0129d3
17866Author: Daniel Stone <daniels@endtroducing.localdomain>
17867Date:   Sun Apr 9 22:20:25 2006 +0300
17868
17869    Coverity #205: Fix potential NULL dereference.
17870
17871commit b6771501feea98d037032f82117b6d4f15779f07
17872Author: Daniel Stone <daniels@endtroducing.localdomain>
17873Date:   Sun Apr 9 22:18:20 2006 +0300
17874
17875    Coverity #209: Fix potential NULL dereference.  (Alan Coopersmith)
17876
17877commit 19b8840af241087bb17b1edabcaa9b28fdd0a1dc
17878Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17879Date:   Fri Apr 7 18:11:52 2006 +0300
17880
17881    Coverity #826: Fix potential memory leak.
17882
17883commit 3a16f262abe48b44ed641525e894bc22e13bf72a
17884Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17885Date:   Fri Apr 7 17:49:41 2006 +0300
17886
17887    Bug #1625: Include keysym.h from Xutil.h.
17888
17889commit 2481b767ae96e2f2503c0390545932c8397b090f
17890Author: Eric Anholt <anholt@leguin.anholt.net>
17891Date:   Wed Apr 5 17:12:15 2006 -0700
17892
17893    Check if visualList == NULL, not nVisualsMatched == 0.  NULL happens in more
17894    cases (allocation failure) than nVisualsMatched == 0.  Noticed from inspection
17895    of Coverity #599, #600.
17896
17897commit 152b17e47d878c2d928eb74581aa69d925a29123
17898Author: Eric Anholt <anholt@leguin.anholt.net>
17899Date:   Wed Apr 5 16:42:26 2006 -0700
17900
17901    Coverity #558: Free newly-allocated Database in error path.
17902
17903commit bc62b99ef36edb34035911c42104be7f6f9d2333
17904Author: Eric Anholt <anholt@leguin.anholt.net>
17905Date:   Wed Apr 5 16:38:52 2006 -0700
17906
17907    Coverity #582: Free newly-allocated region in error path.
17908
17909commit d47f0b3cec1388f7ce60ab2af91df0dea0f221c5
17910Author: Jamey Sharp <jamey@minilop.net>
17911Date:   Fri Mar 31 22:53:07 2006 -0800
17912
17913    Fix buggy interaction with XCB when running out of XIDs.
17914
17915commit e3f452571824d6a875bbf582946de185de9e01e9
17916Author: Jamey Sharp <jamey@minilop.net>
17917Date:   Fri Mar 31 22:52:14 2006 -0800
17918
17919    Add explicit include of Xmd.h to work around bug including both xcb.h and Xmd.h simultaneously.
17920
17921commit 2363b74ca795c1b3a73c9e572532ba5191adec5b
17922Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17923Date:   Fri Mar 17 15:58:39 2006 +0200
17924
17925    Properly clip bounds when only one point is defining an outline.
17926
17927commit e876efb8aa410f2f5c87aaaa7042f847c4ff96f7
17928Merge: be266b20 f71ea0bc
17929Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17930Date:   Sat Mar 11 15:40:35 2006 +0100
17931
17932    Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11
17933
17934commit be266b201dc13530a302a7572283ccd3f32aad87
17935Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17936Date:   Sat Mar 11 15:39:15 2006 +0100
17937
17938    Fix prototype of XConfigureWindow(). Bugzilla #6023.
17939
17940commit f71ea0bc737c5a42e9e022b86e7ec3b4f846d31c
17941Author: Jamey Sharp <jamey@minilop.net>
17942Date:   Fri Mar 3 11:08:41 2006 -0800
17943
17944    Update for XCBSendRequest sequence number API changes.
17945
17946commit a11d1b0ae674320cf9897f6a83ec08c65eca8d9b
17947Author: Jamey Sharp <jamey@minilop.net>
17948Date:   Fri Mar 3 01:42:49 2006 -0800
17949
17950    Use the full_sequence from XCBGenericError/Event for setting last_request_read, and quit replacing _XSetLastRequestRead with an XCB-specific version.
17951
17952commit d8ba4ae7045b227f8b675628b9094dded02f1c08
17953Author: Jamey Sharp <jamey@minilop.net>
17954Date:   Thu Mar 2 23:43:26 2006 -0800
17955
17956    Bugfix: Rely on XCBSendRequest to leave iov in a well-defined state, and place the spare iovecs at the beginning of the array.
17957
17958commit f9afb5a54435c30961306080e9358d4240ecb844
17959Author: Jamey Sharp <jamey@minilop.net>
17960Date:   Thu Mar 2 23:34:19 2006 -0800
17961
17962    assert() that we will not infinite loop or read uninitialized memory.
17963
17964commit d3512ef3aae5b036a8ce6579318108f1ec20ee22
17965Author: Jamey Sharp <jamey@minilop.net>
17966Date:   Thu Mar 2 15:58:52 2006 -0800
17967
17968    Quit relying on XCBSendRequest to pad to 4-byte boundaries and do it ourselves.
17969
17970commit fb590c15a740264ee867d15a2547072e43b21eed
17971Author: Jamey Sharp <jamey@minilop.net>
17972Date:   Thu Mar 2 12:06:04 2006 -0800
17973
17974    Handle errors correctly when Xlib owns the event queue and XCB has the checked error feature.
17975
17976commit 8356ba37d307a9eda895a6bf41ef727bbfc9a695
17977Author: Jamey Sharp <jamey@minilop.net>
17978Date:   Mon Feb 27 11:51:47 2006 -0800
17979
17980    Use the new XCBSendRequest flag, XCB_REQUEST_RAW, to hand a bag-o-bytes down uninterpreted.
17981
17982commit 07bdf1fbbf2418f866df1a2140d514dd3f035139
17983Author: Jamey Sharp <jamey@minilop.net>
17984Date:   Sun Feb 26 15:46:01 2006 -0800
17985
17986    Update for new XCBSendRequest API.
17987
17988commit 409a08cff8347d39e0e6c53c9f380d21f221f5ac
17989Author: Jamey Sharp <jamey@minilop.net>
17990Date:   Thu Feb 23 18:12:31 2006 -0800
17991
17992    Performance fix: Replace calls to XCBGetRequestRead with the new XCBGetQueuedRequestRead. Cuts a lot of syscalls.
17993
17994commit ec30a27341b97620b07dd886f98d1d7664a67685
17995Author: Jamey Sharp <jamey@minilop.net>
17996Date:   Thu Feb 23 18:01:46 2006 -0800
17997
17998    Minor performance fix: Access dpy->xcl->connection directly instead of calling XCBConnectionOfDisplay. It happens a lot.
17999
18000commit 53c471c6a835d5cedeca99f2c97058d196a3fd7e
18001Author: Jamey Sharp <jamey@minilop.net>
18002Date:   Thu Feb 23 11:46:09 2006 -0800
18003
18004    XCBFlush used to return non-positive on failure, and this test did not catch 0. Now it returns boolean: 0 or 1. Testing <= 0 covers both cases. I probably want to switch to a boolean test eventually.
18005
18006commit 41c0121a8718b530feaf7fe315b673d9b8defce2
18007Author: Jamey Sharp <jamey@minilop.net>
18008Date:   Tue Feb 21 21:25:41 2006 -0800
18009
18010    Refactor the code that passes requests down to XCB into a separate issue_complete_request function.
18011
18012commit 67d06e0fe468dca22847aa14d3f917128f89f9cf
18013Author: Jamey Sharp <jamey@minilop.net>
18014Date:   Tue Feb 21 15:33:05 2006 -0800
18015
18016    If we have not actually put the buffer back, deferred invariants may not hold. This is OK.
18017
18018commit 434bf80b4053ff1ba82adf65de1f76b4d3731bf1
18019Author: Jeremy C. Reed <reed@reedmedia.net>
18020Date:   Tue Feb 21 14:10:22 2006 -0800
18021
18022    Set XTHREADLIB correctly for dragonfly platforms.
18023
18024commit b24834762e975bd319f9ab5c7cf790b2a02a9474
18025Author: Jamey Sharp <jamey@minilop.net>
18026Date:   Tue Feb 21 14:03:26 2006 -0800
18027
18028    Sometimes functions other than _XUnlockDisplay call _XPutXCBBuffer. Some invariants appropriate for Unlock are not appropriate otherwise: move them to _XUnlockDisplay.
