ChangeLog revision 15885c33
1commit 695e90ad26f632feb0f58ad94882fb3a263bf114
2Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3Date:   Thu Jun 15 09:14:48 2023 -0700
4
5    libX11 1.8.6
6    
7    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8
9commit 304a654a0d57bf0f00d8998185f0360332cfa36c
10Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11Date:   Sat Jun 10 16:30:07 2023 -0700
12
13    InitExt.c: Add bounds checks for extension request, event, & error codes
14    
15    Fixes CVE-2023-3138: X servers could return values from XQueryExtension
16    that would cause Xlib to write entries out-of-bounds of the arrays to
17    store them, though this would only overwrite other parts of the Display
18    struct, not outside the bounds allocated for that structure.
19    
20    Reported-by: Gregory James DUCK <gjduck@gmail.com>
21    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
22
23commit 71b08b8af20474bb704a11affaa8ea39b06d5ddf
24Author: Alan Coopersmith <alan.coopersmith@oracle.com>
25Date:   Wed May 31 17:45:40 2023 -0700
26
27    libX11 1.8.5
28    
29    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
30
31commit 03ba0140940cc76524d83096a47309f5c398541f
32Author: Benno Schulenberg <bensberg@telfort.nl>
33Date:   Mon May 22 12:02:06 2023 +0200
34
35    nls: use a slash instead of a combining solidus in compose sequences
36    
37    First: combining diacritics like the combining long solidus (`U+0338`)
38    are not meant to be used in compose sequences.  Second: there is just
39    one layout in xkeyboard-config that contains the `U0338` character:
40    the deprecated/obsolete German T3 layout.  So, practically speaking,
41    these compose sequences with `U0338` were untypable.  So, use a slash
42    instead, that almost all layouts have.  This does require that the
43    sequence `<Compose> <less> <slash>` changes its meaning from backslash
44    to not-less-than (`≮`).  This seems like an acceptable sacrifice, as
45    the sequence `<Compose> <slash> <slash>` is a faster/easier sequence
46    for the backslash and most layouts contain a backslash already anyway.
47    
48    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
49
50commit 616701317e25b2c0f509f4d93af5dee6de09012d
51Author: Benno Schulenberg <bensberg@telfort.nl>
52Date:   Mon May 22 11:19:32 2023 +0200
53
54    nls: delete six compose sequences that cannot be typed
55    
56    None of the layouts in xkeyboard-config contain `Ohook` or `Uhook`
57    or `U+01D3` (U with caron) nor their lowercase forms, so these six
58    sequences cannot be typed on any known keyboard.
59    
60    (Also, the target characters of the sequences with hook and horn
61    seem to occur only in Vietnamese, where the horn is part of the
62    vowel and the hook is a tonal mark.  All Vietnamese layouts in
63    xkeyboard-config contain `Ohorn` and `Uhorn`, and then the five
64    tonal marks as dead keys, `dead_hook` among them.  It does not
65    make sense to want to apply a horn to a vowel that already has
66    a tonal mark, nor is it possible with the Vietnamese layouts.)
67    
68    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
69
70commit 35d9621d78cb196b4ccd3ee1baa1ce2a21e4674b
71Author: Benno Schulenberg <bensberg@telfort.nl>
72Date:   Sun May 21 16:17:32 2023 +0200
73
74    nls: sort three sequences alphabetically in their group, like all others
75    
76    Also, consistently form uppercase+lowercase pairs, like for most others.
77
78commit 208d550954c7266fa8093b02a2a97047e1478c00
79Author: Benno Schulenberg <bensberg@telfort.nl>
80Date:   Thu Apr 13 17:25:35 2023 +0200
81
82    nls: make the Amharic compose sequences use the dead-vowel symbols
83    
84    Last year, commit 488b156fe2 changed mistaken ASCII characters to the
85    code points that were actually used in the Ethiopian keyboard layout.
86    
87    But those code points were mistaken too: the intention was not to use
88    the code point U+FE68 in the layout but the raw symbol code 0xfe68 --
89    and so on for the other codes.  The symbols that these raw codes were
90    intended to refer to were `dead_a` and friends.  But these dead vowels
91    were added under different codes: 0xfe80 to 0xfe8b.
92    
93    So, correct the Amharic compose sequences to make use of the dead-vowel
94    symbols that were specifically added for them fifteen years ago.
95    
96    (As soon as this is merged, xkeyboard-config wil be changed too to use
97    the dead vowels in the Ethiopian layout.)
98    
99    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
100
101commit 96cbac89a02220eb21e13ca5fa7c8c5adf77028a
102Author: jmcwilliams403 <jmcwilliams403@gmail.com>
103Date:   Fri Mar 31 09:47:03 2023 -0400
104
105    nls: add compose sequence for capital schwa, and delete a deviant one
106    
107    This sequence has already been defined in the iso8859-9e compose
108    file meant for the Azerbaijani locale, yet for the longest time, only
109    the lowercase form has existed in the en_US.UTF-8 master compose
110    file. This patch adds one sequence for Latin Capital Letter Schwa for
111    the Azerbaijani language.
112    
113    Additionally, the sequence in the pt_BR.UTF-8 locale for capital
114    schwa has been removed as it uses two lowercase e's as part of its
115    sequence, which in practice would make typing both `Multi_key e e`
116    and `Multi_key E E` output capital schwa.
117
118commit a6528d92b3865adb4f9951df91a714f55f56d59e
119Author: Benno Schulenberg <bensberg@telfort.nl>
120Date:   Tue Apr 4 11:35:44 2023 +0200
121
122    XSetScreenSaver.man: restore the part that was accidentally snipped
123    
124    Four years ago, while removing private macros, commit fe46eb2871
125    unintentionally truncated the XSetScreenSaver manpage.  Restore
126    the snipped part (minus trailing spaces and a mistaken singular).
127    
128    This fixes issue #187.
129    
130    Reported-by: Sergio Gelato
131    
132    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
133
134commit eb166af8f9f4184108fe8e0611a09af03d12059e
135Author: jmcwilliams403 <jmcwilliams403@gmail.com>
136Date:   Fri Mar 24 08:37:32 2023 -0400
137
138    NLS: Remove strange dead_cedilla cedi sign sequences
139    
140    These sequences appear to be a leftover artifact of auto-generating
141    dead_currency sequences for cent and the cedi sign, where chaining
142    with ascii "c" vs "ccedilla" would output cent vs the cedi sign
143    respectively, but somehow this ended up creating strange sequences
144    with dead_cedilla followed by cent or the colon sign to output the
145    cedi sign, the latter sequence of which was already removed recently.
146    Additionally, sequences beginninng with dead_cedilla followed by
147    currency have been removed as they also appear to be related to these
148    artifacts. The two remaining sequences beginning with dead_currency
149    followed by dead_cedilla are maintained and moved to the stretch of
150    dead_currency sequences next to ccedilla, and sorted in order of
151    unicode decompositional complexity.
152
153commit 6e889f34679bb96f4d61ef069259c042cb4a9a59
154Author: Benno Schulenberg <bensberg@telfort.nl>
155Date:   Fri Mar 24 09:41:48 2023 +0100
156
157    nls: remove twenty two untypable Greek compose sequences
158    
159    None of the keyboard layouts in xkeyboard-config contains accented
160    Greek characters like `Greek_alphaaccent` or `Greek_iotadieresis`
161    (nor in their Unicode hexadecimal forms, like U03AC or 0x10003CA),
162    so none of these sequences could be typed on an existing layout.
163    
164    Also, the two sequences that started with `Greek_accentdieresis`
165    were plain wrong: making a spacing character into a dead key.
166    
167    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
168
169commit 98be4cb23a5eadc98f81f05c4696f687119152c8
170Author: Benno Schulenberg <bensberg@telfort.nl>
171Date:   Fri Mar 24 09:00:48 2023 +0100
172
173    nls: remove four hundred and sixty untypable Greek compose sequences
174    
175    None of the keyboard layouts in xkeyboard-config contains code points
176    in the range U+1F00..U+1FFF (the Greek Extended block), so none of the
177    removed sequences could be typed on an existing layout.
178    
179    (Of course it is possible that some people have their own private layout
180    that contains some of the Greek Extended characters, but if they do,
181    then they can also define private compose sequences, when needed.)
182    
183    Modern Greek does not use these characters with polytonic diacritics
184    anyway -- they are useful only for typing traditional Greek.
185    
186    The useless sequences were removed with this command:
187    
188      sed -i '/<U1F..>/d' nls/en_US.UTF-8/Compose.pre
189    
190    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
191
192commit ff8706a5eae25b8bafce300527079f68a201d27f
193Author: Benno Schulenberg <bensberg@telfort.nl>
194Date:   Wed Jun 15 11:54:54 2022 +0200
195
196    nls: delete compose sequences that pointlessly mix upper and lower case
197    
198    It's nice to have compose sequences for things like © and ®, but allowing
199    them to be composed also with letters of different case is unneeded.  It
200    is good enough to have sequences of only lowercase and only uppercase.
201    
202    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
203
204commit ca99e338a9b8aad300933b1336f9e3c091392213
205Author: jmcwilliams403 <jmcwilliams403@gmail.com>
206Date:   Sun Mar 12 15:13:27 2023 +0000
207
208    NLS: move dead-caron subscript compositions to the relevant Unicode block
209
210commit dac2985b347cebc0f631e43e41eaeddb68de1676
211Author: jmcwilliams403 <jmcwilliams403@gmail.com>
212Date:   Sun Mar 12 14:49:47 2023 +0000
213
214    NLS: implement the expansion of the six Breton N-graph keysyms
215    
216    These one-keysym-to-multiple-characters sequences will allow fully defining
217    the Breton keyboard layout, without needing custom compose sequences.
218
219commit e45ca7b41dcd3ace7681d6897505f85d374640f2
220Author: Alan Coopersmith <alan.coopersmith@oracle.com>
221Date:   Tue Mar 14 10:22:48 2023 -0700
222
223    gitlab CI: add workflow rules
224    
225    Needed due to the gitlab config changes described in:
226     https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/438
227    
228    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
229
230commit 72eeeb6dedc39f68608312eb0001f77fffccd499
231Author: jmcwilliams403 <jmcwilliams403@gmail.com>
232Date:   Sun Mar 5 13:33:38 2023 +0000
233
234    NLS: move long S compositions to respective blocks
235
236commit 8bbbed75a20556e41658103265974decee974645
237Author: Alan Coopersmith <alan.coopersmith@oracle.com>
238Date:   Sat Mar 4 10:44:45 2023 -0800
239
240    configure: Use LT_INIT from libtool 2 instead of deprecated AC_PROG_LIBTOOL
241    
242    AC_PROG_LIBTOOL was replaced by LT_INIT in libtool 2 in 2008,
243    so it's time to rely on it.
244    
245    Clears autoconf warnings:
246    
247    configure.ac:20: warning: The macro `AC_PROG_LIBTOOL' is obsolete.
248    configure.ac:20: You should run autoupdate.
249    m4/libtool.m4:100: AC_PROG_LIBTOOL is expanded from...
250    configure.ac:20: the top level
251    
252    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
253
254commit 53bf8584e8d7d5d4a4a8114bff26a6f631c7fac1
255Author: Alan Coopersmith <alan.coopersmith@oracle.com>
256Date:   Mon Feb 6 14:09:45 2023 -0800
257
258    configure: replace deprecated AC_HELP_STRING with AS_HELP_STRING
259    
260    This silences an autoconf warning.
261    
262    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
263
264commit 31a9451111fb1e3ae1199d44f94cd0b134ca455d
265Author: Alan Coopersmith <alan.coopersmith@oracle.com>
266Date:   Mon Feb 6 13:13:21 2023 -0800
267
268    configure: raise minimum autoconf requirement to 2.70
269    
270    Needed for builds on NetBSD to work correctly, since it depends on
271    AC_USE_SYSTEM_EXTENSIONS defining _OPENBSD_SOURCE to expose the
272    prototype for reallocarray() in the system headers.
273    
274    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
275
276commit 1ba02f878f83e3b42389547e01b24371f0a50e60
277Author: Alan Coopersmith <alan.coopersmith@oracle.com>
278Date:   Mon Feb 6 13:32:02 2023 -0800
279
280    gitlab CI: Add libtool to required packages
281    
282    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
283
284commit fc5ec8edda1efdf5d66db4e76796e3c407527b3b
285Author: Alan Coopersmith <alan.coopersmith@oracle.com>
286Date:   Sat Feb 4 12:02:29 2023 -0800
287
288    libX11 1.8.4
289    
290    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
291
292commit ba548ed1688d4d5a83913218155d0ddf616ff176
293Author: Alan Coopersmith <alan.coopersmith@oracle.com>
294Date:   Sun Jan 15 11:39:20 2023 -0800
295
296    configure: options summary should say "libX11", not just "X11"
297    
298    We're not configuring the entire window system here, just one library
299    
300    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
301
302commit eb1c272ab5230d548077b9f59aca4b3457c3a8f8
303Author: GaryOderNichts <garyodernichts@gmail.com>
304Date:   Sat Dec 17 16:28:40 2022 +0100
305
306    Fix a9e845 and 797755 Allow X*IfEvent() to reenter libX11
307
308commit 0c7c4ce875ea65160705180ebb68a2c88afb678e
309Author: Benno Schulenberg <bensberg@telfort.nl>
310Date:   Thu Nov 3 17:25:29 2022 +0100
311
312    nls: add two compose sequences for capital B with stroke, for consistency
313    
314    Two compose sequences (with plain ASCII characters) existed for
315    the lowercase b with stroke (ƀ) but not for the capital one (Ƀ).
316    
317    This addresses part of issue #166.
318    
319    Reported-by: J. McWilliams
320    
321    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
322
323commit a80ef097e713aba0330b9cf9f3c1e309084b926f
324Author: Benno Schulenberg <bensberg@telfort.nl>
325Date:   Thu Nov 3 16:48:03 2022 +0100
326
327    nls: delete twelve compose sequences that cannot be typed
328    
329    There is not any keyboard layout (in xkeyboard-config) that contains
330    `obelowdot` or `ubelowdot`, so having compose sequences that use these
331    symbols is pointless.
332    
333    There are some layouts that contain `otilde` (and one that contains
334    `utilde`), but those layouts then do not contain `dead_horn`, so
335    the compose sequences that combine the two symbols are pointless.
336    
337    There are a few layouts that contain U+0256, U+025C, U+025F or U+0279,
338    but those layouts do not contain `dead_hook`, so the compose sequences
339    that combine the latter with one of the former are pointless.
340    
341    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
342
343commit 078cc39f05ddfd6166762f3166129258ddd07dd3
344Author: Benno Schulenberg <bensberg@telfort.nl>
345Date:   Thu Nov 3 16:31:04 2022 +0100
346
347    nls: delete 88 compose sequences that have the accents in the wrong order
348    
349    The various compose sequences with the accents in their proper order
350    (the highest placed accent first) continue to exist.
351    
352    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
353
354commit aa5ef0ba80558f946871f574d38a15673e66221e
355Author: Benno Schulenberg <bensberg@telfort.nl>
356Date:   Thu Nov 3 16:27:51 2022 +0100
357
358    nls: move the compose sequences for Ș and Ț to the fitting Unicode block
359    
360    This addresses a side note of issue #166.
361    
362    Reported-by: J. McWilliams
363    
364    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
365
366commit 82c25879b72f1ba7e9abe00e7bad55add33135cf
367Author: Alan Coopersmith <alan.coopersmith@oracle.com>
368Date:   Wed Jan 4 11:25:48 2023 -0800
369
370    xlibi18n: Update Imake instructions to autoconf equivalent in comments
371    
372    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
373
374commit 88399e01be679bfcc9a5e8922ffe2c47f0e56dee
375Author: Yuxuan Shui <yshuiv7@gmail.com>
376Date:   Tue Jan 3 15:09:28 2023 +0000
377
378    Revert "Update XPutBackEvent() to support clients that put back unpadded events"
379    
380    This reverts commit d6d6cba90215d323567fef13d6565756c9956f60.
381    
382    The reverted commit intended to fix the problem where an unpadded X
383    event struct is passed into XPutBackEvent, by creating a padded struct
384    with _XEventToWire and _XWireToEvent. However, _XWireToEvent updates the
385    last sequence number in Display, which may cause xlib to complain about
386    lost sequence numbers.
387    
388    IMO, the problem that commit tried to solve is a bug in the client
389    library, and workaround it inside Xlib is bad practice, especially given
390    the problem it caused. Plus, the offender cited in the original commit
391    message, freeglut, has already fixed this problem.
392    
393    Fixes: #176 #174
394    
395    Signed-off-by: Yuxuan Shui <yshuiv7@gmail.com>
396
397commit 48487c4b9028e9eec5371742f51e1d64c2e17296
398Author: Alan Coopersmith <alan.coopersmith@oracle.com>
399Date:   Thu Dec 15 14:52:20 2022 -0800
400
401    libX11 1.8.3
402    
403    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
404
405commit d6d6cba90215d323567fef13d6565756c9956f60
406Author: Keith Packard <keithp@keithp.com>
407Date:   Sun Dec 11 10:32:26 2022 -0800
408
409    Update XPutBackEvent() to support clients that put back unpadded events
410    
411    It seems to be common practice of some X11 clients to pass specific event
412    types into APIs that take XEvent*.  For example, freeglut does:
413    
414       XConfigureEvent fakeEvent = {0};
415       ...
416       XPutBackEvent(fgDisplay.Display, (XEvent*)&fakeEvent);
417    
418    This can result in reads overflowing the input event when libX11 does:
419    
420       XEvent store = *event;
421    
422    =================================================================
423    ==75304==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x00016ee4a8e8 at pc 0x000101c54d14 bp 0x00016ee4a0d0 sp 0x00016ee49888
424    READ of size 192 at 0x00016ee4a8e8 thread T0
425        #0 0x101c54d10 in __asan_memcpy+0x1a4 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x3cd10)
426        #1 0x102848a18 in _XPutBackEvent PutBEvent.c:41
427        #2 0x1028490a4 in XPutBackEvent PutBEvent.c:84
428        #3 0x1013295c8 in fgOpenWindow freeglut_window.c:1178
429        #4 0x101321984 in fgCreateWindow freeglut_structure.c:108
430        #5 0x10132b138 in glutCreateWindow freeglut_window.c:1551
431        #6 0x100fb7d94 in main+0x78 (checkeredTriangles:arm64+0x100003d94)
432        #7 0x197de3e4c  (<unknown module>)
433    
434    Address 0x00016ee4a8e8 is located in stack of thread T0 at offset 840 in frame
435        #0 0x1013282f8 in fgOpenWindow freeglut_window.c:1063
436    
437      This frame has 8 object(s):
438        [32, 40) 'title.addr'
439        [64, 176) 'winAttr' (line 1066)
440        [208, 240) 'textProperty' (line 1067)
441        [272, 352) 'sizeHints' (line 1068)
442        [384, 440) 'wmHints' (line 1069)
443        [480, 672) 'eventReturnBuffer' (line 1070)
444        [736, 740) 'num_FBConfigs' (line 1072)
445        [752, 840) 'fakeEvent' (line 1074) <== Memory access at offset 840 overflows this variable
446    
447    This change allows XPutBackEvent() to support such clients without
448    risk of memory read overflow.
449    
450    Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
451    Tested-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
452
453commit 496d9bfeedda89a2211ec665f9214c2693c022f4
454Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
455Date:   Sat Dec 3 16:26:47 2022 -0800
456
457    ximcp: Address warning found by UBSan when growing an empty tree
458    
459    imLcPrs.c:681:52: runtime error: applying zero offset to null pointer
460    
461    Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
462
463commit 7f7bcd7b6f569e9f70e3ddd134924f178e2596b1
464Author: Takao Fujiwara <tfujiwar@redhat.com>
465Date:   Tue Dec 6 16:59:56 2022 +0000
466
467    nls: consecutive cs number in en_US.UTF-8/XLC_LOCALE
468
469commit ba095967349dff583bd0989d0ba78aa334e9bedf
470Author: Ulrich Sibiller <uli42@gmx.de>
471Date:   Wed Nov 30 23:47:29 2022 +0100
472
473    ChkIfEv.c: fix wrong handling of dpy->in_ifevent
474    
475    Is no longer a bool but a counter.
476    
477    Signed-off-by: Ulrich Sibiller <uli42@gmx.de>
478
479commit 70eaf1174e2809d1a6c43868d53f8cd844ef636a
480Author: Ulrich Sibiller <uli42@gmx.de>
481Date:   Wed Nov 30 23:42:02 2022 +0100
482
483    Indentation fixes around recent dpy->in_ifevent changes
484    
485    Use the same indentation as the surrounding code.
486    
487    Signed-off-by: Ulrich Sibiller <uli42@gmx.de>
488    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
489
490commit a9e845809bcaae22496bc8aa3ca252b410d5f39b
491Author: Matthieu Herrb <matthieu@herrb.eu>
492Date:   Fri Nov 11 18:55:23 2022 +0100
493
494    Fix 797755 Allow X*IfEvent() to reenter libX11
495    
496    - the activation logic is reversed
497    - there is also _XInternalLockDisplay() that needs protection
498    - I've found cases (in fvwm2) where the callback calls XCheckIfEvent()
499      recursively. So the flag needs to be a counter.
500    
501    Reviewed-by: Adam Jackson <ajax@redhat.com>
502
503commit bccd787a565d3a88673bfc06574c1939f98d8d72
504Author: Nia Alarie <nia@NetBSD.org>
505Date:   Thu Nov 10 22:31:47 2022 +0100
506
507    Don't use pragma inside a function, it breaks compiling with older GCCs.
508    
509    XKBBind.c:230: error: #pragma GCC diagnostic not allowed inside functions
510    
511    Signed-off-by: Thomas Klausner <wiz@gatalith.at>
512
513commit 696d19d5db7bcb1c1f582c2b1846520e7e0870cb
514Author: Oliver <halting@riseup.net>
515Date:   Fri Nov 11 17:04:00 2022 +0000
516
517    Add XFreeThreads function.
518
519commit b4f24b272c6ef888b6fcfcf80670c196b2e8f755
520Author: Alan Coopersmith <alan.coopersmith@oracle.com>
521Date:   Thu Nov 10 12:18:17 2022 -0800
522
523    libX11 1.8.2
524    
525    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
526
527commit 77629ea5e79fa9b1596438bc36cfc7e0c631af43
528Author: Alan Coopersmith <alan.coopersmith@oracle.com>
529Date:   Thu Nov 3 13:00:50 2022 -0700
530
531    README.md: Add 1.8.2 changes
532    
533    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
534
535commit d1baf2ec4869dea3732d6fe13ba4935cb6caad01
536Author: Alan Coopersmith <alan.coopersmith@oracle.com>
537Date:   Thu Nov 3 11:22:37 2022 -0700
538
539    Remove NEWS file which only covered 2006 & 2007 releases
540    
541    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
542
543commit 1294bfa487bdecfd32621822922fec03203ee188
544Author: Po Lu <luangruo@yahoo.com>
545Date:   Mon Oct 3 08:44:59 2022 +0800
546
547    specs: document change in XIMPreeditCallbacks
548    
549    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
550
551commit 1a7e54609ce4dc0f2d3b933502437a413cf3dea8
552Author: Alan Coopersmith <alan.coopersmith@oracle.com>
553Date:   Thu Nov 3 14:44:22 2022 -0700
554
555    man pages: document XCloseIM frees its argument
556    
557    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
558
559commit a04b84f0f7bb3a78088756759f4b9b8bb363251e
560Author: Alan Coopersmith <alan.coopersmith@oracle.com>
561Date:   Thu Nov 3 20:31:31 2022 +0000
562
563    Copyright & license cleanup
564
565commit 5e41119d9611504ff7e61c6aa4da1df708096801
566Author: Benno Schulenberg <bensberg@telfort.nl>
567Date:   Mon Oct 31 12:42:51 2022 +0100
568
569    nls: reshuffle a few compose sequences, to have similar ones together
570    
571    It makes more sense to have similar sequences grouped together
572    than to rigidly follow the order of ascending Unicode codes.
573
574commit 65d89342f59cfc2eac9bf8010076b937f22554e8
575Author: Benno Schulenberg <bensberg@telfort.nl>
576Date:   Mon Oct 31 12:50:11 2022 +0100
577
578    nls: remove two compose sequences that use deprecated symbols
579    
580    The last few occurrences of `leftcaret` and `rightcaret` were replaced
581    with `less` and `greater` in xkeyboard-config half a year ago.
582
583commit 6baccbae5324e7342cb5176cae82f463a6b34eae
584Author: Benno Schulenberg <bensberg@telfort.nl>
585Date:   Mon Oct 31 12:40:12 2022 +0100
586
587    nls: let `<Multi_key> <minus> <underscore>` compose U+2212 (MINUS SIGN)
588    
589    There was not yet any way to compose the Unicode minus sign, U+2212.
590
591commit 79775575418fd6f8ee1c5e5bbe403df4606fb5b6
592Author: Adam Jackson <ajax@redhat.com>
593Date:   Fri Aug 5 15:19:08 2022 -0400
594
595    Allow X*IfEvent() to reenter libX11
596    
597    The documentation for this family of functions very clearly says not to
598    call into xlib in your predicate function, but historically single
599    threaded apps could get away with it just fine, and now that we've
600    forced thread-safety on the world such apps will now deadlock instead.
601    That's not an acceptable regression even if the app is technically
602    broken. This has been reported with XFCE and FVWM, and Motif's
603    cut-and-paste code has the same bug by inspection, so this does need to
604    be addressed.
605    
606    This change nerfs LockDisplay/UnlockDisplay while inside the critical
607    bit of an IfEvent function. This is still safe in the sense that the
608    display remains locked and no other thread should be able to change it
609    from under us, but the loop that scans the event queue might not be
610    robust against it being modified as a side effect of protocol emitted by
611    the predicate callback. But that's not new, non-XInitThreads'd xlib
612    would have the same caveat.
613    
614    Closes: xorg/lib/libx11#157
615
616commit 0d1d65bdd98966f52bcac4077f94827b20b229dd
617Author: Benno Schulenberg <bensberg@telfort.nl>
618Date:   Fri Oct 21 12:18:25 2022 +0200
619
620    nls: change <Compose> <^> <-> to mean superscript minus instead of macron
621    
622    Several other `<Multi_key> <asciicircum> <symbol>` sequences
623    produce the superscript equivalent of the given symbol.  So,
624    let `<Multi_key> <asciicircum> <minus>` do the same.
625    
626    Also, add two other sequences for producing a plain macron,
627    to compensate a bit the loss of the above sequence.
628    
629    Additionally, make `<Multi_key> <underscore> <minus>` produce
630    a subscript minus, for consistency.
631    
632    This fixes issue #165.
633    
634    Requested-by: J. McWilliams
635    
636    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
637
638commit 310898270787c1a3a6ab50221324f8654dddce8e
639Author: Jan Tojnar <jtojnar@gmail.com>
640Date:   Thu Oct 20 07:10:58 2022 +0000
641
642    nls: Map sr locales to sr_RS compose files
643    
644    Serbian used sr_YU (Yugoslavia) code in the past.
645    Then it was succeeded by sr_CS (Serbia and Montenegro).
646    Finally, it was split into sr_RS (Serbia) and sr_ME (Montenegro).
647    
648    https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd
649    introduced the modern sr_RS and sr_ME codes.
650    
651    Next, https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/40761898692e5063957bfa2518cca3d35b2e354a
652    added the Serbian compose table but only for the legacy sr_CS entry.
653    
654    https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/5cd60398b7787297008b13a848ed3cfbd7ef178d
655    removed the legacy sr_CS entry, the only one pointing to the correct compose file.
656    It also renamed the file to sr_RS, but did not update the compose mapping.
657    
658    Let’s point all Serbian locales to the Compose file again.
659
660commit a35d706cd8dc7b498bca2a3fc4a3732047439ed7
661Author: Alan Coopersmith <alan.coopersmith@oracle.com>
662Date:   Fri Oct 14 13:40:37 2022 -0700
663
664    COPYING: remove notice for the removed UIThrStubs.c
665    
666    Fixes: 701e9e9a ("Use same pthread-stubs as libxcb")
667    
668    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
669
670commit 3b1750e982fedce0d61bface68aadcd5b1ea1fae
671Author: Benno Schulenberg <bensberg@telfort.nl>
672Date:   Fri Oct 14 09:28:42 2022 +0200
673
674    nls: delete two compose sequences with an anomalous post-fixed breve
675    
676    Two years ago, commit b126bfd7fe allowed using also a lowercase `u`
677    wherever an uppercase `U` was used to represent a breve.  But the
678    commit should have limited itself to only the prefixed uses of `U`,
679    as that is how most letters with a breve are composed.
680    
681    Also, group the two compose sequences with an uppercase post-fixed `U`
682    together with the corresponding other post-fixed sequences.
683    
684    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
685
686commit 701e9e9afb88bdc68c84a4611adc533400a3df36
687Author: Alan Coopersmith <alan.coopersmith@oracle.com>
688Date:   Tue Oct 4 11:24:57 2022 -0700
689
690    Use same pthread-stubs as libxcb
691    
692    Avoid conflicts when libX11 calls libxcb and gets its pthread functions
693    overriding our ancient stubs.
694    
695    v2: Keep linking with real threads libraries when thread safety
696        constructor is enabled.
697    
698    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
699
700commit 1d11822601fd24a396b354fa616b04ed3df8b4ef
701Author: Thomas E. Dickey <dickey@invisible-island.net>
702Date:   Tue Oct 4 18:26:17 2022 -0400
703
704    fix a memory leak in XRegisterIMInstantiateCallback
705    
706    Analysis:
707    
708        _XimRegisterIMInstantiateCallback() opens an XIM and closes it using
709        the internal function pointers, but the internal close function does
710        not free the pointer to the XIM (this would be done in XCloseIM()).
711    
712    Report/patch:
713    
714        Date: Mon, 03 Oct 2022 18:47:32 +0800
715        From: Po Lu <luangruo@yahoo.com>
716        To: xorg-devel@lists.x.org
717        Subject: Re: Yet another leak in Xlib
718    
719        For reference, here's how I'm calling XRegisterIMInstantiateCallback:
720    
721        XSetLocaleModifiers ("");
722        XRegisterIMInstantiateCallback (compositor.display,
723                                        XrmGetDatabase (compositor.display),
724                                        (char *) compositor.resource_name,
725                                        (char *) compositor.app_name,
726                                        IMInstantiateCallback, NULL);
727    
728        and XMODIFIERS is:
729    
730            @im=ibus
731    
732    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
733
734commit 1f8fd7ff1cf688ec1d3b34397c6d58110bb0cae5
735Author: Alan Coopersmith <alan.coopersmith@oracle.com>
736Date:   Mon Sep 26 16:01:40 2022 -0700
737
738    If thread-safety-constructor is enabled, link against pthreads not stubs
739    
740    Only really makes a difference if pthreads is not in libc.
741    
742    Fixes: #162
743    
744    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
745    Reviewed-by: Adam Jackson <ajax@redhat.com>
746
747commit bfe5d2dd98f705438b0cdddbbf9c12882da89bae
748Author: Thomas E. Dickey <dickey@invisible-island.net>
749Date:   Sun Sep 11 11:44:41 2022 -0400
750
751    reduce compiler warnings with casts (no object change)
752    
753    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
754
755commit 926f2454bca9a7a7d1f4cf0648b0c193f5cef644
756Author: Thomas E. Dickey <dickey@invisible-island.net>
757Date:   Sun Sep 11 10:35:31 2022 -0400
758
759    use casts to reduce compiler warnings (no object change)
760    
761    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
762
763commit dc553ff908f10832c71dc2b5ba5af7e68589218e
764Author: Mike FABIAN <mfabian@redhat.com>
765Date:   Fri Jul 22 11:39:01 2022 +0200
766
767    Remove KOI8-R character set from en_US.UTF-8/XLC_LOCALE
768    
769    https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/master/src/xlibi18n/lcCT.c#L58
770    has no escape sequence for KOI8-R.
771    That makes Xutf8TextListToTextProperty() sometimes fail when trying to
772    convert to COMPOUND_TEXT and when the KOI8-R charset is tried.
773    
774    This is the cause for this bug:
775    
776    https://github.com/ibus/ibus/issues/2422
777    
778    Removing the KOI8-R charset entry fixes the problem.
779    
780    This commit also fixes a few wrong texts in comments.
781    
782    Signed-off-by: Mike FABIAN <mfabian@redhat.com>
783
784commit 1c04ac514196db5e2255d99635fb6e3c36be330d
785Author: Adam Sampson <ats@offog.org>
786Date:   Sun Jul 4 23:02:04 2021 +0100
787
788    xkb: Fix off-by-one error in XKeycodeToKeysym
789    
790    The code here that made indexes greater than 3 refer to XKB symbol
791    groups had an off-by-one error, so it would always leave out the symbol
792    that should have been at index 4. Rewrite the code to fix this and
793    simplify the logic a bit.
794    
795    Signed-off-by: Adam Sampson <ats@offog.org>
796
797commit ddaacd219609104f138006db6a8f708226db3874
798Author: Thomas E. Dickey <dickey@invisible-island.net>
799Date:   Fri Sep 9 20:17:35 2022 -0400
800
801    reduce compiler warnings for macros BufAlloc, Data and Data32 using casts
802    
803    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
804
805commit 62c3337d89d31e0d3ed807004e73ad711fad3342
806Author: Kirill Chibisov <contact@kchibisov.com>
807Date:   Thu Sep 8 22:50:30 2022 +0000
808
809    ximcp/imRm.c: allow XNSpotLocation with OnTheSpot
810
811commit 1272879074ad9612314858a75cf844783c1963b2
812Author: Alan Coopersmith <alan.coopersmith@oracle.com>
813Date:   Thu Jul 28 17:30:21 2022 -0700
814
815    gitlab CI: stop requiring Signed-off-by in commits
816    
817    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
818
819commit 38033b073f393564acf6a04501927002b2b4aa86
820Author: David H. Gutteridge <david@gutteridge.ca>
821Date:   Mon Jul 25 19:30:19 2022 -0400
822
823    configure.ac: report thread safety constructor build status
824    
825    Signed-off-by: David H. Gutteridge <david@gutteridge.ca>
826
827commit 9a90b3a63d02da92dc8692ed3f6269d3e0c7c8cc
828Author: Alan Coopersmith <alan.coopersmith@oracle.com>
829Date:   Sun Jul 10 11:55:11 2022 -0700
830
831    doc: Document possible return values for XkbGetKeyboard()
832    
833    Fixes: #160
834    
835    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
836
837commit 0beb0cf46ae48b0d4322ff8a0221f47bdb7d2296
838Author: Alan Coopersmith <alan.coopersmith@oracle.com>
839Date:   Sun Jul 10 11:42:19 2022 -0700
840
841    xkb: set num_keys when filling in keys in _XkbReadGetNamesReply()
842    
843    Fixes: #160
844    
845    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
846
847commit 487e9489d1e086cc62c53045206ab6d6d9d02031
848Author: Jon Turney <jon.turney@dronecode.org.uk>
849Date:   Wed Jun 22 16:15:37 2022 +0100
850
851    Fix Win32 build with -fno-common
852    
853    /work/xorg/lib/libX11/src/XlibInt.c:1968: multiple definition of `_Xdebug_p'; .libs/globals.o:globals.c:(.bss+0xc): first defined here
854    
855    Avoid redundant definition of _Xdebug_p in globals.c (which is under the
856    influence of _Xdebug being #defined to _Xdebug_p.
857    
858    This appears to be an ancient hack to work around data exports resolving
859    to the address of the import stub, not the import. (See [1]).
860    
861    (This is probably no longer needed or can be done in a better way, as
862    per the discussion under --enable-auto-import in the ld manpage.)
863    
864    [1] https://cygwin.com/pipermail/cygwin-xfree/2001-May/004606.html
865    
866    Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
867
868commit 35ad99fad2f65ec2c69bd84bdeb8138329092b57
869Author: Benno Schulenberg <bensberg@telfort.nl>
870Date:   Fri Jul 15 11:29:46 2022 +0200
871
872    nls: add four sequences for the Samogitian E with dot above and macron
873    
874    These sequences each produce two code points: the E-with-dot-above and
875    a combining macron.  The XIM input method is required for this to work.
876    
877    (Also add a missing comment for a Unicode block.)
878    
879    This fixes issue #54.
880    
881    Requested-by: Arns Udovīčė
882    
883    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
884
885commit 2732dd20cc2694dd5c32fef292e1f48626011032
886Author: Benno Schulenberg <bensberg@telfort.nl>
887Date:   Wed Jul 6 16:44:50 2022 +0200
888
889    nls: delete eight sequences that pointlessly mix upper and lower case
890    
891    Typing a compose sequence requires some care -- surely the user is able
892    to either keep holding the Shift key or not touch it at all while typing
893    the sequence.  Also, compose sequences are not an infinite resource AND
894    take up space and time -- defining redundant ones is a waste.
895    
896    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
897
898commit ea3cc474e1acfed47b9fb856ea151008fa6cfd30
899Author: Benno Schulenberg <bensberg@telfort.nl>
900Date:   Wed Jul 6 16:36:48 2022 +0200
901
902    nls: add compose sequences for the double-struck capitals ℕ ℤ ℚ ℝ ℂ
903    
904    This allows the user to type the symbols for the five number systems.
905    
906    This fixes the reasonable part of issue #159.
907    
908    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
909
910commit 95a03cde434c447fdb41d8cccbbdd41f20750f32
911Author: Benno Schulenberg <bensberg@telfort.nl>
912Date:   Thu Jul 14 12:00:45 2022 +0200
913
914    nls: delete some pointless and wrong compose sequences meant for Bépo
915    
916    Fourteen years ago, commit 7302984642 added some four hundred compose
917    sequences for the benefit of the French Bépo layout.  But among these
918    four hundred there are several that use symbols that are not available
919    in the Bépo layout and are thus impossible to type.  Some of the used
920    symbols, like Ahook, Ehook and Ohook, are not even present in *any*
921    layout, making these sequences a dead weight in the Compose file.
922    
923    The Amacron and Omacron are available only in the Latvian, Hawaiian,
924    and Maori layouts, and the Omacron also in the Silesian layout.  But
925    the Latvian layouts and the Hawaiian do not contain any dead keys.
926    Only in the Maori and Silesian layouts these sequences with Amacron
927    and Omacron could be typed, but that was not why they were added.
928    
929    More importantly, as James Cloos noted in issue #54, sequences like
930    `<dead_abovedot> <amacron>` for generating `ǡ` (that is: the macron
931    above the dot) are questionable, as in compose sequences generally
932    the first accent typed is the uppermost in the composed character.