18029
18030commit 99b8defd0d5e6993071e21638128c9de2574b37d
18031Author: Jamey Sharp <jamey@minilop.net>
18032Date:   Tue Feb 21 13:04:21 2006 -0800
18033
18034    Execute BeforeFlush hooks on complete buffers, not request-at-a-time.
18035    Traditional Xlib worked this way; I dunno why I changed it.
18036
18037commit 9b01e7849775749182052fe324df9d8e6ceeee99
18038Author: Jamey Sharp <jamey@minilop.net>
18039Date:   Tue Feb 21 12:51:44 2006 -0800
18040
18041    Factor the XCBSendRequest call out of the conditional in _XPutXCBBuffer.
18042
18043commit 7ce7ac882de128955751a5307889db9d712d8a72
18044Author: Jamey Sharp <jamey@minilop.net>
18045Date:   Tue Feb 21 12:44:30 2006 -0800
18046
18047    Quit using a triple-pointer. Almost as if I were a sensible person.
18048
18049commit 35a858be218cdbfa4593d44a67663d5c25297016
18050Author: Jamey Sharp <jamey@minilop.net>
18051Date:   Tue Feb 21 12:21:01 2006 -0800
18052
18053    Remove the XCL_PAD macro.
18054
18055commit e741b70ed2542c5463c57dac44bc37328616733b
18056Author: Jamey Sharp <jamey@minilop.net>
18057Date:   Sun Feb 19 12:29:27 2006 -0800
18058
18059    Update .gitignores for *.o and nls/locale.dir*.
18060
18061commit f25b4b00e1683b0d97dba46dac46d65a9c2270a6
18062Author: Jamey Sharp <jamey@minilop.net>
18063Date:   Sun Feb 19 12:28:41 2006 -0800
18064
18065    Move .cvsignore to .gitignore.
18066
18067commit c7cda56eebaf6ab11403363be14d4948d7d8be38
18068Author: Jamey Sharp <jamey@minilop.net>
18069Date:   Sun Feb 19 11:49:15 2006 -0800
18070
18071    Land XCB support on X.org HEAD.
18072
18073commit 881467b3032261791ef5ec61b3879bb68d0a3d8c
18074Author: Jamey Sharp <jamey@minilop.net>
18075Date:   Sun Feb 19 02:14:11 2006 +0000
18076
18077    Merge the X11 and BIGREQS package sets: they were used in exactly the same places.
18078
18079commit 6b0158dfad714db5b89c04dbea3aedeafa0fb146
18080Author: Jamey Sharp <jamey@minilop.net>
18081Date:   Tue Feb 14 19:37:36 2006 +0000
18082
18083    Refactor _XFlush and _XSend code that sets dpy->synchandler to _XSeqSyncFunction into a new function, _XSetSeqSyncFunction. It makes the patch for XCB cleaner, but is arguably a good idea anyway.
18084
18085commit 010c3acbb3a6993d39274f42d88c00849acb0fb0
18086Author: Alan Coopersmith <alan.coopersmith@sun.com>
18087Date:   Sun Feb 12 18:19:17 2006 +0000
18088
18089    Bug #5628 <https://bugs.freedesktop.org/show_bug.cgi?id=5628> Shadow pages not created correctly when MANDIR & MANSUFFIX don't match.
18090
18091commit fe8c01c80263457f01b70dc1511b2bd9466b7c96
18092Author: Alan Coopersmith <alan.coopersmith@sun.com>
18093Date:   Fri Feb 3 23:34:43 2006 +0000
18094
18095    Fix typo in .TH line
18096
18097commit b091c217f3c3f60dde78b09a95c150df6c83d7ba
18098Author: Kevin E Martin <kem@kem.org>
18099Date:   Thu Dec 15 00:24:28 2005 +0000
18100
18101    Update package version number for final X11R7 release candidate.
18102
18103commit e7c04e0e65a2a0c70c6ad29ec2d6f4350fd81c2a
18104Author: Kevin E Martin <kem@kem.org>
18105Date:   Tue Dec 6 22:48:41 2005 +0000
18106
18107    Change *man_SOURCES ==> *man_PRE to fix autotools warnings.
18108
18109commit b5c495854d5270e64e6d588388ffa906bfcaac22
18110Author: Kevin E Martin <kem@kem.org>
18111Date:   Sat Dec 3 05:49:42 2005 +0000
18112
18113    Update package version number for X11R7 RC3 release.
18114
18115commit 9e96dbc343c7f27ff47607acd75378ab23903e2a
18116Author: Kevin E Martin <kem@kem.org>
18117Date:   Sat Dec 3 04:41:47 2005 +0000
18118
18119    Add check and cflags for malloc(0) returning NULL.
18120
18121commit 649c37b47909620ccafde3e983de8321cddd74ce
18122Author: Alan Coopersmith <alan.coopersmith@sun.com>
18123Date:   Mon Nov 28 22:03:04 2005 +0000
18124
18125    Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir)
18126
18127commit 8d1500df66d796ebff2b0d8c02205e5fa6796d4a
18128Author: Alan Coopersmith <alan.coopersmith@sun.com>
18129Date:   Wed Nov 23 22:33:06 2005 +0000
18130
18131    Bug #5003 <https://bugs.freedesktop.org/show_bug.cgi?id=5003> Patch #3763 <https://bugs.freedesktop.org/attachment.cgi?id=3763> Xorg code misuses S_IF* macros
18132
18133commit 93cf3747f9ae8d30bd485b41c5ff10397f68f078
18134Author: Eric Anholt <anholt@freebsd.org>
18135Date:   Sun Nov 20 23:17:39 2005 +0000
18136
18137    Add/improve libs .cvsignores.
18138
18139commit fa1f4a08112bfa14d3758f4702733dd3892966c3
18140Author: Kevin E Martin <kem@kem.org>
18141Date:   Sat Nov 19 07:15:39 2005 +0000
18142
18143    Update pkgconfig files to separate library build-time dependencies from application build-time dependencies, and update package deps to work with separate build roots.
18144
18145commit 7012f9b56c594cf40855ba0dbf93af1263417a8c
18146Author: Kevin E Martin <kem@kem.org>
18147Date:   Mon Nov 14 21:51:07 2005 +0000
18148
18149    Fix xf86bigfont pkgconfig dep typo.
18150
18151commit 60a72f47951f46b4e0505d9903a94af3b6bed8ca
18152Author: Kevin E Martin <kem@kem.org>
18153Date:   Wed Nov 9 21:19:12 2005 +0000
18154
18155    Update package version number for X11R7 RC2 release.
18156
18157commit 6bb0c3796b6e1beddc376a896e865704886e1462
18158Author: Kean Johnson <kean@armory.com>
18159Date:   Tue Nov 8 06:33:25 2005 +0000
18160
18161    See ChangeLog entry 2005-11-07 for details.
18162
18163commit 14be0098ad90c3e68bd2d21b00ffabb76f1fd780
18164Author: Kevin E Martin <kem@kem.org>
18165Date:   Tue Nov 1 15:11:50 2005 +0000
18166
18167    Update pkgcheck dependencies to work with separate build roots.
18168
18169commit 1bf71462a972e0fc56de63f5c7fd613b37fc70f1
18170Author: Donnie Berkholz <spyderous@gentoo.org>
18171Date:   Fri Oct 28 10:44:03 2005 +0000
18172
18173    Revert that, it's redundant. But it is worth noting that --disable-xkb is broken, if anyone cares.
18174
18175commit be627a39fe373e2e81fdc263780e70a271d9e0c5
18176Author: Donnie Berkholz <spyderous@gentoo.org>
18177Date:   Fri Oct 28 08:28:08 2005 +0000
18178
18179    Add dependency on inputproto for XI.h.
18180
18181commit 5fae4cb456cb03fb70cd065dbc2ca94c8ed99082
18182Author: Adam Jackson <ajax@nwnk.net>
18183Date:   Fri Oct 21 18:44:24 2005 +0000
18184
18185    Bug #4736: Error decoding for Damage extension. (Jonathan Lennox)
18186
18187commit 1171fa9dc77413f0e90933a565ec07068052afb4
18188Author: Kevin E Martin <kem@kem.org>
18189Date:   Wed Oct 19 02:48:08 2005 +0000
18190
18191    Update package version number for RC1 release.
18192
18193commit 2a2d905706308b9d5a1c16af1067fb390f43850c
18194Author: Alan Coopersmith <alan.coopersmith@sun.com>
18195Date:   Tue Oct 18 00:00:08 2005 +0000
18196
18197    Use @LIB_MAN_SUFFIX@ instead of $(LIB_MAN_SUFFIX) in macro substitutions to work better with BSD make
18198
18199commit 66d35b6971fb26762392a2a8e2c47db46c11116a
18200Author: Alan Coopersmith <alan.coopersmith@sun.com>
18201Date:   Mon Oct 17 21:13:15 2005 +0000
18202
18203    Rename .shadows.DONE to shadows.DONE to avoid some make's thinking it's a suffix rule (reported by Matthieu Herrb)
18204
18205commit a316995a17c084e98ef1b7f25d287c2c08b6d749
18206Author: Donnie Berkholz <spyderous@gentoo.org>
18207Date:   Sun Oct 16 03:03:35 2005 +0000
18208
18209    Change '==' to portable '='.
18210
18211commit b76a072530e884bcbea6ed8fed5aef39361dcfc0
18212Author: Alan Coopersmith <alan.coopersmith@sun.com>
18213Date:   Wed Oct 12 00:04:50 2005 +0000
18214
18215    configure.ac Use XORG_MAN_SECTIONS instead of custom man section configuration. Add shadow man pages for man pages that document multiple functions.