933    
934    Reference:
935      https://gitlab.freedesktop.org/xorg/lib/libx11/-/issues/54#note_17321
936    
937    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
938
939commit 9702b836862da6838d725f555e7a1d83faf51740
940Author: Matthieu Herrb <matthieu@herrb.eu>
941Date:   Fri Jul 1 11:13:13 2022 +0200
942
943    Mention that the predicate function is called with the display lock
944    
945    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
946
947commit a5d00f889fadad3f5b11e53fd5c04cf5343fa8a2
948Author: Benno Schulenberg <bensberg@telfort.nl>
949Date:   Fri Jun 24 16:20:52 2022 +0200
950
951    docs: replace three placeholders with something that makes sense
952    
953    Also, uncapitalize two arguments, to match the style of all others.
954    
955    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
956
957commit 9ff8502040e94cb172426d59df722cefa752a401
958Author: Benno Schulenberg <bensberg@telfort.nl>
959Date:   Fri Jun 24 16:07:22 2022 +0200
960
961    docs: remove the unsightly dashes from the overviews of arguments
962    
963    Seventeen months ago, commits 78027fdb7a and 4f15cfc645 removed
964    these dashes from two of the man pages.  Now, remove them all.
965    They are unhelpful and just make one wonder why they are there
966    (probably to function as improvised bullet points).
967    
968    Also remove four leading spaces and a trailing comma.
969    
970    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
971
972commit a7968c47e35703446d67d070234fd9adf07f34b0
973Author: Benno Schulenberg <bensberg@telfort.nl>
974Date:   Thu Jun 23 15:48:05 2022 +0200
975
976    docs: hard-wrap some items so that Table 2 fits within 80 characters
977    
978    (Table 1 hard-wraps the first-column items in the same way.)
979    
980    Also, correct the formatting of the subsequent paragraph.
981    
982    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
983
984commit c601c779ca8f46dc4335180ec2829851fb8bbc13
985Author: Benno Schulenberg <bensberg@telfort.nl>
986Date:   Wed Jun 15 10:55:38 2022 +0200
987
988    nls: delete compose sequences that mix top-row digits with numpad digits
989    
990    Compose sequences for circled numbers, like ⑫  or ㉑, are nice to have,
991    but allowing them to be composed by typing one digit on the top row and
992    the other on the numerical keypad (or the other way around) is over the
993    top.  Remove these absurd sequences.  Keep only the sequences where both
994    digits are either on the top row or on the numerical keypad.
995    
996    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
997
998commit 3a30ada60c5217ada37b143b541c8e6f6284c7fa
999Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1000Date:   Wed Jun 8 17:30:14 2022 -0700
1001
1002    libX11 1.8.1
1003    
1004    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1005
1006commit 70f7403fd3bf362fc50b8de4a2e26300f757c68e
1007Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1008Date:   Tue May 3 14:45:05 2022 -0700
1009
1010    configure.ac: Fix --enable-thread-safety-constructor
1011    
1012    Prior to this, --enable-thread-safety-constructor would disable it,
1013    while --disable worked as expected, and no option left it enabled as
1014    expected by default.  This also fixes the --help text to be displayed.
1015    
1016    Fixes: afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9
1017    Reported-by: @igor.v.kovalenko
1018    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1019    Reviewed-by: Adam Jackson <ajax@redhat.com>
1020
1021commit d0da5a1e0fb214ff911ecff86d35f6673ff70023
1022Author: Adam Jackson <ajax@redhat.com>
1023Date:   Fri Apr 29 11:27:44 2022 -0400
1024
1025    libX11 1.8
1026    
1027    Minor number bump to reflect the thread safety default change.
1028    
1029    Signed-off-by: Adam Jackson <ajax@redhat.com>
1030
1031commit afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9
1032Author: Adam Jackson <ajax@redhat.com>
1033Date:   Tue Mar 22 18:24:29 2022 -0400
1034
1035    global: call XInitThreads() from the library's constructor
1036    
1037    There is really no point in not being thread safe, I measured, all you
1038    can see happen is noop performance gets like twice as slow and you have
1039    thread safety bugs. And we're using xcb as the transport which means we
1040    should expect threads in our clients anyway. Just do it.
1041    
1042    This assumes your compiler understands __attribute__((constructor)). If
1043    this is not your compiler, you can disable this with the appropriate
1044    configure flag, but be aware you're asking for bugs.
1045    
1046    Signed-off-by: Adam Jackson <ajax@redhat.com>
1047
1048commit 1ab48f3cc966751d86e434808180beb2dc76c4fa
1049Author: Benno Schulenberg <bensberg@telfort.nl>
1050Date:   Sun Apr 24 15:06:45 2022 +0200
1051
1052    docs: add release notes for versions 1.7.3, 1.7.4, and 1.7.5
1053    
1054    They were forgotten when those releases were made.
1055    
1056    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1057
1058commit 35d1513bc7fdd3ac6f5807feb601efc34ac19163
1059Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1060Date:   Sat Apr 16 13:04:23 2022 -0700
1061
1062    XkbOpenDisplay: Add _Xconst qualifier to display name argument
1063    
1064    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1065
1066commit c88ceaad4aaa0439e5cba13c1f5737ad54566e0b
1067Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1068Date:   Sat Apr 16 12:59:39 2022 -0700
1069
1070    XKBgeom.h: Add _Xconst qualifier to char * arguments in functions
1071    
1072    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1073
1074commit 0ce4425826074b12a58ced432f9d6fb346474737
1075Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1076Date:   Sat Apr 16 12:20:51 2022 -0700
1077
1078    XKBgeom.h: Remove XKB_IN_SERVER section
1079    
1080    The Xserver made its own copy of this file in 2008, and the API's are
1081    no longer the same between the server and client forks.
1082    
1083    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1084
1085commit be8c01946ab1a1d4b79fdc5358541d630f14dc0a
1086Author: Benno Schulenberg <bensberg@telfort.nl>
1087Date:   Wed Apr 20 09:29:09 2022 +0200
1088
1089    lcCT: use the correct index for checking the second byte
1090    
1091    (Not that it will make any difference, as the checking of these
1092    high bits looks like an excess of precaution.)
1093    
1094    This fixes issue #134.
1095    
1096    Reported-by: Rafał Mikrut
1097    
1098    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1099
1100commit 488b156fe2cc8aca6946a49236ec7b7698fceda4
1101Author: Benno Schulenberg <bensberg@telfort.nl>
1102Date:   Fri Feb 25 16:35:46 2022 +0100
1103
1104    nls: make the Ethiopian compose sequences actually work
1105    
1106    In the Ethiopian keyboard layout, the dead vowel keys do not produce <e>
1107    and <u> and so on, but instead produce <U+FE69> and <U+FE75> and so on,
1108    so the compose sequences should use those latter code points.
1109    
1110    Also, include the basic compose sequences from en_US.UTF-8/Compose,
1111    so that, when switching to a different layout in the Ethiopian locale,
1112    all the usual compose sequences work too.
1113    
1114    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1115
1116commit e6f98a3dbd82928e10adaf9d034dc574af1de6e5
1117Author: Benno Schulenberg <bensberg@telfort.nl>
1118Date:   Mon Apr 18 17:24:55 2022 +0200
1119
1120    nls: add Multi-key sequences for abovedot, diaeresis, and ogonek
1121    
1122    These accents by themselves could only be produced when one had a
1123    dead key for them, not with the help of the Multi key.
1124    
1125    [Note that the sequences <dead_acute> <space> for apostrophe (')
1126    and <dead_diaeresis> <space> for double quote (") are anomalies,
1127    as normally <dead_accent> <space> produces the accent itself.]
1128    
1129    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1130
1131commit d323704e7c57f9ac7fdfea3dc9fa73bf943e72d3
1132Author: Benno Schulenberg <bensberg@telfort.nl>
1133Date:   Mon Apr 18 16:57:35 2022 +0200
1134
1135    nls: remove eight pointless self-producing compose sequences
1136    
1137    Compose sequences are meant to produce certain symbols by combining
1138    certain different symbols, not to produce a symbol with the help of
1139    the symbol itself.
1140    
1141    This fixes issue #59.
1142    
1143    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1144
1145commit 3b62c0322c112c03597d850197dbce22d1830bb7
1146Author: Benno Schulenberg <bensberg@telfort.nl>
1147Date:   Fri Feb 25 14:12:02 2022 +0100
1148
1149    nls: remove misplaced sequences from Ethiopian and Greek compose files
1150    
1151    The Khmer digraphs and Arabic ligatures have nothing to do with
1152    Amharic or Greek.
1153    
1154    (Also rewrap a comment and correct two others.)
1155    
1156    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1157
1158commit 479af78f2e5caaf9b98a35ae5ff4ccdd2f040289
1159Author: Benno Schulenberg <bensberg@telfort.nl>
1160Date:   Fri Feb 25 14:05:35 2022 +0100
1161
1162    nls: remove redundant Khmer digraphs, already defined in included file
1163    
1164    This fixes issue #160.
1165    
1166    Reported-by: Mike Fabian
1167    
1168    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1169
1170commit 6d7d08726f4b0c517041842b27cd7e66e8f371eb
1171Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1172Date:   Sun Apr 10 14:51:55 2022 -0700
1173
1174    XkbUpdateKeyTypeVirtualMods: always initialize mask
1175    
1176    XkbVirtualModsToReal should only fail to set mask if the server does
1177    not support XKB, but it still made Oracle Parfait complain:
1178    
1179    Error: Uninitialised memory
1180       Uninitialised memory variable [uninitialised-mem-var] (CWE 457):
1181          Possible access to uninitialised memory referenced by variable 'mask'
1182            at line 863 of lib/libX11/src/xkb/XKBMisc.c in function 'XkbUpdateKeyTypeVirtualMods'.
1183            Path in callee avoiding write at line 862
1184              mask allocated at line 860
1185    
1186    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1187
1188commit 9ac6859c20be2fc5e70c2908de60c6e466ec04e1
1189Author: Matt Turner <mattst88@gmail.com>
1190Date:   Sun Apr 3 10:15:30 2022 -0700
1191
1192    libX11 1.7.5
1193    
1194    Signed-off-by: Matt Turner <mattst88@gmail.com>
1195
1196commit 76d1cc3c1ce943c6ff81dc8c62a1d1b30fabf02e
1197Author: Simon McVittie <smcv@debian.org>
1198Date:   Sun Apr 3 14:23:36 2022 +0100
1199
1200    Don't try to destroy NULL condition variables
1201    
1202    This avoids a segfault during error-unwinding if an invalid display name
1203    is passed to XOpenDisplay().
1204    
1205    Fixes: 8a368d80 "Fix two memory leaks in _XFreeX11XCBStructure()"
1206    Resolves: #155
1207    Signed-off-by: Simon McVittie <smcv@debian.org>
1208
1209commit 80b30d12519ac54fc5b2b05500521d9c8269be9c
1210Author: Matt Turner <mattst88@gmail.com>
1211Date:   Wed Mar 30 13:54:13 2022 -0700
1212
1213    libX11 1.7.4
1214    
1215    Signed-off-by: Matt Turner <mattst88@gmail.com>
1216
1217commit 8a368d808fec166b5fb3dfe6312aab22c7ee20af
1218Author: Hodong <hodong@yozmos.com>
1219Date:   Thu Jan 20 00:57:41 2022 +0900
1220
1221    Fix two memory leaks in _XFreeX11XCBStructure()
1222    
1223    Even when XCloseDisplay() was called, some memory was leaked.
1224    
1225    XCloseDisplay() calls _XFreeDisplayStructure(), which calls
1226    _XFreeX11XCBStructure().
1227    
1228    However, _XFreeX11XCBStructure() did not destroy the condition variables,
1229    resulting in the leaking of some 40 bytes.
1230    
1231    Signed-off-by: Hodong <hodong@yozmos.com>
1232
1233commit da97120f2322882a17f14e5d7da00c1e772679e8
1234Author: Adam Jackson <ajax@redhat.com>
1235Date:   Tue Mar 22 14:32:49 2022 -0400
1236
1237    xcb: Clarify the XInitThreads error message
1238    
1239    XInitThreads has been called if _Xglobal_lock != NULL, we may as well
1240    check that before printing a misleading error message.
1241    
1242    Signed-off-by: Adam Jackson <ajax@redhat.com>
1243
1244commit 257927c51b08242aa5bf239346717fc817b2b286
1245Author: Giovanni Mascellani <gmascellani@codeweavers.com>
1246Date:   Fri Feb 4 10:49:25 2022 +0100
1247
1248    xcb_io: Allow jumps backwards when widening the request number.
1249    
1250    Request numbers are not always seen in the numeric order by widen(),
1251    for example due to Mesa directly calling _XError(). When this happens,
1252    widen() adds 2^32 to the reported widened number, triggering failed
1253    assertions and bad behavior.
1254    
1255    With this commit, wrapping of the lower dword is detected in a more
1256    robust way, by requiring that a skip of at least 2^31 is seen.
1257    
1258    This fixes issue #152.
1259    
1260    Signed-off-by: Giovanni Mascellani <gmascellani@codeweavers.com>
1261
1262commit 918063298cb893bee98040c9dca45ccdb2864773
1263Author: Benno Schulenberg <bensberg@telfort.nl>
1264Date:   Thu Mar 3 12:00:45 2022 +0100
1265
1266    nls: add a compose sequence for the peace symbol
1267    
1268    As the thing in the circle looks a bit like an upside-down Y,
1269    use <O> <Y> as the sequence, similar to <O> <A> for anarchism.
1270    
1271    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1272
1273commit 4f9eb32709dca705f02ca771bbd47d6c08aba5e0
1274Author: Benno Schulenberg <bensberg@telfort.nl>
1275Date:   Fri Feb 25 12:45:06 2022 +0100
1276
1277    nls: remove obsolete locale aliases with uppercase in the language code
1278    
1279    (These aliases may have been useful on some systems twenty years ago,
1280    but nowadays all should have settled on lowercase language code plus
1281    uppercase country code.)
1282    
1283    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1284
1285commit ba0d3b437deb2212bc7fc7c6f1c603fa6ced443c
1286Author: Benno Schulenberg <bensberg@telfort.nl>
1287Date:   Fri Feb 25 12:35:28 2022 +0100
1288
1289    nls: map eo.UTF-8 directly to en_US.UTF-8 instead of using a relay alias
1290    
1291    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1292
1293commit cfddb79981f23650feb0f9a0f4442c9cad584b33
1294Author: Benno Schulenberg <bensberg@telfort.nl>
1295Date:   Fri Feb 25 12:12:05 2022 +0100
1296
1297    nls: remove the mistaken eo_EO names -- EO is not a valid country code
1298    
1299    (Debian has had several of these lines commented out for years.)
1300    
1301    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1302
1303commit eb5885245a1ba6955d058a03e31894c23b9f5490
1304Author: iyzana <iyzana@protonmail.com>
1305Date:   Mon Feb 21 13:20:23 2022 +0100
1306
1307    nls: add locale alias from eo.UTF-8 to eo_XX.UTF-8
1308    
1309    There is no matching locale in the locale.dir and compose.dir files for
1310    the locale eo.UTF-8. Setting it as the locale resulted in compose files
1311    not being loaded.
1312    
1313    Signed-off-by: iyzana <iyzana@protonmail.com>
1314
1315commit 24df23dff9d8c1cf5e918a5f040b852bdf2cfb84
1316Author: Benno Schulenberg <bensberg@telfort.nl>
1317Date:   Thu Feb 3 17:19:05 2022 +0100
1318
1319    nls: group the compose sequences for tilde, degree, and cedilla together
1320    
1321    Also, correct the comment for the double quote character ("), and move
1322    the soft hyphen and the interrobangs to a special punctuation group.
1323    
1324    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1325
1326commit c0191f020f94907770937bc8ae67ea8acd7d63be
1327Author: Benno Schulenberg <bensberg@telfort.nl>
1328Date:   Thu Feb 3 16:33:05 2022 +0100
1329
1330    nls: use shape-based compose sequences for hammer-and-sickle instead
1331    
1332    Using a misspelled abbreviation (CCCP) of an obsolete country's name
1333    to compose a fairly general communist symbol does not seem right.
1334    
1335        https://en.wikipedia.org/wiki/Hammer_and_sickle
1336    
1337    Many compose sequences instead combine characters that together look
1338    similar to the target character.  Do the same here: question mark (?)
1339    plus backslash (\) look a bit like ☭ .
1340    
1341    In the bargain, this fixes issue #63.
1342    
1343    Reported-by: Marc Mezzarobba
1344    
1345    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1346
1347commit 582dc6f89e1f9288710a55cb2b8fbf2af99d7616
1348Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1349Date:   Sun Feb 20 11:49:42 2022 -0800
1350
1351    gitlab CI: enable Static Application Security Testing (SAST)
1352    
1353    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1354
1355commit 336e8f72559b489396c5cd503c4586ef800c2a8f
1356Author: Benno Schulenberg <bensberg@telfort.nl>
1357Date:   Sun Feb 20 15:06:02 2022 +0100
1358
1359    nls: add compose sequences for leftward, upward and downward double arrow
1360    
1361    These sequences follow the logic of: the arrow shaft (=) first, the
1362    arrow head (< or > or ^ or v) second.  This is so because the sequence
1363    <= is already taken for ≤, and there is no obvious ASCII character for
1364    a double vertical line, so we have to make do with a horizontal one.
1365    
1366    Since commit 6101b967b6 from eight years ago, there is a compose sequence
1367    for the rightward double arrow, but not for the other three directions.
1368    
1369    This fixes issue #138.
1370    
1371    Requested-by: Mélanie Chauvel
1372    
1373    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1374
1375commit 623b77d4f30b47258a40f89262e5aa5d25e95fa7
1376Author: Benno Schulenberg <bensberg@telfort.nl>
1377Date:   Mon Feb 14 11:33:25 2022 +0100
1378
1379    imDefLkup: verify that a pointer isn't NULL before using it
1380    
1381    It is possible for _XimICOfXICID() to return NULL, so it is necessary
1382    to check this isn't actually the case before dereferencing the pointer.
1383    All other callers of _XimICOfXICID() do this check too.
1384    
1385    (The check itself is ugly, but it follows the style of the code in the
1386    rest of the module.)
1387    
1388    Fixes issue #45.
1389    
1390    Reported-by: Bhavi Dhingra
1391    
1392    Original-patch-by: Bhavi Dhingra
1393    
1394    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1395
1396commit b83938f7f97aa6210b00835002e6e7d614b2374a
1397Author: Christopher Chavez <chrischavez@gmx.us>
1398Date:   Mon Feb 14 01:00:45 2022 +0000
1399
1400    Xlib.h: fix spelling in comment
1401    
1402    Signed-off-by: Christopher Chavez <chrischavez@gmx.us>
1403
1404commit 4791055682552b4cc8e775479700fb06980d0b59
1405Author: Christopher Chavez <chrischavez@gmx.us>
1406Date:   Sun Feb 13 05:26:14 2022 +0000
1407
1408    Xlib.h: spelling fix in comment
1409
1410commit 9304e645bdcdd3a71ff449a1157c4815e6304d7a
1411Author: Denis Drakhnia <numas13@gmail.com>
1412Date:   Wed Feb 2 11:54:43 2022 +0200
1413
1414    nls: Add ru_UA.utf8 locale alias.
1415    
1416    Signed-off-by: Denis Drakhnia <numas13@gmail.com>
1417
1418commit 9ef2551f0c201f9163b66d17189a98076b5be07a
1419Author: Benno Schulenberg <bensberg@telfort.nl>
1420Date:   Wed Feb 2 16:50:35 2022 +0100
1421
1422    nls: do some manual adjustment of whitespace in the compose sequences
1423    
1424    This achieves a more consistent vertical alignment (per group) of the
1425    target character and the trailing comments.
1426    
1427    (Most blocks were aligned in some fashion, except the "Greek Extended"
1428    block, which has lines of greatly varying lengths.)
1429    
1430    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1431
1432commit 0dad289ead6eb3412027c9365192441cc539d747
1433Author: Benno Schulenberg <bensberg@telfort.nl>
1434Date:   Wed Feb 2 14:08:22 2022 +0100
1435
1436    nls: among the compose sequences, replace all runs of spaces with tabs
1437    
1438    It was done with this script:
1439    
1440        cat nls/en_US.UTF-8/Compose.pre |
1441        sed 's!\(^<.\{6,13\}>\)\s*:!\1\t\t\t\t:!' |
1442        sed 's!\(^<.\{14,21\}>\)\s*:!\1\t\t\t:!' |
1443        sed 's!\(^<.\{22,29\}>\)\s*:!\1\t\t:!' |
1444        sed 's!\(^<.\{30,37\}>\)\s*:!\1\t:!' |
1445        sed 's!\(: ".*"\)\s*!\1\t!' |
1446        sed 's!\(\s*# \)! # !' > trimmed &&
1447        mv  trimmed  nls/en_US.UTF-8/Compose.pre
1448    
1449    This saves 37 kilobytes of whitespace.
1450    
1451    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1452
1453commit fc79e39c236501357e1fdbe8964882864eeef5d3
1454Author: Benno Schulenberg <bensberg@telfort.nl>
1455Date:   Wed Feb 2 14:06:29 2022 +0100
1456
1457    docs: add missing word, wrap line, fix typo, and trim an excess space
1458    
1459    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1460
1461commit 93446455c46dfc019984ab822ffa662cf6e2949c
1462Author: Benno Schulenberg <bensberg@telfort.nl>
1463Date:   Thu Jan 27 15:01:18 2022 +0100
1464
1465    KeyBind: reshuffle two ifs into the general order of ascending codes
1466    
1467    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1468
1469commit 93f15f5f29378ec4cc1dd011b155104e9a52b058
1470Author: Benno Schulenberg <bensberg@telfort.nl>
1471Date:   Thu Jan 27 14:27:41 2022 +0100
1472
1473    KeyBind: slightly speed up some case conversions by adding missing returns
1474    
1475    This avoids pointlessly tumbling through several more ifs when the
1476    conversion has aleady been done.
1477    
1478    Also remove two redundant conditions (as lower codes have already
1479    been handled) and fold two other conditions together.
1480    
1481    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1482
1483commit 317c7abbcc848011393d1336b9d020af8cabb000
1484Author: Benno Schulenberg <bensberg@telfort.nl>
1485Date:   Thu Jan 27 13:54:29 2022 +0100
1486
1487    KeyBind: update the Greek case-conversion tables to Unicode Data 14.0
1488    
1489    A few symbols, like Heta and Sampi and dotted lunate Sigma, have been
1490    assigned code points since Unicode Data 4.0, and need their entries.
1491    
1492    This fixes issue #132.
1493    
1494    Reported-by: Ray Vine
1495    
1496    Original-patch-by: Ray Vine
1497    
1498    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1499
1500commit 402b843fa72166d9ceefd887f5cea4728cbc6995
1501Author: Benno Schulenberg <bensberg@telfort.nl>
1502Date:   Wed Jan 26 17:20:28 2022 +0100
1503
1504    remove a commented-out code fragment, and remove a stray blank line
1505    
1506    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1507
1508commit fbbc3cf8b8b03f74f778772d4703fc4fec242282
1509Author: Benno Schulenberg <bensberg@telfort.nl>
1510Date:   Wed Jan 26 16:55:50 2022 +0100
1511
1512    XKBCvt: remove an if that never gets triggered
1513    
1514    The 'if' at the beginning of _XkbHandleSpecialSym() allows only symbols
1515    from the numeric keypad and some control symbols to pass -- XK_hyphen
1516    is not among them, so the check for XK_hyphen in the later 'if' will
1517    always be false.
1518    
1519    (The corresponding conversion in _XTranslateKeySym() in KeyBind.c was
1520    removed in March 1994.)
1521    
1522    (Also, several keyboard layouts nowadays contain the 'hyphen' symbol,
1523    allowing the user to enter soft hyphens into a document.  So we really
1524    don't want to remap this symbol.)
1525    
1526    Fixes issue #48.
1527    
1528    Reported-by: Bhavi Dhingra
1529    
1530    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1531
1532commit 6786744f636b3fc545157a13cb2f6169b5be9fa5
1533Author: Benno Schulenberg <bensberg@telfort.nl>
1534Date:   Sun Jan 23 17:12:03 2022 +0100
1535
1536    nls: add the es_CU locale name, for Cuba
1537    
1538    It was added to glibc ten years ago.
1539    
1540    This fixes issue #57.
1541    
1542    Reported-by: Dominique Michel
1543    
1544    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1545
1546commit d241d8af7f24e95949c2cf7ce791e35df7665e98
1547Author: Benno Schulenberg <bensberg@telfort.nl>
1548Date:   Sun Jan 23 16:44:03 2022 +0100
1549
1550    nls: remove some obsolete Norwegian and Yugoslavian locale names
1551    
1552    They have not existed in glibc for more than fifteen years.
1553    
1554    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1555
1556commit 32dd7c03e92e01015744867b9b3148c250cc159a
1557Author: Benno Schulenberg <bensberg@telfort.nl>
1558Date:   Sat Jan 22 17:56:32 2022 +0100
1559
1560    nls: put some entries in their alphabetical order, and adjust alignment
1561    
1562    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1563
1564commit 46a066f25103bb909259baea25de8dd5c4a2b338
1565Author: Benno Schulenberg <bensberg@telfort.nl>
1566Date:   Sat Jan 22 18:00:20 2022 +0100
1567
1568    nls: fix a typo (be_BG => bg_BG) and drop an obsolete comment
1569    
1570    The typo was only partially fixed in commit 22a5255b80 sixteen years ago,
1571    and the internet shows that there was at least one user frustrated that
1572    bg_BG.utf8 didn't work but bg_BG.UTF-8 did.
1573    
1574    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1575
1576commit 84efc0aa598c8f7d6dd9a49e7450e85c8bc4773b
1577Author: Benno Schulenberg <bensberg@telfort.nl>
1578Date:   Wed Jan 19 15:32:15 2022 +0100
1579
1580    nls: remove the compose sequences with the deviant KP_Space symbol
1581    
1582    Using this KP_Space symbol as a stand-in for the symbol 2 doesn't make
1583    sense.  It looks like a mistake, or as if someone had a broken keyboard
1584    and used KP_Space as a substitute for 2.
1585    
1586    Also, no keyboard layout in the last fifteen years has contained the
1587    KP_Space symbol, so I don't see how anyone could type it.
1588    
1589    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1590
1591commit b76e661ca223489da9027b0ab1ca1586efd0a280
1592Author: Benno Schulenberg <bensberg@telfort.nl>
1593Date:   Wed Jan 19 13:49:56 2022 +0100
1594
1595    nls: move the block of Jamo to its logical place (ascending Unicodes)
1596    
1597    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1598
1599commit 21f82fb9c69f8fdbf0bdc1aac6f8c11c2c6d55d1
1600Author: Benno Schulenberg <bensberg@telfort.nl>
1601Date:   Wed Jan 19 13:40:56 2022 +0100
1602
1603    nls: delete some ineffective compose sequences (that were commented out)
1604    
1605    They have never worked; there is no point in keeping them in the file.
1606    
1607    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1608
1609commit 6cc31eafa8095d2117ff1906d7c14c1c01173248
1610Author: Benno Schulenberg <bensberg@telfort.nl>
1611Date:   Wed Jan 19 12:18:21 2022 +0100
1612
1613    nls: group some compose sequences better, and add some comments
1614    
1615    Also, align a few trailing comments more consistently, and change
1616    some comments so they will be excluded from the installed file --
1617    there is no need for those comments there.
1618    
1619    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1620
1621commit d60ede7843612f3c2967f83d629bf8272ba09ff1
1622Author: ArenaL5 <arenal5@protonmail.com>
1623Date:   Thu Jan 13 01:13:56 2022 +0100
1624
1625    nls: add a compose sequence for the copyleft symbol
1626    
1627    Probably not used anywhere yet, let alone seriously, as stated in
1628    https://en.wikipedia.org/wiki/Copyleft#Symbol
1629    
1630    Signed-off-by: ArenaL5 <arenal5@protonmail.com>
1631
1632commit df4dcdf9125c595f6df3a302356647cb8df38553
1633Author: Benno Schulenberg <bensberg@telfort.nl>
1634Date:   Mon Jan 17 17:31:08 2022 +0100
1635
1636    nls: remove redundant compose sequences with deprecated Hebrew symbols
1637    
1638    The compose sequences with the valid Hebrew key symbols are right there.
1639    
1640    Also, no keyboard layout in the past seventeen years has contained these
1641    deprecated symbols on any key.
1642    
1643    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1644
1645commit 4268472bd11a07a799716fb5833ea46cad3dd479
1646Author: Benno Schulenberg <bensberg@telfort.nl>
1647Date:   Sun Jan 16 17:51:29 2022 +0100
1648
1649    nls: remove redundant compose sequences with the deprecated 'underbar'
1650    
1651    The key symbol 'underbar' is equivalent to 'underscore' -- the mirror
1652    compose sequences with the latter symbol continue to exist.
1653    
1654    Also, no keyboard layout in the past fifteen years has contained the
1655    symbol 'underbar' on any key -- except a Telugu layout since a few
1656    years, but that will be corrected soon.
1657    
1658    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1659
1660commit 86e9749ef37189486c205ee7f77e43fed655ac70
1661Author: ArenaL5 <arenal5@protonmail.com>
1662Date:   Sat Jan 15 16:20:56 2022 +0100
1663
1664    nls: add a compose sequence for the middle dot (·) using dead_circumflex
1665    
1666    This makes the middle dot compose sequences consistent with others that
1667    begin with <dead_circumflex> and <Multi_key> <asciicircum>, e.g.
1668    superscripted 3 (³)
1669    
1670    Signed-off-by: ArenaL5 <arenal5@protonmail.com>
1671
1672commit 1adb531cd5be2b6489c2a8106e850e7418fae056
1673Author: Benno Schulenberg <bensberg@telfort.nl>
1674Date:   Fri Jan 14 13:06:10 2022 +0100
1675
1676    docs: unwrap the comments of structure elements in the man pages
1677    
1678    It doesn't matter that these lines get wider than 80 columns, as it's
1679    easy enough to make the pager scroll a bit to the right, or to have a
1680    terminal that is a 100 or more columns wide.  It looks better to not
1681    hard-wrap these comments.
1682    
1683    Also, fix the layout of two table elements.
1684    
1685    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1686
1687commit da330a1644f10f9669c8dd9ee06489c40f9f0bd1
1688Author: Benno Schulenberg <bensberg@telfort.nl>
1689Date:   Fri Jan 14 12:33:23 2022 +0100
1690
1691    docs: unwrap some mistakenly wrapped lines, and correct a comment
1692    
1693    The wrapped lines messed up the table.  Also, add some blank lines
1694    to make the table more readable.
1695    
1696    Elsewhere, add spaces around an operator (for consistency),
1697    and correct a copy-paste error.
1698    
1699    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1700
1701commit b9e0346abebb0c9e5c80c780bba4372419e63bde
1702Author: Benno Schulenberg <bensberg@telfort.nl>
1703Date:   Fri Jan 14 15:57:56 2022 +0100
1704
1705    XKBMAlloc: use the correct first-key variable when extending a key range
1706    
1707    When extending the range of changed virtual-modifier keys, one should
1708    pass the two variables relevant to that range.
1709    
1710    These two mistakes were found by grepping for 'modmap_key' and then
1711    looking for asymmetries between the handling of the plain modmap thing
1712    and the virtual modmap thing.  Inspired by issue #76.
1713    
1714    (I suspect these mistakes were at the base of some intractable bugs
1715    reported against xkeyboard-config where virtual modifiers refused
1716    to work and had to be worked around by making other changes too.)
1717    
1718    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1719
1720commit f15d2a01f75304bc2d9dac109d9a5a47d1ce28e0
1721Author: Benno Schulenberg <bensberg@telfort.nl>
1722Date:   Wed Jan 12 11:34:23 2022 +0100
1723
1724    lcUTF8: remove two unneeded ifs and two elses that are never reached
1725    
1726    When there is a break at the end of the while loop, it means that the
1727    loop will finish the first time that _XlcGetCharSetWithSide() returns
1728    a non-NULL result, which means that the first '(last_charset == NULL)'
1729    will always be true and the else part is redundant.
1730    
1731    Fixes issue #46.
1732    
1733    Reported-by: Bhavi Dhingra
1734    
1735    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
1736
1737commit 033f37d507801645a8251fd33c25ebd145d22cce
1738Author: Antti Savolainen <antti.savo@gmail.com>
1739Date:   Wed Jan 12 16:10:21 2022 +0000
1740
1741    nls: add two compose sequences for an upwards arrow using dead_circumflex
1742    
1743    Signed-off-by: Antti Savolainen <antti.savo@gmail.com>
1744
1745commit b7d724df615311c35418afc733e1fcc68c740aba
1746Author: Hodong <hodong@nimfsoft.com>
1747Date:   Fri Jan 7 10:41:15 2022 +0000
1748
1749    specs/XIM: fix a missing "EXT" in the documentation
1750    
1751    There was an "EXT" in the original text, but it seems to be missing.
1752    
1753    See: https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/4e66da0783b2e5e3b288aaecd3c89396ed425c20/specs/XIM/xim.ms#L693
1754    
1755    Signed-off-by: Hodong Kim <hodong@nimfsoft.com>
1756
1757commit 4c96f3567a8d045ee57b886fddc9618b71282530
1758Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1759Date:   Thu Dec 9 19:51:52 2021 -0800
1760
1761    libX11 1.7.3.1
1762    
1763    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1764
1765commit c1710589bb00d6f442a383fc41f0d6496b436c88
1766Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1767Date:   Tue Dec 7 08:29:43 2021 -0800
1768
1769    specs/XIM: Fix "encording" typos
1770    
1771    Fixes: #145
1772    
1773    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1774
1775commit e30771b4bbdd84fc7ff64a1cda483a46edb5707b
1776Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1777Date:   Mon Dec 6 17:04:14 2021 -0800
1778
1779    libX11 1.7.3
1780    
1781    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1782
1783commit 934417b1e89fe715b9cf6c0d62a3867fa1f5bccb
1784Author: Liang Chang <l-chang@users.sourceforge.net>
1785Date:   Tue Sep 7 18:54:10 2021 +0800
1786
1787    Make Xutf8DrawString works correctly with ISO10646 fonts.
1788    
1789    Signed-off-by: Liang Chang <l-chang@users.sourceforge.net>
1790
1791commit 3ad36c5d3ddca3e4c97faaf87b72a53954d1f473
1792Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1793Date:   Thu Dec 2 17:48:55 2021 -0800
1794
1795    Build xz tarballs instead of bzip2
1796    
1797    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1798
1799commit 93a050c3ad2d2264d3880db3791387b1a9bf2e9e
1800Author: Tatu Frisk <tatu.frisk@ge.com>
1801Date:   Tue Mar 14 14:41:27 2017 +0200
1802
1803    Fix hanging issue in _XReply
1804    
1805    Assume event queue is empty if another thread is blocking waiting for event.
1806    
1807    If one thread was blocking waiting for an event and another thread sent a
1808    reply to the X server, both threads got blocked until an event was
1809    received.
1810    
1811    Signed-off-by: Tatu Frisk <tatu.frisk@ge.com>
1812    Signed-off-by: Jose Alarcon <jose.alarcon@ge.com>
1813
1814commit 8382253010267b93c9aeffc769e3f5e43d698094
1815Author: Matthieu Herrb <matthieu@herrb.eu>
1816Date:   Mon Aug 30 21:47:09 2021 +0200
1817
1818    Avoid NULL pointer deref. Fixes issue #47.
1819    
1820    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
1821
1822commit 58d7b540c453601faf942c6f19f8277a5c5f78cc
1823Author: Peter Hutterer <peter.hutterer@who-t.net>
1824Date:   Wed Dec 1 14:31:16 2021 +1000
1825
1826    gitlab CI: update to use the ci-fairy image
1827    
1828    Let's use the pre-prepared image instead of building our own.
1829    
1830    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1831
1832commit 150154c994829f9f3734a560f99f70b21d86b49c
1833Author: Hodong <hodong@nimfsoft.com>
1834Date:   Sun Nov 14 06:36:42 2021 +0000
1835
1836    Fix a typo
1837
1838commit 2356e59ff24f8d1b25cdc4dffc5171c65dc2b86e
1839Author: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
1840Date:   Sun Sep 5 10:25:38 2021 +0300
1841
1842    libX11: do not crash in GetResReq() macro
1843    
1844    When _XGetRequest() detects that requested length exceeds remaining display
1845    output buffer capacity it would return NULL. GetResReq() macro obtains "req"
1846    pointer from a call to _XGetRequest() and then proceeds to assign request id
1847    through "req" pointer which leads to NULL pointer dereference in this case.
1848    
1849    Fix this by checking if "req" is valid before assigning request id.
1850    
1851    Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
1852
1853commit e92efc63acd7b377faa9e534f4bf52aaa86be2a9
1854Author: Peter Hutterer <peter.hutterer@who-t.net>
1855Date:   Tue Jul 27 11:46:19 2021 +1000
1856
1857    makekeys: handle the new _EVDEVK xorgproto symbols
1858    
1859    These keys are all defined through a macro in the form:
1860       #define XF86XK_BrightnessAuto                _EVDEVK(0x0F4)
1861    
1862    The _EVDEVK macro is simply an offset of 0x10081000.
1863    Let's parse these lines correctly so those keysyms end up in our
1864    hashtables.
1865    
1866    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
1867
1868commit d01d23374107f6fc55511f02559cf75be7bdf448
1869Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
1870Date:   Wed Jun 16 12:17:04 2021 +0100
1871
1872    Avoid undefined behaviour after realloc()
1873    
1874    Adding the offset between the realloc result and the old allocation to
1875    update pointers into the new allocation is undefined behaviour: the
1876    old pointers are no longer valid after realloc() according to the C
1877    standard. While this works on almost all architectures and compilers,
1878    it causes  problems on architectures that track pointer bounds (e.g.
1879    CHERI or Arm's Morello): the value_list pointers will still have the
1880    bounds of the previous allocation and therefore any dereference will
1881    result in a run-time trap.
1882    
1883    I found this due to a crash (dereferencing an invalid capability) while
1884    trying to run `xev` over SSH on a CHERI-RISC-V system. With these two
1885    realloc changes, and https://gitlab.freedesktop.org/xorg/proto/xorgproto/-/merge_requests/41
1886    I am able to succesfully run `xev` compiled for CHERI-RISC-V.
1887    
1888    Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
1889
1890commit 1c845834a32421abb59408aee259ac0bef054d30
1891Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
1892Date:   Tue Jun 15 08:46:53 2021 +0100
1893
1894    Fix cross-compilation from macOS
1895    
1896    We can't use `LC_CTYPE=C sed` there since /usr/bin/sed is not compatible
1897    with the expressions in nls/ (`sed: RE error: illegal byte sequence`).
1898    To fix this use $(SED) instead which autotools will set to a GNU
1899    version of sed (usually /usr/local/bin/gsed) on macOS.