18216
18217commit 8ee5c1429af98206e05a0536f87c0f21a529cdf0
18218Author: Eric Anholt <anholt@freebsd.org>
18219Date:   Tue Oct 11 02:18:36 2005 +0000
18220
18221    Add appropriate pthread libs/flags for FreeBSD, fixing the build of ico and probably others.
18222
18223commit 41ff3b9d1f194a7b56437b650d5f589225c078c6
18224Author: Alan Coopersmith <alan.coopersmith@sun.com>
18225Date:   Sun Oct 9 22:28:39 2005 +0000
18226
18227    Bug #3021 <https://bugs.freedesktop.org/show_bug.cgi?id=3021> Requests and Errors for XFixes are not in XErrorDB (Jonathan Lennox)
18228
18229commit 91ed79852e790049ab54e68f288afb3c953194c9
18230Author: Kevin E Martin <kem@kem.org>
18231Date:   Fri Oct 7 15:00:00 2005 +0000
18232
18233    Clean up generated files
18234    Add missing dist tarball files
18235
18236commit 440399b470c97b159a530602fff11c315aca8d97
18237Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18238Date:   Wed Oct 5 18:27:10 2005 +0000
18239
18240    Add el_GR.UTF-8 compose file
18241
18242commit 010f0647e25ac617d0f92c8d2b8dda684da545db
18243Author: Daniel Stone <daniel@fooishbar.org>
18244Date:   Fri Sep 30 07:52:46 2005 +0000
18245
18246    Bug #2609: Add Kyrgyz locale (Ilyas Bakirov).
18247
18248commit 3ef2fb67bd8905b208ad7eb790c3843e14cea7ed
18249Author: Daniel Stone <daniel@fooishbar.org>
18250Date:   Fri Sep 30 07:47:55 2005 +0000
18251
18252    Bug #1640: Kinyarwanda locale support. (Steve Murphy)
18253
18254commit 4ae0decabe0960870df0ec165f495166c10a053c
18255Author: Daniel Stone <daniel@fooishbar.org>
18256Date:   Fri Sep 30 07:40:03 2005 +0000
18257
18258    Bug #2268: Add South African locales (Dwayne Bailey). Some whitespace cleanups, as the parser is a little touchy.
18259
18260commit 3f79eb4c99844f618f1889741d1631c2ffe5385f
18261Author: Daniel Stone <daniel@fooishbar.org>
18262Date:   Fri Sep 30 07:11:19 2005 +0000
18263
18264    Generate locale.alias and compose.dir exactly like we do in the monolith, which fixes non-UTF-8 locales in particular (so we get foo: bar, as well as foo bar). Switch to generating locale.dir the same way.
18265
18266commit d1237d1483ff972c76a0ac344ec97d5280db0007
18267Author: Alan Coopersmith <alan.coopersmith@sun.com>
18268Date:   Thu Sep 29 21:27:12 2005 +0000
18269
18270    Add Xcms.txt to lib/X11
18271
18272commit 4a86f299693f7376cbe98175f0b0c44d691802b3
18273Author: Alan Coopersmith <alan.coopersmith@sun.com>
18274Date:   Sat Sep 24 20:11:06 2005 +0000
18275
18276    Add XQueryExtension.man
18277
18278commit a057a66e2041d45198a13a4ece7c07068f76f21b
18279Author: Alan Coopersmith <alan.coopersmith@sun.com>
18280Date:   Sat Sep 24 00:16:32 2005 +0000
18281
18282    Add XTHREAD_CFLAGS for platforms that need special defines like
18283    -D_REENTRANT or -D_POSIX_whatever to get re-entrant function definitions. Set XDMCP_LIBS correctly for later libXdmcp tests.
18284
18285commit 3e920a65a7c376ad63eae2240fd06904d25d18bf
18286Author: Alan Coopersmith <alan.coopersmith@sun.com>
18287Date:   Fri Sep 2 23:00:30 2005 +0000
18288
18289    Issue an error if XTRANS macros were not found when generating configure from configure.ac to flag errors early, instead of when people wonder why libX11 can't talk to an Xserver because it has no transports defined
18290
18291commit 3eb9f2d693af89d04e2fd92492c8205dce332c9c
18292Author: Kristian Høgsberg <krh@redhat.com>
18293Date:   Thu Sep 1 19:24:13 2005 +0000
18294
18295    Use $(X11_LOCALEDATADIR) instead of @X11_LOCALEDATADIR@ so this install destination can be overridden at make install time.
18296
18297commit 7afa64325183b78d2d6a4862821f8b3e9866105c
18298Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18299Date:   Sun Aug 28 19:45:48 2005 +0000
18300
18301    OpenBSD needs -lpthread for threaded applications too.
18302
18303commit 2b2f3d3877cb7927f196d01a5df6a27bf8d0518a
18304Author: Daniel Stone <daniel@fooishbar.org>
18305Date:   Fri Aug 26 05:16:46 2005 +0000
18306
18307    Define ERRORDB/KEYSYMDB to XERRORDB/XKEYSYMDB if the former is undefined but the latter is. (Gerte Hoogewerf)
18308
18309commit 07066da0902df91c71f2adb81d1a17ec29165553
18310Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18311Date:   Sun Aug 21 15:45:04 2005 +0000
18312
18313    update
18314
18315commit de44d8b111f57bd2f015e085fd8298c5f2a15ef3
18316Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18317Date:   Sun Aug 21 15:38:39 2005 +0000
18318
18319    Threads support for BSD systems:
18320    - need to check for gewtpwuid_r to define mtsafeapi
18321    - build UIThrstubs if needed.
18322
18323commit 60217fdb918bafb2082519efe5cba3b13ad3082a
18324Author: Adam Jackson <ajax@nwnk.net>
18325Date:   Wed Aug 17 19:46:08 2005 +0000
18326
18327    Add xthreadlib variable to x11.pc. Bump to 0.99.1.
18328
18329commit e1f4c6f5e36c1511f66fa1fac76520fd97eecbad
18330Author: Alan Coopersmith <alan.coopersmith@sun.com>
18331Date:   Wed Aug 17 01:27:08 2005 +0000
18332
18333    Fix more broken multi-line .ds macros. Remove extraneous ;'s .
18334
18335commit 1909786f4a7d686369edcfc05a938df115fab37c
18336Author: Alan Coopersmith <alan.coopersmith@sun.com>
18337Date:   Tue Aug 16 19:23:15 2005 +0000
18338
18339    Bugzilla #4112 <https://bugs.freedesktop.org/show_bug.cgi?id=4112> Patch #2687 <https://bugs.freedesktop.org/attachment.cgi?id=2897> Fix multi-line macros in XPutImage man page. (Debian bug #323210, fix by David Mart?nez Moreno)
18340
18341commit 83406d69c62070d2eeef23eb47f1ca887f711ee5
18342Author: Alan Coopersmith <alan.coopersmith@sun.com>
18343Date:   Mon Aug 15 19:53:37 2005 +0000
18344
18345    Move RAWCPP macro to xorg-macros.m4 so other modules can use it Add check for whether or not RAWCPP needs -traditional instead of hardcoding it, so non-gcc cpp's can be used
18346
18347commit 5bb43de17de8e71d967488a713bf2b3448533444
18348Author: Alan Coopersmith <alan.coopersmith@sun.com>
18349Date:   Sat Aug 6 18:59:49 2005 +0000
18350
18351    Typo fix in output message
18352
18353commit afe34b95862bb3c06cdbe724cb5ec3001a4a5215
18354Author: Alan Coopersmith <alan.coopersmith@sun.com>
18355Date:   Thu Aug 4 02:55:49 2005 +0000
18356
18357    //bugs.freedesktop.org/show_bug.cgi?id=1887> Patch #3005 <https://bugs.freedesktop.org/attachment.cgi?id=3005> libX11 locale defs severely adrift from glibc: adding new aliases (From Debian via Nathanael Nerode)
18358
18359commit 3979a0b88edf6475ce5cfaa386e18ef980bda13c
18360Author: Alan Coopersmith <alan.coopersmith@sun.com>
18361Date:   Thu Aug 4 02:51:30 2005 +0000
18362
18363    //bugs.freedesktop.org/show_bug.cgi?id=1887> Patch #3002 <https://bugs.freedesktop.org/attachment.cgi?id=3002> libX11 locale defs severely adrift from glibc: locales bugfix for bs_BA (From Debian via Nathanael Nerode)
18364
18365commit 34b454df192f4563499c453ccdb8c079f4a20cbe
18366Author: Alan Coopersmith <alan.coopersmith@sun.com>
18367Date:   Sat Jul 30 20:30:46 2005 +0000
18368
18369    Include config.h so Xtrans knows which transport types to build code for
18370
18371commit cd9c9936b49c125eda779b99887d7e6ae4cf56cd
18372Author: Alan Coopersmith <alan.coopersmith@sun.com>
18373Date:   Sat Jul 30 19:15:16 2005 +0000
18374
18375    Add -D flags to clear various warnings (Stefan Dirsch)
18376
18377commit e7fef67b4531faddd805d8f2157903006d3117ed
18378Author: Kevin E Martin <kem@kem.org>
18379Date:   Fri Jul 29 21:22:50 2005 +0000
18380
18381    Various changes preparing packages for RC0:
18382    - Verify and update package version numbers as needed
18383    - Implement versioning scheme
18384    - Change bug address to point to bugzilla bug entry form
18385    - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to reenable it)
18386    - Fix makedepend to use pkgconfig and pass distcheck
18387    - Update build script to build macros first
18388    - Update modular Xorg version
18389
18390commit 2ebb00244928237088e68325b1032b3550455ce9
18391Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18392Date:   Sat Jul 23 20:19:31 2005 +0000
18393
18394    remove orphan TAB at begin of line
18395
18396commit dd7a9cdecda73e024ca84c5b9a22b18688038d94
18397Author: Kevin E Martin <kem@kem.org>
18398Date:   Sat Jul 23 18:09:39 2005 +0000
18399
18400    Modify modular libs to use Xregion.h instead of region.h
18401
18402commit 6d635a88d91647b1b63611c3591f74916f88cd1c
18403Author: Kevin E Martin <kem@kem.org>
18404Date:   Sat Jul 23 18:06:16 2005 +0000
18405
18406    lib/Xrender/Picture.c Change region.h to Xregion.h and modify internal references to include <X11/Xregion.h>.
18407
18408commit 36283f50fd9748733ae84cb7fb52ca8d9e661c15
18409Author: Daniel Stone <daniel@fooishbar.org>
18410Date:   Sat Jul 16 06:25:35 2005 +0000
18411
18412    Set soversion to 6.2.0.
18413
18414commit 75fd5ae6e4683b9b9dcc13bc2f0faf223610a74f
18415Author: Keith Packard <keithp@keithp.com>
18416Date:   Fri Jul 15 04:27:32 2005 +0000
18417
18418    Add missing Makefile.am to lib/X11/modules
18419
18420commit 6d84a8b1329af1fcfe86b198f1a6e7dd6ff616c3
18421Author: Keith Packard <keithp@keithp.com>
18422Date:   Fri Jul 15 04:08:51 2005 +0000
18423
18424    Move i18n modules to top-level so they can be built in the right order (before xlib for non-loadable, after xlib for loadable).
18425    Link i18n modules against xlib to resolve Xlib symbols used by them.
18426
18427commit 45f40126a73295345bb5eb187b1167874842ab6e
18428Author: Alexander Gottwald <ago@freedesktop.org>
18429Date:   Thu Jul 14 19:50:00 2005 +0000
18430
18431    Add $(top_srcdir)/src to include list
18432
18433commit 0aed7d91f5928d09d541617aad03709b5090658d
18434Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18435Date:   Thu Jul 14 17:04:49 2005 +0000
18436
18437    Build fix for non-GNU make.
18438
18439commit b79422ccb02ab44548d1038956ab0cd4e2638645
18440Author: Adam Jackson <ajax@nwnk.net>
18441Date:   Thu Jul 14 15:12:44 2005 +0000
18442
18443    typo fixes (Matthieu Herrb)
18444
18445commit 419304cde2fda19457c667870edefc0b227651b3
18446Author: Alan Coopersmith <alan.coopersmith@sun.com>
18447Date:   Wed Jul 13 02:41:36 2005 +0000
18448
18449    Add missing backslashes to xlocale_la_SOURCES
18450
18451commit d14cc5c8964c4539b57c6cb51ef653292e410b79
18452Author: Keith Packard <keithp@keithp.com>
18453Date:   Mon Jul 11 20:32:55 2005 +0000
18454
18455    Ammend AM_CFLAGS with all needed includes; cannot figure out an easy way to automate this.
18456
18457commit 424c2d8905eb2ad0a2df15b1da2f96140bfbcffb
18458Author: Lars Knoll <lars@trolltech.com>
18459Date:   Mon Jul 11 15:24:32 2005 +0000
18460
18461    compile
18462
18463commit de9784eb1bde46efc316da279e3da27c6cc288a7
18464Author: Keith Packard <keithp@keithp.com>
18465Date:   Mon Jul 11 09:26:40 2005 +0000
18466
18467    Ignore built man page files
18468
18469commit 0c258c36d1523113790c599b16d2947d7aa6469a
18470Author: Keith Packard <keithp@keithp.com>
18471Date:   Mon Jul 11 09:18:31 2005 +0000
18472
18473    Minor changes to help modular Xlib build i18n modules
18474
18475commit 6e752ea1203b786423e40f43340bce15ca3de0f0
18476Author: Keith Packard <keithp@keithp.com>
18477Date:   Mon Jul 11 08:29:18 2005 +0000
18478
18479    Enable loadable i18n modules, making them configurable on the configure command line.
18480    Clean up conditionals for XKB, XCMS, XLOCALEDIR
18481    Create new lib directory for locale modules in ${X11_LIBDIR}/locale/lib. Add this to the default XLOCALEDIR search path.
18482    Create separate X11_LOCALEDATADIR variable pointing at ${datadir}/X11/locale for installing locale data.
18483    Split out xcms, xkb, xlibi18n sources from main xlib bits so they can be conditionally included more easily. Lots of source files have been moved with this step; the result seems like it might be easier to maintain.
18484    Display message at end of configure script with selected options.
18485    Fix manual building with cpp to add -traditional in cpprules.in. This isn't conditionalized at all, so it will break on systems not using GNU cpp.
18486
18487commit b46cf0d879f1dbf92dcf5a0305d18986c766ed84
18488Author: Daniel Stone <daniel@fooishbar.org>
18489Date:   Sun Jul 10 22:37:33 2005 +0000
18490
18491    Fix segfault when _XimProtoCreateIC() fails to create a context; Debian #239991. (Chung-chieh Shan)
18492
18493commit 6f2132b18e61ca9755e4b45550f3f5097dd1fbb7
18494Author: Alan Coopersmith <alan.coopersmith@sun.com>
18495Date:   Sat Jul 9 20:06:04 2005 +0000
18496
18497    Set __libmansuffix__ & __xorgversion__ correctly when cpp processing man pages
18498
18499commit 3939ac4410446b46071c9d714f4270c12bf904fe
18500Author: Alan Coopersmith <alan.coopersmith@sun.com>
18501Date:   Sat Jul 9 18:44:14 2005 +0000
18502
18503    - Since all but one line of all the nls/*/Makefile.am files are identical, move common bits to nls/localerules.in for easier updating and use automake includes to include in all the nls/*/Makefile.am files
18504    - Don't assume $(CPP) can take gcc-only -traditional flag
18505    - CPP process man pages as is done in the monolithic tree
18506
18507commit 19ba9d0df86c688319377467254b9ea9c4b0eccc
18508Author: Keith Packard <keithp@keithp.com>
18509Date:   Sat Jul 9 06:01:49 2005 +0000
18510
18511    Clean up .cvsignore files
18512
18513commit f4bba6fa9b9ce637be7662754750054567e9250a
18514Author: Adam Jackson <ajax@nwnk.net>
18515Date:   Sat Jul 9 02:53:29 2005 +0000
18516
18517    typo fix. i suck.