1900    
1901    Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk>
1902
1903commit 401f58f8ba258d4e7ce56a8f756595b72e544c15
1904Author: Lev Korol <epicatsupercell@gmail.com>
1905Date:   Sun Apr 11 21:57:45 2021 +0300
1906
1907    nls: Add en_IL locale
1908    
1909    Signed-off-by: Lev Korol <epicatsupercell@gmail.com>
1910
1911commit f906fe8e9769e4313294b68e61c402610ade69da
1912Author: Alan Coopersmith <alan.coopersmith@oracle.com>
1913Date:   Sat Jun 5 09:12:13 2021 -0700
1914
1915    libX11 1.7.2
1916    
1917    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
1918
1919commit 51b73ac0acda65005c8a9f17ca4ea7281b00ca84
1920Author: Tobias Stoeckmann <tobias@stoeckmann.org>
1921Date:   Sun May 23 15:05:17 2021 +0200
1922
1923    Protect against overly long strings
1924    
1925    Checking against upper limit of USHRT_MAX must happen before truncating
1926    size_t to int. On 64 bit systems with strings larger than 2 GB this
1927    could otherwise lead to negative ints or ints smaller than USHRT_MAX.
1928    
1929    In XParseColor this could lead to out of boundary access with strings
1930    starting with a # (color sequence). A modulo 12 operation is performed
1931    to validate the string length, but with an overflown length, the for
1932    loop would eventually read behind terminating '\0' character.
1933    
1934    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
1935
1936commit ab2f59530b16bdfbf023b8e025c7c8aba3b6fd0c
1937Author: Matthieu Herrb <matthieu@herrb.eu>
1938Date:   Sun May 30 16:33:48 2021 +0200
1939
1940    Check for NULL strings before getting their lengths
1941    
1942    Problem reported by Karsten Trulsen
1943    
1944    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
1945
1946commit a8216e8653a246db664e1443ecf707b27de65067
1947Author: Matthieu Herrb <matthieu@herrb.eu>
1948Date:   Sat May 22 17:35:47 2021 +0200
1949
1950    include <limits.h> always, not if HAVE_CONFIG_H is set.
1951    
1952    via Matthew Green from the NetBSD project
1953    
1954    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
1955
1956commit 6953a586df4819143c4d55e011b3a5e5377981b8
1957Author: Matthieu Herrb <matthieu.herrb@laas.fr>
1958Date:   Tue May 18 15:27:58 2021 +0200
1959
1960    Version 1.7.1
1961    
1962    Release notes in README.md, version bump in configure.ac
1963
1964commit 8d2e02ae650f00c4a53deb625211a0527126c605
1965Author: Matthieu Herrb <matthieu@herrb.eu>
1966Date:   Fri Feb 19 15:30:39 2021 +0100
1967
1968    Reject string longer than USHRT_MAX before sending them on the wire
1969    
1970    The X protocol uses CARD16 values to represent the length so
1971    this would overflow.
1972    
1973    CVE-2021-31535
1974    
1975    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
1976
1977commit 838ea5a5a0267c25b20c095c9a70684edeeefba4
1978Author: Gaurav Ujjwal <gujjwal00@gmail.com>
1979Date:   Sun May 9 05:04:36 2021 +0000
1980
1981    Fix out-of-bound access in KeySymToUcs4()
1982    
1983    Array `keysym_to_unicode_590_5fe` is only valid for range  [0x590, 0x5fe] but current lower-bound is checked against 0x589.
1984    
1985    So invalid values from 0x58a to 0x58f are being allowed by current check.
1986    
1987    If any of these invalid value is passed as `keysym`,    `keysym - 0x590` would underflow.
1988    
1989    Signed-off-by: Gaurav Ujjwal <gujjwal00@gmail.com>
1990
1991commit 32491b02c76ac22f99dc433a32466e22bdc181a7
1992Author: Christopher Chavez <chrischavez@gmx.us>
1993Date:   Mon May 3 15:46:44 2021 +0000
1994
1995    Xlib.h: spelling fix in comment
1996
1997commit 5faa8dc0b364f55f19034c370436074646d9ad54
1998Author: Benno Schulenberg <bensberg@telfort.nl>
1999Date:   Tue Nov 24 18:54:24 2020 +0100
2000
2001    nls: adjust three comments about the APL compose sequences
2002    
2003    Commit 0bbc0d5e605e (from eight years ago) removed the lines that two
2004    of these comments referred to.  Without those lines, the comments don't
2005    make sense any more.  Reword and shorten them.
2006    
2007    Also reword a comment about two sequences that don't work.
2008    
2009    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2010
2011commit b126bfd7fe990b8c1df088f55c31d9b3fd440cde
2012Author: Benno Schulenberg <bensberg@telfort.nl>
2013Date:   Tue Nov 24 12:23:32 2020 +0100
2014
2015    nls: allow composing all breved letters also with a lowercase "u"
2016    
2017    The letters ă and ŭ can already be composed with "u a" and "u u", but
2018    ĕ, ğ, ĭ, and ŏ can be composed only with an uppercase U.  Emancipate
2019    the latter four and understand also a lowercase "u" to mean 'breve'.
2020    
2021    (Yesterday I needed ğ and was annoyed that "u g" did not work.)
2022    
2023    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2024
2025commit 78027fdb7aa07621a3b7cdf37050bae39012f2fb
2026Author: Walter Harms <wharms@bfs.de>
2027Date:   Sat Nov 28 21:05:33 2020 +0100
2028
2029    fix same roff code
2030    
2031    Signed-off-by: Walter Harms <wharms@bfs.de>
2032
2033commit 4f15cfc64548098fef3ddd24f7359b76cf1bbda7
2034Author: Walter Harms <wharms@bfs.de>
2035Date:   Sat Nov 28 20:56:35 2020 +0100
2036
2037    Fix some roff code add see also
2038    
2039    Signed-off-by: Walter Harms <wharms@bfs.de>
2040
2041commit daa106927f78e6c5c9c36558a4b951d7cb58a8d1
2042Author: Walter Harms <wharms@bfs.de>
2043Date:   Sat Nov 28 17:49:25 2020 +0100
2044
2045    fix broken nroff coding for code comments
2046    
2047    the comments /* */ are code as /\(**  */ that does not work.
2048    the coding in other X11 man pages is /\&* */ so we do the same here.
2049
2050commit 7bdeae239b17db002964ac73d06f43f216d1cf86
2051Author: Walter Harms <wharms@bfs.de>
2052Date:   Fri Nov 27 22:43:21 2020 +0100
2053
2054    FIX: warning: macro `hN' not defined
2055    
2056    Signed-off-by: Walter Harms <wharms@bfs.de>
2057
2058commit b7ec67d3ea78ab68857328d8f72d923324e68cc2
2059Author: Walter Harms <wharms@bfs.de>
2060Date:   Fri Nov 27 22:26:15 2020 +0100
2061
2062    FIX: warning: macro `IN' not defined
2063    
2064    just remove an other dead macro use.
2065
2066commit 7ca3ceb9175a829a40c347cb3b2e98341ec5a2df
2067Author: Walter Harms <wharms@bfs.de>
2068Date:   Fri Nov 27 21:58:04 2020 +0100
2069
2070    fix warning: macro `s' not defined
2071    
2072    this is caused by bad nroff coding, fix some more issues on the fly
2073
2074commit 14fb4e535626ba934cad5bc4308f511524cd1103
2075Author: Walter Harms <wharms@bfs.de>
2076Date:   Fri Nov 27 20:04:22 2020 +0100
2077
2078    FIX: warning: macro `hN' not defined
2079    
2080    this was found by checking man pages with
2081     groff -t -mandoc -Z  -wmac -Tutf8 $FILE >/dev/null
2082    
2083    In most cases .hN could be replaced with .BR
2084    
2085    Signed-off-by: Walter Harms <wharms@bfs.de>
2086
2087commit cb03da4447aa53e8e5382f124917d89b50148036
2088Author: Walter Harms <wharms@bfs.de>
2089Date:   Fri Nov 27 19:00:00 2020 +0100
2090
2091    FIX: warning: macro `Pn' not defined
2092    
2093    The missing macro is found via:
2094    roff -t -mandoc -Z  -wmac -Tutf8 XAnyEvent.man >/dev/null
2095    
2096    To fix the problem the macro is replaced with .RB.
2097    
2098    Signed-off-by: Walter Harms <wharms@bfs.de>
2099
2100commit cc9f8878f2cbe17c7b4035b4ff4352b52ece38e0
2101Author: Benno Schulenberg <bensberg@telfort.nl>
2102Date:   Wed Nov 25 16:58:27 2020 +0100
2103
2104    nls: add 'C.utf8' as an alias for 'en_US.UTF-8'
2105    
2106    The normal form is 'C.UTF-8', but 'C.utf8' has been seen in the wild.
2107    
2108    Fixes #102.
2109    
2110    Reported-by: Tomas Korbar
2111    
2112    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2113
2114commit ca8115186f810eccb7d86b0979980eff3ba95f0b
2115Author: Keith Packard <keithp@keithp.com>
2116Date:   Thu Nov 19 11:30:41 2020 -0800
2117
2118    Version 1.7.0
2119    
2120    Release notes in README.md, version bump in configure.ac
2121    
2122    Signed-off-by: Keith Packard <keithp@keithp.com>
2123
2124commit 325ac177fb6d38f4c7689d9ce059792eaef388ef
2125Author: Keith Packard <keithp@keithp.com>
2126Date:   Mon Nov 16 15:48:40 2020 -0800
2127
2128    Update libX11 .so version to 6.4.0
2129    
2130    There have been API additions since the previous release, so the
2131    .so version needs to change.
2132    
2133    Signed-off-by: Keith Packard <keithp@keithp.com>
2134
2135commit 103e2e11519984aa5746c06e7d90d1f5bc8174a3
2136Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2137Date:   Sat Oct 31 09:24:01 2020 -0700
2138
2139    Don't leave dangling pointers in Free functions
2140    
2141    While these are mostly called during teardown of larger structures
2142    that are about to themselves be freed, there's no guarantee that
2143    will always be the case, so try to be safer here.
2144    
2145    [ This bug was found by the Parfait 4.0 bug checking tool.
2146      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
2147    
2148    v2: Deduplicate & simplify pointer clearing in _XFreeEventCookies
2149        as suggested by @keithp
2150    
2151    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2152
2153commit 7d70e302218e2ec8518548ddcdc02d828e7889af
2154Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2155Date:   Sat Oct 31 08:56:34 2020 -0700
2156
2157    XimProto.h: wrap XIM_HEADER_SIZE definition in parens
2158    
2159    Resolves parfait warning of potential macro misinterpretation if
2160    expanded in the midst of other arithmetic operations with higher
2161    precedence.
2162    
2163    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2164
2165commit a3c0b5dbd6b12ae64bc78b11795647a7f6df0c7a
2166Author: Keith Packard <keithp@keithp.com>
2167Date:   Tue Nov 17 13:15:18 2020 -0800
2168
2169    Copy locale modifiers when creating XimInstCallback [v2]
2170    
2171    Locale modifiers may be freed whenever XSetLocaleModifiers gets
2172    called, even if the locale hasn't changed. This means that we cannot
2173    save a pointer to those modifiers in the XimInstCallback record and
2174    must, instead, make a copy of them instead.
2175    
2176    This fixes a problem uncovered when running wish under libasan as
2177    follows (on current Debian unstable):
2178    
2179            $ LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libasan.so.6 wish
2180    
2181    Reported-by: Vittorio Zecca <zeccav@gmail.com>
2182    Signed-off-by: Keith Packard <keithp@keithp.com>
2183    
2184    v2:
2185            Remove incorrect 'else' token found by @alanc
2186
2187commit 960e2e0cfac12c3477c672d0d40818a0dc74aca5
2188Author: Peter Hutterer <peter.hutterer@who-t.net>
2189Date:   Mon Nov 16 09:34:57 2020 +1000
2190
2191    gitlab CI: add a basic build test
2192    
2193    Using Arch as base distribution here because we can expect our dependencies to
2194    be up-to-date. We rely on the Arch for our dependencies rather than building
2195    those from git (notably: xorg-macros, xtrans and libxcb).
2196    
2197    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2198
2199commit dbb55e1a5e82870466b095097d9e46046680ec25
2200Author: Frediano Ziglio <fziglio@redhat.com>
2201Date:   Wed Jan 29 09:06:54 2020 +0000
2202
2203    Fix poll_for_response race condition
2204    
2205    In poll_for_response is it possible that event replies are skipped
2206    and a more up to date message reply is returned.
2207    This will cause next poll_for_event call to fail aborting the program.
2208    
2209    This was proved using some slow ssh tunnel or using some program
2210    to slow down server replies (I used a combination of xtrace and strace).
2211    
2212    How the race happens:
2213    - program enters into poll_for_response;
2214    - poll_for_event is called but the server didn't still send the reply;
2215    - pending_requests is not NULL because we send a request (see call
2216      to  append_pending_request in _XSend);
2217    - xcb_poll_for_reply64 is called from poll_for_response;
2218    - xcb_poll_for_reply64 will read from server, at this point
2219      server reply with an event (say sequence N) and the reply to our
2220      last request (say sequence N+1);
2221    - xcb_poll_for_reply64 returns the reply for the request we asked;
2222    - last_request_read is set to N+1 sequence in poll_for_response;
2223    - poll_for_response returns the response to the request;
2224    - poll_for_event is called (for instance from another poll_for_response);
2225    - event with sequence N is retrieved;
2226    - the N sequence is widen, however, as the "new" number computed from
2227      last_request_read is less than N the number is widened to N + 2^32
2228      (assuming last_request_read is still contained in 32 bit);
2229    - poll_for_event enters the nested if statement as req is NULL;
2230    - we compare the widen N (which now does not fit into 32 bit) with
2231      request (which fits into 32 bit) hitting the throw_thread_fail_assert.
2232    
2233    To avoid the race condition and to avoid the sequence to go back
2234    I check again for new events after getting the response and
2235    return this last event if present saving the reply to return it
2236    later.
2237    
2238    To test the race and the fix it's helpful to add a delay (I used a
2239    "usleep(5000)") before calling xcb_poll_for_reply64.
2240    
2241    Original patch written by Frediano Ziglio, see
2242    https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/34
2243    
2244    Reworked primarily for readability by Peter Hutterer, see
2245    https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/53
2246    
2247    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2248
2249commit 30ccef3a48029bf4fc31d4abda2d2778d0ad6277
2250Author: Keith Packard <keithp@keithp.com>
2251Date:   Sat Nov 7 22:22:47 2020 -0800
2252
2253    Avoid recursing through _XError due to sequence adjustment
2254    
2255    This patch is based on research done by Dmitry Osipenko to uncover the
2256    cause of a large class of Xlib lockups.
2257    
2258    _XError must unlock and re-lock the display around the call to the
2259    user error handler function. When re-locking the display, two
2260    functions are called to ensure that the display is ready to generate a request:
2261    
2262        _XIDHandler(dpy);
2263        _XSeqSyncFunction(dpy);
2264    
2265    The first ensures that there is at least one XID available to use
2266    (possibly calling _xcb_generate_id to do so). The second makes sure a
2267    reply is received at least every 65535 requests to keep sequence
2268    numbers in sync (possibly generating a GetInputFocus request and
2269    synchronously awaiting the reply).
2270    
2271    If the second of these does generate a GetInputFocus request and wait
2272    for the reply, then a pending error will cause recursion into _XError,
2273    which deadlocks the display.
2274    
2275    One seemingly easy fix is to have _XError avoid those calls by
2276    invoking InternalLockDisplay instead of LockDisplay. That function
2277    does everything that LockDisplay does *except* call those final two
2278    functions which may end up receiving an error.
2279    
2280    However, that doesn't protect the system from applications which call
2281    some legal Xlib function from within their error handler. Any Xlib
2282    function which cannot generate protocol or wait for events is valid,
2283    including many which invoke LockDisplay.
2284    
2285    What we need to do is make LockDisplay skip these two function calls
2286    precisely when it is called from within the _XError context for the
2287    same display.
2288    
2289    This patch accomplishes this by creating a list of threads in the
2290    display which are in _XError, and then having LockDisplay check the
2291    current thread against those list elements.
2292    
2293    Inspired-by: Dmitry Osipenko <digetx@gmail.com>
2294    Signed-off-by: Keith Packard <keithp@keithp.com>
2295    Tested-by: Dmitry Osipenko <digetx@gmail.com>
2296    Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
2297
2298commit c9c4d6efbf92ab51695e2e740319503221d68eed
2299Author: Benno Schulenberg <bensberg@telfort.nl>
2300Date:   Fri Nov 13 11:34:04 2020 +0100
2301
2302    nls: improve the comments for the Serbian compose sequences
2303    
2304    Also put an extra space before the lone combining characters
2305    so they have some room to breathe.
2306    
2307    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2308
2309commit 3497dae3d9c52a252110b2a0983b82000be0fe8e
2310Author: Benno Schulenberg <bensberg@telfort.nl>
2311Date:   Fri Nov 13 11:23:09 2020 +0100
2312
2313    nls: remove mistaken Serbian compose sequences with combining diacritics
2314    
2315    Combining characters are not dead keys -- they have an immediate effect
2316    and combine with the preceding character.  So they cannot be used in
2317    compose sequences.
2318    
2319    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2320
2321commit 5cd60398b7787297008b13a848ed3cfbd7ef178d
2322Author: Benno Schulenberg <bensberg@telfort.nl>
2323Date:   Mon Nov 2 16:34:39 2020 +0100
2324
2325    nls: rename the obsolete sr_CS locale to sr_RS
2326    
2327    Fixes #107, for the most part.
2328    
2329    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2330
2331commit 7cf1729590534e257b851899cf03421b65c538e3
2332Author: Keith Packard <keithp@keithp.com>
2333Date:   Sat Nov 7 17:02:56 2020 -0800
2334
2335    Don't attempt to unload locale context data
2336    
2337    Most locale context users call _XlcCurrentLC, which returns a pointer
2338    which never needs to be passed to _XCloseLC, meaning it has unbounded
2339    lifetime, so that locale data can never be freed.
2340    
2341    Remove all reference counting and just leave all locales that were
2342    ever used in memory.
2343    
2344    Signed-off-by: Keith Packard <keithp@keithp.com>
2345    Acked-by: Martin Peres <martin.peres@mupuf.org>
2346
2347commit 87a8fd8051b823636c3c4f3c37c4cfd11fdb880d
2348Author: Jacek Caban <jacek@codeweavers.com>
2349Date:   Mon Aug 14 19:20:19 2017 +0200
2350
2351    Don't cache last lcd in _XlcCurrentLC.
2352    
2353    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678
2354    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538
2355    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088
2356    
2357    The way it's currently cached is not thread safe. As long as locale doesn't change, the same object is reused anyway.
2358    
2359    Signed-off-by: Jacek Caban <jacek@codeweavers.com>
2360    Signed-off-by: Keith Packard <keithp@keithp.com>
2361    Acked-by: Martin Peres <martin.peres@mupuf.org>
2362
2363commit 7a38c09b82ee2522d987c3da4af8af186e562c61
2364Author: Keith Packard <keithp@keithp.com>
2365Date:   Sat Nov 7 16:59:37 2020 -0800
2366
2367    Don't cache converters in _Xlcmbtowc and Xlcwctomb
2368    
2369    These functions were caching encoding conversion functions in static
2370    variables which is not thread safe. Let the conversion loader do its
2371    job and cache locale to converters there. It's less efficient, but
2372    it's also (now) thread safe.
2373    
2374    Signed-off-by: Keith Packard <keithp@keithp.com>
2375    Acked-by: Martin Peres <martin.peres@mupuf.org>
2376
2377commit b52ba5c209165fc2c533b77b86147b31763299f3
2378Author: Jacek Caban <jacek@codeweavers.com>
2379Date:   Mon Aug 14 19:20:20 2017 +0200
2380
2381    Make conv_list thread safe.
2382    
2383    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678
2384    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538
2385    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088
2386    Signed-off-by: Jacek Caban <jacek@codeweavers.com>
2387    Signed-off-by: Keith Packard <keithp@keithp.com>
2388    Acked-by: Martin Peres <martin.peres@mupuf.org>
2389
2390commit 8b5ce3a9b1219c81711944880e5a1e1cfad3d956
2391Author: Antti Savolainen <antti.savo@gmail.com>
2392Date:   Wed Sep 16 07:47:26 2020 +0000
2393
2394    Added DOWNWARD ARROW support
2395
2396commit 7138826dccba9d538dfb142ce9af1a40d64ea849
2397Author: Antti Savolainen <antti.savo@gmail.com>
2398Date:   Tue Sep 8 00:54:23 2020 +0300
2399
2400    Added support for UPWARDS ARROW and ALMOST EQUAL TO
2401    
2402    Signed-off-by: Antti Savolainen <antti.savo@gmail.com>
2403
2404commit 56d59299f7433cdeb88bac970b06eb7df44df5bb
2405Author: Jonathan Belsewir <jonbel@jonasund.de>
2406Date:   Thu Nov 5 22:57:50 2020 +0000
2407
2408    Add less and greater to characters that are difficult to access on some keyboards
2409
2410commit 5caf45f8d3bebad5b72ac88545adc156ccfd754d
2411Author: Benno Schulenberg <bensberg@telfort.nl>
2412Date:   Fri Oct 23 17:12:22 2020 +0200
2413
2414    nls: Allow all letters with a caron to be composed also with v.
2415    
2416    It was strange that the accented letters Ž and ž can be composed with
2417    sequences that start with "v" ("v Z" and "v z"), but not Č and č and
2418    Š and š (and other letters with a caron).  For these letters, compose
2419    sequences that start with a "c" had to be used, which was frustrating
2420    because it is hard to remember that "c" stands for "caron", AND the
2421    graphically more obvious "v" is right next to it.
2422    
2423    (Unfortunately, the sequence "v l" is already taken for vertical line.
2424    Maybe the compose sequences for vertical line could be reduced to just
2425    "V L" and "L V"?)
2426    
2427    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2428
2429commit 62de4ac9025e2cb3f62a890241c06ebc75acce62
2430Author: Benno Schulenberg <bensberg@telfort.nl>
2431Date:   Sun Nov 1 20:23:55 2020 +0100
2432
2433    nls: vertically align the target column of aliases
2434    
2435    Also improve the grammar of the initial comment.
2436    
2437    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2438
2439commit f5b26a16ef959fc6860e9cf58b97a042600900b4
2440Author: Benno Schulenberg <bensberg@telfort.nl>
2441Date:   Sun Nov 1 20:17:05 2020 +0100
2442
2443    nls: remove some twenty aliases that have been obsolete for fifteen years
2444    
2445    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2446
2447commit 3e61f3b49660e081108f06bc514366f1784cb2e0
2448Author: Benno Schulenberg <bensberg@telfort.nl>
2449Date:   Sun Nov 1 19:50:56 2020 +0100
2450
2451    nls: remove country-specific aliases for two constructed languages
2452    
2453    These artificial languages are meant to be international and are
2454    thus not specific to any country.  If one would want to support
2455    aliases like ia_FR or ia_CH, then one would also have to support
2456    ia_AU, ia_DE, ia_ES, et cetera, et cetera.  That would be silly.
2457    
2458    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2459
2460commit 3f9c43adde22bb2ab4bef5837bc8401af9bb537c
2461Author: Benno Schulenberg <bensberg@telfort.nl>
2462Date:   Sun Nov 1 17:28:24 2020 +0100
2463
2464    nls: remove four aliases where the name is identical to the definition
2465    
2466    They were found with:
2467    
2468        while read one two; do
2469            if [[ $one == $two: ]]; then echo $two; fi;
2470        done  <nls/locale.alias.pre
2471    
2472    Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2473
2474commit 9f9c5365816bdd036fd80d54b22e86764ea4b7a7
2475Author: Carlos Garnacho <carlosg@gnome.org>
2476Date:   Fri Jun 14 17:55:14 2019 +0200
2477
2478    Add XSetIOErrorExitHandler() function
2479    
2480    This function complements XSetIOErrorHandler(), allowing to override
2481    the default behavior that trusts on I/O errors never coming back
2482    (i.e. exit()ing the process).
2483    
2484    This is meant as a mechanism for Wayland compositors (that are too
2485    a X11 client + compositing manager) to unfasten seatbelts and jump
2486    through the car window. It might get lucky and land on a stack of
2487    pillows.
2488    
2489    In consequence, some functions labeled as _X_NORETURN can as a
2490    matter of fact return. So those hints were removed.
2491    
2492    Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
2493    Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
2494
2495commit 4cb758019e374fa7c022fe79d28444e13441717b
2496Author: Carlos Garnacho <carlosg@gnome.org>
2497Date:   Fri Jun 14 17:54:47 2019 +0200
2498
2499    Prepare for _XIOError() possibly returning
2500    
2501    Ensure current state is cut short on _XIOError(), possible reentrancy
2502    should be skipped through the XlibDisplayIOError flag checks.
2503    
2504    Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
2505    Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
2506
2507commit a37986f7a7d4ae9b8292b132dbb0da2885ea919c
2508Author: Ross Burton <ross.burton@intel.com>
2509Date:   Wed Jan 22 17:11:23 2020 +0000
2510
2511    cpprules.in: squash whitespace in generated files
2512    
2513    CPP is used to generate files, but as cpp reads files from the build host the
2514    output has a number of blank lines at the beginning which varies depending on
2515    what GCC and friends is used.
2516    
2517    Pathalogical example:
2518    
2519     $ cpp -undef -traditional /dev/null
2520     # 1 "/dev/null"
2521     # 1 "<built-in>"
2522     # 1 "<command-line>"
2523     # 31 "<command-line>"
2524     # 1 "/usr/include/stdc-predef.h" 1 3 4
2525    
2526     # 17 "/usr/include/stdc-predef.h" 3 4
2527    
2528     [ 40 blank line ]
2529    
2530     # 32 "<command-line>" 2
2531     # 1 "/dev/null"
2532    
2533    So depending on the content of stdc-predef.h and what other headers CPP will
2534    load, the amount of whitespace in the generates files varies. This can result in
2535    differences in reproducible environments, and file conflicts in multilib
2536    environments.
2537    
2538    As whitespace is irrelevant to these machine-readable files, extend the sed to
2539    just delete blank lines.
2540
2541commit 16192ce2a7e462e09f95aa45ce9eaa180e483c43
2542Author: Carmina16 <mistresssilvara@hotmail.com>
2543Date:   Sun Jun 16 02:54:25 2019 +0000
2544
2545    Adding ia and ie locales
2546
2547commit caa71668af7fd3ebdd56353c8f0ab90824773969
2548Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2549Date:   Mon Sep 28 16:16:07 2020 -0700
2550
2551    Use __builtin_popcountl if available to replace Ones() in GetImage.c
2552    
2553    If the compiler knows of a better algorithm for counting the number of
2554    bits set in a word for the target CPU, let it use that, instead of the
2555    classic algorithm optimized for PDP-6.
2556    
2557    Based on libXext commit 490a25e6f8a4d2482af4364c700b68ad11a4d10b
2558    
2559    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2560
2561commit 1a9269a95c353e3de9f4d33bab9f1363d84e79db
2562Author: Krzesimir Nowak <qdlacz@gmail.com>
2563Date:   Mon Jun 1 12:58:30 2020 +0000
2564
2565    Fix leaking modifiers string.
2566    
2567    Reported by valgrind:
2568    
2569    ```
2570    ==118175== 17 bytes in 1 blocks are definitely lost in loss record 13 of 1,675
2571    ==118175==    at 0x483A809: malloc (vg_replace_malloc.c:307)
2572    ==118175==    by 0x5CD1B46: _XlcDefaultMapModifiers (in /usr/lib64/libX11.so.6.3.0)
2573    ==118175==    by 0x5CD1F1A: XSetLocaleModifiers (in /usr/lib64/libX11.so.6.3.0)
2574    ==118175==    by 0x496841C: X11_InitKeyboard (SDL_x11keyboard.c:324)
2575    ==118175==    by 0x496F0CA: X11_VideoInit (SDL_x11video.c:455)
2576    ==118175==    by 0x494747B: SDL_VideoInit_REAL (SDL_video.c:532)
2577    ==118175==    by 0x489E886: SDL_InitSubSystem_REAL (SDL.c:206)
2578    ==118175==    by 0x402634: main (fade.cc:35)
2579    ```
2580
2581commit d127217f26df1bf7566c1f372d8b5329a06754ea
2582Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2583Date:   Sun Aug 16 17:03:24 2020 -0700
2584
2585    Exempt XLookupKeysym from XKeycodeToKeysym deprecation warnings
2586    
2587    Gets rid of:
2588    src/xkb/XKBBind.c: In function ‘XLookupKeysym’:
2589    src/xkb/XKBBind.c:234:5: warning: ‘XKeycodeToKeysym’ is deprecated
2590    [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wdeprecated-declarations-Wdeprecated-declarations]
2591      234 |     return XKeycodeToKeysym(dpy, event->keycode, col);
2592          |     ^~~~~~
2593    src/xkb/XKBBind.c:96:1: note: declared here
2594       96 | XKeycodeToKeysym(Display *dpy,
2595          | ^~~~~~~~~~~~~~~~
2596    
2597    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2598
2599commit a64eb5f478c6ca84ef3457a01f9a7c3db8ecf1e5
2600Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2601Date:   Sun Aug 16 16:50:43 2020 -0700
2602
2603    lcUTF8: Increase string buffer size from 20 to 32 bytes
2604    
2605    While we don't expect large enough ints to need it, we don't
2606    enforce a maximum size, so gcc assumes the worst and warns:
2607    
2608    ../../../src/xlibi18n/lcUTF8.c: In function ‘create_tofontcs_conv’:
2609    ../../../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=]
2610     1736 |  snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
2611          |                                  ^~~~~~~~~~~~~
2612    ../../../src/xlibi18n/lcUTF8.c:1736:2: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20
2613     1736 |  snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
2614          |  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2615    ../../../src/xlibi18n/lcUTF8.c:1739:46: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
2616     1739 |      snprintf(buf, sizeof(buf), "fs%d.charset", i);
2617          |                                              ^
2618    ../../../src/xlibi18n/lcUTF8.c:1739:6: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20
2619     1739 |      snprintf(buf, sizeof(buf), "fs%d.charset", i);
2620          |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2621    ../../../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=]
2622     1754 |         snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
2623          |                                         ^~~~~~~~~~~~~
2624    ../../../src/xlibi18n/lcUTF8.c:1754:9: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20
2625     1754 |         snprintf(buf, sizeof(buf), "fs%d.charset.name", i);
2626          |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2627    ../../../src/xlibi18n/lcUTF8.c:1757:53: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
2628     1757 |             snprintf(buf, sizeof(buf), "fs%d.charset", i);
2629          |                                                     ^
2630    ../../../src/xlibi18n/lcUTF8.c:1757:13: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20
2631     1757 |             snprintf(buf, sizeof(buf), "fs%d.charset", i);
2632          |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2633    
2634    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2635
2636commit 54925250ad9c825bea671dcade5c913bad88abee
2637Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2638Date:   Sun Aug 16 16:33:59 2020 -0700
2639
2640    i18n: use memcpy instead of strncpy on unterminated char arrays
2641    
2642    Avoids gcc warnings that we're using strncpy wrong to copy a known-length
2643    set of characters without a terminating '\0' to a buffer whose length we
2644    are checking separately.  (Should also be imperceptibly faster since we
2645    no longer check if each byte is '\0' when we already know it won't be.)
2646    
2647    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2648
2649commit 8754868dbbbc5fa096282df0504aa7880488eeb9
2650Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2651Date:   Sun Aug 16 16:02:01 2020 -0700
2652
2653    Mark FreeBlueTblElements label unused in LRGB.c
2654    
2655    Quiets gcc 10.2 warning of:
2656    src/xcms/LRGB.c: In function ‘LINEAR_RGB_InitSCCData’:
2657    src/xcms/LRGB.c:798:1: warning: label ‘FreeBlueTblElements’ defined
2658     but not used
2659    [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wunused-label-Wunused-label]
2660      798 | FreeBlueTblElements:
2661          | ^~~~~~~~~~~~~~~~~~~
2662    
2663    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2664
2665commit 52b0cd67073f2f4665884b5a7decfedeb3b7ca1e
2666Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2667Date:   Sun Aug 16 14:54:47 2020 -0700
2668
2669    Raise minimum required xproto version to 7.0.25 (released 2013-11-23)
2670    
2671    Allows us to depend on _X_COLD directly instead of having to check for it.
2672    
2673    (Since we also use _X_UNUSED, 7.0.22 or later was implicitly required
2674     already but not checked for.)
2675    
2676    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2677
2678commit deafb99fd8faf8b044eaee13d072440ce4db76af
2679Author: Maya Rashish <maya@NetBSD.org>
2680Date:   Sun Aug 23 21:19:33 2020 +0300
2681
2682    Avoid the use of "register" keyword in public headers.
2683    
2684    This causes issues when compiling code for C++17.
2685
2686commit 505420662577749e36640db48f6b6b9ae0236e09
2687Author: Matthieu Herrb <matthieu@herrb.eu>
2688Date:   Mon Aug 24 15:42:25 2020 +0200
2689
2690    libX11 1.6.12
2691    
2692    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2693
2694commit acdaaadcb3d85c61fd43669fc5dddf0f8c3f911d
2695Author: Matthieu Herrb <matthieu@herrb.eu>
2696Date:   Thu Aug 13 18:02:58 2020 +0200
2697
2698    Fix an integer overflow in init_om()
2699    
2700    CVE-2020-14363
2701    
2702    This can lead to a double free later, as reported by Jayden Rivers.
2703    
2704    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2705
2706commit d15c24c8b44be5e4054c8ecd0ff9dcf2c8e18e5b
2707Author: Niclas Zeising <zeising@daemonic.se>
2708Date:   Tue Aug 11 13:44:38 2020 +0200
2709
2710    Fix input clients connecting to server
2711    
2712    Fix a bug where some input clients can't connect to the input server.
2713    This fixes #117.
2714    
2715    FreeBSD bugzilla reference:
2716    https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248549
2717    
2718    Signed-off-by: Niclas Zeising <zeising@daemonic.se>
2719
2720commit 6dd618e5ea8e7558dd8e99a2777ab7b69211789c
2721Author: Maya Rashish <maya@NetBSD.org>
2722Date:   Sat Aug 15 00:48:56 2020 +0300
2723
2724    Avoid the use of "register" keyword in XkbTranslateKeySym.
2725    
2726    This causes issues when compiling code for C++17.
2727    While here, make function prototype match the header with regards
2728    to removal of another register keyword.
2729
2730commit 780d222343098b4bfe3ec529b4a0dc2aadea19bf
2731Author: Christopher Chavez <chrischavez@gmx.us>
2732Date:   Mon Aug 10 17:08:39 2020 +0000
2733
2734    Fix typo GCCLipYOrigin -> GCClipYOrigin in XCreateGC() manpage
2735
2736commit 29a8251a456bb7dc10146b4713f499a624d6dacd
2737Author: Felix Yan <felixonmars@archlinux.org>
2738Date:   Thu Aug 6 16:03:38 2020 +0000
2739
2740    Correct a typo in GetStCmap.c
2741
2742commit 0d8f038db83f68724a84ef9d4ccd0d5e33ec6a70
2743Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2744Date:   Thu Aug 6 08:07:57 2020 -0700
2745
2746    libX11 1.6.11
2747    
2748    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2749
2750commit 93fce3f4e79cbc737d6468a4f68ba3de1b83953b
2751Author: Yichao Yu <yyc1992@gmail.com>
2752Date:   Sun Aug 2 13:43:58 2020 -0400
2753
2754    Fix size calculation in `_XimAttributeToValue`.
2755    
2756    The check here guards the read below.
2757    For `XimType_XIMStyles`, these are `num` of `CARD32` and for `XimType_XIMHotKeyTriggers`
2758    these are `num` of `XIMTRIGGERKEY` ref[1] which is defined as 3 x `CARD32`.
2759    (There are data after the `XIMTRIGGERKEY` according to the spec but they are not read by this
2760    function and doesn't need to be checked.)
2761    
2762    The old code here used the native datatype size instead of the wire protocol size causing
2763    the check to always fail.
2764    
2765    Also fix the size calculation for the header (size). It is 2 x CARD16 for both types
2766    despite the unused `CARD16` for `XimType_XIMStyles`.
2767    
2768    [1] https://www.x.org/releases/X11R7.6/doc/libX11/specs/XIM/xim.html#Input_Method_Styles
2769    
2770    This fixes a regression caused by 388b303c62aa35a245f1704211a023440ad2c488 in 1.6.10.
2771    
2772    Fix #116
2773
2774commit 9949364ea761ab7efa2a9930ec7718f966a740c0
2775Author: Matthieu Herrb <matthieu@herrb.eu>
2776Date:   Thu Jul 30 21:36:22 2020 +0200
2777
2778    libX11 1.6.10
2779    
2780    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2781
2782commit 1703b9f3435079d3c6021e1ee2ec34fd4978103d
2783Author: Matthieu Herrb <matthieu@herrb.eu>
2784Date:   Fri Jul 24 21:09:10 2020 +0200
2785
2786    Change the data_len parameter of _XimAttributeToValue() to CARD16
2787    
2788    It's coming from a length in the protocol (unsigned) and passed
2789    to functions that expect unsigned int parameters (_XCopyToArg()
2790    and memcpy()).
2791    
2792    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2793    Reviewed-by: Todd Carson <toc@daybefore.net>
2794
2795commit 1a566c9e00e5f35c1f9e7f3d741a02e5170852b2
2796Author: Todd Carson <tc@daybefore.net>
2797Date:   Fri Jul 24 20:59:32 2020 +0200
2798
2799    Zero out buffers in functions
2800    
2801    It looks like uninitialized stack or heap memory can leak
2802    out via padding bytes.
2803    
2804    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2805    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
2806
2807commit 2fcfcc49f3b1be854bb9085993a01d17c62acf60
2808Author: Todd Carson <tc@daybefore.net>
2809Date:   Fri Jul 24 20:22:44 2020 +0200
2810
2811    Fix more unchecked lengths
2812    
2813    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2814    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
2815
2816commit 388b303c62aa35a245f1704211a023440ad2c488
2817Author: Todd Carson <tc@daybefore.net>
2818Date:   Fri Jul 24 19:36:51 2020 +0200
2819
2820    fix integer overflows in _XimAttributeToValue()
2821    
2822    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2823    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
2824
2825commit 0e6561efcfaa0ae7b5c74eac7e064b76d687544e
2826Author: Todd Carson <tc@daybefore.net>
2827Date:   Fri Jul 24 19:33:30 2020 +0200
2828
2829    Fix signed length values in _XimGetAttributeID()
2830    
2831    The lengths are unsigned according to the specification. Passing
2832    negative values can lead to data corruption.
2833    
2834    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
2835    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
2836
2837commit 2b7598221d87049d03e9a95fcb541c37c8728184
2838Author: Alan Coopersmith <alan.coopersmith@oracle.com>
2839Date:   Wed Jul 22 15:19:58 2020 -0700
2840
2841    Fix spelling/wording issues
2842    
2843    Found by using:
2844        codespell --builtin clear,rare,usage,informal,code,names
2845    
2846    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2847
2848commit f9e3378ebcfc79ff00591b8617a68e97580a741e
2849Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
2850Date:   Sun Jul 19 22:19:24 2020 +0200
2851
2852    xlibi18n: Fix converting horizline keysyms to UCS
2853    
2854    U+23BA - U+23BD are meant to represent the scan lines, and U+2500 is
2855    unified with scan line 5.