18518
18519commit 3305da61a61695e24c1ea11d3f59dddb52873d47
18520Author: Adam Jackson <ajax@nwnk.net>
18521Date:   Sat Jul 9 02:41:18 2005 +0000
18522
18523    Check for bigfont proto headers via pkgconfig (Arwed von Merkatz)
18524
18525commit afdae2e8ccb93de2987b5a5e850695af15ece7e0
18526Author: Keith Packard <keithp@keithp.com>
18527Date:   Fri Jul 8 07:11:56 2005 +0000
18528
18529    Create and use XlibConf.h to match modular build which needs XTHREADS defined when building extensions
18530
18531commit 86fa88dc594ca2093030acf5c84973ee2b33eb4c
18532Author: Keith Packard <keithp@keithp.com>
18533Date:   Fri Jul 8 06:57:06 2005 +0000
18534
18535    Create and install XlibConf.h to provide external users of Xlibint.h the defines necessary to correctly use the library. Xlibint.h should #include this new file.
18536    Add a bunch of .cvsignore files
18537
18538commit 30f6ffedeaf899e699f0c6b14c9471ce9bcc430c
18539Author: Chris Less <clee@c133.org>
18540Date:   Mon Jul 4 23:01:48 2005 +0000
18541
18542    Fixing bug #380 - add a man page for XQueryExtension, XListExtensions, and XFreeExtensionList.
18543
18544commit 55328bb6caabd37885cd19d7a5821ed425daf321
18545Author: Eric Anholt <anholt@freebsd.org>
18546Date:   Sun Jul 3 10:17:38 2005 +0000
18547
18548    Avoid a gmake-specific construct, and instead just write the name of the current directory into each Makefile.am.
18549
18550commit 726538ba21b631bfb0b8094a5546acdaf57379ff
18551Author: Daniel Stone <daniel@fooishbar.org>
18552Date:   Sun Jul 3 07:37:33 2005 +0000
18553
18554    Fix more include paths; add dix-config.h to XKB code.
18555
18556commit 8c7677138e70e14eca0255f2168583f0ecc21994
18557Author: Daniel Stone <daniel@fooishbar.org>
18558Date:   Sun Jul 3 07:00:55 2005 +0000
18559
18560    Add Xtrans definitions (FONT_t, TRANS_CLIENT) to clean up warnings.
18561    Add XSERV_t, TRANS_SERVER, TRANS_REOPEN to quash warnings.
18562    Add #include <dix-config.h> or <xorg-config.h>, as appropriate, to all source files in the xserver/xorg tree, predicated on defines of HAVE_{DIX,XORG}_CONFIG_H. Change all Xfont includes to <X11/fonts/foo.h>.
18563
18564commit 197697c92a63091a4cc3cc04dcb7fa29d2655758
18565Author: Daniel Stone <daniel@fooishbar.org>
18566Date:   Fri Jul 1 22:13:35 2005 +0000
18567
18568    Fix objdir != srcdir, as well as make distcheck.
18569    Don't attempt to create Compose.pre files; formatting fixes.
18570    Added if not already present.
18571
18572commit c162d60ad8f124563f94a2a266de59373936266c
18573Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18574Date:   Wed Jun 22 22:46:31 2005 +0000
18575
18576    Apply these patches from Theo van Klaveren:
18577    lib-dmx.patch lib-FS.patch lib-X11.patch lib-XRes.patch lib-XScrnSaver.patch lib-xtrans.patch
18578    to make the libraries distcheck.
18579
18580commit 9ee8abdab03ea605a6327118ab7dacab6adf8876
18581Author: Alan Coopersmith <alan.coopersmith@sun.com>
18582Date:   Sat Jun 18 07:48:43 2005 +0000
18583
18584    Move Secure RPC flags from X11/configure.ac to xtrans/xtrans.m4 since multiple modules will need them
18585
18586commit 12afc57b7d455781eee305e9ed6a899ceec8729e
18587Author: Daniel Stone <daniel@fooishbar.org>
18588Date:   Wed Jun 15 16:50:47 2005 +0000
18589
18590    Typo fix to locale/error/keysym location declarations.
18591
18592commit 9a895777e30762f61b98d25be3b5d8b1169baa17
18593Author: Daniel Stone <daniel@fooishbar.org>
18594Date:   Wed Jun 15 13:37:43 2005 +0000
18595
18596    Typo in ImUtil.h commit -- I AM CAPTAIN SKILL.
18597
18598commit add49285663684875ab7a5c58ec7a2cf8b775f67
18599Author: Daniel Stone <daniel@fooishbar.org>
18600Date:   Wed Jun 15 13:32:35 2005 +0000
18601
18602    Define locations for XErrorDB, XKeysymDB, and locale data in configure.ac. Add AC_DEFINE_DIR macro from autoconf-archive.cryp.to towards this end.
18603    Move ImUtil.h from src/ to include/X11/.
18604
18605commit 845dfc6b42b950890866ee4df27761e086f50dca
18606Author: Daniel Stone <daniel@fooishbar.org>
18607Date:   Wed Jun 15 13:27:48 2005 +0000
18608
18609    Move ImUtil.h from src/ to include/X11/. Additionally, copy Cmap.h as a distribution file.
18610
18611commit bba117f0d98f62cfb060d0fab97b407a3a0bfda9
18612Author: Daniel Stone <daniel@fooishbar.org>
18613Date:   Fri Jun 10 14:11:36 2005 +0000
18614
18615    Remove pointless include of Xlib.h.
18616    Fix #include path to bigreqstr.h.
18617
18618commit 1a0de49da1274882bab05b0f7240936b37955e5c
18619Author: Alexander Gottwald <ago@freedesktop.org>
18620Date:   Thu Jun 9 21:30:15 2005 +0000
18621
18622    Use $(srcdir) for Compose.pre and XLC_LOCALE.pre
18623
18624commit af4f0f302644ebfbb0ca9f4016a4aee85c973d37
18625Author: Alexander Gottwald <ago@freedesktop.org>
18626Date:   Thu Jun 9 15:55:33 2005 +0000
18627
18628    Replace <X11/transport.c> with <X11/Xtrans/transport.c>
18629
18630commit fd5f58e0baf692e34b9b622286f18762cc2500d3
18631Author: Alexander Gottwald <ago@freedesktop.org>
18632Date:   Thu Jun 9 15:52:02 2005 +0000
18633
18634    Replace <X11/Xtrans.h> with <X11/Xtrans/Xtrans.h>
18635    Copy Xtrans.h to exports/include/X11/Xtrans only
18636
18637commit 44538f9940f969d46c0e5e4b201c684cde2ba611
18638Author: Alan Coopersmith <alan.coopersmith@sun.com>
18639Date:   Sun Jun 5 03:29:33 2005 +0000
18640
18641    Port Imake flags to autoconf tests & --enable-* flags: HASSETUGID, HASGETRESUID, NO_XLOCALEDIR, HAS_SHM and XF86BIGFONT
18642
18643commit 588e30e9ec65fa6205a34be650b79d5e2243edec
18644Author: Alan Coopersmith <alan.coopersmith@sun.com>
18645Date:   Sat Jun 4 22:53:21 2005 +0000
18646
18647    Add --enable-secure-rpc flag and checks for needed functions for Secure RPC ("SUN-DES-1") authentication method
18648
18649commit a547afee2ef49cc41bbb67f9cff5a52a283c0854
18650Author: Alan Coopersmith <alan.coopersmith@sun.com>
18651Date:   Sat Jun 4 21:20:20 2005 +0000
18652
18653    Bug #3436 <https://bugs.freedesktop.org/show_bug.cgi?id=3436> Conndis.c uses incorrect type for args to XdmcpWrap() (Mike Harris - mharris@www.linux.org.uk)
18654
18655commit ca93c761559ae464189c7ea7cf11c6a6679f2431
18656Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18657Date:   Sat May 28 01:02:32 2005 +0000
18658
18659    Don't use $< in explicit rules. This only works with GNU make. <https://bugs.freedesktop.org/show_bug.cgi?id=3383>
18660
18661commit 09ebb349359e3dd9131fa2fa8b07559faa173654
18662Author: Alan Coopersmith <alan.coopersmith@sun.com>
18663Date:   Sun May 22 19:05:11 2005 +0000
18664
18665    Convert man pages to long file names in lib/X11, lib/Xt, & lib/Xext
18666
18667commit 761219b1ef9befc350c8a35b6f96d047e5f008cc
18668Author: Alan Coopersmith <alan.coopersmith@sun.com>
18669Date:   Sat May 21 23:07:48 2005 +0000
18670
18671    xtrans:
18672    Create autoconf macro XTRANS_CONNECTION_FLAGS to provide standard set of --enable flags for which transports to support and common place to update for required libraries for platforms that need certain libs for certain transports
18673    ICE:
18674    Add ICE_t #define required by Xtrans headers. Replace static defines of LOCALCONN & UNIXCONN with new XTRANS_CONNECTION_FLAGS macro.
18675    X11:
18676    Moved transport type checks to new macro XTRANS_CONNECTION_FLAGS in xtrans.m4 in xtrans module so they can be shared by all modules using xtrans.
18677
18678commit 4b1ba6eb975b547b808f4d5c8825a3261de5e02a
18679Author: Alan Coopersmith <alan.coopersmith@sun.com>
18680Date:   Sat May 21 04:26:12 2005 +0000
18681
18682    Quote $ac_cv_search_* variables to prevent errors from test when they are set to "none required" (as happens on Solaris since dlopen is in libc)
18683    Comment out "override CC = gcc" line as it breaks builds with non-GNU makes, and its incorrect to force a specific compiler. Change LINK line from gcc to $(CC).
18684
18685commit 1d425d5e2092dd18a7dd599b37ed9af61cf59819
18686Author: Adam Jackson <ajax@nwnk.net>
18687Date:   Thu May 19 00:22:32 2005 +0000
18688
18689    revert last change, didn't do right thing at all, sorry for the noise
18690
18691commit 1b0c46c1ae61d751dd3ca96de8e2c3fe21c5f4f2
18692Author: Adam Jackson <ajax@nwnk.net>
18693Date:   Thu May 19 00:10:07 2005 +0000
18694
18695    Require automake 1.7 in AM_INIT_AUTOMAKE
18696
18697commit cd4657c175dbab6aaca36f18a0ca92c95b5567dd
18698Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18699Date:   Tue May 17 22:32:09 2005 +0000
18700
18701    - Check for xproto as its CFLAGS are needed in the .pc file
18702
18703commit 46e8d8a65430dd87c10b066b5cff99a689c22241
18704Author: Egbert Eich <eich-at-freedesktop-dot-org>
18705Date:   Tue May 17 08:10:10 2005 +0000
18706
18707    gcc4 allows to check if sentinels are correct (a sentinel is the terminating element in a varargs list). A sentinel needs to be NULL, not 0 - which doesn't make a difference on 32bit but matters on 64bit. Furthermore it can be told that functions have a printf-like format string and argument list so that they can verify that both match. To use these features certain attributes need to be set - which are compiler specific. To do this we define macros which are expanded depending on the compiler version. For now we put those in include/Xfuncproto.h (the XFree86 DDX layer contains a file compiler.h which however is not visible outside the DDX) (Bugzilla #3268).
18708
18709commit 23198d2bfbf0049b2630235cd4d4a4ffba7ec6c1
18710Author:  <ssp@aware-of-vacuity.boston.redhat.com>
18711Date:   Mon May 16 22:35:27 2005 +0000
18712
18713    Make Xdmcp unconditionally required, require various protocol modules.
18714    Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com>
18715    Check for kbproto if using XKB.
18716    Require xextproto rather than xextensions
18717    Remove the entries from the xlibs tree, as they are not relevant here.
18718
18719commit 8bd3aea84ce54b8b76a898f3ae00e2b499c14a5e
18720Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18721Date:   Mon May 16 21:48:36 2005 +0000
18722
18723    Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com>
18724    Check for kbproto if using XKB.
18725
18726commit 9b1fa9ca3ed852ed40860f137511683f4bdae6fa
18727Author:  <ssp@aware-of-vacuity.boston.redhat.com>
18728Date:   Mon May 16 21:27:35 2005 +0000
18729
18730    Require xextproto rather than xextensions
18731    Remove the entries from the xlibs tree, as they are not relevant here.
18732
18733commit 7eee605e3aeed549d1053325a03027c5e8cbf71b
18734Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18735Date:   Fri May 13 22:53:36 2005 +0000
18736
18737    - For now put xtrans in X11/Xtrans/X11, since libX11 is looking for it in <X11/...>
18738    - For Xcomposite and Xdamage, don't link the build system out of the xc tree
18739    - Link the public X11 headers into their own directory
18740    - Add links to XKeysymDB and XErrorDB
18741    - Add links to all the Xlib man pages
18742    - Add links to the lcUniConv subdirectory
18743    - Conditionally include config.h in Xlib source
18744
18745commit 6769ccda88caf27d1441d335ef2b318a047a612b
18746Author: Alan Coopersmith <alan.coopersmith@sun.com>
18747Date:   Mon Mar 21 04:58:21 2005 +0000
18748
18749    xc/lib/X11/ErrDes.c
18750    //bugs.freedesktop.org/show_bug.cgi?id=132) Patch #2168 (https://bugs.freedesktop.org/attachment.cgi?id=2168) Replace a couple of BUFSIZE uses with better values to check against. Fixes by Stuart Anderson <anderson@netsweng.com>
18751
18752commit 3b9e8ece93b916c55a82df53e85f097418edf471
18753Author: Roland Mainz <roland.mainz@nrubsig.org>
18754Date:   Sat Mar 19 22:04:55 2005 +0000
18755
18756    xc/nls/Compose/iso8859-2
18757    xc/nls/Compose/iso8859-3
18758    xc/nls/Compose/iso8859-9
18759    //bugs.freedesktop.org/show_bug.cgi?id=2592) attachment #2156 (https://bugs.freedesktop.org/attachment.cgi?id=2156) Fix a couple of typos in ISO8859-* Compose files ("asciicircum" instead "of asciicirum"). Patch by Matthias Hopf <mhopf@suse.de>.
18760
18761commit 0ce5950a08b9ab23ca8a32effdd40c421e92df84
18762Author: Alan Coopersmith <alan.coopersmith@sun.com>
18763Date:   Tue Mar 8 02:53:36 2005 +0000
18764
18765    Bugzilla Bug 2006 (https://bugs.freedesktop.org/show_bug.cgi?id=2006) Patch #2031 (https://bugs.freedesktop.org/attachment.cgi?id=2031) XEmbed client doesn't receive key events from XIM: Use | to set a bit, not &. Patch by Hidetoshi Tajima <hidetoshi.tajima@sun.com>.
18766
18767commit df341cd2f2f263f13323e8c0936ea8aa0d7fbba1
18768Author: Roland Mainz <roland.mainz@nrubsig.org>
18769Date:   Tue Feb 1 03:12:28 2005 +0000
18770
18771    xc/nls/Compose/pt_BR.UTF-8
18772    //bugs.freedesktop.org/show_bug.cgi?id=2400) attachment #1762 (https://bugs.freedesktop.org/attachment.cgi?id=1762): Fix build bustage caused by broken patch for brazillian locale support (see Bugzilla #1896). Patch by Kevin E. Martin <kem@freedesktop.org>.
18773
18774commit 483d3973f56dda51f9251d3ea808aaf7f3b48c10
18775Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18776Date:   Fri Jan 28 18:31:31 2005 +0000
18777
18778    cleaned up boundary-case handling for Uxxxx Unicode keysym notation
18779
18780commit f234188a4c1e6b655aef0e3957ccad20d4c5847f
18781Author: Roland Mainz <roland.mainz@nrubsig.org>
18782Date:   Wed Jan 19 01:53:55 2005 +0000
18783
18784    xc/nls/compose.dir
18785    xc/nls/locale.dir
18786    xc/nls/Compose/Imakefile
18787    xc/nls/Compose/pt_BR.UTF-8
18788    xc/nls/XI18N_OBJS/Imakefile
18789    xc/nls/XI18N_OBJS/pt_BR.UTF-8
18790    xc/nls/XLC_LOCALE/Imakefile
18791    xc/nls/XLC_LOCALE/pt_BR.UTF-8
18792    //bugs.freedesktop.org/show_bug.cgi?id=1896) attachment #1675 (https://bugs.freedesktop.org/attachment.cgi?id=1675): Add support for pt_BR.UTF-8 locale. Patch by Gustavo Noronha Silva, Branden Robinson, Julien Lafon.