2856    
2857    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
2858
2859commit 4385a84c4a79c1b60bf12839899a1544198de975
2860Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
2861Date:   Thu Jun 4 02:19:13 2020 +0200
2862
2863    Braille: Fix default lookup
2864    
2865    cbb59d172 ('Braille: Fix typing quickly') broke the default lookup that
2866    translates Braille keysym patterns to Braille Unicode patterns since it
2867    rightfully clears brl_committing, but then we do not have it any more to
2868    fill brl_committed.
2869    
2870    This change saves the committed pattern so we can return it in the
2871    default lookup.
2872    
2873    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
2874
2875commit fb65b67169105c82c243f265e0f993cf72efdacd
2876Author: Marko Myllynen <myllynen@redhat.com>
2877Date:   Fri Feb 28 13:59:36 2020 +0200
2878
2879    Update Finnish compose sequences for SFS 5966:2019 standard
2880    
2881    U0219, U0218, U021B, U021A are now part of the standard.
2882    
2883    Fixes #70 which was printed incorrectly in the earlier standard version.
2884
2885commit 07724ce046348a21cc82a874853b14b9a285df75
2886Author: Tristan Miller <psychonaut@nothingisreal.com>
2887Date:   Wed Sep 25 13:34:01 2019 +0200
2888
2889    Add acute accent compose sequences for Russian vowels.
2890    
2891    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/104
2892    
2893    Signed-off-by: Tristan Miller <psychonaut@nothingisreal.com>
2894
2895commit 3ec9264d01a3a66c0fa0145a7c1db06a64be1507
2896Author: Thomas E. Dickey <dickey@invisible-island.net>
2897Date:   Sat Apr 25 14:40:54 2020 -0400
2898
2899    document an ancient performance problem (see Debian #954845).
2900    
2901    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
2902
2903commit 791c93816ce0520d54925b0b3ccbc29823370041
2904Author: Thomas E. Dickey <dickey@invisible-island.net>
2905Date:   Sat Apr 25 14:22:11 2020 -0400
2906
2907    update/correct function prototypes in manpages to match headers
2908    
2909    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
2910
2911commit 844271305a28e87031562dfeb43defbea02960a2
2912Author: Thomas E. Dickey <dickey@invisible-island.net>
2913Date:   Tue Apr 21 20:53:35 2020 -0400
2914
2915    reduce gcc-normal warnings using casts (no object change)
2916    
2917    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
2918
2919commit 53b59ddfc114a55d26e9717a1b4822f1b18c40ca
2920Author: Thomas E. Dickey <dickey@invisible-island.net>
2921Date:   Tue Apr 21 18:13:22 2020 -0400
2922
2923    reduce gcc-normal warnings using casts (no object change)
2924    
2925    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
2926
2927commit 68c72a7341b114277ab232f2499ee3bd035af8a0
2928Author: Adam Jackson <ajax@redhat.com>
2929Date:   Wed Dec 11 11:53:11 2019 -0500
2930
2931    Fix XTS regression in XCopyColormapAndFree
2932    
2933    XCopyColormapAndFree/5 threw an assertion:
2934    
2935        520|4 5 00014017 1 2|Assertion XCopyColormapAndFree-5.(A)
2936        520|4 5 00014017 1 3|When a colourmap argument does not name a valid colourmap,
2937        520|4 5 00014017 1 4|then a BadColor error occurs.
2938        520|4 5 00014017 1 5|METH: Create a bad colourmap by creating and freeing a colourmap.
2939        520|4 5 00014017 1 6|METH: Call test function using bad colourmap as the colourmap argument.
2940        520|4 5 00014017 1 7|METH: Verify that a BadColor error occurs.
2941        520|4 5 00014017 1 8|unexpected signal 6 (SIGABRT) received
2942        220|4 5 2 15:05:53|UNRESOLVED
2943        410|4 5 1 15:05:53|IC End
2944        510|4|system 0: Abandoning testset: caught unexpected signal 11 (SIGSEGV)
2945    
2946    More specifically:
2947    
2948        lt-XCopyColormapAndFree: xcb_io.c:533: _XAllocID: Assertion `ret != inval_id' failed.
2949    
2950    This bug was introduced (by following my advice, d'oh) in:
2951    
2952        commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d
2953        Author: Tapani Pälli <tapani.palli@intel.com>
2954        Date:   Mon May 13 08:29:49 2019 +0300
2955    
2956            Protect colormap add/removal with display lock
2957    
2958    In that patch we moved the call to _XcmsCopyCmapRecAndFree inside the
2959    display lock. The problem is said routine has side effects, including
2960    trying to implicitly create a colormap in some cases. Since we don't run
2961    the XID handler until SyncHandle() we would see inconsistent internal
2962    xlib state, triggering the above assert.
2963    
2964    Fix this by dropping and re-taking the display lock before calling into
2965    XCMS.
2966    
2967    Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2968
2969commit 7f46a38139f66fda734f3a6c445b84ea89c8f310
2970Author: Alex Henrie <alexhenrie24@gmail.com>
2971Date:   Wed Jan 1 14:54:06 2020 -0700
2972
2973    Handle small final sigma in XConvertCase
2974    
2975    lowercase: GREEK SMALL LETTER FINAL SIGMA (U+03C2)
2976    uppercase: GREEK CAPITAL LETTER SIGMA (U+03A3)
2977    
2978    This mapping was correct in UCSConvertCase, but the "legacy" mapping
2979    must also be correct for Caps Lock to work with the final sigma key.
2980    
2981    https://gitlab.freedesktop.org/xorg/lib/libx11/issues/5
2982    
2983    Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
2984
2985commit a48787d3d155c8a09cead345f5055da30c1db537
2986Author: Peter Hutterer <peter.hutterer@who-t.net>
2987Date:   Wed Dec 11 14:12:27 2019 +1000
2988
2989    Handle ssharp in XConvertCase()
2990    
2991    lowercase: LATIN SMALL LETTER SHARP S (U+00DF)
2992    uppercase: LATIN CAPITAL LETTER SHARP S (U+1E9E)
2993    
2994    The uppercase sharp s (XK_ssharp) is a relatively recent addition to unicode
2995    but was added to the relevant keyboard layouts in xkeyboard-config-2.25
2996    (d1411e5e95c)
2997    https://gitlab.freedesktop.org/xkeyboard-config/xkeyboard-config/issues/144
2998    
2999    Alas, the CapsLock behavior was broken on the finnish layout (maybe others).
3000    This was due to xkbcomp using XConvertCase() to determine whether a key
3001    requires the type FOUR_LEVEL_ALPHABETIC or FOUR_LEVEL_SEMIALPHABETIC.
3002    
3003    Let's make this function return the right lower/upper symbols for the sharp s
3004    and hope that the world won't get any worse because of it.
3005    
3006    https://gitlab.freedesktop.org/xorg/lib/libx11/issues/110
3007    
3008    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
3009
3010commit b8766a438461e20b5c6f7a6111b703d4cb71501f
3011Author: Raul Fernandes <rgfernandes@gmail.com>
3012Date:   Wed Oct 9 18:39:26 2019 +0000
3013
3014    Use memcmp and memcpy
3015
3016commit db7cca17ad7807e92a928da9d4c68a00f4836da2
3017Author: Adam Jackson <ajax@redhat.com>
3018Date:   Wed Oct 9 13:41:29 2019 -0400
3019
3020    libX11 1.6.9
3021    
3022    Signed-off-by: Adam Jackson <ajax@redhat.com>
3023
3024commit f5ba2c632a4c84279bdd48c687271292f5dc4f87
3025Author: Dmitry Osipenko <digetx@gmail.com>
3026Date:   Sat May 18 19:10:32 2019 +0300
3027
3028    Fix lockup in _XReply() caused by recursive synchronization
3029    
3030    This patch is based on a suggestion made by Uli Schlachter in a comment
3031    to the bug report https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93.
3032    
3033    Explanation of the bug (given by Uli Schlachter as well):
3034    
3035    An error was received and handled. Since there was an error callback set,
3036    Xlib unlocks the display, runs the error callback, and then locks the display
3037    again. This goes through _XLockDisplay and then calls _XSeqSyncFunction.
3038    On this "lock the thing"-path, Xlib notices that sequence numbers are close to
3039    wrap-around and tries to send a GetInputFocus request. However, the earlier
3040    calls already registered themselves as "we are handling replies/errors, do
3041    not interfere!" and so the code here waits for "that other thread" to be done
3042    before it continues. Only that there is no other thread, but it is this thread
3043    itself and thus a deadlock follows.
3044    
3045    The bug is relatively easy to reproduce on any desktop environment by
3046    using actively a touchscreen input that supports multitouch, i.e. practically
3047    all mobile devices are affected.
3048    
3049    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93
3050    Suggested-by: Uli Schlachter <psychon@znc.in>
3051    Tested-by: Dmitry Osipenko <digetx@gmail.com>
3052    Reported-by: Dmitry Osipenko <digetx@gmail.com>
3053    Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
3054
3055commit 1f1ca0863fd81fd0538b104eb9d9841ddbc7c905
3056Author: Adam Jackson <ajax@redhat.com>
3057Date:   Fri Sep 13 16:44:32 2019 -0400
3058
3059    xkb: Provide <X11/extensions/XKBgeom.h> ourselves
3060    
3061    This header refers to libX11 types, it doesn't belong in xorgproto and
3062    has been moved to legacy in 2019.2. Supply it ourselves.
3063
3064commit edc7680ed5a03cedb5facf14693823455e12c29c
3065Author: Ross Burton <ross.burton@intel.com>
3066Date:   Tue Aug 6 14:53:43 2019 +0100
3067
3068    src/util/Makefile: explicitly reset LINK to not use libtool
3069    
3070    Simply looking at libtool redefines LINK globally to use libtool, which when
3071    you're trying to cross-compile to Windows can cause complications.
3072    
3073    As in src/util/ we're simply building a small binary for the build host, reset
3074    LINK to the automake default so that the traditional compile/link steps occur
3075    without libtool.
3076    
3077    Also remove -all-static from LDFLAGS as that is a libtool-specific argument
3078    intended to solve this problem.
3079    
3080    Closes: #100
3081    Signed-off-by: Ross Burton <ross.burton@intel.com>
3082
3083commit f111bf3ac091a3a6d1f094172621d9d033c3833c
3084Author: Thomas E. Dickey <dickey@invisible-island.net>
3085Date:   Tue Jul 30 16:53:27 2019 -0400
3086
3087    improve some formatting
3088    
3089    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3090
3091commit 1d0a6fe9126c7ce61c6648cabf43b2de4d6d3ba9
3092Author: Thomas E. Dickey <dickey@invisible-island.net>
3093Date:   Tue Jul 30 15:49:09 2019 -0400
3094
3095    fix a substitution error from recent commit, e.g, "s/^\.EE/XDe/"
3096    
3097    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3098
3099commit 8fe8aafe7e9beffaff0460fe08354dfcde1efe98
3100Author: Thomas E. Dickey <dickey@invisible-island.net>
3101Date:   Tue Jul 30 05:07:40 2019 -0400
3102
3103    split lines at sentence endings
3104    
3105    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3106
3107commit 6f2cbcea5d6dcd1df7ef309481cad1663284c061
3108Author: Thomas E. Dickey <dickey@invisible-island.net>
3109Date:   Tue Jul 30 04:52:25 2019 -0400
3110
3111    trim trailing whitespace from manpages
3112    
3113    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3114
3115commit 4abf482d8e6b0b1ae84e0f528e71066afd6d18b5
3116Author: Thomas E. Dickey <dickey@invisible-island.net>
3117Date:   Tue Jul 30 04:36:24 2019 -0400
3118
3119    another fake-quote fix
3120    
3121    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3122
3123commit c0f0651f2b4dfa9fef86d1374689220125631993
3124Author: Thomas E. Dickey <dickey@invisible-island.net>
3125Date:   Tue Jul 30 04:34:54 2019 -0400
3126
3127    the last commit overlooked some fake-quote pairs
3128    
3129    Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net>
3130
3131commit ef31143c9decda54b3ab2937d3300fcf5b493fd6
3132Author: Walter Harms <wharms@bfs.de>
3133Date:   Mon Jul 29 18:50:19 2019 +0200
3134
3135    fix ``fake quotes'' in text
3136    
3137    to remove the fake quotes replace them with propper
3138    predefined macros \*(lq and \*(rq. this will allow
3139    nroff to choose the propper characters when using ps etc.
3140    
3141    Signed-off-by: Walter Harms <wharms@bfs.de>
3142
3143commit 705027ea9b17d6552388d2067736ab4b0bf9dfb5
3144Author: Walter Harms <wharms@bfs.de>
3145Date:   Mon Jul 29 18:27:12 2019 +0200
3146
3147    get rid of ``fake quotes''
3148    
3149    these fake quotes are inside a comment and can be replaced with
3150    some double quotes
3151    
3152    Signed-off-by: Walter Harms <wharms@bfs.de>
3153
3154commit fe46eb287106ad05538e896783a5cf877a4138df
3155Author: Walter Harms <wharms@bfs.de>
3156Date:   Mon Jul 29 17:57:02 2019 +0200
3157
3158    remove all private macro defines
3159    
3160    after converting everything to st. man page macros there is
3161    no need to maintain X11 private nroff macros, so remove them.
3162    
3163    Signed-off-by: Walter Harms <wharms@bfs.de>
3164
3165commit b7f09ae19c9a098ecd545b4d4c45a6afe39a8764
3166Author: Walter Harms <wharms@bfs.de>
3167Date:   Mon Jul 29 17:47:35 2019 +0200
3168
3169    Replace home-grown .Ds .De macro with man page .EX/.EE macro
3170    
3171    Same pages use the man page .EX/.EE macro. Replace all occurences
3172    of .De/.Ds with the std. macros to make the code better to maintain.
3173    
3174    Signed-off-by: Walter Harms <wharms@bfs.de>
3175
3176commit 951e772bc8d89bc62c4bbac14adc5d4f5027419f
3177Author: Walter Harms <wharms@bfs.de>
3178Date:   Wed Jul 3 18:23:51 2019 +0200
3179
3180    fix TBL format
3181    
3182    Signed-off-by: Walter Harms <wharms@bfs.de>
3183
3184commit 4eb167f5aca4cca05a230c728ffdb4fd382a4e1e
3185Author: Walter Harms <wharms@bfs.de>
3186Date:   Wed Jul 3 18:18:09 2019 +0200
3187
3188    replace home grown .ZN with std, .B and .BR
3189    
3190    Replace the home grown macro .ZN with std. macros
3191    from man macro paket. So we can get rid of the
3192    definition an get a clean header.
3193    
3194    Signed-off-by: Walter Harms <wharms@bfs.de>
3195
3196commit 08e6b36292ebf01ba2dcc4a48ec371c6f5d13458
3197Author: Walter Harms <wharms@bfs.de>
3198Date:   Sat Jun 22 18:33:48 2019 +0200
3199
3200    remove in-text macros
3201    
3202    remove in-text macros an fix a few typos on the way.
3203    hopefuly that will make maintenance more easy.
3204    
3205    Signed-off-by: Walter Harms <wharms@bfs.de>
3206
3207commit cb0e894fe9927b07bd9d97c067fd0d007accb105
3208Author: Walter Harms <wharms@bfs.de>
3209Date:   Fri Jun 21 18:01:16 2019 +0200
3210
3211    note that we can handle kbd==NULL
3212    
3213    Signed-off-by: Walter Harms <wharms@bfs.de>
3214
3215commit 0327c427d62f671eced067c6d9b69f4e216a8cac
3216Author: Adam Jackson <ajax@redhat.com>
3217Date:   Mon Jun 17 13:36:08 2019 -0400
3218
3219    makekeys: Detach ourselves from X headers entirely
3220    
3221    Subsequent to a121b7b0c210efe10bf93453b29050282324c906 we are no longer
3222    building makekeys with enough -I/foo/bar to find the X11 headers, so if
3223    they're not in a system include path, things fail. Since this utility is
3224    only needed at build time, there's no real reason to demand the X
3225    headers be installed for both the build and target machines if cross-
3226    compiling, we can just assume a vaguely ANSI environment instead.
3227    
3228    Tested-by: Niclas Zeising <zeising@daemonic.se>
3229    Reviewed-by: Keith Packard <keithp@keithp.com>
3230    Reviewed-by: Matt Turner <mattst88@gmail.com>
3231
3232commit 57f0d421a1320090d3a4948181fe228d86cc8977
3233Author: Matt Turner <mattst88@gmail.com>
3234Date:   Mon Jun 17 10:45:45 2019 -0400
3235
3236    libX11 1.6.8
3237    
3238    Signed-off-by: Matt Turner <mattst88@gmail.com>
3239
3240commit 5464b302bc0c4a055d3eac06c6ab15e209ff7668
3241Author: Matt Turner <mattst88@gmail.com>
3242Date:   Sun Jun 9 14:00:15 2019 -0700
3243
3244    Use AC_SYS_LARGEFILE
3245    
3246    ... and include config.h in makekeys.c to get the definition of
3247    _FILE_OFFSET_BITS. Without it, libX11 can fail to build on a file
3248    system with 64-bit inode numbers.
3249    
3250    Bug: https://bugs.gentoo.org/550502
3251    Bug: https://bugs.gentoo.org/616140
3252    Signed-off-by: Matt Turner <mattst88@gmail.com>
3253
3254commit 33b9148a833ab2ea46e44871da3c031bf1b78f77
3255Author: Ran Benita <ran234@gmail.com>
3256Date:   Thu Feb 13 20:55:08 2014 +0200
3257
3258    Compose.man: fix escaped hexadecimal char description
3259    
3260    The man page says:
3261        Strings may be direct text encoded in the locale for which the
3262        compose file is to be used, or an escaped octal or hexadecimal
3263        character code.   Octal codes are specified as "\123" and
3264        hexadecimal codes as "\0x123a".
3265    
3266    But the grammar in the parser and the implementation say:
3267        ESCAPED_CHAR  ::= ('\\' | '\"' | OCTAL | HEX )
3268        HEX           ::= '\' (x|X) HEX_CHAR [HEX_CHAR]]
3269        HEX_CHAR      ::= (0|1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|a|b|c|d|e|f)
3270    
3271    So "\0x123a" -> "\x3a".
3272    
3273    Signed-off-by: Ran Benita <ran234@gmail.com>
3274    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3275
3276commit d9b2cc35edf01e4f751a9ccb5ac185745184cec0
3277Author: ojab <ojab@ojab.ru>
3278Date:   Mon Apr 24 20:12:17 2017 +0000
3279
3280    Compose sequences for rouble sign
3281    
3282    Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout.
3283    
3284    Signed-off-by: Slava Kardakov <ojab@ojab.ru>
3285    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3286
3287commit 3f211616f6414bd9958566c0ca137e87a2b6df95
3288Author: Pavel Labath <pavelo@centrum.sk>
3289Date:   Fri Nov 30 20:45:04 2012 +0100
3290
3291    Fix a leak in XCreateFontSet
3292    
3293    a simple snippet like XFreeFontSet(d, XCreateFontSet(d, ...)) will generate lots of memory leaks,
3294    as evidenced by the following valgrind output:
3295    ==983== HEAP SUMMARY:
3296    ==983==     in use at exit: 39,409 bytes in 341 blocks
3297    ==983==   total heap usage: 4,795 allocs, 4,454 frees, 489,086 bytes allocated
3298    ==983==
3299    ==983== 1,688 (136 direct, 1,552 indirect) bytes in 1 blocks are definitely lost in loss record
3300    40 of 46
3301    ==983==    at 0x4C2B042: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
3302    ==983==    by 0x56D5A93: add_codeset.clone.9 (in /usr/lib64/libX11.so.6.3.0)
3303    ==983==    by 0x56D5FE0: load_generic (in /usr/lib64/libX11.so.6.3.0)
3304    ==983==    by 0x56D7612: initialize (in /usr/lib64/libX11.so.6.3.0)
3305    ==983==    by 0x56D7E75: _XlcCreateLC (in /usr/lib64/libX11.so.6.3.0)
3306    ==983==    by 0x56F9A5F: _XlcUtf8Loader (in /usr/lib64/libX11.so.6.3.0)
3307    ==983==    by 0x56DF815: _XOpenLC (in /usr/lib64/libX11.so.6.3.0)
3308    ==983==    by 0x56B255A: XOpenOM (in /usr/lib64/libX11.so.6.3.0)
3309    ==983==    by 0x56A665A: XCreateFontSet (in /usr/lib64/libX11.so.6.3.0)
3310    ==983==    by 0x4FCA80: conky::x11_output::create_gc() (x11.cc:746)
3311    ==983==    by 0x4FC3B4: conky::x11_output::use_own_window() (x11.cc:602)
3312    ==983==    by 0x4FAD42: conky::priv::own_window_setting::set(bool const&, bool) (x11.cc:92)
3313    ==983==
3314    ==983== LEAK SUMMARY:
3315    ==983==    definitely lost: 136 bytes in 1 blocks
3316    ==983==    indirectly lost: 1,552 bytes in 34 blocks
3317    ==983==      possibly lost: 0 bytes in 0 blocks
3318    ==983==    still reachable: 37,721 bytes in 306 blocks
3319    ==983==         suppressed: 0 bytes in 0 blocks
3320    
3321    This patch makes the leak dissappear (Well, at least the "definitely lost part". The "still
3322    reachable" thingy remains). After some analysis, I've discovered that the XLCd structure is
3323    destroyed improperly. The "constructor" is in lcGeneric.c, but the structure is destroyed using
3324    code from lcPublic.c. I've found that changing the destructor call to _XlcDestroyLC executes the
3325    correct code path, and I'm pretty sure this is correct (the object was constructed using
3326    _XlcCreateLC, it make sense to destroy it using its conterpart).
3327    
3328    So far I haven't observed any strange behaviour on my system caused by this change (although, I'm
3329    not sure, how many programs actually use this function).
3330    
3331    Signed-off-by: Pavel Labath <pavelo@centrum.sk>
3332    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3333
3334commit 4645e219133458781e3fb48eaea6a74cccb1b9aa
3335Author: Jon Turney <jon.turney@dronecode.org.uk>
3336Date:   Tue Apr 30 14:39:06 2019 +0100
3337
3338    Avoid using libtool wrapper for makekeys
3339    
3340    For Windows targets, libtool uses a wrapper executable, not a wrapper
3341    script (see [1]), which it compiles with the host compiler.  This
3342    doesn't work when cross-compiling.
3343    
3344    Since we don't actually need to link with anything, use the libtool flag
3345    -all-static to tell it to stay completely out of this.
3346    
3347    [1] https://www.gnu.org/software/libtool/manual/html_node/Wrapper-executables.html
3348
3349commit 6886d9ba06826785a8ccb312c04ea82b91bb6a25
3350Author: Jon Turney <jon.turney@dronecode.org.uk>
3351Date:   Wed Apr 24 12:50:41 2019 +0100
3352
3353    Use EXEEXT_FOR_BUILD for makekeys
3354    
3355    Use EXEXT_FOR_BUILD, to fix cross-compiling where EXEEXT differs from
3356    EXEEXT_FOR_BUILD, such as when building for Windows from unix.
3357    
3358    (Note: As written, this assumes EXEEXT_FOR_BUILD is always empty when
3359    cross-compiling.  There could be some elaborate autodetection for
3360    EXEXT_FOR_BUILD, but for the moment, if you are cross-compiling from
3361    Windows to Unix, you'll need to set EXEEXT_FOR_BUILD explicity...)
3362
3363commit a121b7b0c210efe10bf93453b29050282324c906
3364Author: Jon Turney <jon.turney@dronecode.org.uk>
3365Date:   Tue Apr 30 14:30:41 2019 +0100
3366
3367    Remove makekeys dependency on X headers
3368    
3369    This is the patch from https://bugs.freedesktop.org/show_bug.cgi?id=6669
3370    by Pierre Ossman, reworked for master.
3371    
3372    Avoid using LIBS (which are for host, but we don't need) and rewrite
3373    makekeys slightly to avoid needing to include any X headers, which
3374    avoids potentially having -I with host paths in CFLAGS, which can cause
3375    standard headers e.g. stdio.h for the host to also be used, which can
3376    break things...
3377
3378commit cbb59d1727cb1bfcff91b54fd55f4139e360b7ab
3379Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
3380Date:   Fri Jun 7 23:35:33 2019 +0200
3381
3382    Braille: Fix typing quickly
3383    
3384    On finishing releasing Braille keys, we should clear the just-commited
3385    pattern, to reset the state to initial state, and avoid having to wait for
3386    0.3s before typing the next pattern.
3387    
3388    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3389    Tested-by: Jean-Philippe Mengual <jpmengual@hypra.fr>
3390
3391commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d
3392Author: Tapani Pälli  <tapani.palli@intel.com>
3393Date:   Mon May 13 08:29:49 2019 +0300
3394
3395    Protect colormap add/removal with display lock
3396    
3397    This fixes a bug where concurrent threads call XCreateColormap and
3398    XFreeColormap corrupting a linked list where colormap structures
3399    are stored.
3400    
3401    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/94
3402    
3403    v2: handle XCopyColormapAndFree also (Adam Jackson)
3404    
3405    Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
3406
3407commit fea80d03c381a4de379863c0d41f6d4c00f15e1e
3408Author: Jon Turney <jon.turney@dronecode.org.uk>
3409Date:   Wed Mar 13 13:12:20 2019 +0000
3410
3411    Add autoconf checks for headers we include for FIONREAD
3412    
3413    Add autoconf checks for the extra headers we include to define FIONREAD.
3414    This needs sys/socket.h on Cygwin, and none of the alternatives on
3415    Windows.
3416
3417commit 87c77a1e6d7034536e9d25ce24a667ebf53486a7
3418Author: Matthias Dieter Wallnöfer  <matthias.wallnoefer@lugbz.org>
3419Date:   Mon Mar 11 14:19:52 2019 +0000
3420
3421    locale.alias.pre - add the de_IT locale
3422    
3423    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3424
3425commit 568f77f414072506bbe48646637c872987c97e63
3426Author: Matthias Dieter Wallnöfer  <matthias.wallnoefer@lugbz.org>
3427Date:   Mon Mar 11 14:17:10 2019 +0000
3428
3429    locale.dir.pre - add the de_IT locale
3430    
3431    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3432
3433commit 38c0c92feaef7dc4a72920b1dd79872e6578086c
3434Author: Matthias Dieter Wallnöfer  <matthias.wallnoefer@lugbz.org>
3435Date:   Mon Mar 11 14:12:07 2019 +0000
3436
3437    compose.dir.pre - add de_IT locale
3438    
3439    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3440
3441commit 49f835611ba6ab82e68b7b851507931c142db7a5
3442Author: Matthias Dieter Wallnöfer  <matthias.wallnoefer@lugbz.org>
3443Date:   Tue Jan 29 15:34:34 2019 +0000
3444
3445    Update locale.alias.pre - introduce the de_LI aliases (equivalent to de_CH)
3446    
3447    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3448
3449commit ed669234d0cd5933618aa5fd2da697b59845b9be
3450Author: Matthias Dieter Wallnöfer  <matthias.wallnoefer@lugbz.org>
3451Date:   Tue Jan 29 15:28:18 2019 +0000
3452
3453    locale.dir.pre - fix the de_LI latin15 locale
3454    
3455    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3456
3457commit 82b5c1454a49333b4d94984d40db79cff386d37f
3458Author: Josh Triplett <josh@joshtriplett.org>
3459Date:   Sun Feb 3 08:31:41 2019 +0100
3460
3461    compose.dir.pre: Give the C.UTF-8 locale a full set of UTF-8 compose sequences
3462    
3463    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
3464
3465commit 603c3ab25a692194805fd4b0c62afac8093897fe
3466Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3467Date:   Sat Mar 9 20:58:07 2019 -0800
3468
3469    Add missing close paren to XrmGetStringDatabase synopsis in man page
3470    
3471    Fixes: commit 0642e99ac05afb0c
3472    
3473    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3474
3475commit 6590cc41e3c5c5c836a49a2c82774d7cc9ab3c85
3476Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3477Date:   Sat Feb 23 11:12:27 2019 -0800
3478
3479    Include reallocarray.h in tarballs so "make distcheck" passes
3480    
3481    Fixes: commit bcf7b5aa06c23aee
3482    
3483    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3484
3485commit a4a7df06351ce9565d298a84547ff71f2d220fbf
3486Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3487Date:   Sat Feb 23 11:04:41 2019 -0800
3488
3489    XlibInt.c: include headers needed for ioctl(...FIONREAD...) on Solaris
3490    
3491    Fixes: commit 5538b3e4ae6dee
3492    
3493    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3494
3495commit 38ec04de7e537dde6b7b1eaca46b1a24fcdbc044
3496Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3497Date:   Sat Feb 23 10:48:26 2019 -0800
3498
3499    specs/XKB: Fix misprint in documentation of XkbFreeGeomOverlayKeys()
3500    
3501    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/86
3502    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23550
3503    
3504    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3505
3506commit 74f2f26811ae82b15dede31504b0eb541f7a3293
3507Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3508Date:   Sat Feb 23 10:46:05 2019 -0800
3509
3510    specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayKeys()
3511    
3512    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/85
3513    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23549
3514    
3515    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3516
3517commit 66b039a41690f076e962ac6bd46cca017360a7d1
3518Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3519Date:   Sat Feb 23 10:44:15 2019 -0800
3520
3521    specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayRows()
3522    
3523    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/84
3524    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23548
3525    
3526    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3527
3528commit cdea0e840315171f8390adc63a2397ddf111eea4
3529Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3530Date:   Sat Feb 23 10:38:08 2019 -0800
3531
3532    specs/XKB: Fix misprint in documentation of XkbGetNamedGeometry()
3533    
3534    Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/82
3535    aka: https://bugs.freedesktop.org/show_bug.cgi?id=23520
3536    
3537    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3538
3539commit c72d17ada98723cd9e03ddf06680a0e54b2c3089
3540Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3541Date:   Sat Feb 23 10:30:32 2019 -0800
3542
3543    XkbAllocGeomOverlayKey man page should be named XkbAllocGeomOverlayKeys
3544    
3545    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3546
3547commit 05002b8aef9544c8545e0823a02cf14f335d2b68
3548Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3549Date:   Sun Feb 17 15:13:25 2019 -0800
3550
3551    Fix man page typo: XkbAllBellNotifyMask -> XkbAllBellEventsMask
3552    
3553    Reported-by: Daniel Hahler <git@thequod.de>
3554    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3555
3556commit 5538b3e4ae6dee32c47db9dfc85b07bbe7b90f6c
3557Author: Adam Jackson <ajax@redhat.com>
3558Date:   Fri Mar 24 11:07:36 2017 -0400
3559
3560    _XDefaultIOError: Do better at detecting explicit shutdown
3561    
3562    Currently, when the X server crashes or a client is disconnected with
3563    XKillClient, you get a somewhat confusing error message from libX11
3564    along the lines of:
3565    
3566    XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
3567          after 98 requests (40 known processed) with 0 events remaining.
3568    
3569    What's happening here is the previous recvmsg has thrown EAGAIN, since
3570    the socket is non-blocking. In this case, check whether the socket has
3571    any more data to read, and if not treat it like EPIPE.
3572    
3573    Signed-off-by: Adam Jackson <ajax@redhat.com>
3574
3575commit 6d2cde9633b5ee020cb60caea1cf61e090b86dd2
3576Author: Adam Jackson <ajax@redhat.com>
3577Date:   Fri Mar 24 11:07:35 2017 -0400
3578
3579    _XDefaultIOError: Reformat to be less ugly
3580    
3581    Signed-off-by: Adam Jackson <ajax@redhat.com>
3582    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3583
3584commit 003e30a66a249f5c70b30d1c187385124cd4cdad
3585Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3586Date:   Tue Jan 1 14:34:04 2019 -0800
3587
3588    Avoid use-after-free in _XimProtoSetIMValues()
3589    
3590    Fixes gitlab issue #49
3591    
3592    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3593
3594commit 30656fd66ad47aeaca0bfc237a818f189554f07e
3595Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3596Date:   Tue Jan 1 12:21:04 2019 -0800
3597
3598    Fix implicit conversion warnings in _XlcCreateDefaultCharSet
3599    
3600    lcCharSet.c:187:50: warning: implicit conversion changes signedness:
3601          'int' to 'unsigned long' [-Wsign-conversion]
3602        tmp = Xmalloc(name_len + 1 + ct_sequence_len + 1);
3603              ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
3604    ../../include/X11/Xlibint.h:453:32: note: expanded from macro 'Xmalloc'
3605                           ~~~~~~  ^~~~
3606    lcCharSet.c:192:31: warning: implicit conversion changes signedness:
3607          'int' to 'unsigned long' [-Wsign-conversion]
3608        memcpy(tmp, name, name_len+1);
3609        ~~~~~~            ~~~~~~~~^~
3610    lcCharSet.c:216:45: warning: implicit conversion changes signedness:
3611          'int' to 'unsigned long' [-Wsign-conversion]
3612        memcpy(tmp, ct_sequence, ct_sequence_len+1);
3613        ~~~~~~                   ~~~~~~~~~~~~~~~^~
3614    lcCharSet.c:183:16: warning: implicit conversion loses integer precision:
3615          'unsigned long' to 'int' [-Wshorten-64-to-32]
3616        name_len = strlen(name);
3617                 ~ ^~~~~~~~~~~~
3618    lcCharSet.c:184:23: warning: implicit conversion loses integer precision:
3619          'unsigned long' to 'int' [-Wshorten-64-to-32]
3620        ct_sequence_len = strlen(ct_sequence);
3621                        ~ ^~~~~~~~~~~~~~~~~~~
3622    lcCharSet.c:198:37: warning: implicit conversion loses integer precision:
3623         'long' to 'unsigned int' [-Wshorten-64-to-32]
3624            unsigned int length = colon - charset->name;
3625                         ~~~~~~   ~~~~~~^~~~~~~~~~~~~~~
3626    
3627    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3628
3629commit 2e6300905784fa8adfcea019262ea58b3e1ed4ad
3630Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3631Date:   Tue Jan 1 11:41:47 2019 -0800
3632
3633    Remove no-longer-used name variable in _XGetAtomName
3634    
3635    Fixes gcc warning:
3636    GetAtomNm.c: In function ‘_XGetAtomName’:
3637    GetAtomNm.c:39:11: warning: unused variable ‘name’ [-Wunused-variable]
3638         char *name;
3639               ^~~~
3640    
3641    Introduced by commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96
3642    
3643    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3644
3645commit f077871c2a12cf098b0c50572e112c6dc5e04de0
3646Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3647Date:   Sat Dec 29 11:39:29 2018 -0800
3648
3649    Fix out-of-tree builds after commit 8a62e26515e528bb47917ec4a55cc2b492419eeb
3650    
3651    In-tree builds found reallocarray.h in $(top_builddir)/src but the
3652    out-of-tree build didn't find it at all.
3653    
3654    Reported-by: Emmanuele Bassi <ebassi@gmail.com> from GNOME continuous integration pipeline
3655    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3656
3657commit ba7f7cc783672784bebc6e69349b55c6c75c1973
3658Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3659Date:   Sun Sep 27 18:35:27 2015 -0700
3660
3661    Convert src/xlibi18n to use reallocarray()
3662    
3663    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3664
3665commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96
3666Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3667Date:   Sat Sep 26 11:12:47 2015 -0700
3668
3669    Replace Xmalloc+strcpy pairs with strdup calls
3670    
3671    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3672
3673commit 258a8ced681dc1bc50396be7439fce23f9807e2a
3674Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3675Date:   Sun Jul 19 11:58:50 2015 -0700
3676
3677    Replace _XkbTypedRealloc & _XkbClearElems with new _XkbResizeArray
3678    
3679    Makes resizing & clearing more consistent and gets rid of some weird
3680    quirks like always subtracting 1 from the size passed to _XkbClearElems
3681    so it could always add 1 to the size passed in.
3682    
3683    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3684
3685commit f94273e2f4284040306b888e5ca0381ee00591af
3686Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3687Date:   Sun Jul 19 09:47:29 2015 -0700
3688
3689    Add some missing allocation failure checks in src/xcms
3690    
3691    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3692
3693commit 76850471b4a5a6d30f5660234e87ec2772e7b426
3694Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3695Date:   Sun Jul 19 09:33:21 2015 -0700
3696
3697    Convert src/xcms to use reallocarray()
3698    
3699    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3700
3701commit 8a62e26515e528bb47917ec4a55cc2b492419eeb
3702Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3703Date:   Sun Jul 19 09:24:02 2015 -0700
3704
3705    Convert src/XKB to use reallocarray()
3706    
3707    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3708
3709commit 6d1dc1f6169ebf0ba71785d461bd98129c65c862
3710Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3711Date:   Thu Jun 4 23:30:43 2015 -0700
3712
3713    Convert main src directory to use reallocarray()
3714
3715commit bcf7b5aa06c23aee00af7999b58cb96a8571ac42
3716Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3717Date:   Thu Jun 4 23:18:23 2015 -0700
3718
3719    Import reallocarray() from OpenBSD
3720    
3721    Wrapper for realloc() that checks for overflow when multiplying
3722    arguments together, so we don't have to add overflow checks to
3723    every single call.  For documentation on usage, see:
3724    http://www.openbsd.org/cgi-bin/man.cgi/OpenBSD-current/man3/calloc.3
3725    
3726    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3727
3728commit 9bdfe9c9af11d77d66e5ff651e82b20e695cb460
3729Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3730Date:   Mon Nov 19 21:26:23 2018 -0800
3731
3732    Update README for gitlab migration
3733    
3734    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3735
3736commit c8bcc19970a2cc4ad337de1338c63a07325eefcd
3737Author: Albert Astals Cid <aacid@kde.org>
3738Date:   Fri Nov 9 22:24:52 2018 +0100
3739
3740    locale.dir: Add ast_ES.UTF-8
3741    
3742    With this patch xev properly reports
3743       XLookupString gives 2 bytes: (c2 b4) "´"
3744    for the dead_acute key when using the Asturian locale
3745
3746commit 117bb310b44ce03802c91a70949eb58dcc702a3c
3747Author: Albert Astals Cid <aacid@kde.org>
3748Date:   Fri Nov 9 22:18:43 2018 +0100
3749
3750    compose.dir: Add ast_ES.UTF-8
3751    
3752    With this patch one can properly type dead keys like á when using the
3753    Asturian locale
3754
3755commit d7e22252d68258a63ca537827c99d0ac0bc4b9d8
3756Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3757Date:   Sat Nov 10 13:30:47 2018 -0800
3758
3759    Remove documentation of obsolete B16 & B32 tags in specs/libX11
3760    
3761    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3762
3763commit f3c978476e0be6813268af494efb7ac507451116
3764Author: Matt Turner <mattst88@gmail.com>
3765Date:   Tue Oct 9 10:26:05 2018 -0400
3766
3767    libX11 1.6.7
3768    
3769    Signed-off-by: Matt Turner <mattst88@gmail.com>
3770
3771commit 823a0f8a820247b6c1e092f679b49cbdc2ea5c95
3772Author: Michel Dänzer  <michel.daenzer@amd.com>
3773Date:   Fri Sep 28 17:24:17 2018 +0200
3774
3775    poll_for_event: Allow using xcb_poll_for_queued_event
3776    
3777    It avoids reading from the display connection again in cases where that
3778    was already done.
3779    
3780    Suggested-by: Uli Schlachter <psychon@znc.in>
3781    Reviewed-by: Uli Schlachter <psychon@znc.in>
3782
3783commit 406afe4b0f1b655c0db19bbc9a0c48da9a46acf5
3784Author: Michel Dänzer  <michel.daenzer@amd.com>
3785Date:   Tue Sep 25 17:10:58 2018 +0200
3786
3787    poll_for_response: Call poll_for_event again if xcb_poll_for_reply fails
3788    
3789    If xcb_poll_for_reply fails to find a reply, poll_for_response would
3790    always return NULL. However, xcb_poll_for_reply may have read events
3791    from the display connection while looking for a reply. In that case,
3792    returning NULL from poll_for_response is wrong and can result in the
3793    client hanging, e.g. because it returns to waiting for the display
3794    connection file descriptor becoming readable after XPending incorrectly
3795    returned 0 pending events.
3796    
3797    The solution is to call poll_for_event again after xcb_poll_for_reply
3798    returned 0. This will return the first of any events read by
3799    xcb_poll_for_reply.
3800    
3801    Fixes issue #79.
3802    
3803    Reported-by: Yuxuan Shui <yshuiv7@gmail.com>
3804    Bugzilla: https://bugs.freedesktop.org/108008
3805    Bugzilla: https://bugs.freedesktop.org/107992
3806    Reviewed-by: Adam Jackson <ajax@redhat.com>
3807
3808commit d0416863d5bf75af54ce81f6c30d4c1476b5e04f
3809Author: Bhavi Dhingra <b.dhingra@samsung.com>
3810Date:   Mon Sep 28 08:33:40 2015 +0000
3811
3812    XcmsLookupColor: fully initialize XColor structs passed to _XColor_to_XcmsRGB
3813    
3814    Fixes https://gitlab.freedesktop.org/xorg/lib/libx11/issues/44
3815    aka https://bugs.freedesktop.org/show_bug.cgi?id=92154
3816    
3817    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3818    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3819
3820commit 733f64bfeb311c1d040b2f751bfdef9c9d0f89ef
3821Author: Matthieu Herrb <matthieu@herrb.eu>
3822Date:   Tue Aug 21 16:54:50 2018 +0200
3823
3824    libX11 1.6.6
3825    
3826    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3827
3828commit 173704243fbcbea0b6de2068b4fb403c7408fb54
3829Author: Matthieu Herrb <matthieu@herrb.eu>
3830Date:   Tue Aug 21 16:53:40 2018 +0200
3831
3832    Remove statement with no effect.
3833    
3834    Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
3835
3836commit e83722768fd5c467ef61fa159e8c6278770b45c2
3837Author: Tobias Stoeckmann <tobias@stoeckmann.org>
3838Date:   Fri Jul 27 16:38:00 2018 +0200
3839
3840    Fixed crash on invalid reply (CVE-2018-14598).
3841    
3842    If the server sends a reply in which even the first string would
3843    overflow the transmitted bytes, list[0] (or flist[0]) will be set to
3844    NULL and a count of 0 is returned.
3845    
3846    If the resulting list is freed with XFreeExtensionList or
3847    XFreeFontPath later on, the first Xfree call:
3848    
3849        Xfree (list[0]-1)
3850     turns into
3851        Xfree (NULL-1)
3852    
3853    which will most likely trigger a segmentation fault.
3854    
3855    I have modified the code to return NULL if the first string would
3856    overflow, thus protecting the freeing functions later on.
3857    
3858    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
3859
3860commit dbf72805fd9d7b1846fe9a11b46f3994bfc27fea
3861Author: Tobias Stoeckmann <tobias@stoeckmann.org>
3862Date:   Fri Jul 27 16:37:17 2018 +0200
3863
3864    Fixed out of boundary write (CVE-2018-14600).
3865    
3866    The length value is interpreted as signed char on many systems
3867    (depending on default signedness of char), which can lead to an out of
3868    boundary write up to 128 bytes in front of the allocated storage, but
3869    limited to NUL byte(s).
3870    
3871    Casting the length value to unsigned char fixes the problem and allows
3872    string values with up to 255 characters.
3873    
3874    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
3875
3876commit b469da1430cdcee06e31c6251b83aede072a1ff0
3877Author: Tobias Stoeckmann <tobias@stoeckmann.org>
3878Date:   Fri Jul 27 16:36:34 2018 +0200
3879
3880    Fixed off-by-one writes (CVE-2018-14599).
3881    
3882    The functions XGetFontPath, XListExtensions, and XListFonts are
3883    vulnerable to an off-by-one override on malicious server responses.
3884    
3885    The server replies consist of chunks consisting of a length byte
3886    followed by actual string, which is not NUL-terminated.
3887    
3888    While parsing the response, the length byte is overridden with '\0',
3889    thus the memory area can be used as storage of C strings later on. To
3890    be able to NUL-terminate the last string, the buffer is reserved with
3891    an additional byte of space.
3892    
3893    For a boundary check, the variable chend (end of ch) was introduced,
3894    pointing at the end of the buffer which ch initially points to.
3895    Unfortunately there is a difference in handling "the end of ch".
3896    
3897    While chend points at the first byte that must not be written to,
3898    the for-loop uses chend as the last byte that can be written to.
3899    
3900    Therefore, an off-by-one can occur.
3901    
3902    I have refactored the code so chend actually points to the last byte
3903    that can be written to without an out of boundary access. As it is not
3904    possible to achieve "ch + length < chend" and "ch + length + 1 > chend"
3905    with the corrected chend meaning, I removed the inner if-check.
3906    
3907    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
3908
3909commit d81da209fd4d0c2c9ad0596a8078e58864479d0d
3910Author: Tobias Stoeckmann <tobias@stoeckmann.org>
3911Date:   Tue Jul 3 22:31:37 2018 +0200
3912
3913    Validation of server response in XListHosts.
3914    
3915    If a server sends an incorrect length in its response, a client is prone
3916    to perform an out of boundary read while processing the data.
3917    
3918    The length field of xHostEntry is used to specify the amount of bytes
3919    used to represent the address. It is 16 bit, which means that it is not
3920    possible to perform an arbitrary memory access, but it might be enough
3921    to read sensitive information, e.g. malloc-related pointers and offsets.
3922    
3923    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
3924    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3925
3926commit b676e62377483df77bcb6472d26b24f901323fa9
3927Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
3928Date:   Wed Jun 13 15:46:58 2018 +0200
3929
3930    XkbOpenDisplay.3: fix typo
3931    
3932    XkbOpenDisplay returns a pointer to Display, not a Display.
3933    
3934    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
3935
3936commit 9c5845ff0dc080ff10bd68af4fc40fcd805728ca
3937Author: Martin Natano <natano@natano.net>
3938Date:   Sat Oct 8 19:57:50 2016 +0200
3939
3940    Don't rebuild ks_tables.h if nothing changed.
3941    
3942    ks_tables.h is always considered out of date due to the forced rebuild
3943    of the makekeys util. This means the file is also rebuilt during 'make
3944    install', which is usually performed as root, which can to lead
3945    permission problems later on.
3946    
3947    Signed-off-by: Martin Natano <natano@natano.net>
3948    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3949
3950commit 796f754cba6d75b676a0fc39b97802198fceda4f
3951Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3952Date:   Sat May 5 14:43:30 2018 -0700
3953
3954    Change fall through comment in lcDB.c to match gcc's requirements
3955    
3956    Needs to match one of the regexps shown under
3957    https://gcc.gnu.org/onlinedocs/gcc-7.3.0/gcc/Warning-Options.html#index-Wimplicit-fallthrough
3958    
3959    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3960
3961commit 82ca6308757126fa7ffc6588f1e5d8e3be04251b
3962Author: Alan Coopersmith <alan.coopersmith@oracle.com>
3963Date:   Sat Mar 24 19:45:14 2018 -0700
3964
3965    Use size_t for buffer sizes in SetHints.c
3966    
3967    These variables store values returned from strlen() as a size_t
3968    and are passed to Xmalloc, which expects a size_t, so lets stop
3969    converting back and forth to int along the way.
3970    
3971    Reported by: Konstantin SKliarov
3972    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3973    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
3974
3975commit 50a5a98984097d614227d22a49265e12b172cac7
3976Author: Bhavi Dhingra <b.dhingra@samsung.com>
3977Date:   Tue Jul 5 11:37:50 2016 +0530
3978
3979    Fix possible memory leak in cmsProp.c:140
3980    
3981    https://bugs.freedesktop.org/show_bug.cgi?id=96814
3982    
3983    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3984    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
3985
3986commit a9dafdd57c71473fa3a2ec4887e973e4e9876d83
3987Author: Michal Srb <msrb@suse.com>
3988Date:   Thu Mar 15 09:50:58 2018 +0100
3989
3990    Use flexible array member instead of fake size.
3991    
3992    The _XimCacheStruct structure is followed in memory by two strings containing
3993    fname and encoding. The memory was accessed using the last member of the
3994    structure `char fname[1]`. That is a lie, prohibits us from using sizeof and
3995    confuses checkers. Lets declare it properly as a flexible array, so compilers
3996    don't complain about writing past that array. As bonus we can replace the
3997    XOffsetOf with regular sizeof.
3998    
3999    Fixes GCC8 error:
4000      In function 'strcpy',
4001          inlined from '_XimWriteCachedDefaultTree' at imLcIm.c:479:5,
4002          inlined from '_XimCreateDefaultTree' at imLcIm.c:616:2,
4003          inlined from '_XimLocalOpenIM' at imLcIm.c:700:5:
4004      /usr/include/bits/string_fortified.h:90:10: error: '__builtin_strcpy'
4005      forming offset 2 is out of the bounds [0, 1] [-Werror=array-bounds]
4006         return __builtin___strcpy_chk (__dest, __src, __bos (__dest));
4007    
4008    Caused by this line seemingly writing past the fname[1] array:
4009      imLcIm.c:479:  strcpy (m->fname+strlen(name)+1, encoding);
4010    
4011    Reviewed-by: Keith Packard <keithp@keithp.com>
4012    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4013
4014commit 34f4464f69a4d6ff0d1042500a62f9a5ed7f3647
4015Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4016Date:   Tue Mar 6 11:42:27 2018 -0800
4017
4018    If XGetImage fails to create image, don't dereference it to bounds check
4019    
4020    Reported by gcc 7.3:
4021    
4022    GetImage.c:110:25: warning: potential null pointer dereference [-Wnull-dereference]
4023      if (planes < 1 || image->height < 1 || image->bytes_per_line < 1 ||
4024                        ~~~~~^~~~~~~~
4025    
4026    Introduced by 8ea762f94f4c942d898fdeb590a1630c83235c17 in Xlib 1.6.4
4027    
4028    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4029    Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
4030
4031commit e835a9dcc3362b5e92893be756dd7ae361e64ced
4032Author: wharms <wharms@bfs.de>
4033Date:   Sun Sep 3 14:17:45 2017 +0200
4034
4035    silence gcc warning assignment discards 'const' qualifier from pointer target type
4036
4037commit 36a1ac0253fea82ff79cc52ba56c5691cfd07a3b
4038Author: wharms <wharms@bfs.de>
4039Date:   Sun Aug 20 21:51:57 2017 +0200
4040
4041    remove empty line
4042
4043commit e02dfe54f32b4165351d2712a9d2e0584906a3ce
4044Author: wharms <wharms@bfs.de>
4045Date:   Sun Aug 20 21:50:33 2017 +0200
4046
4047    add _X_UNUSED to avoid unused variable warnings
4048
4049commit 2911c39cecd63ed3747072a5eeeb9eedffc881e9
4050Author: walter harms <wharms@bfs.de>
4051Date:   Sat Jun 4 17:19:59 2016 +0200
4052
4053    Fixes: warning: variable 'req' set but not,used
4054    
4055    Fixes: warning: variable 'req' set but not used [-Wunused-but-set-variable]
4056           by marking req _X_UNUSED
4057            Solution was discussed on xorg-devel ML
4058           Peter Hutter, Alan Coopersmith
4059            Re: [PATCH libX11 3/5] fix: warning: pointer targets in passing argument 2 of '_XSend' differ in signedness [-Wpointer-sign]
4060    
4061    Signed-off-by: harms wharms@bfs.de
4062
4063commit bf82ec0402479fd8399d69e7d62fc17d7956699a
4064Author: walter harms <wharms@bfs.de>
4065Date:   Sat Jun 4 17:22:07 2016 +0200
4066
4067    mark _XDefaultIOError as no_return
4068    
4069    mark _XDefaultIOError as no_return. No one comes back from exit() ...
4070    
4071    Signed-off-by: harms wharms@bfs.de
4072
4073commit 9abe8380074edea3ac4e72466ec47e921ca05e47
4074Author: walter harms <wharms@bfs.de>
4075Date:   Sat Jun 4 17:21:52 2016 +0200
4076
4077    no need to check XFree arguments
4078    
4079    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
4080    
4081    Signed-off-by: harms wharms@bfs.de
4082
4083commit 433477fcb7e07d0c26a22ba78aae88827ed1f440
4084Author: walter harms <wharms@bfs.de>
4085Date:   Tue Apr 26 16:32:20 2016 +0200
4086
4087    fix memleak in error path
4088    
4089    free all mem on error
4090    Signed-off-by: walter harms <wharms@bfs.de>
4091
4092commit ed9f0d34abc645eee56e21863f23acb4d0bb8e9a
4093Author: walter harms <wharms@bfs.de>
4094Date:   Tue Apr 26 17:58:16 2016 +0200
4095
4096    fix memleak in error path
4097    
4098    V2: remove unneeded NULL (reported by eric.engestrom@imgtec.com)
4099    
4100    fix mem leak in error path
4101    Signed-off-by: walter harms <wharms@bfs.de>
4102
4103commit 7c78fc57693afa94cf26170f0f6276e3b7374ed0
4104Author: walter harms <wharms@bfs.de>
4105Date:   Tue Apr 26 16:34:11 2016 +0200
4106
4107    no need to check args for Xfree()
4108    
4109    simplify code
4110    
4111    Signed-off-by: walter harms <wharms@bfs.de>
4112
4113commit c1c14af441ae73d1a8e67a971fafcf967e45ac48
4114Author: walter harms <wharms@bfs.de>
4115Date:   Tue Apr 26 16:23:46 2016 +0200
4116
4117    remove stray extern
4118    
4119    remove stray extern
4120    
4121    Signed-off-by: walter harms <wharms@bfs.de>
4122
4123commit 714921f041a245dc5f37a689268b584226a2ccb9
4124Author: walter harms <wharms@bfs.de>
4125Date:   Mon Apr 11 18:26:52 2016 +0200
4126
4127    no need to check argument for _XkbFree()
4128    
4129    simplify code by removing unneeded checks
4130    
4131    Signed-off-by: walter harms <wharms@bfs.de>
4132
4133commit d02c2466f65063a03c97dbcee05071c12a3676e6
4134Author: walter harms <wharms@bfs.de>
4135Date:   Mon Apr 11 18:22:38 2016 +0200
4136
4137    fix more shadow warning
4138    
4139    Signed-off-by: walter harms <wharms@bfs.de>
4140
4141commit 0355c3926d5372f9762f235071dbd94a89bbbdad
4142Author: walter harms <wharms@bfs.de>
4143Date:   Thu Mar 31 19:16:33 2016 +0200
4144
4145    fix shadow char_size
4146    
4147    Signed-off-by: walter harms <wharms@bfs.de>
4148
4149commit 916dffadf052135df3398651be873c353da629e1
4150Author: walter harms <wharms@bfs.de>
4151Date:   Thu Mar 31 19:14:32 2016 +0200
4152
4153    remove argument check for free() adjust one inden
4154    
4155    Signed-off-by: walter harms <wharms@bfs.de>
4156
4157commit 6ec901ebca3fea6a762e22090dc35b1b90911133
4158Author: walter harms <wharms@bfs.de>
4159Date:   Thu Mar 31 19:12:17 2016 +0200
4160
4161    _XIOError(dpy); will never return so remore dead
4162    
4163    Signed-off-by: walter harms <wharms@bfs.de>
4164
4165commit 83107a677b2ed458e4d62ea4a601e8181d3683d8
4166Author: walter harms <wharms@bfs.de>
4167Date:   Thu Mar 31 19:10:49 2016 +0200
4168
4169    fix shadow warning
4170    
4171    Signed-off-by: walter harms <wharms@bfs.de>
4172
4173commit 4fe66b1c5112b07bd09e28bbc021911d08a9621f
4174Author: Ryan C. Gordon <icculus@icculus.org>
4175Date:   Wed Aug 2 02:41:03 2017 -0400
4176
4177    Valgrind fix for XStoreColor and XStoreColors.
4178    
4179    If the "pad" field isn't set, Valgrind will report it as uninitialized
4180    memory accesses when the struct is copied into the Display's send buffer.
4181    
4182    In practice, this is (probably) harmless, but Valgrind is correct in
4183    believing it's a bug.
4184    
4185    https://bugs.freedesktop.org/attachment.cgi?id=133189
4186    
4187    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4188    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4189
4190commit 7d2010fec25c2f52b873ad0572479eb43128b038
4191Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4192Date:   Fri Apr 7 00:13:03 2017 -0700
4193
4194    Improve table formatting in XkbChangeControls & XkbKeyNumGroups man pages
4195    
4196    Includes fix for Solaris Bug 24564279: "XkbKeyNumGroups.3x11 man page
4197    contains some malformed text" caused by extra whitespace after .TE macros
4198    
4199    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4200
4201commit b856d5d929047d1ea169814d56e43784ea404c83
4202Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4203Date:   Thu Mar 16 00:05:49 2017 -0700
4204
4205    Clarify state parameter to XkbSetNamedDeviceIndicator
4206    
4207    Checking a Bool to see if it's NULL does not work well in C.
4208    Also reported in https://bugs.freedesktop.org/show_bug.cgi?id=251
4209    
4210    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4211    Reviewed-by: Adam Jackson <ajax@redhat.com>
4212
4213commit c6dadd4cebd994aafb37a58b3adbaa82507c2d18
4214Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4215Date:   Wed Mar 15 23:50:26 2017 -0700
4216
4217    Make Xkb{Get,Set}NamedIndicator spec & manpages match code
4218    
4219    The XKB Library spec and the man pages for XkbGetNamedIndicator &
4220    XkbSetNamedIndicator included a device_spec argument neither function
4221    takes, and do not include the XkbGetNamedDeviceIndicator &
4222    XkbSetNamedDeviceIndicator variants that do take it (along with two
4223    other arguments).
4224    
4225    This updates them to match the interfaces the code has provided for
4226    decades.
4227    
4228    This has been reported multiple times, so this fixes:
4229    https://bugs.freedesktop.org/show_bug.cgi?id=251
4230    https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=729812
4231    Sun Bug 4528016 XkbSetNamedIndicator & XkbGetNamedIndicator man pages are wrong
4232      (filed: alan.coopersmith@sun.com 2001-11-15 - now aka Oracle bug 15087506)
4233    X.Org Group Defect Id #9418
4234    
4235    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4236    Reviewed-by: Adam Jackson <ajax@redhat.com>
4237
4238commit 2d20890e7ffd3ee88a9ceb25cdd2ac1fe7aaceb6
4239Author: Arthur Huillet <ahuillet@nvidia.com>
4240Date:   Wed Feb 1 15:02:41 2017 +0100
4241
4242    _XDefaultError: set XlibDisplayIOError flag before calling exit
4243    
4244    _XReply isn't reentrant, and it can lead to deadlocks when the default error
4245    handler is called: _XDefaultError calls exit(1). It is called indirectly by
4246    _XReply when a X protocol error comes in that isn't filtered/handled by an
4247    extension or the application. This means that if the application (or one of its
4248    loaded shared libraries such as the NVIDIA OpenGL driver) has registered any
4249    _fini destructor, _fini will get called while still on the call stack of
4250    _XReply. If the destructor interacts with the X server and calls _XReply, it
4251    will hit a deadlock, looping on the following in _XReply:
4252    
4253        ConditionWait(dpy, dpy->xcb->reply_notify);
4254    
4255    It is legal for an application to make Xlib calls during _fini, and that is
4256    useful for an OpenGL driver to avoid resource leaks on the X server side, for
4257    example in the dlopen/dlclose case. However, the driver can not readily tell
4258    whether its _fini is being called because Xlib called exit, or for another
4259    reason (dlclose), so it is hard to cleanly work around this issue in the driver.
4260    
4261    This change makes it so _XReply effectively becomes a no-op when called after
4262    _XDefaultError was called, as though an XIOError had happened. The dpy
4263    connection isn't broken at that point, but any call to _XReply is going to hang.
4264    This is a bit of a kludge, because the more correct solution would be to make
4265    _XReply reentrant, maybe by broadcasting the reply_notify condition before
4266    calling the default error handler. However, such a change would carry a grater
4267    risk of introducing regressions in Xlib.
4268    
4269    This change will drop some valid requests on the floor, but this should not
4270    matter, as it will only do so in the case where the application is dying: X will
4271    clean up after it once exit() is done running. There is the case of
4272    XSetCloseDownMode(RETAIN_PERMANENT), but an application using that and wishing
4273    to clean up resources in _fini would currently be hitting a deadlock, which is
4274    hardly a better situation.
4275    
4276    Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
4277    Reviewed-by: Jamey Sharp <jamey@minilop.net>
4278
4279commit 42f4d7af9cf6d1dbfa575552e057328b054a20c9
4280Author: Matt Turner <mattst88@gmail.com>
4281Date:   Sat Feb 25 21:54:22 2017 -0800
4282
4283    libX11 1.6.5
4284    
4285    Signed-off-by: Matt Turner <mattst88@gmail.com>
4286
4287commit a0da5835e8078445947c828fe2d86c1a31439012
4288Author: Adam Jackson <ajax@redhat.com>
4289Date:   Tue Feb 14 15:33:29 2017 -0500
4290
4291    Revert "Compose sequences for rouble sign"
4292    
4293    This reverts commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0.
4294    
4295    Reported to break 'make check':
4296    
4297    https://lists.freedesktop.org/archives/xorg-devel/2017-February/052720.html
4298
4299commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0
4300Author: Mihail Konev <k.mvc@ya.ru>
4301Date:   Fri Feb 10 18:48:18 2017 +0500
4302
4303    Compose sequences for rouble sign
4304    
4305    Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout.
4306    Also add Cyrillic sequences for hryvnia sign.
4307    
4308    Submitted-by: Victor V. Kustov <coyote@bks.tv>
4309    Reviewed-by: Victor V. Kustov <coyote@bks.tv>
4310    Signed-off-by: Mihail Konev <k.mvc@ya.ru>
4311
4312commit 23d9623c661694aba8cf1e8f277dffa7a86cf065
4313Author: Petr Písař < <petr.pisar@atlas.cz>
4314Date:   Sun Oct 30 12:49:11 2016 +0100
4315
4316    Revert cs_CZ.UTF-8 XLC_LOCALE to en_US.UTF-8
4317    
4318    The cs_CZ.UTF-8/XLC_LOCALE is an empty file leading to unsupported cs_CZ.UTF-8
4319    locale and reporting this error:
4320    
4321    Warning: locale not supported by Xlib, locale set to C
4322    
4323    Therefore this patch reverts to the en_US.UTF-8 definition file that was used
4324    before. This patch also deduplicates the cs_CZ.UTF-8 entry.
4325    
4326    <https://bugs.freedesktop.org/show_bug.cgi?id=98219>
4327    
4328    This reverts commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51
4329    
4330    Signed-off-by: Julien Cristau <jcristau@debian.org>
4331
4332commit 71b0929ebc1f0f877f63e3f6de260f529daa6c69
4333Author: Mihail Konev <k.mvc@ya.ru>
4334Date:   Thu Jan 26 13:52:49 2017 +1000
4335
4336    autogen: add default patch prefix
4337    
4338    Signed-off-by: Mihail Konev <k.mvc@ya.ru>
4339
4340commit 2979011bc170c55894b9185b26376f8efc6db7d4
4341Author: Emil Velikov <emil.l.velikov@gmail.com>
4342Date:   Mon Mar 9 12:00:52 2015 +0000
4343
4344    autogen.sh: use quoted string variables
4345    
4346    Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent
4347    fall-outs, when they contain space.
4348    
4349    Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
4350    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
4351    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4352
4353commit 4a0082a1b6affa65d38294f0e13511525cd8ad15
4354Author: Peter Hutterer <peter.hutterer@who-t.net>
4355Date:   Tue Jan 24 10:32:07 2017 +1000
4356
4357    autogen.sh: use exec instead of waiting for configure to finish
4358    
4359    Syncs the invocation of configure with the one from the server.
4360    
4361    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4362    Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
4363
4364commit c74b070f2712c95f0db7c320a10232b0e5c83049
4365Author: Julien Cristau <jcristau@debian.org>
4366Date:   Sat Jan 7 16:20:31 2017 +0100
4367
4368    Fix wrong Xfree in XListFonts failure path
4369    
4370    'ch' gets moved inside the allocated buffer as we're looping through
4371    fonts, so keep a reference to the start of the buffer so we can pass
4372    that to Xfree in the failure case.
4373    
4374    Fixes: commit 20a3f99eba5001925b8b313da3accb7900eb1927 "Plug a memory leak"
4375    
4376    Signed-off-by: Julien Cristau <jcristau@debian.org>
4377    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4378    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4379
4380commit 663f47075fe67bab4f99bc5d186c49175d4a4334
4381Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4382Date:   Sun Jan 1 20:57:58 2017 -0800
4383
4384    specs/libX11: Update Portability Considerations for the 21st century
4385    
4386    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4387    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4388
4389commit 28f4b989b52fcf45c7e75a878d8d9c2583cd6b3c
4390Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4391Date:   Sun Jan 1 20:39:53 2017 -0800
4392
4393    specs/libX11: Fix broken synopsis for Data/Data16/Data32
4394    
4395    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4396
4397commit 382561951e3460b09c21a1a23748cde0315fbb19
4398Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4399Date:   Sun Jan 1 20:31:54 2017 -0800
4400
4401    specs/libX11: Add missing parameter types for XGetWindowProperty()
4402    
4403    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4404
4405commit 2beaecdb66965b861d6b790d151ba947f65f0a22
4406Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4407Date:   Sun Jan 1 20:16:08 2017 -0800
4408
4409    specs/libX11: Make paramdef spacing more consistent
4410    
4411    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4412
4413commit 4c436c6c14cfd4c397b011563bf13c2872861133
4414Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4415Date:   Sun Jan 1 20:02:53 2017 -0800
4416
4417    specs/libX11: Fix paramdef entries listing multiple parameters
4418    
4419    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4420
4421commit 1728b1a8a4718315da8e5c9cbc2c04bb75c74107
4422Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4423Date:   Sun Jan 1 19:38:40 2017 -0800
4424
4425    specs/libX11: More synopsis fixes
4426    
4427    Mostly transforming macro definitions and functions taking void arguments
4428    from undecorated <para> tags to use <funcsynopsis> tags to get decorations.
4429    
4430    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4431
4432commit f0dc83db7e3a3d4a76c0f9d24763b80f01c893a9
4433Author: Lucien Gentis <lucien.gentis@waika9.com>
4434Date:   Wed Jun 1 17:02:47 2016 +0200
4435
4436    Typos in "Xlib - C Language X Interface" document - Chapter 02
4437    
4438    This patch fixes typos and lack of tags in "Xlib - C Language X Interface" document - Chapter 02.
4439    
4440    Signed-off-by: Lucien Gentis <lucien.gentis@waika9.com>
4441    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4442    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4443
4444commit 20a3f99eba5001925b8b313da3accb7900eb1927
4445Author: Emilio Pozuelo Monfort <pochu@debian.org>
4446Date:   Tue Oct 25 21:30:15 2016 +0200
4447
4448    Plug a memory leak
4449    
4450    This was introduced in 8ea762f.
4451    
4452    Reported-by: Julien Cristau <jcristau@debian.org>
4453    Signed-off-by: Emilio Pozuelo Monfort <pochu@debian.org>
4454    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4455
4456commit 8f349feac24aacc958bd816afcc52380764e3d92
4457Author: Matthieu Herrb <matthieu.herrb@laas.fr>
4458Date:   Tue Oct 4 21:01:39 2016 +0200
4459
4460    libX11 1.6.4
4461    
4462    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
4463
4464commit 8ea762f94f4c942d898fdeb590a1630c83235c17
4465Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4466Date:   Sun Sep 25 21:25:25 2016 +0200
4467
4468    Validation of server responses in XGetImage()
4469    
4470    Check if enough bytes were received for specified image type and
4471    geometry. Otherwise GetPixel and other functions could trigger an
4472    out of boundary read later on.
4473    
4474    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4475    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4476
4477commit 8c29f1607a31dac0911e45a0dd3d74173822b3c9
4478Author: Tobias Stoeckmann <tobias@stoeckmann.org>
4479Date:   Sun Sep 25 21:22:57 2016 +0200
4480
4481    The validation of server responses avoids out of boundary accesses.
4482    
4483    v2: FontNames.c  return a NULL list whenever a single
4484    length field from the server is incohent.
4485    
4486    Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
4487    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4488
4489commit 78851f6a03130e3c720b60c3cbf96f8eb216d741
4490Author: walter harms <wharms@bfs.de>
4491Date:   Mon Aug 15 19:18:14 2016 +0200
4492
4493    XFree will accept NULL as argument
4494    
4495    since Xfree is a define for free():
4496      Xlibint.h:#define Xfree(ptr) free((ptr))
4497    
4498    Xfree will accept NULL and do nothing.
4499    
4500    Signed-off-by: Hans de Goede <hdegoede@redhat.com>
4501
4502commit 83adf3d1e3d0d6602244381334f75c216da4ab6e
4503Author: Matthew D. Fuller <fullermd@over-yonder.net>
4504Date:   Sat Jun 4 11:24:01 2016 -0500
4505
4506    Fixup param specification for XChangeProperty()
4507    
4508    Signed-off-by: Matthew D. Fuller <fullermd@over-yonder.net>
4509    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4510
4511commit 3129c757f9da8586ab8b8654a56c8f687cc9ef5c
4512Author: Mats Blakstad <mats.gbproject@gmail.com>
4513Date:   Sun Feb 28 13:22:03 2016 -0500
4514
4515     New compose keys for local languages in Togo
4516    
4517    Signed-off-by: James Cloos <cloos@jhcloos.com>
4518
4519commit e1011b9e2f6c82255959cf3cc1d8cda402ded0a9
4520Author: Daniel Albers <daniel@lbe.rs>
4521Date:   Wed Mar 9 14:35:48 2016 +0100
4522
4523    Add Compose sequence for U+1F4A9.
4524    
4525    Signed-off-by: Daniel Albers <daniel@lbe.rs>
4526
4527commit 6d7bb040c928485f2557c2c914b95cffb2354179
4528Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4529Date:   Sat Feb 6 14:18:32 2016 -0800
4530
4531    xcms: use size_t for pointer offsets passed to strncmp
4532    
4533    instead of converting to int and back
4534    
4535    Fixes clang warnings of the form:
4536    HVC.c:190:43: warning: implicit conversion changes signedness: 'int' to
4537          'unsigned long' [-Wsign-conversion]
4538              if (strncmp(spec, _XcmsTekHVC_prefix, n) != 0) {
4539                  ~~~~~~~
4540    
4541    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4542
4543commit a9266804eed38a83897ab5f0f9f8a8ab82a98882
4544Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4545Date:   Sat Feb 6 13:32:44 2016 -0800
4546
4547    xcms: use unsigned indexes when looping through unsigned values
4548    
4549    Clears many gcc warnings of the form:
4550    
4551    uvY.c: In function ‘XcmsCIEuvYToCIEXYZ’:
4552    uvY.c:263:19: warning: comparison between signed and unsigned integer
4553      expressions [-Wsign-compare]
4554         for (i = 0; i < nColors; i++, pColor++) {
4555                       ^
4556    
4557    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4558
4559commit 0ee0d383b4488b7b90d8bd50b75c371e0dc0d397
4560Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4561Date:   Sat Feb 6 13:01:25 2016 -0800
4562
4563    xcms: use size_t for strlen/sizeof values instead of converting to int & back
4564    
4565    Fixes gcc warnings of the form:
4566    
4567    IdOfPr.c: In function ‘XcmsFormatOfPrefix’:
4568    IdOfPr.c:69:32: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
4569         if ((len = strlen(prefix)) >= sizeof(string_buf)) {
4570                                    ^
4571    IdOfPr.c:83:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
4572       if (len >= sizeof(string_buf)) Xfree(string_lowered);
4573               ^
4574    IdOfPr.c:97:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
4575       if (len >= sizeof(string_buf)) Xfree(string_lowered);
4576               ^
4577    IdOfPr.c:104:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
4578         if (len >= sizeof(string_buf)) Xfree(string_lowered);
4579                 ^
4580    
4581    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4582
4583commit 4de6ed3e7b1833c52c9d58ab74d59d57ca2a9f0d
4584Author: Dominik Muth <nxdomainuser-muth@yahoo.com>
4585Date:   Thu Mar 26 07:52:58 2015 +0100
4586
4587    Xlib.h: Fix macros imitating C functions.
4588    
4589    The basic rule "put parantheses around macro parameters" should be
4590    observed where possible. Otherwise code like
4591    
4592        ConnectionNumber(foo = bar);
4593    
4594    fails to compile. (It obviously passes if ConnectionNumber is a C
4595    function.) There are several other macros amended for the same reason.
4596    
4597    This bug appeared while building http://ioccc.org/1993/cmills.c, so
4598    historically it was not present.
4599    
4600    Signed-off-by: Dominik Muth <muth@nxdomain.no-ip.biz>
4601    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4602    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4603
4604commit 3706b0f2b14cc97578a6bee620266edca2722ebf
4605Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4606Date:   Fri Nov 15 18:03:25 2013 -0800
4607
4608    Don't need to link libX11-xcb against libX11
4609    
4610    libX11-xcb only accesses data structures defined in X11 headers,
4611    it doesn't call any functions or reference any global variables
4612    in libX11 itself.  (Seems to have been left from previous XCL
4613    implementation.)
4614    
4615    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4616
4617commit eddf1bbd18872b286a9f939140f0cd9ba4e93804
4618Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4619Date:   Fri Jan 22 11:44:25 2016 -0800
4620
4621    Stop checking for preferred order of local transports
4622    
4623    Removes --with-local-transport-order=... flag to configure.
4624    
4625    Code which used this ordered list was removed in commit 15e5eaf6289
4626    which outsourced X11 connection handling & authentication to libxcb.
4627    
4628    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4629    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4630
4631commit 1a66c1e964ff8d11382313404f48b5a3d5ed8be8
4632Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4633Date:   Fri Jan 22 09:39:28 2016 -0800
4634
4635    Stop checking XTRANS_SECURE_RPC_FLAGS since we no longer use them
4636    
4637    Removes --enable-secure-rpc & --disable-secure-rpc flags to configure
4638    
4639    Code that used SECURE_RPC definitions was removed in commit 15e5eaf6289
4640    which outsourced X11 connection handling & authentication to libxcb.
4641    
4642    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4643    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
4644
4645commit 7eb724dc24505f1591ef32620fa63f079b540646
4646Author: Olivier Fourdan <ofourdan@redhat.com>
4647Date:   Thu Jan 21 11:54:19 2016 +0100
4648
4649    XKB: fix XkbGetKeyboardByName with Xming server
4650    
4651    XkbGetKeyboardByName relies on flags to read the data from the server.
4652    
4653    If the X server sends us the wrong flags or if a subreply is smaller
4654    than it should be, XkbGetKeyboardByName will not read all the available
4655    data and leave data in the buffer, which will cause the next _XReply()
4656    to fail with:
4657    
4658    [xcb] Extra reply data still left in queue
4659    [xcb] This is most likely caused by a broken X extension library
4660    [xcb] Aborting, sorry about that.
4661    xcb_io.c:576: _XReply: Assertion `!xcb_xlib_extra_reply_data_left' failed.
4662    Aborted
4663    
4664    Check if there is some extra data left at the end of
4665    XkbGetKeyboardByName() and discard that data if any is found.
4666    
4667    Many thanks to Peter Hutterer <peter.hutterer@who-t.net> for finding the
4668    root cause of the issue and Adam Jackson <ajax@redhat.com> for helping
4669    with the analysis!
4670    
4671    Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
4672    Reviewed-by: Daniel Stone <daniels@collabora.com>
4673    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
4674    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4675
4676commit 43ba0a68d3d17b496ec1f48d44921122ddd7d7d9
4677Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4678Date:   Sat Dec 19 18:03:41 2015 -0800
4679
4680    lcPubWrap: replace malloc(strlen) + strcpy with strdup
4681    
4682    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4683    Reviewed-by: Adam Jackson <ajax@redhat.com>
4684
4685commit 6fc95cb12b70c5a67cb4fc5e5749f9f1ec741e2a
4686Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4687Date:   Sat Dec 19 10:21:04 2015 -0800
4688
4689    XlcDL.c: reduce code duplication
4690    
4691    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4692    Reviewed-by: Adam Jackson <ajax@redhat.com>
4693
4694commit f7ecc0856be58608881d2086954cb71857ad64e1
4695Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4696Date:   Sat Dec 19 10:19:25 2015 -0800
4697
4698    XlcDL.c: replace strcpy+strcat sequences with snprintf
4699    
4700    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4701    Reviewed-by: Adam Jackson <ajax@redhat.com>
4702
4703commit 522989b34398bd6a6ea144c4af0ba69d6dc4faea
4704Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4705Date:   Sat Dec 19 10:05:42 2015 -0800
4706
4707    XDefaultOMIF: Remove comments referring to ancient Sun bug ids
4708    
4709    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4710    Reviewed-by: Adam Jackson <ajax@redhat.com>
4711
4712commit b738a104ae80e4270dd1d215ad0c6a80016982c2
4713Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4714Date:   Sat Dec 19 10:00:22 2015 -0800
4715
4716    XDefaultOMIF: additional code simplification
4717    
4718    Don't need to test for a case that we already returned for, don't need
4719    to store a count that will only ever be 1 if we didn't return, don't
4720    need to increment pointers to allow storing more than one item when we
4721    can only ever possibly do one.
4722    
4723    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4724    Reviewed-by: Adam Jackson <ajax@redhat.com>
4725
4726commit 31011cf100419269eae7409581c784638be503cf
4727Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4728Date:   Sat Dec 19 09:46:31 2015 -0800
4729
4730    XDefaultOMIF: replace strlen+Xmalloc+strcpy with strdup
4731    
4732    Code seems to have been originally written to handle appending multiple
4733    strings, but only ever operates on a single string.