18793
18794commit 7448ea7ef425d35cfc31eb41d46f4d879774f376
18795Author: Egbert Eich <eich-at-freedesktop-dot-org>
18796Date:   Fri Jan 14 18:03:09 2005 +0000
18797
18798    Made some security enhancements:
18799    - no writing past end of buffer caused by bogus locale.
18800    - explicitely add a \0 character at end of string. (Bugzilla #2262)
18801
18802commit 5557d47fcf22a6f3adf327691158f2270e3d5094
18803Author: Roland Mainz <roland.mainz@nrubsig.org>
18804Date:   Fri Jan 14 04:59:05 2005 +0000
18805
18806    xc/nls/compose.dir
18807    xc/nls/locale.alias
18808    xc/nls/locale.dir
18809    //bugs.freedesktop.org/show_bug.cgi?id=1830) attachment #1674 (https://bugs.freedesktop.org/attachment.cgi?id=1674): Adding support for the bs_BA (bs, bs_BA, bs_BA.iso88592, bs_BA.ISO-8859-2, bs_BA.ISO_8859-2, bs_BA.UTF-8) locale. Patch by Vedran Ljubovic <vljubovic@smartnet.ba>.
18810
18811commit 1fa3737f042f798fa11a9ff5b03f3b3ba3529824
18812Author: Egbert Eich <eich-at-freedesktop-dot-org>
18813Date:   Tue Jan 11 17:37:57 2005 +0000
18814
18815    'Normalize' locale names (ie. remove any '-' and '_' and convert to lower case after the <language>_<territory> part) before matching against locale.alias (Bugzilla #2262). This needs adequate testing that we don't accidentally introduce undesirable side effects.
18816
18817commit 642cd269f94a234aa470a1d43385bd42625d89a1
18818Author: Egbert Eich <eich-at-freedesktop-dot-org>
18819Date:   Tue Dec 14 08:59:20 2004 +0000
18820
18821    Removed #ifdef'ed out code together with the comment explaining why it was #ifdef'ed out.
18822    Fixed typo.
18823    Added comment to a changed that's been committed with one of the previous commits.
18824
18825commit a07ccae36e629741d2e48de7730114d30a975a41
18826Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18827Date:   Sun Dec 12 08:42:50 2004 +0000
18828
18829    Fix missing XChangeProperty() prototype missing in synopsis section.
18830
18831commit c73adb1c8ad806c2f9f0ba32f4b449e7a871501c
18832Author: Alexander Gottwald <ago at freedesktop dot org>
18833Date:   Wed Dec 8 13:42:01 2004 +0000
18834
18835    Bugzilla #1980 (https://bugs.freedesktop.org/show_bug.cgi?id=1980) Handle XERRORDB only on WIN32 platform
18836
18837commit d5e7ab194103e22fd6c5094aaa107d90210bb600
18838Author: Jim Gettys <jg@freedesktop.org>
18839Date:   Thu Dec 2 16:18:16 2004 +0000
18840
18841    fix comment to indicate additional possible mode. i bug 1756 reported by Owen Taylor.
18842
18843commit 03940d7330cb1bbf93d49c650aefb19de457da7c
18844Author: Alexander Gottwald <ago at freedesktop dot org>
18845Date:   Wed Dec 1 13:06:55 2004 +0000
18846
18847    Bugzilla #1980, https://bugs.freedesktop.org/show_bug.cgi?id=1980 Make location of XErrorDB configurable
18848
18849commit 11a03ab908bfcfab8a3492684e8e9320e492d552
18850Author: Alexander Gottwald <ago at freedesktop dot org>
18851Date:   Wed Dec 1 12:42:17 2004 +0000
18852
18853    Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory
18854
18855commit f3d83ee153f42e8899b844377e6b842d93411e62
18856Author: Alexander Gottwald <ago@freedesktop.org>
18857Date:   Mon Nov 15 15:06:54 2004 +0000
18858
18859    Bufzilla #1802, http://freedesktop.org/bugzilla/show_bug.cgi?id=1802 Added mingw (Win32) port
18860
18861commit b798ea11911ac58a8e6e7d15a2a643b023859749
18862Author: Alexander Gottwald <ago at freedesktop dot org>
18863Date:   Mon Nov 15 13:29:56 2004 +0000
18864
18865    Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory
18866
18867commit 436108cd6c84053698e5ca629096f59b34f50c2a
18868Author: Kristian Høgsberg <krh@redhat.com>
18869Date:   Thu Nov 11 15:37:01 2004 +0000
18870
18871    Fix #1818
18872
18873commit bf2e6ef66ba55f90efa4a4ba8c8b6d3ec0d1531c
18874Author: Roland Mainz <roland.mainz@nrubsig.org>
18875Date:   Tue Nov 9 00:56:56 2004 +0000
18876
18877    xc/nls/compose.dir
18878    xc/nls/locale.alias
18879    xc/nls/locale.dir
18880    //freedesktop.org/bugzilla/show_bug.cgi?id=1544): Adding support for the si_LK (si, sinhala, si_LK, si_LK.UTF-8) locale. Patch by Anuradha Ratnaweera <gnu.slash.linux@gmail.com>.
18881
18882commit 443890ceefbd6dafe68e30d103ec4f9d316ed655
18883Author: Roland Mainz <roland.mainz@nrubsig.org>
18884Date:   Fri Nov 5 00:58:49 2004 +0000
18885
18886    xc/nls/compose.dir
18887    xc/nls/locale.alias
18888    xc/nls/locale.dir
18889    Adding some of the major indic locales (bn_IN.UTF-8, bn_IN.utf8, gu_IN.UTF-8, gu_IN.utf8, pa_IN.UTF-8, pa_IN.utf8) to X. Patch by Leon Ho <llch@redhat.com>.
18890
18891commit af7467ec734321f127b957921cce7792902b6794
18892Author: Egbert Eich <eich-at-freedesktop-dot-org>
18893Date:   Mon Oct 18 17:29:03 2004 +0000
18894
18895    Correcting font encodings for GB18030, GBK and BIG5-HKSCS. Adding nls support for those encodings (Bugzilla 1573, James Su).
18896
18897commit 1b900b59cb24fe8be3db0d434b04d462c0eeb206
18898Author: Egbert Eich <eich-at-freedesktop-dot-xorg>
18899Date:   Mon Oct 18 14:21:45 2004 +0000
18900
18901    Made handling of DevelDrivers for x86-64 more conformant to other platforms.
18902    Compress all font encodings (Stefan Dirsch).
18903    Fixed warnings.
18904    Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and Root mode (bugzilla #1580, James Su).
18905    Added another compose key combination for the Euro symbol (Stefan Dirsch).
18906    Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian).
18907    Changed default encoding for ru from KOI8-R to ISO8859-5 (Mike Fabian). This is the encoding that is also used by glibc. We may break other libcs - lets see who complains.
18908    Added explanation for DESTDIR to install to a different directory than /.
18909    Added some early bailouts to atiprobe if PCI structure pointer is NULL to prevent sig11.
18910    XV support doesn't depend on 2D acceleration any more. This patch removes this limitation from the NSC driver. This is a patch that I have committed to XFree86 a while ago but never ported over to X.Org. Matthieu Herrb contributed some build fixes.
18911    Fixing SetDPMSTimers() so that DPMS*Time == 0 disables the appropriate timer. This takes advantage of the fact that TimerSet() with a timeout argument 0 behaves like TimerCanel().
18912    Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink).
18913    Create missing links for backward compatibility to XFree86 (Stefan Dirsch).
18914    Changed comment to mention xorg.
18915    Changed cursor for the 'kill' action from XC_draped_box to XC_Pirate. If you don't like it we can change it back (original author unkown).