4734    
4735    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4736    Reviewed-by: Adam Jackson <ajax@redhat.com>
4737
4738commit c27c46d5e22bbf60fb5608eaabe584b7fdeb0b09
4739Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4740Date:   Sat Dec 19 09:20:55 2015 -0800
4741
4742    Use strdup instead of Xmalloc+strcpy in _XDefaultOpenIM
4743    
4744    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4745    Reviewed-by: Adam Jackson <ajax@redhat.com>
4746
4747commit 4359dfabc04af082872d2bc2d5b52e26d6d93290
4748Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4749Date:   Fri Dec 4 22:20:53 2015 -0800
4750
4751    Delete #if 0 hunks of code
4752    
4753    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4754
4755commit a2f9dfac286f37e54eb47d4736cc3f0150224a84
4756Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4757Date:   Thu Dec 3 23:38:07 2015 -0800
4758
4759    Bug 93183: _XDefaultOpenIM memory leaks in out-of-memory error paths
4760    
4761    Rework code to store allocations directly into XIM struct instead of
4762    temporary local variables, so we can use _XCloseIM to unwind instead
4763    of duplicating it, and consistently jump to error handler on failure,
4764    instead of sometimes leaking and sometimes freeing.
4765    
4766    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93183
4767    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4768
4769commit 07a97b3944467dce085a1efd24706cc851d2caf2
4770Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4771Date:   Thu Dec 3 23:19:48 2015 -0800
4772
4773    Bug 93184: read_EncodingInfo invalid free
4774    
4775    Free the correct bits of memory if we run out and need to unwind
4776    
4777    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93184
4778    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4779
4780commit 11118e9eb3705fcbe42b6a68d4a8aa86ab0211f1
4781Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4782Date:   Sat Nov 28 13:18:11 2015 -0800
4783
4784    Remove unused definition of XCONN_CHECK_FREQ
4785    
4786    The only use of XCONN_CHECK_FREQ was removed in commit 15e5eaf62897b3179
4787    when we dropped the old Xlib connection handling in favor of xcb's.
4788    
4789    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4790    Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
4791
4792commit 5f0da8311a61498edf073cc877f5b467bfd5f863
4793Author: James Cloos <cloos@jhcloos.com>
4794Date:   Thu Dec 3 18:24:44 2015 -0500
4795
4796    Fix another missing update in cf4d5989383a
4797    
4798    Reported in:
4799    
4800       https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7
4801    
4802    Signed-off-by: James Cloos <cloos@jhcloos.com>
4803
4804commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51
4805Author: James Cloos <cloos@jhcloos.com>
4806Date:   Thu Dec 3 18:15:40 2015 -0500
4807
4808    Fix missing update in cf4d5989383a
4809    
4810    Reported in:
4811    
4812       https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7
4813    
4814    Signed-off-by: James Cloos <cloos@jhcloos.com>
4815
4816commit dbcb847a08c44d99e4e1de2ba777d63238fb0e03
4817Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4818Date:   Sun Sep 27 18:38:32 2015 -0700
4819
4820    Get rid of some extraneous ; at the end of C source lines
4821    
4822    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4823    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
4824
4825commit 121a1bad334459f66f78bfca6df53dc841cf97f8
4826Author: Gunnar Hjalmarsson <gunnarhj@ubuntu.com>
4827Date:   Wed Sep 23 11:44:55 2015 -0400
4828
4829    Add compose file for pt_PT similar to pt_BR
4830    
4831    This is a forward of the Ubuntu bug https://launchpad.net/bugs/518056
4832    
4833    One of the conclusions from the discussion on that bug report, which
4834    basically is about typing the ccedilla character easily on a non-
4835    Portuguese keyboard, is that X11 should include a compose file for
4836    pt_PT.UTF-8 similar to the file for pt_BR.UTF-8.
4837    
4838    FDO bug: https://bugs.freedesktop.org/show_bug.cgi?id=90300
4839    
4840    Signed-off-by: Gunnar Hjalmarsson <gunnarhj@ubuntu.com>
4841    Signed-off-by: James Cloos <cloos@jhcloos.com>
4842
4843commit 3f41d8a7f82eb5ffbd5c5d36472cf7043186b904
4844Author: Julien Cristau <jcristau@debian.org>
4845Date:   Fri May 1 13:50:15 2015 +0200
4846
4847    Mark _XNextRequest as hidden
4848    
4849    It's only used inside XNextRequest(), so doesn't need to be exported.
4850    
4851    Signed-off-by: Julien Cristau <jcristau@debian.org>
4852    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4853
4854commit a72d2d06c002b644b7040a0a9936c8525e092ba8
4855Author: Christian Linhart <chris@demorecorder.com>
4856Date:   Mon Sep 7 17:17:32 2015 +0200
4857
4858    fix for Xlib 32-bit request number issues
4859    
4860    Make use of the new 64-bit sequence number API in XCB 1.11.1 to avoid
4861    the 32-bit sequence number wrap in libX11.
4862    
4863    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71338
4864    Signed-off-by: Christian Linhart <chris@demorecorder.com>
4865    Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
4866    Reviewed-by: Adam Jackson <ajax@redhat.com>
4867
4868commit 58af066a764305c506efea7065ef7679369a1a98
4869Author: Thomas Klausner <wiz@NetBSD.org>
4870Date:   Sun Jul 19 10:23:21 2015 +0200
4871
4872    Ignore test-driver (used by newer autoconf).
4873    
4874    Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
4875    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4876
4877commit 80b9a346b9ba200fa4652560282e80d249519287
4878Author: Thomas Klausner <wiz@NetBSD.org>
4879Date:   Sun Jul 19 10:22:45 2015 +0200
4880
4881    Do not return() after exit().
4882    
4883    Signed-off-by: Thomas Klausner <wiz@NetBSD.org>
4884    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4885
4886commit c827edcd1c4a7f920aa25208083b5b58d60d2b44
4887Author: Ross Burton <ross.burton@intel.com>
4888Date:   Mon May 18 14:49:01 2015 +0100
4889
4890    Add missing NULL checks to ICWrap
4891    
4892    ICWrap.c dereferences the xim parameter passed in from client code without a
4893    NULL check.  I have seen mplayer trigger this resulting in a segfault.  In this
4894    case mplayer had called XOpenIM and NULL was returned which was later passed
4895    into XCreateIC.
4896    
4897    Patch originally by Drew Moseley <drew_moseley@mentor.com>.
4898    
4899    Signed-off-by: Ross Burton <ross.burton@intel.com>
4900    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4901    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4902
4903commit 26e0d2de294f8adf1ce65f1dbff0b59af41a00b9
4904Author: Alan Coopersmith <alan.coopersmith@oracle.com>
4905Date:   Thu Jun 4 20:51:17 2015 -0700
4906
4907    Replace Xmalloc+memset pairs with Xcalloc calls
4908    
4909    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4910
4911commit f0286b2770ece10aef5e2e8c004260217f12fd25
4912Author: Bhavi Dhingra <b.dhingra@samsung.com>
4913Date:   Thu Jun 4 19:07:12 2015 -0700
4914
4915    omGeneric.c: Correct the parameter usage of sizeof
4916    
4917    Incorrect parameter usage with sizeof. Earlier passed argument FontData
4918    will be 4 bytes always as its a pointer hence the change is needed and
4919    FontDataRec should be used for memset.
4920    
4921    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4922    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4923
4924commit 47da70d75f9e48e800719c0db752f9ccd2d77aea
4925Author: Peter Hutterer <peter.hutterer@who-t.net>
4926Date:   Tue May 19 12:30:22 2015 +1000
4927
4928    Fix three "use of uninitialized variable" coverity warnings
4929    
4930    False positive, if rlen/nbytes are unset we quit early before using it. Still,
4931    initialize it so we don't have to deal with these warnings again.
4932    
4933    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4934    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
4935    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4936
4937commit 19a30f17f30e9ae9641a7c0634fc52134208b060
4938Author: Peter Hutterer <peter.hutterer@who-t.net>
4939Date:   Mon May 18 07:56:22 2015 +1000
4940
4941    Fix an indentation issue
4942    
4943    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4944    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
4945    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4946
4947commit 013ccece124b990217ad3bcf2c41688e8fda1df8
4948Author: Peter Hutterer <peter.hutterer@who-t.net>
4949Date:   Mon May 18 07:55:17 2015 +1000
4950
4951    Fix potential memory leak
4952    
4953    If we hit the depth limit, filename leaks. Move the depth check up before we
4954    allocate filename.
4955    Introduced in 226622349a4b1e16064649d4444a34fb4be4f464.
4956    
4957    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
4958    Reviewed-by: Hans de Goede <hdegoede@redhat.com>
4959    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
4960
4961commit d3415d1f052530760b4617db45affcb984cfe35c
4962Author: Mike FABIAN <mfabian@redhat.com>
4963Date:   Mon Apr 20 17:59:30 2015 +0200
4964
4965    Fix spelling mistake introduced by 748d47e69f5c12d8557d56a8a8ec166588da7b93
4966    
4967    Sorry, my patch to fix the spelling mistakes in the ks_IN and sd_IN
4968    locales fixed it only partly, I introduced a new spelling mistake
4969    in the sd_IN locales. This patch fixes this.
4970    
4971    Signed-off-by: James Cloos <cloos@jhcloos.com>
4972
4973commit 748d47e69f5c12d8557d56a8a8ec166588da7b93
4974Author: Mike FABIAN <mfabian@redhat.com>
4975Date:   Wed Feb 19 11:46:45 2014 +0100
4976
4977    fix spelling mistakes in ks_IN and sd_IN devanagari locales
4978    
4979    The codeset must be *before* the modifier.
4980    
4981    See also: http://pubs.opengroup.org/onlinepubs/7908799/xbd/envvar.html
4982    
4983    opengroup> The syntax for these environment variables is thus defined as:
4984    opengroup>
4985    opengroup> [language[_territory][.codeset][@modifier]]
4986    
4987    Signed-off-by: James Cloos <cloos@jhcloos.com>
4988
4989commit c64fe5553aa4738f9d1d74a795f5651fbb7b1b09
4990Author: Mike FABIAN <mfabian@redhat.com>
4991Date:   Wed Feb 19 11:50:55 2014 +0100
4992
4993    add be_BY.UTF-8@latin and sr_RS.UTF-8@latin to locale.dir
4994    
4995    See also: https://bugzilla.redhat.com/show_bug.cgi?id=1066910
4996    
4997    If these are not in locale.dir,
4998    
4999        $ LANG=sr_RS.UTF-8@latin xterm
5000    
5001    and
5002    
5003        $ LANG=sr_RS@latin xterm
5004    
5005    give the warning:
5006    
5007        Warning: locale not supported by Xlib, locale set to C
5008    
5009    and some programs (like xmms) fail to find translations for Serbian
5010    in Latin because of this.
5011    
5012    Signed-off-by: James Cloos <cloos@jhcloos.com>
5013
5014commit c85be01b006126c4407eebd1eb6e01a17312b7b4
5015Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5016Date:   Sun Mar 22 16:46:45 2015 -0700
5017
5018    Move Compose \ o / to be with other emoji compose sequences
5019    
5020    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5021
5022commit 5a499ca7b064bf7e6a4fcc169f22862dce0c60c5
5023Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5024Date:   Mon Mar 9 15:28:29 2015 -0700
5025
5026    libX11 1.6.3
5027    
5028    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5029
5030commit c8e19b393defd53f046ddc2da3a16881221b3c34
5031Author: Joonas Javanainen <joonas.javanainen@gmail.com>
5032Date:   Thu Feb 5 17:31:04 2015 +0200
5033
5034    Fix XErrorEvent struct field order in man page
5035    
5036    In the man page the field "resourceid" was in a different place than
5037    in the actual struct layout in Xlib.h
5038    
5039    Signed-off-by: Joonas Javanainen <joonas.javanainen@gmail.com>
5040    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5041
5042commit ddf3b09bb262d01b56fbaade421ac85b0e60a69f
5043Author: Ran Benita <ran234@gmail.com>
5044Date:   Tue Feb 3 13:23:50 2015 -0500
5045
5046    compose: fix the description of modifiers in compose sequences
5047    
5048    The Compose format has a feature which allows specifying certain
5049    modifiers must (or must not) be present with a given keysym in the
5050    sequence.
5051    
5052    The grammar in imLcPrs.c and the Compose man page both do not match what
5053    the code actually does (see the handling of the variables
5054    `modifier_mask` and `modifier` in parseline() in imLcPrs.c, which are
5055    eventually matched as `ev->state & modifier_mask == modifier`).
5056    
5057    Also explicitly list the accepted modifier names, since they are
5058    not standard (e.g. "Ctrl" instead of "Control").
5059    
5060    Signed-off-by: Ran Benita <ran234@gmail.com>
5061    Signed-off-by: James Cloos <cloos@jhcloos.com>
5062
5063commit 129f13f385c50e3d8b53ea7441b17386b0f36aeb
5064Author: Ran Benita <ran234@gmail.com>
5065Date:   Tue Feb 3 13:23:49 2015 -0500
5066
5067    xkb: fix misleading comment about consumed modifiers
5068    
5069    In the spec and the man page the `mods_rtrn` argument is described as
5070    "backfilled with unconsumed modifiers" but actually it is backfilled
5071    with the *consumed* modifiers. This is also mentioned a few lines below
5072    in each case.
5073    
5074    Signed-off-by: Ran Benita <ran234@gmail.com>
5075    Signed-off-by: James Cloos <cloos@jhcloos.com>
5076
5077commit 446f5f7f41317a85a0cd0efa5e6a1b37bc99fba2
5078Author: Ingo Schwarze <schwarze@usta.de>
5079Date:   Tue Dec 9 10:44:13 2014 +0100
5080
5081    Fix pasto in XkbGetKeyBehaviors(3) manual
5082    
5083    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
5084    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5085
5086commit f3831dde6972e4da9e018c6a5f4013d8756a5e78
5087Author: Benno Schulenberg <bensberg@justemail.net>
5088Date:   Sun Nov 23 21:35:36 2014 +0100
5089
5090    nls: Sorting compose sequences rigorously in mirroring pairs, as is custom.
5091    
5092    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5093
5094commit a51681b60c84109fe19f5d449e13080522499324
5095Author: Benno Schulenberg <bensberg@justemail.net>
5096Date:   Thu Sep 19 16:42:01 2013 +0200
5097
5098    nls: Remove a duplicate locale name, and sort some others more strictly.
5099    
5100    Also improve the grammar of the initial comment.
5101    
5102    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5103    Signed-off-by: James Cloos <cloos@jhcloos.com>
5104
5105commit 426b7f850f5376db96a4b12420ee141603fcc3cd
5106Author: Benno Schulenberg <bensberg@justemail.net>
5107Date:   Fri Sep 20 12:37:29 2013 +0200
5108
5109    nls: Add a comment to the block of accented Hebrew letters.
5110    
5111    And align them in a nicer manner.
5112    
5113    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5114    Signed-off-by: James Cloos <cloos@jhcloos.com>
5115
5116commit 7474c6f1ee78dd097b1d0b4c7e3e4ea41317e335
5117Author: Benno Schulenberg <bensberg@justemail.net>
5118Date:   Fri Sep 20 12:05:09 2013 +0200
5119
5120    nls: Add one lowercase compose variant for ®, to mirror those for ©.
5121    
5122    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5123    Signed-off-by: James Cloos <cloos@jhcloos.com>
5124
5125commit 18dcd13514fa538afefa78c93523d9dbd4688e74
5126Author: Benno Schulenberg <bensberg@justemail.net>
5127Date:   Sat Sep 7 20:10:43 2013 +0200
5128
5129    nls: Adding the visual composing characters to two comment lines.
5130    
5131    The lines around them also contain the characters in the comments.
5132    
5133    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5134    Signed-off-by: James Cloos <cloos@jhcloos.com>
5135
5136commit 18d8307575af748913d5da17e3de45da2a22ede4
5137Author: Benno Schulenberg <bensberg@justemail.net>
5138Date:   Sat Sep 7 20:09:32 2013 +0200
5139
5140    nls: Grouping a lone superscript minus together with its mates.
5141    
5142    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5143    Signed-off-by: James Cloos <cloos@jhcloos.com>
5144
5145commit 33301cc45e6a2b8aa841ed6325547af970f8c4db
5146Author: Benno Schulenberg <bensberg@justemail.net>
5147Date:   Sat Sep 7 19:53:38 2013 +0200
5148
5149    nls: Grouping the compose sequences for Dstroke/dstroke together.
5150    
5151    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5152    Signed-off-by: James Cloos <cloos@jhcloos.com>
5153
5154commit 78fe1becb6c18fa33c0d5f04005b45d856f8952a
5155Author: Gioele Barabucci <gioele@svario.it>
5156Date:   Sun Sep 21 23:05:55 2014 +0200
5157
5158    Add compose sequence for U+20B9 INDIAN RUPEE SIGN
5159    
5160    The compose sequence for the new Indian Rupee sign is modelled after
5161    the sequence for the Euro sign.
5162    
5163    Signed-off-by: Gioele Barabucci <gioele@svario.it>
5164    Signed-off-by: James Cloos <cloos@jhcloos.com>
5165
5166commit 6101b967b641355dd863fd1ce52c6a7d58bcbe68
5167Author: Gabriel Souza Franco <gabrielfrancosouza@gmail.com>
5168Date:   Thu Jul 31 22:23:28 2014 -0300
5169
5170    Add double-arrow compose sequence
5171    
5172    Signed-off-by: Gabriel Souza Franco <gabrielfrancosouza@gmail.com>
5173    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5174    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5175
5176commit 368a6401c6a3275d3497fec38a3dcbc38cd9df60
5177Author: James Cloos <cloos@jhcloos.com>
5178Date:   Fri Aug 1 18:30:42 2014 -0400
5179
5180    Add cs_CZ.UTF-8 locale to configure.ac
5181    
5182    Commit cf4d5989383a should have included this.
5183    
5184    Reported-by: Colin Harrison <colin.harrison@virgin.net>
5185    Signed-off-by: James Cloos <cloos@jhcloos.com>
5186
5187commit cf4d5989383acc4ed1b7eebadde9f380f2129766
5188Author: James Cloos <cloos@jhcloos.com>
5189Date:   Tue Jul 29 15:02:56 2014 -0400
5190
5191    Add nls for cs_CZ.UTF-8
5192    
5193    Based on the iso8859-2 compose, and a bug report by Vladimír Marek,
5194    override the en_US.UTF-8 use of <dead_caron> <u> to enter »ǔ« instead
5195    to enter »ů«, and likewise for the majuscule, for the Czech locale.
5196    
5197    This evidently is the norm for Czech keyboards.
5198    
5199    Fixes bz#81875.
5200    
5201    The XI18N_OBJS and XLC_LOCALE.pre files are empty, as they are for
5202    several other locales.  That may require an update.
5203    
5204    Reported-by:  Vladimír Marek <vlmarek@volny.cz>
5205    Signed-off-by: James Cloos <cloos@jhcloos.com>
5206
5207commit d9f569572bd14db31921471e7b877523b5cf1e4c
5208Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5209Date:   Sat Jul 26 12:17:47 2014 -0700
5210
5211    Fix source paths for out-of-tree lintlib builds
5212    
5213    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5214
5215commit 0885cad1e4a9ed57266582be320be55259c881bf
5216Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5217Date:   Sun Jul 20 09:21:20 2014 -0700
5218
5219    specs/XKB: Markup fractions as <{super,sub}script> instead of <emphasis>
5220    
5221    Matches the way they were styled in original doc, before conversion
5222    to DocBook.
5223    
5224    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5225
5226commit d0a9e9d56bb003315787201ee525b4d00fd54e06
5227Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5228Date:   Sat Jul 19 23:55:47 2014 -0700
5229
5230    specs/XKB: acknowledge my contributions
5231    
5232    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5233
5234commit 9c0be82017f513e2eb63d59b095f1cf1955f2e2b
5235Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5236Date:   Sat Jul 19 23:53:48 2014 -0700
5237
5238    specs/XKB: Trim leading spaces off text lines
5239    
5240    perl -i -p -e 's{/\*(\S)}{/* \1}g;' *.xml
5241    
5242    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5243
5244commit 75b0b10990f38d966c6fcc821bf15e58c5a90c91
5245Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5246Date:   Sat Jul 19 20:30:55 2014 -0700
5247
5248    specs/XKB: Add olinks to libX11 for "X Library Functions Affected by Xkb"
5249    
5250    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5251
5252commit 5009621799444e9d1d284719f871d00be13e7330
5253Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5254Date:   Sat Jul 19 12:34:28 2014 -0700
5255
5256    specs/XKB: Fix miscelleanous typos & spelling errors
5257    
5258    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5259
5260commit ac219bf1522a592bd3e3283b1a6ea3dfd2c3c48c
5261Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5262Date:   Sat Jul 19 00:46:41 2014 -0700
5263
5264    specs/XKB: add some more indexterms by hand
5265    
5266    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5267
5268commit 5c3aa4c69e65ecf2e56d5e26f3833fb5d31973c2
5269Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5270Date:   Sat Jul 19 00:48:02 2014 -0700
5271
5272    specs/XKB: fixup various formatting issues in <programlisting>s
5273    
5274    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5275
5276commit 70c648ff852fd9bc784967cfc77ea70bd7f14c8d
5277Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5278Date:   Fri Jul 18 23:56:29 2014 -0700
5279
5280    specs/XKB: fixup various formatting issues in <literallayout>s
5281    
5282    Including translating some that are really just lists into
5283    <simplelist> markup.
5284    
5285    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5286
5287commit 94b56774784ac00b9db02403aecea10bb0814c10
5288Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5289Date:   Fri Jul 18 23:18:52 2014 -0700
5290
5291    specs/XKB: add some more links by hand
5292    
5293    random bits where a link looked handy
5294    
5295    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5296
5297commit 7a15a934cdb07ed1b991bd0ef633f32ee00b1833
5298Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5299Date:   Fri Jul 18 23:16:31 2014 -0700
5300
5301    specs/XKB: add links for terms in definition list under figure 1.1
5302    
5303    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5304
5305commit 92b86fc3c50fbb7ab2e36af10a2fb3fe6284f58c
5306Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5307Date:   Fri Jul 18 22:52:16 2014 -0700
5308
5309    specs/XKB: add links to more tables listing section references
5310    
5311    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5312
5313commit 06a4483a05053b4f8d8c0d4cc0513c68ea912676
5314Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5315Date:   Fri Jul 18 22:49:34 2014 -0700
5316
5317    specs/XKB: Table 4.1: remove page numbers & unnecessary para tags
5318    
5319    Page numbers refer to old doc format, didn't translate to new one
5320    
5321    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5322
5323commit 61bd55c6d1ab211b89d604bd05555dc417f6e53d
5324Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5325Date:   Sat Jul 19 11:55:51 2014 -0700
5326
5327    specs/XKB: remove unwanted white space around C -&gt; struct references
5328    
5329    perl -i -0 -p -e 's{\s*-&gt;\s*}{-&gt;}g' *xml
5330    
5331    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5332
5333commit 2be0cc0b2abbcc98cfd150210dea415a04787251
5334Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5335Date:   Fri Jul 18 22:26:20 2014 -0700
5336
5337    specs/XKB: replace -&gt; with &rarr; when used as arrow, not in C structs
5338    
5339    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5340
5341commit bf1f3d6f6f995303624679ae546f507c70967dc0
5342Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5343Date:   Fri Jul 18 22:21:10 2014 -0700
5344
5345    specs/XKB: replace =&gt; with &rArr; for double arrows
5346    
5347    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5348
5349commit c7ee427fc0a72abd4a4f147ab16a5d1128a6a2ba
5350Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5351Date:   Sat Jul 19 11:33:35 2014 -0700
5352
5353    specs/XKB: Markup some ranges with &ndash; instead of -
5354
5355commit ec4075303c6c0d1d64bfe378e585968f9a137da7
5356Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5357Date:   Fri Jul 18 22:10:45 2014 -0700
5358
5359    specs/XKB: Markup some subtractions with &minus; instead of -
5360    
5361    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5362
5363commit 441a267e461132a38abed205245f028686526f1d
5364Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5365Date:   Fri Jul 18 22:04:40 2014 -0700
5366
5367    specs/XKB: make sure all files have DOCTYPEs so standard entities work
5368    
5369    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5370
5371commit 88fd70bee410f290b4f540405fdc7ecd85c26f25
5372Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5373Date:   Fri Jul 18 21:55:41 2014 -0700
5374
5375    specs/XKB: Markup quoted terms as <quote> instead of with ""
5376    
5377    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5378
5379commit 3576587ff10334a8f48c34b4fe5b7e829dec9a1b
5380Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5381Date:   Sat Jul 19 09:01:26 2014 -0700
5382
5383    specs/XKB: fixup newlines between tags and punctuation
5384    
5385    Get rid of unwanted whitespace before punctuation by moving them to the
5386    lines with the tags, instead of before & after.
5387    
5388    perl -i -0 -p -e 's{\>\s*\n([\.,;:])}{>\1\n}g' *xml
5389    
5390    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5391
5392commit 6d5ec492cd28c206423337f926503349702af5a6
5393Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5394Date:   Fri Jul 18 21:29:33 2014 -0700
5395
5396    specs/XKB: fixup newlines between tags and parens
5397    
5398    Get rid of unwanted whitespace inside parens by moving them to the
5399    lines with the tags, instead of before & after.
5400    
5401    perl -i -0 -p \
5402         -e 's{(?<!--) \(\s*\n\<}{\n(<}g;' \
5403         -e 's{\>\s*\n\)([\.,;]?)(?! [^\n]*--)}{>)\1\n}g' *xml
5404    
5405    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5406
5407commit 59d688f4c787250e0b401a92b1db0437d8c60f2d
5408Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5409Date:   Fri Jul 18 21:09:24 2014 -0700
5410
5411    specs/XKB: Markup key terms as <firstterm> instead of <emphasis>
5412    
5413    Also add <indexterm> entries for most of them, to make their definitions
5414    or introductions easy to find from the index.
5415    
5416    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5417
5418commit 861f3087ee0f501362a67501f384c2ca4c7bfe73
5419Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5420Date:   Thu Jul 10 20:00:53 2014 -0700
5421
5422    specs/XKB: Manual fixup of type markup
5423    
5424    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5425
5426commit 6b96259dabe52701fd1bcaa0625b574180c4e769
5427Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5428Date:   Thu Jul 10 15:00:30 2014 -0700
5429
5430    specs/XKB: Manual fixup of parameter markup
5431    
5432    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5433
5434commit 5526dce6812a84102f556fdde8f2b52b21c8bcdc
5435Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5436Date:   Thu Jul 10 13:40:40 2014 -0700
5437
5438    specs/XKB: Manual fixup of struct name/field markup
5439    
5440    Handles typos that caused the scripts to miss matches, misnamed structs, etc.
5441    
5442    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5443
5444commit c36ee1a4db4e7876526190b8ab6b0da5867f76f7
5445Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5446Date:   Wed Jul 9 23:26:37 2014 -0700
5447
5448    specs/XKB: Manual fixup of symbol name markup
5449    
5450    Handles typos that caused the scripts to miss matches, misnamed masks, etc.
5451    
5452    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5453
5454commit eb1453a0c69606b8af96b90ddccf1b93a069fb35
5455Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5456Date:   Wed Jul 9 00:03:23 2014 -0700
5457
5458    specs/XKB: Manual fixup of function name markup
5459    
5460    Handles typos that caused the scripts to miss matches, functions like
5461    malloc & free from other libraries, function name patterns, etc.
5462    
5463    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5464
5465commit 252d99c87b60ac6f3f2b36f292f9b3880daabe26
5466Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5467Date:   Thu Jul 10 19:42:31 2014 -0700
5468
5469    specs/XKB: Markup protocol requests as <systemitem> instead of <emphasis>
5470    
5471    No great fit in DocBook, so follow what we used in Xlib spec.
5472    
5473    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5474
5475commit ed60df10aad15057577d5714c955d22d2a446e51
5476Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5477Date:   Thu Jul 10 15:45:57 2014 -0700
5478
5479    specs/XKB: Markup keyboard keys as <keycap> instead of <emphasis>
5480    
5481    Also uses <guilabel> for LED names/labels, for lack of a better fit
5482    in DocBook.
5483    
5484    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5485
5486commit f57b91ee497414083cc1bf481d28eb9ad9f965fb
5487Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5488Date:   Thu Jul 10 15:41:19 2014 -0700
5489
5490    specs/XKB: Markup characters & strings as <literal> instead of <emphasis>
5491    
5492    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5493
5494commit cfd4279c9b59d7e87c9f7c67692c87973adb7667
5495Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5496Date:   Thu Jul 10 11:08:14 2014 -0700
5497
5498    specs/XKB: Markup structs as <struct{name,field}> instead of <emphasis>
5499    
5500    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5501
5502commit a014bb4cc013b0d1b76524b7868e860c7e7ebc79
5503Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5504Date:   Wed Jul 9 23:30:54 2014 -0700
5505
5506    specs/XKB: Markup symbol names in table entries too
5507    
5508    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5509
5510commit 907f7ad7a7a977bf4f19daa3143d47c4d07ca33d
5511Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5512Date:   Tue Jul 8 23:35:48 2014 -0700
5513
5514    specs/XKB: Markup *Ptr as <type> instead of <emphasis>
5515    
5516    perl -i -p -e 's{<emphasis>(\w*Ptr)</emphasis>}{<type>\1</type>}g' *xml
5517    
5518    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5519
5520commit 9f6c00629fd4b713082cc11f9150f7aafd272c89
5521Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5522Date:   Tue Jul 8 23:34:29 2014 -0700
5523
5524    specs/XKB: Markup *Rec as <structname> instead of <emphasis>
5525    
5526    perl -i -p -e \
5527     's{<emphasis>(\w*Rec)</emphasis>}{<structname>\1</structname>}g' *xml
5528    
5529    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5530
5531commit 83839e37802b8b752c77859a95de60ad757feb79
5532Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5533Date:   Tue Jul 8 23:29:49 2014 -0700
5534
5535    specs/XKB: Markup XKB macros as <symbol> instead of <emphasis>
5536    
5537    Performed via:
5538     perl -n -e 'printf "s{<emphasis>\\s*%s\\s*</emphasis>}{<symbol>%s</symbol>};\n", $1, $1 if m{^#define\s+([^\s\(]*)}' \
5539      /usr/include/X11/extensions/XK*h /usr/include/X11/XKBlib.h \
5540      | sort -u > xkb-defines.pl
5541     perl -i -p -f xkb-defines.pl *.xml
5542    
5543    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5544
5545commit 9e397ed37ce4cc70621de347de3a795df88d4506
5546Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5547Date:   Tue Jul 8 22:53:12 2014 -0700
5548
5549    specs/XKB: Use ° instead of o for degrees.
5550    
5551    Conversion from framemaker turned superscripted "o" into plain "o".
5552    
5553    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5554
5555commit 3b8364c21f5119105a2c14ae8cc75a11494cb7a7
5556Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5557Date:   Tue Jul 8 22:46:03 2014 -0700
5558
5559    specs/XKB: Markup keysyms as <keysym> instead of <emphasis>
5560    
5561    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5562
5563commit 33bef065683c8f910f3722730503c0c0699ee8be
5564Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5565Date:   Tue Jul 8 22:26:56 2014 -0700
5566
5567    specs/XKB: Markup NULL as <symbol> instead of <emphasis>
5568    
5569    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5570
5571commit f10aa1e09468bd28454d85ac8ab55d9dc7178029
5572Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5573Date:   Tue Jul 8 22:22:29 2014 -0700
5574
5575    specs/XKB: Markup function args as <parameter> instead of <emphasis>
5576    
5577    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5578
5579commit 79ba58cceb69521fcb313c69233cf93a9fcb177f
5580Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5581Date:   Mon Jul 7 23:56:13 2014 -0700
5582
5583    specs/XKB: Convert remaining error names to errorname tags
5584    
5585    Most were caught by applying libX11 lists, but BadKeyboard & XKB*_Bad*
5586    are XKB-specific.   (Plus some were badly split across tag boundaries.)
5587    
5588    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5589
5590commit 42b2f5388c399949ece377f9cc9c479c06964972
5591Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5592Date:   Mon Jul 7 23:16:56 2014 -0700
5593
5594    specs/XKB: manually fixup some more emphasis tagging mismatches
5595    
5596    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5597
5598commit 3b2f47d44a55d93c65455ff183f3b47da04b1de1
5599Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5600Date:   Mon Jul 7 23:02:31 2014 -0700
5601
5602    specs/XKB: re-normalize <emphasis> layout in xml files
5603    
5604    Same script as before, just with <!-- xref --> comments out of the way
5605    now.
5606    
5607    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5608
5609commit ce95f152eda509263874c53fb7c6b4a6bdab2c29
5610Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5611Date:   Mon Jul 7 22:59:25 2014 -0700
5612
5613    specs/XKB: Remove remaining xref comments
5614    
5615    All the places marked by these have been turned into the appropriate
5616    link, xref, or olink tags now.
5617    
5618    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5619
5620commit 6590b66e19af8dff68888ac403ac82f5d585b4e2
5621Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5622Date:   Mon Jul 7 22:56:36 2014 -0700
5623
5624    specs/XKB: make olinks to xkbproto for references in section 10.13
5625    
5626    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5627
5628commit fcda446877a62e7443d7bc704ba3610e90d1e755
5629Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5630Date:   Mon Jul 7 22:22:28 2014 -0700
5631
5632    specs/XKB: Add <figure> tags and make Figure references link to them
5633    
5634    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5635
5636commit 087a2338476719e340dc3d5af0df6fdc4a26ce7a
5637Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5638Date:   Mon Jul 7 21:30:01 2014 -0700
5639
5640    specs/XKB: Turn Table references into links
5641    
5642    Adds id attributes to all table tags so we can link to them
5643    
5644    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5645
5646commit 135fa07b74cb50172c6a75768d499cd87ddb336e
5647Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5648Date:   Sun Jul 6 21:29:59 2014 -0700
5649
5650    specs/XKB: Turn section references into xref links
5651    
5652    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5653
5654commit 53e931d79926af8a3996253efd8b5f6c21d9e5d7
5655Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5656Date:   Sun Jul 6 21:17:18 2014 -0700
5657
5658    specs/XKB: Turn Chapter references into xref links
5659    
5660    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5661
5662commit b7f00ce5bcb0c00696bb82503ab548e14f04d17d
5663Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5664Date:   Sun Jul 6 20:40:18 2014 -0700
5665
5666    specs/XKB: Apply <emphasis> to semantic tag transformations from Xlib spec
5667    
5668    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5669
5670commit b00a7ddff2744238fbfe31c2298b02028a45a6ff
5671Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5672Date:   Sun Jul 6 20:34:51 2014 -0700
5673
5674    specs/XKB: Markup function names as <function> instead of <emphasis>
5675    
5676    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5677
5678commit bfbb58b7679221cb5c9212665209ea9099ad079a
5679Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5680Date:   Sun Jul 6 20:25:46 2014 -0700
5681
5682    specs/XKB: normalize <emphasis> layout in xml files
5683    
5684    - Stop placing <emphasis> on empty space, commas, and periods.
5685    - Move periods & commas after closing </emphasis> tag
5686    - move <emphasis> open & close tags to same line, instead of mirroring
5687      nroff layout.
5688    
5689    Simplifies automating further transformations of these tags.
5690    
5691    Performed via:
5692     perl -i -0 -p \
5693        -e 's{<emphasis>(\s*)</emphasis>}{}msg;' \
5694        -e 's{<emphasis>([\s\.,]*)</emphasis>\s*}{\1}msg;' \
5695        -e 's{\n([\.,])\s*}{\1\n}msg;' \
5696        -e 's{([^\.])([\.,])\s*</emphasis>}{\1</emphasis>\2}msg;' \
5697        -e 's{\s*<emphasis>\n\s*}{\n<emphasis>}msg;' *xml
5698    
5699    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5700
5701commit b16ee69a0103109a661a88140a1765dcd7bda634
5702Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5703Date:   Sun Jul 6 15:57:41 2014 -0700
5704
5705    specs/XKB: Convert to funcsynopsis+variablelist instead of informaltable
5706    
5707    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5708
5709commit b41d43d4cf0c0a1a049a171ee8cf6fd8a3ee4335
5710Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5711Date:   Sun Jul 6 19:29:38 2014 -0700
5712
5713    specs/XKB: Add index
5714    
5715    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5716
5717commit 72ae1d793be078db521dda60af578ece71f364de
5718Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5719Date:   Sun Jul 6 13:21:40 2014 -0700
5720
5721    specs/XKB: Fix various markup issues in functiondecl tables
5722    
5723    - Merge some functionargdecl entries incorrectly split across rows
5724    - Add missing parameter name markup to some functionargdecls
5725    - Add missing function prototype markup to a functiondecl
5726    - Remove stray emphasis tags in a functiondecl
5727    
5728    Allows them to correctly convert to funcsynopsis markup in next step.
5729    
5730    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5731
5732commit 9fdb973012de80ac60dbc59c39162f4e839fc5a4
5733Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5734Date:   Sun Jul 6 12:15:23 2014 -0700
5735
5736    specs/XKB: Convert header filenames to filename tags
5737    
5738    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5739
5740commit 5525e8433f93bce464412f27cffa203ea628f368
5741Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5742Date:   Tue Jul 8 00:05:32 2014 -0700
5743
5744    specs/libX11: disengender a user reference
5745    
5746    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5747
5748commit d8679eae9317b389ad4acb0430360ee0663e2af8
5749Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5750Date:   Fri Jul 11 18:41:42 2014 -0700
5751
5752    specs/libX11: Correct value of IconicState to match Xutil.h
5753    
5754    Xutil.h has always had a value of 3 for IconicState, since 2 was
5755    previously used for the long-obsolete ZoomState, so make the spec
5756    match what programs have used for decades.
5757    
5758    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5759    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5760
5761commit 7ce2b0f12a48fb832f457cbafb0e1144ef557f9a
5762Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5763Date:   Fri Jul 11 10:34:08 2014 -0700
5764
5765    Use C99 named initializers to fill in events passed to XSendEvent
5766    
5767    Forces compiler to zero-fill unset fields in the struct (fixing bug 81236)
5768    and allows optimizer to order field initialization to best fit cache layout
5769    or other considerations.
5770    
5771    Before & after output of gcc -S on AMD64 shows insertion of "rep stosq"
5772    instructions to rapidly zero-fill structs.
5773    
5774    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5775    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5776
5777commit 169805e1dc8743b37b00e24cf3a5eb8748f733ad
5778Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5779Date:   Sun Jul 6 11:13:49 2014 -0700
5780
5781    Fix validation of ctrls parameter to XkbGetPerClientControls()
5782    
5783    Nothing in the XKB spec states that the memory pointed to by ctrls has to
5784    be initialized to any given value when passed to the function, only that
5785    it is set by the function to the values returned by the X server:
5786    http://www.x.org/releases/X11R7.7/doc/libX11/XKB/xkblib.html#The_Miscellaneous_Per_client_Controls
5787    
5788    The check for the incoming value seems to be copied from
5789    XkbSetPerClientControls without explanation.
5790    
5791    Instead change it to checking if ctrls is non-NULL, since there's no
5792    point asking the X server to return a value the caller won't even see.
5793    
5794    Found while investigating report from cppcheck-1.65:
5795    [src/xkb/XKB.c:699] -> [src/xkb/XKB.c:719]: (warning) Possible null pointer
5796     dereference: ctrls - otherwise it is redundant to check it against null.
5797    
5798    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5799
5800commit 1e362fac92c6688fb42b195ccad16d7a337a34c1
5801Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5802Date:   Sun Jul 6 10:54:57 2014 -0700
5803
5804    Fix map->num_types check in XkbAddKeyType()
5805    
5806    Check is intended to ensure we allocate at least XkbNumRequiredTypes
5807    in map, but was accidentally marked with a ! causing the wrong check.
5808    
5809    Reported-by: Harms <wharms@bfs,de>
5810    Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
5811    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5812
5813commit ff9a5c199251a84fa59d14fd48dadb3f8920b54b
5814Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5815Date:   Sun Jul 6 15:08:21 2014 -0700
5816
5817    specs/libX11: Add missing spaces to 'unsignedint' & 'unsignedlong' types
5818    
5819    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5820
5821commit a06ea86773568926c36ae650b188fc818d540db7
5822Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5823Date:   Sun Jul 6 15:04:27 2014 -0700
5824
5825    specs/libX11: Fix height & width in parameter lists to be two separate entries
5826    
5827    "unsigned int width, unsigned int height", not a single parameter "height"
5828    of type "unsignedintwidth,".
5829    
5830    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5831
5832commit e4db5e503682b3304fe82e4b17b419a8e0f0a9f2
5833Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5834Date:   Sun Jul 6 14:38:10 2014 -0700
5835
5836    specs/libX11: Fix x & y in parameter lists to be two separate parameters
5837    
5838    "int x, int y" not a single parameter y of type "intx"
5839    
5840    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5841
5842commit aa8bda0db2c6d82515b90ceb4a7d6403e38895e9
5843Author: walter harms <wharms@bfs.de>
5844Date:   Sat Jun 7 12:03:17 2014 +0200
5845
5846    lcDefConv.c: fix use before check
5847    
5848    * Do not use variables before checked for NULL.
5849    * remove some superfluid spaces (Mark Kettenis)
5850    
5851    Signed-off-by: Harms <wharms@bfs,de>
5852    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5853    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5854
5855commit c0670e5d3ae330e611ecb05303d579a4f8a3d114
5856Author: Alan Coopersmith <alan.coopersmith@oracle.com>
5857Date:   Mon Jun 16 15:16:10 2014 -0700
5858
5859    Start adding Unicode 7.0 support to compose table
5860    
5861    New characters defined in http://www.unicode.org/charts/PDF/U1F300.pdf
5862    
5863    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5864
5865commit a4679baaa18142576d42d423afe816447f08336c
5866Author: walter harms <wharms@bfs.de>
5867Date:   Sat Jun 7 11:54:34 2014 +0200
5868
5869    rm redundant null checks
5870    
5871    remove more redundant NULL checks
5872    note that _XkbFree() is really Xfree()
5873    
5874    Signed-off-by: Harms <wharms@bfs,de>
5875    Reviewed-by: Rémi Cardona <remi@gentoo.org>
5876    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5877
5878commit 602d7f5030fe93b2fe7f29fb7310deb6f50cb6df
5879Author: walter harms <wharms@bfs.de>
5880Date:   Sat Jun 7 15:17:27 2014 +0200
5881
5882    libX11: rm redundante NULL checks
5883    
5884    This patch removes the last remaining  NULL checks for Xfree()
5885    
5886    Signed-off-by: Harms <wharms@bfs,de>
5887    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
5888    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
5889
5890commit d81fed46144d089bdfa1d916a28dffc9ebffe1e4
5891Author: walter harms <wharms@bfs.de>
5892Date:   Fri Jun 6 22:53:05 2014 +0200
5893
5894    Remove more redundant null checks before Xfree()
5895    
5896    Signed-off-by: Harms <wharms@bfs,de>
5897    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5898    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5899
5900commit 0b7fd7dbec136bae317bd9a329309eaa089beee3
5901Author: walter harms <wharms@bfs.de>
5902Date:   Thu Jun 5 18:37:40 2014 +0200
5903
5904    Remove redundant null checks before free
5905    
5906    This patch removes some redundant null checks before free.
5907    It should not change the code otherwise. Be aware that this
5908    is only the first series.
5909    
5910    Signed-off-by: Harms <wharms@bfs,de>
5911    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5912    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5913
5914commit 7d452fad5068ba92b964e92bc46708046f4044aa
5915Author: walter harms <wharms@bfs.de>
5916Date:   Wed Jun 4 17:12:31 2014 +0200
5917
5918    libX11/lcUTF8.c fix: dereferenced before check
5919    
5920    * Do not use variables before checked for NULL.
5921    
5922    Signed-off-by: Harms <wharms@bfs,de>
5923    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5924    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5925
5926commit a3808f51517a720e7ff738208af60865779dd6ef
5927Author: walter harms <wharms@bfs.de>
5928Date:   Wed Jun 4 17:10:20 2014 +0200
5929
5930    libX11/XKBNames.c fix: dereferenced before check
5931    
5932    * Do not use variables before checked for NULL.
5933    
5934    Signed-off-by: Harms <wharms@bfs,de>
5935    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5936    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5937
5938commit b3c9f6a17e430aabe16aecbe097f7312a0f6ff78
5939Author: walter harms <wharms@bfs.de>
5940Date:   Wed Jun 4 17:08:12 2014 +0200
5941
5942    libX11/lcGenConv.c fix: dereferenced before check
5943    
5944    * Do not use variables before checked for NULL.
5945    
5946    Signed-off-by: Harms <wharms@bfs,de>
5947    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5948    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5949
5950commit c6bc28d433243c32b3f74955f1478451b4fd27b5
5951Author: Andreas Schwab <schwab@linux-m68k.org>
5952Date:   Sun Jan 19 16:59:13 2014 +0100
5953
5954    Restore lost tabs in sed commands
5955    
5956    Signed-off-by: Andreas Schwab <schwab@linux-m68k.org>
5957    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
5958
5959commit e3dc0d17339e61eaf0b51b8907510984e3bf23cb
5960Author: Benno Schulenberg <bensberg@justemail.net>
5961Date:   Thu Sep 19 11:01:11 2013 +0200
5962
5963    nls: Transform Brazilian compose file to an include plus three overrides.
5964    
5965    Signed-off-by: James Cloos <cloos@jhcloos.com>
5966
5967commit 16c87dda4da2271aaecc5d8b6fe6ecd072cc584c
5968Author: Benno Schulenberg <bensberg@justemail.net>
5969Date:   Fri Sep 6 12:10:01 2013 +0200
5970
5971    nls: Adding the missing compose sequences with <comma> for O with ogonek.
5972    
5973    All other vowels with ogoneks can already be composed using <comma>.
5974    
5975    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5976    Signed-off-by: James Cloos <cloos@jhcloos.com>
5977
5978commit 655b60f48376069750b151c46da836fdd411c83b
5979Author: Benno Schulenberg <bensberg@justemail.net>
5980Date:   Fri Sep 6 12:04:24 2013 +0200
5981
5982    nls: Adding the missing postfix sequences for composing vowels with ogoneks.
5983    
5984    It existed for lowercase i, but not for uppercase I nor the other vowels.
5985    
5986    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
5987    Signed-off-by: James Cloos <cloos@jhcloos.com>
5988
5989commit 7f8f9a36ef901f31279c385caf960a22daeb33fe
5990Author: Owen W. Taylor <otaylor@fishsoup.net>
5991Date:   Fri May 9 18:21:05 2014 -0400
5992
5993    Fix XNextRequest() after direct usage of XCB
5994    
5995    When XCB owns the X socket, dpy->request is not updated, so
5996    NextRequest() and XNextRequest() return the wrong value. There's
5997    nothing we can do to fix NextRequest() while retaining ABI compat,
5998    but change XNextRequest() to grab the socket back from XCB,
5999    updating dpy->request.
6000    
6001    Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net>
6002    Reviewed-by: Uli Schlachter <psychon@znc.in>
6003    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6004
6005commit 0f9e734ea96556fe750a4baf354d42d5a87bcd14
6006Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6007Date:   Sun May 4 11:54:59 2014 -0700
6008
6009    Add missing .TE tags to end tables in Xkb man pages
6010    
6011    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6012
6013commit 280274e5292e013b43e552274111fab434f5ed4e
6014Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6015Date:   Sun May 4 11:23:17 2014 -0700
6016
6017    XCreateGC.man: simplify table to work with Solaris tbl
6018    
6019    Having every table cell be a text diversion (T{...T}) was too much for
6020    Solaris tbl to handle, and thus "man XCreateGC" would print the error
6021    /usr/man/man3x11/XCreateGC.3x11: line 402: Too many text block diversions
6022    tbl quits
6023    and not display the table of mask bits or any text in the man page after
6024    that table.   Since the #define column doesn't need special handling,
6025    making it not use text diversions brings the table under the tbl limit.
6026    
6027    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6028
6029commit 93bb325a54025dd15f4744abce54b358960420f9
6030Author: James Cloos <cloos@jhcloos.com>
6031Date:   Tue May 20 17:30:10 2014 -0400
6032
6033    Revert "nls: Adding compose sequences (with <parenleft> first) that GTK also has."
6034    
6035    Parenleft is already in use for sequences of the form <(> <letter> <)>
6036    to generate circled letters.
6037    
6038    Eg, <Multikey> <parenleft> <a> <parenright> generates ⓐ.
6039    
6040    This reverts commit f020235f4bd91fb6eade82f8c9f7b85a57981768.
6041    
6042    Signed-off-by: James Cloos <cloos@jhcloos.com>
6043
6044commit 060707851be918f2f507a26d17b016f764ddf2b4
6045Author: Benno Schulenberg <bensberg@justemail.net>
6046Date:   Fri Sep 6 11:42:59 2013 +0200
6047
6048    nls: Adding accessible compose sequences for Ș and Ț (with comma below).
6049    
6050    Compose sequences with <dead_belowcomma> exist, but very few keyboard
6051    layouts contain that symbol.  So a more usual character is needed to be
6052    able to easily compose Ș, ș, Ț and ț.  The semicolon is normally only
6053    used for composing letters with ogoneks -- but only vowels take ogoneks,
6054    so the character is free for consonants, and thus <semicolon> is used
6055    here to compose commas below.  It is somewhat fitting, because on most
6056    Romanian keyboards the Ș is placed on that key, and the Ț next to it.
6057    
6058    (Oh -- the more obvious sequences with <comma> were already taken for
6059    composing S and T with cedillas.)
6060    
6061    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6062    Signed-off-by: James Cloos <cloos@jhcloos.com>
6063
6064commit ca435c2f753aa2961fb35ac448cdb2cc77112755
6065Author: Benno Schulenberg <bensberg@justemail.net>
6066Date:   Tue Sep 3 21:10:33 2013 +0200
6067
6068    nls: Ordering some compose sequences in a more customary way.
6069    
6070    The custom seems to be: pairing the ones that have only the sequence of
6071    two keys reversed, and putting the one with the diacritic first first.
6072    
6073    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6074    Signed-off-by: James Cloos <cloos@jhcloos.com>
6075
6076commit f020235f4bd91fb6eade82f8c9f7b85a57981768
6077Author: Benno Schulenberg <bensberg@justemail.net>
6078Date:   Tue Sep 3 20:44:42 2013 +0200
6079
6080    nls: Adding compose sequences (with <parenleft> first) that GTK also has.
6081    
6082    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6083    Signed-off-by: James Cloos <cloos@jhcloos.com>
6084
6085commit bda0b3b5bd19154206dc40166364e73d4b6b1374
6086Author: Benno Schulenberg <bensberg@justemail.net>
6087Date:   Tue Sep 3 11:14:16 2013 +0200
6088
6089    nls: Allowing Romanian Ă and ă to be composed also with lowercase <u>.
6090    
6091    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6092    Signed-off-by: James Cloos <cloos@jhcloos.com>
6093
6094commit 8be4610939b833587954957f5963eb4191b43d19
6095Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6096Date:   Thu Mar 13 23:22:48 2014 -0700
6097
6098    Fix "follwing" typo in en_US.UTF-8/Compose comment
6099    
6100    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6101
6102commit 6f30e9034f29c3ae6ad7e617b3d5e903aa107b6a
6103Author: Reuben Thomas <rrt@sc3d.org>
6104Date:   Mon Jan 27 14:18:24 2014 +0000
6105
6106    en_US.UTF-8/Compose: Fix apparent copy-paste-o, changing capital to small A.
6107    
6108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6109
6110commit d6bd988bc00494914b38b95ee5df77ac4f32f19f
6111Author: Peter Hutterer <peter.hutterer@who-t.net>
6112Date:   Mon Mar 3 12:38:48 2014 +1000
6113
6114    man: fix man page for XkbGetMap
6115    
6116    Returned structure must be freed with XkbFreeKeyboard().
6117    
6118    Reported-by: Morten Bøgeskov <mb@dbc.dk>
6119    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
6120
6121commit a6dcf2201a05adbff54122df05a1e6325936abb6
6122Author: Ran Benita <ran234@gmail.com>
6123Date:   Tue Feb 11 13:26:16 2014 +0200
6124
6125    Remove dead USE_OWN_COMPOSE-protected code
6126    
6127    The build doesn't provide any way to define this option. It also refers
6128    to files (imComp.h) and functions (e.g. XimCompInitTables(),
6129    XimCompProcessSym()) which are not found anywhere, and the ordinary
6130    Compose implementation in xim doesn't use any of it.
6131    
6132    Signed-off-by: Ran Benita <ran234@gmail.com>
6133    Reviewed-by: Julien Cristau <jcristau@debian.org>
6134    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6135
6136commit b64bee2ddb7b96f00713a8b8435f11ad9ac1c9e0
6137Author: Ran Benita <ran234@gmail.com>
6138Date:   Sun Feb 16 15:24:58 2014 +0200
6139
6140    nls: always use XCOMM instead of # for comments in Compose.pre files
6141    
6142    Lines starting with # are considered as preprocessor directives in the
6143    .pre files.
6144    
6145    Fixes warnings like:
6146    <stdin>:3:0: error: invalid preprocessing directive #Khmer
6147    
6148    Signed-off-by: Ran Benita <ran234@gmail.com>
6149    Signed-off-by: James Cloos <cloos@jhcloos.com>
6150
6151commit 470e2289a3ebc59c5a35e54e1adeb0f261d5bf88
6152Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6153Date:   Thu Feb 6 13:48:08 2014 -0800
6154
6155    Fix typos in Xrm.c comments
6156    
6157    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6158
6159commit aacf95dacc7c598e7297894580d4d655593813b2
6160Author: Marko Myllynen <myllynen@redhat.com>
6161Date:   Mon Jan 13 16:43:18 2014 +0200
6162
6163    Annotate Finnish Compose map with Unicode code points
6164    
6165    Scripted annotation, no functional changes.
6166    
6167    Signed-off-by: James Cloos <cloos@jhcloos.com>
6168
6169commit 20fdccd81b54678376d49e00edfebbbe94951f07
6170Author: Teemu Likonen <tlikonen@iki.fi>
6171Date:   Fri Jul 20 19:21:04 2012 +0300
6172
6173    Fix "RING ABOVE" key in the Finnish compose file
6174    
6175    The Finnish keyboard standard defines that <dead_abovering> <space> must
6176    insert the character U+02DA RING ABOVE. Currently the Finnish Compose
6177    file inserts U+00B0 DEGREE SIGN even though the line's comment says
6178    "RING ABOVE". This commit changes the character to U+02DA RING ABOVE.
6179    
6180    Signed-off-by: Teemu Likonen <tlikonen@iki.fi>
6181    Signed-off-by: James Cloos <cloos@jhcloos.com>
6182
6183commit 8757e2ac8e04f2932ff437127f3e2ae9ac20c1d7
6184Author: Ran Benita <ran234@gmail.com>
6185Date:   Wed Jan 29 02:11:47 2014 +0200
6186
6187    nls: remove duplicate 'ohorn' and 'uhorn' compose sequences
6188    
6189    Since <Ohorn> == <U01A0> and <ohorn> == <U01A1>, when translated to
6190    keysyms:
6191    
6192     #define XK_Ohorn                      0x10001a0  /* U+01A0 LATIN CAPITAL LETTER O WITH HORN */
6193     #define XK_ohorn                      0x10001a1  /* U+01A1 LATIN SMALL LETTER O WITH HORN */
6194    
6195    (and similarly for uhorn), there is no need to have both names. Remove
6196    the unicode literal ones.
6197    
6198    Signed-off-by: Ran Benita <ran234@gmail.com>
6199    Signed-off-by: James Cloos <cloos@jhcloos.com>
6200
6201commit b98998cb3bea7cb3005f2e9d5bc5332d14b1d5d3
6202Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6203Date:   Sat Jan 18 22:35:18 2014 -0800
6204
6205    Add RANDR 1.4 requests & events to XErrorDB
6206    
6207    Checked against randrproto.txt & randr.h
6208    
6209    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6210
6211commit 321392ded15a7ee9d177d4ebe8846336ba76741c
6212Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6213Date:   Fri Jan 3 20:04:33 2014 -0800
6214
6215    Remove unused ETEST & ESZTEST macros from XlibInt.c
6216    
6217    Left behind when 15e5eaf62897 removed support for building without XCB.
6218    
6219    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6220    Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
6221
6222commit 1ffc0c5503d4f419fdbc765243832a53491bf5bc
6223Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6224Date:   Thu Dec 26 09:26:13 2013 -0800
6225
6226    _XkbReadGeomOverlay: check for NULL first, then use pointer
6227    
6228    Flagged by cppcheck 1.62:
6229    [lib/libX11/src/xkb/XKBGeom.c:479] -> [lib/libX11/src/xkb/XKBGeom.c:480]:
6230     (warning) Possible null pointer dereference: row - otherwise it is
6231     redundant to check it against null.
6232    
6233    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6234
6235commit ddf5f130cc29bb3bf8b9c757dcbac31bc56e9379
6236Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6237Date:   Thu Dec 26 09:22:49 2013 -0800
6238
6239    XkbSelectEventDetails: remove unnecessary assignments
6240    
6241    clear & selectAll are set to 0 already a few lines earlier,
6242    affectWhich is set to XkbMapNotifyMask a few lines later.
6243    None are used between the other assignments and the removed ones.
6244    
6245    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6246
6247commit 3d69b0a83e62f8f6fbdd952fc49cdbdf8825e1e6
6248Author: Jon TURNEY <jon.turney@dronecode.org.uk>
6249Date:   Thu Oct 7 18:46:08 2010 +0100
6250
6251    Don't try so hard to find a matching font with the given encoding
6252    
6253    See http://sourceware.org/bugzilla/show_bug.cgi?id=10948
6254    
6255    Currently, if the locale is UTF-8, no CJK fonts are installed, and someone
6256    does XCreateFontSet() with a font name of "*", we end up asking the server
6257    to list the (non-existent) fonts 11 times for each CJK encoding, which can
6258    take a while.
6259    
6260    A * wildcard can match multiple components in a XLFD name in XListFonts(),
6261    so there's no need to try adding more than one to get a match.
6262    
6263    We do try once with a leading '*-' in case the fontname isn't a full
6264    well-formed XLFD name, maybe even that isn't needed?
6265    
6266    (See also http://invisible-island.net/xterm/xterm.faq.html#slow_menus)
6267    
6268    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
6269    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6270
6271commit 7e163300735d4bcd3386b86eec112acdad139c59
6272Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6273Date:   Mon Dec 2 21:51:27 2013 -0800
6274
6275    unifdef -UISC
6276    
6277    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6278
6279commit 1e43c262d13cab2b759665f9f13bdedbc7afbfd4
6280Author: Benno Schulenberg <bensberg@justemail.net>
6281Date:   Thu Sep 19 13:20:05 2013 +0200
6282
6283    nls: Fix transposed locale identifier for Khmer.
6284    
6285    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6286    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6287    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6288
6289commit 0e45f64766c0557c8e99a979c70ca6f55664dae7
6290Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6291Date:   Sat Nov 16 20:21:54 2013 -0800
6292
6293    Drop X_LOCALE fallback for OS'es without setlocale()
6294    
6295    C89 or bust!   This was documented as being needed for "only Lynx,
6296    Linux-libc5, OS/2" and has never been enabled in modular builds,
6297    since none of those platforms have had anyone step up to add support
6298    since the X11R7 conversion to autotools.
6299    
6300    Mostly performed with unifdef -UX_LOCALE, followed by removal of files
6301    left without any purpose, and manual cleanup of remaining references.
6302    
6303    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6304    Reviewed-by: Adam Jackson <ajax@redhat.com>
6305
6306commit 6cb02b166361200da35ba14f52cd9aaa493eb0ea
6307Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6308Date:   Wed Oct 23 10:37:53 2013 -0700
6309
6310    Xcms file parsing should not require the impossible to succeed
6311    
6312    The field2 helper function, to split lines from Xcms.txt files into
6313    two tab delimited fields, contained a check:
6314    
6315        if ((*pBuf != '\n') || (*pBuf != '\0')) {
6316            return(XcmsFailure);
6317    
6318    which would cause it to return failure unless *pBuf had a value that
6319    was simultaneously equal to both \n & \0, and no one wants to live in
6320    a world where that could ever be true.
6321    
6322    This has gone unnoticed since 1991, since this only caused lines
6323    in Xcms.txt that started with whitespace to be rejected, but now
6324    gcc -Wlogicalop has brought it to our attention, and
6325    https://bugs.freedesktop.org/show_bug.cgi?id=70803 was filed.
6326    
6327    Now that we see it, and cannot unsee it, we change it to use the
6328    same logic as the check at other points in this function, to return
6329    failure only if we hit \n or \0 before we find the first non-whitespace
6330    character, so that lines starting with whitespace will have the space
6331    skipped over to get to the color name to be defined.
6332    
6333    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6334    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
6335
6336commit 18a5278b008e9faa59b346fcab18a8d74b875fda
6337Author: Gaetan Nadon <memsize@videotron.ca>
6338Date:   Sat Sep 28 17:33:52 2013 -0400
6339
6340    makekeys: don't need to use target-specific CFLAGS
6341    
6342    It's the only thing built in that directory, so we can use AM_CFLAGS
6343    and AM_CPPFLAGS as usual.
6344    
6345    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6346    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
6347
6348commit 5dcb40f28d59587597d2ff6e6ac64c71cfe6ff7b
6349Author: James Cloos <cloos@jhcloos.com>
6350Date:   Tue Sep 17 12:50:42 2013 -0400
6351
6352    nls/en_US.UTF-8/Compose.pre: Fix typo.
6353    
6354    Fix typo added in 215ce6a67863, s/actute/acute/.
6355    
6356    Fixes bug #69476.  Reported by Jean Krohn.
6357    
6358    Signed-off-by: James Cloos <cloos@jhcloos.com>
6359
6360commit cb107760df33ffc8630677e66e2e50aa37950a5c
6361Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6362Date:   Sun Sep 8 18:37:01 2013 -0700
6363
6364    libX11 1.6.2
6365
6366commit 215ce6a67863de7acfd6dd3562b4fd97ef87b411
6367Author: Benno Schulenberg <bensberg@justemail.net>
6368Date:   Sun Sep 1 12:38:30 2013 +0200
6369
6370    nls: Adding more accessible compose sequences for J́ and j́.
6371    
6372    Few keyboards have an <acute> key, so this adds the much more
6373    accessible and usual compose sequences with <apostrophe>, ánd
6374    the most comfortable ones with <dead_acute>.
6375    
6376    Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
6377    Signed-off-by: James Cloos <cloos@jhcloos.com>
6378
6379commit e9b14d10d0258bfcc273ff8bc84cd349dccda62c
6380Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6381Date:   Sat Aug 24 17:27:43 2013 -0700
6382
6383    Bug 68413 - [Bisected]Error in `xterm': realloc(): invalid next size
6384    
6385    Pass *new* size to realloc, not old size.
6386    
6387    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6388
6389commit c2b8e30790c21d6386767265263b3294ce1b1f9a
6390Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6391Date:   Fri Aug 16 21:04:02 2013 -0700
6392
6393    Stop checking for HAVE_DIX_CONFIG_H on the client side
6394    
6395    Leftover from when these XKB files were shared with the server sources
6396    and could be compiled in either the client or server, with the different
6397    autoconf config files in each.
6398    
6399    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6400
6401commit 84276609b2f0aec74fb464c428c7db5714b0fcfc
6402Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6403Date:   Fri Aug 16 18:27:28 2013 -0700
6404
6405    Rearrange some variable declarations & initializations in XKB
6406    
6407    Little things noticed during XKB restyling that seemed to make the
6408    code easier to read.
6409    
6410    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6411
6412commit b90b7e859cf45ec76921fa21bbfc1f3840d6e8d1
6413Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6414Date:   Sun Aug 11 13:29:33 2013 -0700
6415
6416    Reindent XKB code to X.Org standard style
6417    
6418    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6419
6420commit c0a0f78eb49c2e4ad956209de77475c85b9314ea
6421Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6422Date:   Fri Aug 16 18:14:14 2013 -0700
6423
6424    Fix overflow checks in _XkbReadKeySyms when key_sym_map is already created
6425    
6426    We were checking to make sure that the largest keysym value was within
6427    the range of the allocated buffer, but checking against different limits
6428    in the not-yet-allocated vs. the already-allocated branches.
6429    
6430    The check should be the same in both, and reflect the size used for the
6431    allocation, which is based on the maximum key code value, so we move it
6432    to be a common check, before we branch, instead of duplicating in each
6433    branch.
6434    
6435    map->key_sym_map is an array of XkbSymMapRec structs, [0..max_key_code]
6436    map->syms is the array for which num_syms is recorded, hence is not the
6437    right value to check for ensuring our key_sym_map accesses are in range.
6438    
6439    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6440    Reported-by: Barry Kauler <bkauler@gmail.com>
6441    Tested-by: Barry Kauler <bkauler@gmail.com>
6442
6443commit bea6cbd027973142fc64532274e1d16861b47190
6444Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6445Date:   Sun Aug 11 17:02:21 2013 -0700
6446
6447    Remove long unused src/udcInf.c
6448    
6449    I can find no record of what this file was for.  Neither the X11R6.8.2
6450    monolith Imakefile nor any modular release Makefile.am have ever built
6451    it and nothing else references it.
6452    
6453    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6454
6455commit bf3501e0395abe890acfea98fdd9f50a6966f118
6456Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6457Date:   Sun Aug 11 00:07:33 2013 -0700
6458
6459    Remove unnecessary casts of pointers to (char *) in calls to Xfree()
6460    
6461    Left one cast behind that is necessary to change from const char *
6462    to char * in src/xlibi18n/lcCharSet.c.
6463    
6464    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6465
6466commit 6ead9dd92ab90aabd9f0e328d59597e6b5bc09d3
6467Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6468Date:   Sat Aug 10 23:57:55 2013 -0700
6469
6470    Don't cast sizeof() results to unsigned when passing to Xmalloc/Xcalloc
6471    
6472    sizeof() returns size_t, malloc() & calloc() expect sizes in size_t,
6473    don't strip down to unsigned int and re-expand unnecessarily.
6474    
6475    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6476
6477commit 25a7a329def672fc8d26078538173777850c6390
6478Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6479Date:   Sat Aug 10 23:51:08 2013 -0700
6480
6481    Remove even more casts of return values from Xmalloc/Xrealloc
6482    
6483    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6484
6485commit f8fa16092a148b74ca35b4beb182053352606f2f
6486Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6487Date:   Sat Aug 10 23:05:13 2013 -0700
6488
6489    xlibi18n: fix argsize argument to _XlcParsePath
6490    
6491    The array is defined as having NUM_LOCALEDIR entries, so use that
6492    instead of hardcoded 256 value (the other two calls already did this).
6493    
6494    Reported by parfait:
6495       Buffer overflow (CWE 120): In pointer dereference of argv[argc] with index argc
6496          Pointer size is 64 elements (of 8 bytes each), index is 255
6497            at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'.
6498            called at line 178 in function '_XlcParsePath' with argv = argv.
6499            called at line 722 in function '_XlcLocaleLibDirName' with argv = args, argsize = 256.
6500            at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'.
6501            called at line 178 in function '_XlcParsePath' with argv = argv.
6502            called at line 638 in function '_XlcLocaleDirName' with argv = args, argsize = 256.
6503    
6504    [ This bug was found by the Parfait 1.2.0 bug checking tool.
6505      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
6506    
6507    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6508
6509commit e7d46c6452c0b90fd66ae9f538546b968e0dd608
6510Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6511Date:   Sat Aug 10 22:32:42 2013 -0700
6512
6513    i18n modules: Fix some const cast warnings
6514    
6515    imRm.c: In function '_XimSetICMode':
6516    imRm.c:2419:37: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6517    imRm.c:2420:30: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6518    
6519    lcGenConv.c: In function 'byteM_parse_codeset':
6520    lcGenConv.c:345:13: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6521    
6522    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6523
6524commit cbd86eccf175dc82a5cbcea54c8bd21ce18b70c0
6525Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6526Date:   Sat Aug 10 22:21:54 2013 -0700
6527
6528    xlibi18n: Fix a bunch of const cast warnings
6529    
6530    Add const qualifiers to casts where needed, remove other casts that
6531    are no longer needed.
6532    
6533    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6534
6535commit eb3676113fc2dd0f34d92b89beb81b3f61569aa1
6536Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6537Date:   Sat Aug 10 22:18:00 2013 -0700
6538
6539    Fix const handling in XSetLocaleModifiers
6540    
6541    Instead of reusing the input parameter to store the output, make a
6542    result variable instead, so that there's less const confusion.
6543    
6544    Fixes gcc warnings:
6545    lcWrap.c: In function 'XSetLocaleModifiers':
6546    lcWrap.c:87:18: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6547    lcWrap.c:91:25: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6548    lcWrap.c:93:12: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
6549    
6550    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6551
6552commit 8ebbffa98563960910152e4f2e31cb032375d871
6553Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6554Date:   Sat Aug 10 21:46:37 2013 -0700
6555
6556    Constify lc_name argument to _XlcLocaleDirName() & _XlcLocaleLibDirName()
6557    
6558    Makes code considerably less crufty and clears gcc warnings:
6559    XlcDL.c: In function '_XlcDynamicLoad':
6560    XlcDL.c:384:44: warning: cast discards '__attribute__((const))' qualifier
6561     from pointer target type [-Wcast-qual]
6562    XlcDL.c:386:51: warning: cast discards '__attribute__((const))' qualifier
6563     from pointer target type [-Wcast-qual]
6564    
6565    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6566
6567commit 07e4e864107b38c2f393564fdacc90f4e858f23f
6568Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6569Date:   Sat Aug 10 13:37:53 2013 -0700
6570
6571    init_om: remove unneeded extra copy of string to local buffer
6572    
6573    Strings from the supported_charset_list[] were being copied one by
6574    one to a stack buffer, and then strdup called on that buffer.
6575    
6576    Instead, just strdup the original string, without the local copy,
6577    and use a more traditional for loop, so it's easier to figure out
6578    what the code is doing (cleaning up a gcc const-cast warning in
6579    the process).
6580    
6581    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6582    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6583
6584commit 1cec14dad904ba21a861f4af131be5982ecb83dd
6585Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6586Date:   Sat Aug 10 12:34:53 2013 -0700
6587
6588    Delete unused XKB_INSURE_SIZE macro from XKBlibint.h
6589    
6590    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6591    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6592
6593commit 5f32182c7c4045540ff3833c48ee24a3a25726e2
6594Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6595Date:   Sat Aug 10 12:19:17 2013 -0700
6596
6597    miRegionOp(): ensure region size is not updated if realloc fails
6598    
6599    This function performs operations on a region, and when finished,
6600    checks to see if it should compact the rectangle list.  If the
6601    number of rectangles for which memory is allocated in the list is
6602    more than twice the number used, it tries to shrink.   realloc()
6603    should not fail in this case, but if it does, might as well keep
6604    the correct value for the number of allocated rectangles, so we
6605    don't try to grow it unnecessarily later if adding to the region.
6606    
6607    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6608
6609commit bd2a0b5a187798bb2e2f05dc5062ca79e37075dd
6610Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6611Date:   Sat Aug 10 12:19:17 2013 -0700
6612
6613    miRegionCopy(): handle realloc failure better
6614    
6615    Zero out the region size when freeing the region so callers don't think
6616    there's anything there.   (Pointer is already set to NULL from the realloc
6617    result itself.)  Return 0 to the callers, and have them cascade that back
6618    to their callers to indicate failure, instead of their usual return value
6619    of 1 on success.
6620    
6621    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6622
6623commit 5dc8b5385d513bbda88697c2372db750d23f46d4
6624Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6625Date:   Sat Aug 10 11:27:22 2013 -0700
6626
6627    Avoid memory leak/corruption if realloc fails in Xregion.h:MEMCHECK macro
6628    
6629    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6630
6631commit 453c4ee436ef32d91501d7736d7a91c1aeafc565
6632Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6633Date:   Sat Aug 10 12:07:51 2013 -0700
6634
6635    Avoid memory leak/corruption if realloc fails in imLcPrs.c:parseline()
6636    
6637    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6638
6639commit b3fea74ec5b7d4f83755a52a8d49c564b71c6d12
6640Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6641Date:   Sat Aug 10 12:30:39 2013 -0700
6642
6643    lcDB.c: ensure buffer size is updated correctly if realloc fails
6644    
6645    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6646
6647commit 43bb822c714a73c3b2d15e621ffb3333cd10da8c
6648Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6649Date:   Sat Aug 10 11:07:47 2013 -0700
6650
6651    Avoid memory leak/corruption if realloc fails in XlcDL.c:resolve_object()
6652    
6653    Previously, if realloc failed to increase the size, we'd still
6654    record that we had allocated the larger size, but the pointer
6655    to it would be NULL, causing future calls to be broken, and the
6656    previous allocation to be lost/leaked.
6657    
6658    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6659    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6660
6661commit 5d47a39978e92bb34ec928b1b15d71c0c2434870
6662Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6663Date:   Fri Aug 9 23:33:03 2013 -0700
6664
6665    omGeneric.c: convert sprintf calls to snprintf
6666    
6667    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6668    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6669
6670commit 88a27a2aa9b7d35cb79b16334ea3413e572b724a
6671Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6672Date:   Fri Aug 9 23:30:30 2013 -0700
6673
6674    ximcp/imRm.c: convert sprintf calls to snprintf
6675    
6676    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6677    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6678
6679commit 4fadae243fb485628c9a137f5da3489ed6214b21
6680Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6681Date:   Fri Aug 9 23:02:12 2013 -0700
6682
6683    xlibi18n: convert sprintf calls to snprintf
6684    
6685    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6686    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6687
6688commit 36a7edf0e5edfc5ef4ff2c3a8b4fa3dc4796e854
6689Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6690Date:   Fri Aug 9 23:02:12 2013 -0700
6691
6692    lcfile: skip over any null entries in args list
6693    
6694    Previous code seemed to assume that printf("%s", NULL) would result
6695    in a 0-length string, not "(null)" or similar, but since there's no
6696    point looking for files in "(null)/filepath...", instead we just
6697    skip over NULL entries in search paths when generating file names.
6698    
6699    In the *DirName() functions, this effectively just moves the "bail on
6700    NULL in arg[i]" check up from the later code that assigned it to targetdir
6701    and then bailed if that was NULL.
6702    
6703    Not sure how there ever could be a NULL in arg[i], given the current
6704    implementation of XlcParsePath, but it's easy enough to check once and
6705    reject up front instead of on every reference.
6706    
6707    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6708
6709commit ee0824f24392d5ca3d5fd5f5ed8d78c0d892f7c0
6710Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6711Date:   Fri Aug 9 22:00:09 2013 -0700
6712
6713    Fix file leak on malloc error in XlcDL.c:resolve_object()
6714    
6715    File Leak: Leaked File fp
6716       at line 219 of lib/libX11/src/xlibi18n/XlcDL.c in function 'resolve_object'.
6717          fp initialized at line 198 with fopen
6718    
6719    [ This bug was found by the Parfait 1.2.0 bug checking tool.
6720      http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ]
6721    
6722    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6723    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
6724
6725commit 9b291044a240e5b9b031ed814e0c84e53a1c3084
6726Author: Julien Cristau <jcristau@debian.org>
6727Date:   Sat Jun 15 18:02:21 2013 +0200
6728
6729    Add missing locales to configure.ac
6730    
6731    Commits 40761898692e5063957bfa2518cca3d35b2e354a and
6732    f198c6aa98f88ff285d903175a3c4c0fd33a4575 added two new locales
6733    (sr_CS.UTF-8 and km_KH.UTF-8), but didn't list them in configure.ac,
6734    meaning they're not included in tarballs.
6735    
6736    Signed-off-by: Julien Cristau <jcristau@debian.org>
6737    Reviewed-by: James Cloos <cloos@jhcloos.com>
6738
6739commit 8f58e54a5f46c3cd4897a23b89950f4800ae38d4
6740Author: ISHIKAWA,chiaki <ishikawa@yk.rim.or.jp>
6741Date:   Tue Dec 18 15:28:05 2012 +0000
6742
6743    Fix bogus timestamp generated by XIM
6744    
6745    Fix bogus timestamp generted by XIM due to uninitialized
6746    data field. Also set appropriate serial, too.
6747    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39367
6748    
6749    Signed-off-by: Chiaki ISHIKAWA <ishikawa@yk.rim.or.jp>
6750    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6751
6752commit e7fd6f0eda57300df4d6b695b7064610ca5dec57
6753Author: Egbert Eich <eich@freedesktop.org>
6754Date:   Thu Jun 16 18:47:49 2011 +0200
6755
6756    XIM: Fix sync problem on focus change.
6757    
6758    XSetICFocus() and XUnsetICFocus() are both asynchronous events.
6759    This is a pretty stupid idea: those functions may undo certain
6760    settings on the client side for which requests from the server
6761    may still be in the queue unprocessed. Thus things may be set
6762    in the wrong order ie instead of set -> unest it will be unset -> set.
6763    Moreover there is no way for either the client or the server to
6764    cause the event queue to be flushed - which is pretty bad as
6765    XIM is bidirectional.
6766    The scenario is as follows:
6767    Two ICs are created:
6768            ic1 = XCreateIC(im,
6769                XNInputStyle, XIMPreeditCallbacks | XIMStatusCallbacks,
6770                XNClientWindow, window,
6771                XNPreeditAttributes, preedit_attr,
6772                XNStatusAttributes, status_attr,
6773                NULL);
6774            ic2 = XCreateIC(im, XNInputStyle,
6775                           XIMPreeditNothing | XIMStatusNothing,
6776                           XNClientWindow, window, NULL);
6777    Then the focus is removed from ic2:
6778            XUnsetICFocus(ic2);
6779    If SCIM is used as the input server it will send a bunch of requests
6780    following an XCreateIC(). One of the requests registers a key release
6781    filter. XUnsetICFocus() unsets both key press and release filters.