18916    Added 'pic' to the man page rendering command pipeline (Werner Fink).
18917    Added missing return value (Stefan Dirsch, Roland Mainz)
18918
18919commit 1ba103c3cad2329db3a31f88d7798b913affc570
18920Author: Eric Anholt <anholt@freebsd.org>
18921Date:   Wed Oct 13 07:38:00 2004 +0000
18922
18923    Use attributes instead of pragmas for weak symbols on gcc 2.95 as well. Using pragmas may result in the symbols being undefined with big -O. (FreeBSD ports/69708, Masakazu HIGAKI <higamasa@dream.com>)
18924
18925commit d80237db627bf71ad5615ff4ba05e6ab436aa740
18926Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
18927Date:   Fri Oct 8 22:57:56 2004 +0000
18928
18929    Fri Oct 8 18:53:11 2004 Soeren Sandmann <sandmann@redhat.com>
18930    Move iso10646 last so the "fallback" fonts will actually be used if they are better matches.
18931
18932commit d1f76d17ecf418396627b1a58319f2b9b255548a
18933Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18934Date:   Sun Sep 26 22:54:57 2004 +0000
18935
18936    bug fix for previous patch
18937
18938commit 48932d9b71d10e15812f47d5b842ab6aa8dac625
18939Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18940Date:   Sun Sep 26 20:46:17 2004 +0000
18941
18942    The big keysym cleanup, to bring implementation in line with the recent revision of Appendix A of the protocol spec. (Markus Kuhn)
18943
18944commit 2e02a95dcc43dd3ec7bbaf4675ffb94f5074f543
18945Author: Egbert Eich <eich@freedesktop.org>
18946Date:   Wed Sep 15 09:05:22 2004 +0000
18947
18948    Unregistering events in XSelectInput() when unregistering IM filter callbacks may be a bad idea as others may be interested in this event. Removed the call to XSelectInput() altogether as we are in root window anyway (Lubos Lunak).
18949    Fix size of a variable that gets assigned the value of SmartScheduleTime (long) to long. This should help to prevent smart scheduler lockup on 64 bit systems due to overruns (Andreas Schwab).
18950
18951commit 2d3afb68a104a80a21ee622b9abb9c95e83505d3
18952Author: Egbert Eich <eich@freedesktop.org>
18953Date:   Tue Aug 31 11:37:03 2004 +0000
18954
18955    Fixed some lockups in XIM code when the application is running with multi thread support. These lockups occur deep down in XFilterEvents() which itself locks when another Xlib function gets called that also locks. This fixes two instances by separating those Xlib functions into an internal (non-locking) call and a locking wrapper that is used as an external function. There may be several other such instances therefore another more general patch is eventually required (Bugzilla #1182).
18956
18957commit e689746c8d0e21e9011e8b91a3071d235d3a2a74
18958Author: Kevin E Martin <kem@kem.org>
18959Date:   Thu Aug 19 06:48:06 2004 +0000
18960
18961    Fix header file to #ifdef the XKB keysyms when they are used. This fixes the X test suite build failure.
18962
18963commit d558a53a6f57eecfcaadce5141fe3a08860defcb
18964Author: Keith Packard <keithp@keithp.com>
18965Date:   Sat Aug 14 07:12:36 2004 +0000
18966
18967    Use XLIB_SKIP_ARGB_VISUALS environment variable to disable all depth 32 visuals. Necessary to keep Flash from crashing.
18968    Must call ValidateGC/ValidatePicture on "real" GC/Picture to ensure pCompositeClip is set correctly.
18969    Need to take the composite clip from the "real" GC/Picture and turn it into the clientClip for the backing version.
18970    Adjust pixmap screen origin to account for drawable->x/y Change debugging output a bit (disabled by default)
18971
18972commit 85c2d81f299ed3444658011b7d6fb0a7ab8a6f55
18973Author: Alexander Gottwald <ago@freedesktop.org>
18974Date:   Fri Aug 13 16:28:19 2004 +0000
18975
18976    Set most significant bit to be a one. (Bug #1024, Kensuke Matsuzaki)
18977    Fix conversion from sjis and euc. (Bug #1024, Toshio Takabe)
18978
18979commit c4d56e4e288d4e48b84b021a61638f46e9a45e27
18980Author: Adam Jackson <ajax@nwnk.net>
18981Date:   Wed Aug 11 05:25:13 2004 +0000
18982
18983    Bug #372: Prevent a crash in XPolygonRegion when called with a bogus point count. Reported by Andreas Luik.
18984
18985commit 92487437173f600f208d825f65756d3ad14a4f7e
18986Author: Kevin E Martin <kem@kem.org>
18987Date:   Mon Aug 9 22:37:22 2004 +0000
18988
18989    Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, Harold L. Hunt II, Alexander Gottwald).
18990    Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, David Dawes).
18991    Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson).
18992    HP-PA build fix (Bug #828, Guy Martin, Paul Anderson).
18993    Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine).
18994    Fix attempt to read video ROM before enabling it (Bug #843, Ivan Kokshaysky, Mike A. Harris).
18995    Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. Harris).
18996    Clarify xset man page description of how to use the keyboard repeat rate settings (Bug #846, Mike A. Harris).
18997    Fix problem where print-screen key would get remapped to sys-req in certain keymaps, which broke GNOME printscreen functionality (Bug #847, Owen Taylor).
18998    Fix several render problems:
18999    - MMIO mode support
19000    - Hang on IGP chips
19001    - VT switching hang
19002    - 3D render corruption (Bug #922, Hui Yu).
19003
19004commit 55c2ee568e7d3903258286a13bdf96ce5348ffda
19005Author: Matthieu Herrb <matthieu.herrb@laas.fr>
19006Date:   Tue Jul 27 06:06:05 2004 +0000
19007
19008    - remove remaining AMOEBA references.
19009    - remove unused file.
19010
19011commit 6e884b12911eedfb003e90a3829ce66f7fc9cf2d
19012Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk>
19013Date:   Tue Jul 20 17:48:09 2004 +0000
19014
19015    Tue Jul 20 19:38:06 2004 Soeren Sandmann <sandmann@daimi.au.dk>
19016    Set font_data->xlfd_data to NULL after XFree(). (#837, patch from Bastien Nocera).
19017    Tue Jul 20 18:23:32 2004 Soeren Sandmann <sandmann@daimi.au.dk>
19018    Use /dev/urandom on Linux.(#761).
19019
19020commit 6f0bc97aa4e6de5a4b001f40ac10795cfdf09fc1
19021Author: Eric Anholt <anholt@freebsd.org>
19022Date:   Fri May 28 23:26:44 2004 +0000
19023
19024    Forced commit to note repocopy from xc/lib/XThrStub, will be connected to the build after this.
19025
19026commit aa7010c43ae9f39fb84b5ff155f76117c9e527a0
19027Author: Egbert Eich <eich@pdx.freedesktop.org>
19028Date:   Mon May 24 19:02:11 2004 +0000
19029
19030    Improve 'uniqueness' of authorization cookie sent by client for XDM-AUTHORIZATION-1. Old 'uniquness' consisted of the PID of the client, a time stamp (in seconds) and a number obtained by starting to count down from 0xffff. When a client did an XOpenDisplay() then execv'ed a child and did XOpenDisplay() again within the same second, the cookie was identical to the previous one (as the PID did not change but the static 'count down' variable was reinitialized) and thus refused by the server.
19031
19032commit 720702da29769d80ad1254d92edbad5b30f8a8da
19033Author: Alan Coopersmith <alan.coopersmith@sun.com>
19034Date:   Sat May 22 03:47:42 2004 +0000
19035
19036    Bugzilla #658: XStringToKeysym fails for Greek_IOTAdiaeresis (Fixes VSW5 testcase XStringToKeysym-7 failure)
19037
19038commit 3aed873292424b497d9a7dcee2975b95bf5ac966
19039Author: Alan Coopersmith <alan.coopersmith@sun.com>
19040Date:   Sat Apr 24 23:39:25 2004 +0000
19041
19042    XOpenDisplay should try tcp connection if local connections fail (aka Sun bug id #4624183).
19043    Also includes fix from NetBSD Problem Report #25098 (Michael van Elst) (Xlib segfaults with IPv6 if compiled with HASXDMAUTH).
19044    xc/config/cf/sunLib.tmpl
19045    xc/lib/FS/Imakefile Add missing shared library dependencies for Solaris
19046
19047commit c6349f43193b74a3c09945f3093a871b0157ba47
19048Author: Egbert Eich <eich@freedesktop.org>
19049Date:   Fri Apr 23 18:42:09 2004 +0000
19050
19051    Merging XORG-CURRENT into trunk
19052
19053commit c3c4ddc682950a01b80825021f3e2503ab01ea7f
19054Author: Kaleb Keithley <kaleb@freedesktop.org>
19055Date:   Tue Nov 25 19:28:07 2003 +0000
19056
19057    Initial revision
19058
19059commit dc4268a7dadc8da0d561757a68461246728613d3
19060Author: Kaleb Keithley <kaleb@freedesktop.org>
19061Date:   Fri Nov 14 16:48:47 2003 +0000
19062
19063    Initial revision
19064
19065commit deae12c6b683898f5213992d561a59d4ea889cca
19066Author: Kaleb Keithley <kaleb@freedesktop.org>
19067Date:   Fri Nov 14 15:54:30 2003 +0000
19068
19069    R6.6 is the Xorg base-line
19070