6782    Since it is asynchronous, the input server requests to register key
6783    press and release filters may not have been processed, when XUnsetICFocus()
6784    is called. Since there is no explicite way for client programs to enforce
6785    the request queue to be flushed explicitely before an X[Set/Unset]ICFocus()
6786    call it would be safest to make those two calls synchronous in the sense
6787    that they ensure the request queue has been handled before they execute.
6788    The easiest way to do this from Xlib is thru a call to XGetICValues()
6789    which sends a request to the server and subsequently reads the queue
6790    from the server to the client. This will cause all outstanding requests
6791    in the queue to be read and handled.
6792    This is an ugly hack and this could be fixed directly in the client,
6793    however it seems to be easier to fix Xlib than to fix numerous clients.
6794    This problem arose since there is no well documented way how to handle
6795    and synchronize XIM requests and not all input servers send requests
6796    when an IC is created.
6797    This has been discussed extensively in:
6798     https://bugzilla.novell.com/show_bug.cgi?id=221326
6799    
6800    Signed-off-by: Egbert Eich <eich@freedesktop.org>
6801
6802commit 26ec7d3821bc19debc73c8c3e42e6e33ef6f856e
6803Author: Egbert Eich <eich@freedesktop.org>
6804Date:   Thu Jun 16 17:28:39 2011 +0200
6805
6806    XIM: Fix race on focus change: set 'FABRICATED' only when keyev filters in place.
6807    
6808    When synthesized key events are sent on commit XIM sets the 'fabricated'
6809    flag so that the keypress handler knows that these were not real events.
6810    This also happens when committing due to the loss of focus. However in this
6811    case the keypress/release filters which consume and unset this flag are no
6812    longer in the filter chain.
6813    So the flag is erronously set when a real keyboard event is received after
6814    focus has been regained. So the first event is wrongly treated as a
6815    fabricated key in the keypress handler which will at the same time reset
6816    the flag so the second key event is treated correctly.
6817    This fix only sets the flag when at least one of the keyboard filters is in
6818    place.
6819    How to reproduce this bug: run scim, choose a Japanese input method start
6820    two instances of xterm: start typing in one xterm (this should pop up an
6821    IM window). Without comitting (hitting 'enter') move focus to the other
6822    xterm, then move focus back. Start typing again. The first character will
6823    be committed immediately without popping up an input window.
6824    With this fix this behavior is gone.
6825    
6826    See also: https://bugzilla.novell.com/show_bug.cgi?id=239698
6827    
6828    Signed-off-by: Egbert Eich <eich@freedesktop.org>
6829
6830commit 44f84223f5e2dd46883fcbd352af2798bfa9aeb6
6831Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6832Date:   Mon Jul 29 21:29:49 2013 -0700
6833
6834    libX11 1.6.1
6835    
6836    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6837
6838commit d19cfaca15826943d3c637ef7fa5db0a23d5feed
6839Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6840Date:   Sat Jul 27 12:19:00 2013 -0700
6841
6842    Fix undefined XCMSDIR error when building lint library
6843    
6844    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6845
6846commit 3083cd43d7dcd59da587975e7cadda681cd8a103
6847Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6848Date:   Sat Jul 27 00:36:08 2013 -0700
6849
6850    Add ku_TR.UTF-8 (Kurdish language, Turkey region) to compose/locale.dir
6851    
6852    Upstreaming from changes originally integrated into OpenSolaris
6853    under Sun bug id 6882572.
6854    
6855    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6856    Reviewed-by: Thomas Klausner <wiz@NetBSD.org>
6857
6858commit 208e586c808e88a2ee819e4450dc27f557afc2bf
6859Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6860Date:   Sat Jul 27 01:03:18 2013 -0700
6861
6862    omGeneric: remove space between struct name & member name
6863    
6864    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6865
6866commit 7db74514e454d3fc4ff70aa08ddac66bfffda4dd
6867Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6868Date:   Tue Jul 23 22:18:46 2013 -0700
6869
6870    Refactor common code from XAddHost & XRemoveHost into single function
6871    
6872    On the Xlib side, the only real difference is the mode flag we send
6873    to the server with the address, so just make that an argument to the
6874    function with the common code for packing the address into the request.
6875    
6876    (Aside from labels, gcc 4.7.2 generates identical code before & after
6877     this change due to inlining, verified via diff of gcc -S output.)
6878    
6879    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6880
6881commit 3292195a64a9ce4f0d27134cd544651ec647e728
6882Author: Alan Coopersmith <alan.coopersmith@oracle.com>
6883Date:   Tue Jul 23 22:11:34 2013 -0700
6884
6885    XSetModifierMapping: Use Data instead of GetReqExtra
6886    
6887    Handle arbitrary length data in the same fashion as other calls,
6888    avoiding need to ensure it fits all in the request buffer.
6889    
6890    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6891
6892commit feb131b18aee31c2c125dc3275b0260940245882
6893Author: Kees Cook <kees@outflux.net>
6894Date:   Sun Jun 9 11:13:43 2013 -0700
6895
6896    libX11: check "req" when calling GetReqExtra
6897    
6898    This fixes the two callers of GetReqExtra to check that "req" is non-NULL
6899    to avoid crashing now that GetReqExtra does internal bounds-checking on
6900    the resulting buffer sizes.
6901    
6902    Additionally updates comment describing return values to use names
6903    instead of only literal values.
6904    
6905    Signed-off-by: Kees Cook <kees@outflux.net>
6906    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6907    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6908
6909commit 54540d7cba0c2bfe9176221c7bca910058d304df
6910Author: Kees Cook <kees@outflux.net>
6911Date:   Sun Jun 9 11:13:42 2013 -0700
6912
6913    libX11: check size of GetReqExtra after XFlush
6914    
6915    Two users of GetReqExtra pass arbitrarily sized allocations from the
6916    caller (ModMap and Host). Adjust _XGetRequest() (called by the GetReqExtra
6917    macro) to double-check the requested length and invalidate "req" when
6918    this happens. Users of GetReqExtra passing lengths greater than the Xlib
6919    buffer size (normally 16K) must check "req" and fail gracefully instead
6920    of crashing.
6921    
6922    Any callers of GetReqExtra that do not check "req" for NULL
6923    will experience this change, in the pathological case, as a NULL
6924    dereference instead of a buffer overflow. This is an improvement, but
6925    the documentation for GetReqExtra has been updated to reflect the need
6926    to check the value of "req" after the call.
6927    
6928    Bug that manifested the problem:
6929    https://bugs.launchpad.net/ubuntu/+source/x11-xserver-utils/+bug/792628
6930    
6931    Signed-off-by: Kees Cook <kees@outflux.net>
6932    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6933    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6934
6935commit 24d3ee0d08f24e23c91d55702f010f73d7b908e5
6936Author: Thomas Klausner <wiz@NetBSD.org>
6937Date:   Tue Jun 25 22:35:29 2013 +0200
6938
6939    Tighten out-of-range comparisons.
6940    
6941    [For all of these, LONG_MAX was the correct value to prevent overflows
6942     for the recent CVEs.   Lowering to INT_MAX catches buggy replies from
6943     the server that 32-bit clients would reject but 64-bit would accept,
6944     so we catch bugs sooner, and really, no sane & working server should
6945     ever report more than 2gb of extension names, font path entries,
6946     key modifier maps, etc.  -alan- ]
6947    
6948    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6949    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6950
6951commit 6d926088d80a08e13e6d6c4ff207b81ad52e667f
6952Author: Thomas Klausner <wiz@NetBSD.org>
6953Date:   Tue Jun 25 18:34:32 2013 +0200
6954
6955    Fix out-of-range comparison in _XF86BigfontQueryFont
6956    
6957    clang complained (correctly):
6958    warning: comparison of constant 768614336404564650 with expression
6959    of type 'CARD32' (aka 'unsigned int') is always true
6960    [-Wtautological-constant-out-of-range-compare]
6961    
6962    [While LONG_MAX is correct, since it's used in size_t math, the
6963     numbers have to be limited to 32-bit range to be usable by 32-bit
6964     clients, and values beyond that range are far more likely to be
6965     bugs in the data from the server than valid numbers of characters
6966     in a font.   -alan- ]
6967    
6968    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6969    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6970
6971commit 383e2b0d029482a0f4c39fe00e15397538576fc1
6972Author: Thomas Klausner <wiz@NetBSD.org>
6973Date:   Tue Jun 25 18:33:56 2013 +0200
6974
6975    Check for symbol existence with #ifdef, not #if
6976    
6977    Reviewed-by: Jamey Sharp <jamey@minilop.net>
6978    
6979    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6980
6981commit 9955d1c8de994a90fe7f2e3187e7362611d7d265
6982Author: Thomas Klausner <wiz@NetBSD.org>
6983Date:   Tue Jun 25 18:33:07 2013 +0200
6984
6985    Use newer callback-based API for XIM.
6986    
6987    Let libX11 load and make available the newer (X11R6) callback-based
6988    API for XIM (expected by emacs).
6989    
6990    This patch updates the files to match the other nls/ files.
6991    
6992    Patch from Ian D. Leroux <idleroux@fastmail.fm> on pkgsrc-users@NetBSD.org
6993    following a hint by Nhat Minh Lê <nhat.minh.le@gmail.com>.
6994    
6995    Reviewed-by: James Cloos <cloos@jhcloos.com>
6996    
6997    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
6998
6999commit a17ceb7100bd36c2db210473ee701deb5d515731
7000Author: Thomas Klausner <wiz@NetBSD.org>
7001Date:   Tue Jun 25 18:31:32 2013 +0200
7002
7003    Stop truncating source to destination length if it is larger.
7004    
7005    It seems useless to do that since the code tests for both source
7006    length and destination to be non-zero. This fixes a cut'n'paste
7007    problem in xterm where the paste length was limited to 1024 (BUFSIZ)
7008    in button.c.
7009    
7010    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7011
7012commit a336db9a0add3ae0783dda6e52459236622a12af
7013Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7014Date:   Mon Jun 24 23:02:05 2013 -0700
7015
7016    Require ANSI C89 pre-processor, drop pre-C89 token pasting support
7017    
7018    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7019
7020commit 9dfb0f3c0a761590bcdc1f3396b1e064da4e18e8
7021Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7022Date:   Fri Jun 7 11:30:11 2013 -0700
7023
7024    troff macro expansion in specs/libX11
7025    
7026    Many of the custom nroff macros (.ds <macro> <contents>) were left
7027    unsubstituted in the nroff->docbook conversion.   This substitution
7028    is now performed, via the following perl script:
7029    
7030    #! /usr/bin/perl -w -i
7031    
7032    use Text::Wrap;
7033    
7034    while ($_ = <>) {
7035        while ($_ =~ m/\((\w+)\b/g) {
7036            my $m = $1;
7037            if (exists $macro{$m}) {
7038                $_ =~ s/\($m/$macro{$m}/;
7039                $_ = wrap('', '', $_);
7040                $_ =~ s/[ \t]+$//;
7041            }
7042        }
7043    
7044        if ($_ =~ /\<!-- .ds (\w+) (.*) -->/) {
7045            my ($m, $s) = ($1, $2);
7046            $macro{$m} = $s;
7047            while ($macro{$m} =~ /\\\s*$/) {
7048                $macro{$m} =~ s/\\\s*$//ms;
7049                $macro{$m} .= <>;
7050                chomp($macro{$m});
7051            }
7052            $macro{$m} =~ s/\\ / /g;
7053        } else {
7054            print $_;
7055        }
7056    }
7057    
7058    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7059
7060commit 20c17bd9ebf767a24643279e45866dddcb57b5ce
7061Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7062Date:   Fri Jun 7 09:27:26 2013 -0700
7063
7064    specs/libX11: correct prototype for XListPixmapFormats/XImageByteOrder
7065    
7066    The XListPixmapFormats arguments was being shown with XImageByteOrder's
7067    name and return types.   Appears to have been a glitch in the nroff ->
7068    docbook conversion.
7069    
7070    Reported-by: ZHANG Zhaolong <zhangzl2013@126.com>
7071    Reviewed-by: Jamey Sharp <jamey@minilop.net>
7072    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7073
7074commit ed3d830243c8a0eefaf24e15b11823272ffe5049
7075Author: Thomas Klausner <wiz@NetBSD.org>
7076Date:   Sun Jun 2 20:49:55 2013 +0200
7077
7078    Deal with the limited range of VAX floating point numbers when compiling for VAX.
7079    
7080    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7081
7082commit 0a48235d63639fb917c44d27c86e928e79fbac66
7083Author: Eric S. Raymond <esr@thyrsus.com>
7084Date:   Thu Jun 6 16:43:56 2013 -0400
7085
7086    Remove call to undefined macro.
7087
7088commit 9e4719b9b719f2f8d255f6778e2e8c1809e32599
7089Author: Eric S. Raymond <esr@thyrsus.com>
7090Date:   Thu Jun 6 16:42:20 2013 -0400
7091
7092    Remove call to undefined macro.
7093
7094commit 8496122eb00ce6cd5d2308ee54f64b68c378e455
7095Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7096Date:   Mon Jun 3 20:06:43 2013 -0700
7097
7098    Update README to reflect where to find the Xlib specs now
7099    
7100    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7101
7102commit 655d631e86c95b14888758b27ed2836ca3e3ce86
7103Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7104Date:   Mon Jun 3 19:21:06 2013 -0700
7105
7106    libX11 1.6.0
7107    
7108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7109
7110commit 4a89b7ea908554628f374537a79931c8006a2de3
7111Author: Thomas Klausner <wiz@NetBSD.org>
7112Date:   Sun Jun 2 11:49:54 2013 -0700
7113
7114    cmsMath.c: Add missing stdio header for printf(3) in DEBUG build.
7115    
7116    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7117
7118commit 96dcf747f13b26b8e4d17b1bc8605d933c3e1dc6
7119Author: Thomas Klausner <wiz@NetBSD.org>
7120Date:   Sun Jun 2 20:49:48 2013 +0200
7121
7122    XCreateGC man page: Avoid .TS H and .TH macros
7123    
7124    Avoid .TS H and .TH for now as it doesn't alter the output in this case,
7125    and improve the output with mandoc(1).
7126    
7127    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7128
7129commit 95a388158c9d73df7d24016d6a3d61506d7d53a4
7130Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7131Date:   Thu May 23 19:43:35 2013 -0700
7132
7133    libX11 1.5.99.902 (1.6 RC2)
7134    
7135    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7136
7137commit a3bdd2b090915fe0163b062f0e6576fe05dd332e
7138Author: Julien Cristau <jcristau@debian.org>
7139Date:   Thu May 23 20:39:46 2013 +0200
7140
7141    xkb: fix off-by-one in _XkbReadGetNamesReply and _XkbReadVirtualModMap
7142    
7143    The size of the arrays is max_key_code + 1.  This makes these functions
7144    consistent with the other checks added for CVE-2013-1997.
7145    
7146    Also check the XkbGetNames reply when names->keys was just allocated.
7147    
7148    Signed-off-by: Julien Cristau <jcristau@debian.org>
7149    Tested-by: Colin Walters <walters@verbum.org>
7150    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7151
7152commit 7e30056e78e4b7979ff47f102e00327617266019
7153Author: Niveditha Rau <Niveditha.Rau@Oracle.COM>
7154Date:   Fri May 17 15:26:21 2013 -0700
7155
7156    Make sure internal headers include required headers
7157    
7158    Fixes builds with Solaris Studio 12.3 when lint is enabled, since it no
7159    longer ignores *.h files, but complains when they reference undefined
7160    typedefs or macros.
7161    
7162    Signed-off-by: Niveditha Rau <Niveditha.Rau@Oracle.COM>
7163    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7164
7165commit 2820100bf8ba130b94253f415e7fa5ac28bb2037
7166Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7167Date:   Thu May 16 23:05:36 2013 -0700
7168
7169    Free fs->properties in _XF86BigfontQueryFont overflow error path
7170    
7171    Fixes small memory leak introduced in commit 5669a22081
7172    
7173    Reported-by: Julien Cristau <jcristau@debian.org>
7174    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7175
7176commit 3131740513133a9ff7cb12123d29ceb18584fc38
7177Author: Matthieu Herrb <matthieu.herrb@laas.fr>
7178Date:   Wed May 8 19:33:09 2013 +0200
7179
7180    XListFontsWithInfo: Re-decrement flist[0] before calling free() on it.
7181    
7182    Freeing a pointer that wasn't returned by malloc() is undefined
7183    behavior and produces an error with OpenBSD's implementation.
7184    
7185    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7186    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7187    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7188
7189commit 3fe4bea086149f06a142a8f1d575f627ec1e22c7
7190Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7191Date:   Fri Apr 19 14:30:40 2013 -0700
7192
7193    Give GNU & Solaris Studio compilers hints about XEatData branches
7194    
7195    Try to offset the cost of all the recent checks we've added by giving
7196    the compiler a hint that the branches that involve us eating data
7197    are less likely to be used than the ones that process it.
7198    
7199    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7200
7201commit e1b457beb8d4e831ef44279dada6c475cb955738
7202Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7203Date:   Sun Mar 31 12:22:35 2013 -0700
7204
7205    _XkbReadGetMapReply: reject maxKeyCodes smaller than the minKeyCode
7206    
7207    Various other bounds checks in the code assume this is true, so
7208    enforce it when we first get the data from the X server.
7209    
7210    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7211
7212commit 12ad4c6432496897ff000eb7cfecd0fb4b290331
7213Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7214Date:   Sat Mar 16 10:03:13 2013 -0700
7215
7216    Use calloc in XOpenDisplay to initialize structs containing pointers
7217    
7218    Prevents trying to free uninitialized pointers if we have to bail out
7219    partway through setup, such as if we receive a corrupted or incomplete
7220    connection setup block from the server.
7221    
7222    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7223
7224commit d38527e25f8b6e2f1174ecc21260c5c5416f972e
7225Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7226Date:   Thu Mar 7 23:46:05 2013 -0800
7227
7228    Remove more unnecessary casts from Xmalloc/calloc calls
7229    
7230    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7231
7232commit b2c86b582c58f50c7b14da01cf7ebd20ef12a6b2
7233Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7234Date:   Sat Mar 2 16:56:16 2013 -0800
7235
7236    Convert more _XEatData callers to _XEatDataWords
7237    
7238    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7239    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7240
7241commit 192bbb9e2fc45df4e17b35b6d14ea0eb418dbd39
7242Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7243Date:   Sat Mar 9 11:04:37 2013 -0800
7244
7245    Make XGetWindowProperty() always initialize returned values
7246    
7247    Avoids memory corruption and other errors when callers access them
7248    without checking to see if XGetWindowProperty() returned an error value.
7249    
7250    Callers are still required to check for errors, this just reduces the
7251    damage when they don't.
7252    
7253    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7254    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7255    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7256
7257commit db1b1c871da29aa0545182bf888df81627f165a5
7258Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7259Date:   Sat Mar 2 15:08:21 2013 -0800
7260
7261    Avoid overflows in XListExtensions() [CVE-2013-1997 15/15]
7262    
7263    Ensure that when breaking the returned list into individual strings,
7264    we don't walk past the end of allocated memory to write the '\0' bytes
7265    
7266    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7267    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7268
7269commit 8d5936594993921acdfec778dd8f41b555e2543a
7270Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7271Date:   Sat Mar 2 15:08:21 2013 -0800
7272
7273    Avoid overflows in XGetFontPath() [CVE-2013-1997 14/15]
7274    
7275    Ensure that when breaking the returned list into individual strings,
7276    we don't walk past the end of allocated memory to write the '\0' bytes
7277    
7278    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7279    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7280
7281commit 0c404db6a92dc2c198328bf586c02d8abbe02013
7282Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7283Date:   Sat Mar 2 15:08:21 2013 -0800
7284
7285    Avoid overflows in XListFonts() [CVE-2013-1997 13/15]
7286    
7287    Ensure that when breaking the returned list into individual strings,
7288    we don't walk past the end of allocated memory to write the '\0' bytes
7289    
7290    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7291    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7292
7293commit 0b0f5d4358c3de7563d6af03f0d2ce454702a06a
7294Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7295Date:   Sat Mar 2 15:08:21 2013 -0800
7296
7297    integer overflow in XGetModifierMapping() [CVE-2013-1981 13/13]
7298    
7299    Ensure that we don't underallocate when the server claims a very large reply
7300    
7301    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7302    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7303
7304commit a351b8103b2ba78882e1c309e85893ca3abe2073
7305Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7306Date:   Sat Mar 2 15:08:21 2013 -0800
7307
7308    integer overflow in XGetPointerMapping() & XGetKeyboardMapping() [CVE-2013-1981 12/13]
7309    
7310    Ensure that we don't underallocate when the server claims a very large reply
7311    
7312    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7313    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7314
7315commit 833f6b70bc789d33607f6dbfee9e0a4178ec4b59
7316Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7317Date:   Sat Mar 2 15:08:21 2013 -0800
7318
7319    integer overflow in XGetImage() [CVE-2013-1981 11/13]
7320    
7321    Ensure that we don't underallocate when the server claims to have sent a
7322    very large reply.
7323    
7324    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7325    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7326
7327commit 79d8dc08eb98842173ce239b9dd60df0e9e9ae72
7328Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7329Date:   Fri Mar 8 22:25:35 2013 -0800
7330
7331    integer overflow in XGetWindowProperty() [CVE-2013-1981 10/13]
7332    
7333    If the reported number of properties is too large, the calculations
7334    to allocate memory for them may overflow, leaving us returning less
7335    memory to the caller than implied by the value written to *nitems.
7336    
7337    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7338    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7339    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7340
7341commit 164bf4dfe839b1cc75cdeee378a243d04a8200e4
7342Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7343Date:   Sat Mar 2 13:18:48 2013 -0800
7344
7345    integer overflows in TransFileName() [CVE-2013-1981 9/13]
7346    
7347    When trying to process file paths the tokens %H, %L, & %S are expanded
7348    to $HOME, the standard compose file path & the xlocaledir path.
7349    If enough of these tokens are repeated and values like $HOME are set to
7350    very large values, the calculation of the total string size required to
7351    hold the expanded path can overflow, resulting in allocating a smaller
7352    string than the amount of data we'll write to it.
7353    
7354    Simply restrict all of these values, and the total path size to PATH_MAX,
7355    because really, that's all you should need for a filename path.
7356    
7357    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7358    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7359    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7360
7361commit 460e8a223b87d4fa0ea1e97823e998a770e0f2a2
7362Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7363Date:   Fri Mar 1 18:37:37 2013 -0800
7364
7365    integer truncation in _XimParseStringFile() [CVE-2013-1981 8/13]
7366    
7367    Called from _XimCreateDefaultTree() which uses getenv("XCOMPOSEFILE")
7368    to specify filename.
7369    
7370    If the size of off_t is larger than the size of unsigned long (as in
7371    32-bit builds with large file flags), a file larger than 4 gigs could
7372    have its size truncated, leading to data from that file being written
7373    past the end of the undersized buffer allocated for it.
7374    
7375    While configure.ac does not use AC_SYS_LARGEFILE to set large file mode,
7376    builders may have added the large file compilation flags to CFLAGS on
7377    their own.
7378    
7379    size is left limited to an int, because if your Xim file is
7380    larger than 2gb, you're doing it wrong.
7381    
7382    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7383    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7384    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7385
7386commit 226622349a4b1e16064649d4444a34fb4be4f464
7387Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7388Date:   Sat Mar 2 12:39:58 2013 -0800
7389
7390    Unbounded recursion in _XimParseStringFile() when parsing include files [CVE-2013-2004 2/2]
7391    
7392    parseline() can call _XimParseStringFile() which can call parseline()
7393    which can call _XimParseStringFile() which can call parseline() ....
7394    eventually causing recursive stack overflow and crash.
7395    
7396    Limit is set to a include depth of 100 files, which should be enough
7397    for all known use cases, but could be adjusted later if necessary.
7398    
7399    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7400    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7401    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7402
7403commit 236b603d235dc264d1c6250dca09c745458a9088
7404Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7405Date:   Sat Mar 2 12:01:39 2013 -0800
7406
7407    Unbounded recursion in GetDatabase() when parsing include files [CVE-2013-2004 1/2]
7408    
7409    GetIncludeFile() can call GetDatabase() which can call GetIncludeFile()
7410    which can call GetDatabase() which can call GetIncludeFile() ....
7411    eventually causing recursive stack overflow and crash.
7412    
7413    Easily reproduced with a resource file that #includes itself.
7414    
7415    Limit is set to a include depth of 100 files, which should be enough
7416    for all known use cases, but could be adjusted later if necessary.
7417    
7418    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7419    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7420    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7421
7422commit 076428918e6c35f66b9b55c3fa097ff06496d155
7423Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7424Date:   Fri Mar 1 18:37:37 2013 -0800
7425
7426    integer overflow in ReadInFile() in Xrm.c [CVE-2013-1981 7/13]
7427    
7428    Called from XrmGetFileDatabase() which gets called from InitDefaults()
7429    which gets the filename from getenv ("XENVIRONMENT")
7430    
7431    If file is exactly 0xffffffff bytes long (or longer and truncates to
7432    0xffffffff, on implementations where off_t is larger than an int),
7433    then size may be set to a value which overflows causing less memory
7434    to be allocated than is written to by the following read() call.
7435    
7436    size is left limited to an int, because if your Xresources file is
7437    larger than 2gb, you're very definitely doing it wrong.
7438    
7439    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7440    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7441    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7442
7443commit 90fd5abac2faca86f9f100353a3c9c7b89f31484
7444Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7445Date:   Sat Mar 2 11:44:19 2013 -0800
7446
7447    Integer overflows in stringSectionSize() cause buffer overflow in ReadColornameDB() [CVE-2013-1981 6/13]
7448    
7449    LoadColornameDB() calls stringSectionSize() to do a first pass over the
7450    file (which may be provided by the user via XCMSDB environment variable)
7451    to determine how much memory needs to be allocated to read in the file,
7452    then allocates the returned sizes and calls ReadColornameDB() to load the
7453    data from the file into that newly allocated memory.
7454    
7455    If stringSectionSize() overflows the signed ints used to calculate the
7456    file size (say if you have an xcmsdb with ~4 billion lines in or a
7457    combined string length of ~4 gig - which while it may have been
7458    inconceivable when Xlib was written, is quite possible today), then
7459    LoadColornameDB() may allocate a memory buffer much smaller than the
7460    amount of data ReadColornameDB() will write to it.
7461    
7462    The total size is left limited to an int, because if your xcmsdb file
7463    is larger than 2gb, you're doing it wrong.
7464    
7465    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7466    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7467    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7468
7469commit b9ba832401734e1cbd30a930c0d11d850293f3f9
7470Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7471Date:   Sat Mar 2 11:25:25 2013 -0800
7472
7473    unvalidated length in _XimXGetReadData() [CVE-2013-1997 12/15]
7474    
7475    Check the provided buffer size against the amount of data we're going to
7476    write into it, not against the reported length from the ClientMessage.
7477    
7478    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7479    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7480    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7481
7482commit de2e6c322c4aca22856b380f67f8e488e7510015
7483Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7484Date:   Sat Mar 2 11:11:08 2013 -0800
7485
7486    unvalidated index/length in _XkbReadGetNamesReply() [CVE-2013-1997 11/15]
7487    
7488    If the X server returns key name indexes outside the range of the number
7489    of keys it told us to allocate, out of bounds memory writes could occur.
7490    
7491    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7492    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7493    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7494
7495commit 2df882eeb3a70256170127a746a9ba26376599a1
7496Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7497Date:   Sat Mar 2 11:01:04 2013 -0800
7498
7499    unvalidated index in _XkbReadVirtualModMap() [CVE-2013-1997 10/15]
7500    
7501    If the X server returns modifier map indexes outside the range of the number
7502    of keys it told us to allocate, out of bounds memory writes could occur.
7503    
7504    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7505    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7506    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7507
7508commit 4d7c422a37eb9617fb22f8e37527c2b34b105665
7509Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7510Date:   Sat Mar 2 11:04:44 2013 -0800
7511
7512    unvalidated index in _XkbReadExplicitComponents() [CVE-2013-1997 9/15]
7513    
7514    If the X server returns key indexes outside the range of the number of
7515    keys it told us to allocate, out of bounds memory writes could occur.
7516    
7517    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7518    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7519    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7520
7521commit e56a2ada719c5cfac5ed61a52a80ade86c0f5957
7522Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7523Date:   Sat Mar 2 10:51:51 2013 -0800
7524
7525    unvalidated index in _XkbReadModifierMap() [CVE-2013-1997 8/15]
7526    
7527    If the X server returns modifier map indexes outside the range of the number
7528    of keys it told us to allocate, out of bounds memory writes could occur.
7529    
7530    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7531    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7532    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7533
7534commit 06c086e8a1d8374ea9a95ff989f053c96bb1bdca
7535Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7536Date:   Sat Mar 2 10:39:21 2013 -0800
7537
7538    unvalidated index in _XkbReadKeyBehaviors() [CVE-2013-1997 7/15]
7539    
7540    If the X server returns key behavior indexes outside the range of the number
7541    of keys it told us to allocate, out of bounds memory writes could occur.
7542    
7543    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7544    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7545    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7546
7547commit 00626c3830b869259098985afa38933d77ccec72
7548Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7549Date:   Sat Mar 2 09:40:22 2013 -0800
7550
7551    unvalidated index in _XkbReadKeyActions() [CVE-2013-1997 6/15]
7552    
7553    If the X server returns key action indexes outside the range of the number
7554    of keys it told us to allocate, out of bounds memory access could occur.
7555    
7556    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7557    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7558    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7559
7560commit fd7d4956bc7a1c4b5c38661b12777ebee4d685d9
7561Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7562Date:   Sat Mar 2 09:28:33 2013 -0800
7563
7564    unvalidated index in _XkbReadKeySyms() [CVE-2013-1997 5/15]
7565    
7566    If the X server returns keymap indexes outside the range of the number of
7567    keys it told us to allocate, out of bounds memory access could occur.
7568    
7569    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7570    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7571    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7572
7573commit 59ae16a00d18588e98af57d26e442af8ea42b7aa
7574Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7575Date:   Sat Mar 2 09:18:26 2013 -0800
7576
7577    unvalidated indexes in _XkbReadGetGeometryReply() [CVE-2013-1997 4/15]
7578    
7579    If the X server returns color indexes outside the range of the number of
7580    colors it told us to allocate, out of bounds memory access could occur.
7581    
7582    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7583    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7584    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7585
7586commit bff938b9fe1629cbacb726509edfa2a3840b7207
7587Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7588Date:   Sat Mar 2 09:12:47 2013 -0800
7589
7590    unvalidated indexes in _XkbReadGeomShapes() [CVE-2013-1997 3/15]
7591    
7592    If the X server returns shape indexes outside the range of the number
7593    of shapes it told us to allocate, out of bounds memory access could occur.
7594    
7595    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7596    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7597    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7598
7599commit f293659d5a4024bda386305bb7ebeb4647c40934
7600Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7601Date:   Fri Mar 1 22:49:01 2013 -0800
7602
7603    unvalidated index in _XkbReadGetDeviceInfoReply() [CVE-2013-1997 2/15]
7604    
7605    If the X server returns more buttons than are allocated in the XKB
7606    device info structures, out of bounds writes could occur.
7607    
7608    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7609    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7610    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7611
7612commit cddc4e7e3cb4b9b7ad25f8591971a86901c249f2
7613Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7614Date:   Fri Mar 1 19:30:09 2013 -0800
7615
7616    unvalidated lengths in XAllocColorCells() [CVE-2013-1997 1/15]
7617    
7618    If a broken server returned larger than requested values for nPixels or
7619    nMasks, XAllocColorCells would happily overflow the buffers provided by
7620    the caller to write the results into.
7621    
7622    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7623    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7624    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7625
7626commit 2cd62b5eb99ffbb2fce99f3c459455e630b35bf7
7627Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7628Date:   Fri Mar 1 22:49:01 2013 -0800
7629
7630    integer overflow in XListHosts() [CVE-2013-1981 5/13]
7631    
7632    If the reported number of host entries is too large, the calculations
7633    to allocate memory for them may overflow, leaving us writing beyond the
7634    bounds of the allocation.
7635    
7636    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7637    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7638    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7639
7640commit 1f6a3dbf699b85c0ea715ef21de7e7095a714e12
7641Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7642Date:   Fri Mar 1 22:49:01 2013 -0800
7643
7644    integer overflow in XGetMotionEvents() [CVE-2013-1981 4/13]
7645    
7646    If the reported number of motion events is too large, the calculations
7647    to allocate memory for them may overflow, leaving us writing beyond the
7648    bounds of the allocation.
7649    
7650    v2: Ensure nEvents is set to 0 when returning NULL events pointer
7651    
7652    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7653    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7654
7655commit 39515b7c3ba8cae9021bf6695e378ae19487082f
7656Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7657Date:   Fri Mar 1 22:49:01 2013 -0800
7658
7659    integer overflow in XListFontsWithInfo() [CVE-2013-1981 3/13]
7660    
7661    If the reported number of remaining fonts is too large, the calculations
7662    to allocate memory for them may overflow, leaving us writing beyond the
7663    bounds of the allocation.
7664    
7665    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
7666    v3: On error cases, also set values to be returned in pointer args to 0/NULL
7667    
7668    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7669    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7670
7671commit 5669a220816b7d58fcaf0c302ead16fbe5c87817
7672Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7673Date:   Fri Mar 1 21:05:27 2013 -0800
7674
7675    integer overflow in _XF86BigfontQueryFont() [CVE-2013-1981 2/13]
7676    
7677    Similar to _XQueryFont, but with more ways to go wrong and overflow.
7678    Only compiled if libX11 is built with XF86BigFont support.
7679    
7680    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
7681    
7682    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7683    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7684
7685commit 6df8a63d34b7514077188e2062a13774f920c085
7686Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7687Date:   Fri Mar 1 21:05:27 2013 -0800
7688
7689    integer overflow in _XQueryFont() on 32-bit platforms [CVE-2013-1981 1/13]
7690    
7691    If the CARD32 reply.nCharInfos * sizeof(XCharStruct) overflows an
7692    unsigned long, then too small of a buffer will be allocated for the
7693    data copied in from the reply.
7694    
7695    v2: Fix reply_left calculations, check calculated sizes fit in reply_left
7696    
7697    Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
7698    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7699
7700commit 9f5d83706543696fc944c1835a403938c06f2cc5
7701Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7702Date:   Fri Mar 1 20:54:24 2013 -0800
7703
7704    Add _XEatDataWords to discard a given number of 32-bit words of reply data
7705    
7706    Matches the units of the length field in X protocol replies, and provides
7707    a single implementation of overflow checking to avoid having to replicate
7708    those checks in every caller.
7709    
7710    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7711    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7712
7713commit d7f04c340ade3834e603c23d543132e1ee4e0c63
7714Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7715Date:   Sat Mar 2 13:03:55 2013 -0800
7716
7717    Move repeated #ifdef magic to find PATH_MAX into a common header
7718    
7719    Lets stop duplicating the mess all over
7720    
7721    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7722    Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
7723
7724commit f3a553a4e4a55d9d19deda1ea01883e1d5d682b1
7725Author: Pander <pander@users.sourceforge.net>
7726Date:   Tue May 7 18:38:14 2013 -0400
7727
7728    Add compose sequences for J́ and j́.
7729    
7730    The resulting sequences are:
7731    
7732      U+004A LATIN CAPITAL LETTER J  U+0301 COMBINING ACUTE ACCENT
7733      U+006A LATIN SMALL LETTER J    U+0301 COMBINING ACUTE ACCENT
7734    
7735    Used in Dutch, per:
7736    
7737      http://lists.x.org/archives/xorg-devel/2013-February/035514.html
7738      https://nl.wikipedia.org/wiki/Accenttekens_in_de_Nederlandse_spelling
7739    
7740    Signed-off-by: Pander <pander@users.sourceforge.net>
7741    Signed-off-by: James Cloos <cloos@jhcloos.com>
7742
7743commit f49bb2dd6d4ea45c55bd21acc0efe2b764441020
7744Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7745Date:   Sat Mar 16 18:30:56 2013 -0700
7746
7747    Move big request comment in XOpenDisplay to the right place
7748    
7749    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7750
7751commit 3996543c1b2919e97d61a5d70fe1ebd7cd76fc83
7752Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7753Date:   Sat Mar 9 19:16:03 2013 -0800
7754
7755    libX11 1.5.99.901 (1.6 RC1)
7756    
7757    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7758
7759commit f9cd175a471116a616e681fb0ca1a61b3d84a6a0
7760Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7761Date:   Fri Mar 8 22:33:28 2013 -0800
7762
7763    Fix very weird indenting in src/GetFProp.c
7764    
7765    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7766
7767commit e9bd757630368afb374c5d1bcc5d4d85ad3c6c4c
7768Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7769Date:   Fri Mar 8 15:37:33 2013 -0800
7770
7771    XAllocClassHint: Assume calloc sets pointers in allocated memory to NULL
7772    
7773    While the C standard technically allows for the compiler to translate
7774    pointer = 0 or pointer = NULL into something other than filling the
7775    pointer address with 0 bytes, the rest of the Xlib code already assumes
7776    that calloc initializes any pointers in the struct to NULL, and there
7777    are no known systems supported by X.Org where this is not true.
7778    
7779    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7780
7781commit 39547d600a13713e15429f49768e54c3173c828d
7782Author: Karl Tomlinson <xmail@karlt.net>
7783Date:   Mon Feb 18 01:25:34 2013 +0000
7784
7785    MakeBigReq: don't move the last word, already handled by Data32
7786    
7787    MakeBigReq inserts a length field after the first 4 bytes of the request
7788    (after req->length), pushing everything else back by 4 bytes.
7789    
7790    The current memmove moves everything but the first 4 bytes back.
7791    If a request aligns to the end of the buffer pointer when MakeBigReq is
7792    invoked for that request, this runs over the buffer.
7793    Instead, we need to memmove minus the first 4 bytes (which aren't moved),
7794    minus the last 4 bytes (so we still align to the previous tail).
7795    
7796    The 4 bytes that fell out are already handled with Data32, which will
7797    handle the buffermax correctly.
7798    
7799    The case where req->length = 1 was already not functional.
7800    
7801    Reported by Abhishek Arya <inferno@chromium.org>.
7802    
7803    https://bugzilla.mozilla.org/show_bug.cgi?id=803762
7804    
7805    Reviewed-by: Jeff Muizelaar <jmuizelaar@mozilla.com>
7806    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7807    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7808
7809Notes:
7810    CVE-2013-7439
7811
7812commit 3cdb6c3a1646f670afa03d424ec12ac418181d1e
7813Author: Quentin Glidic <sardemff7+git@sardemff7.net>
7814Date:   Tue Jan 15 21:07:17 2013 +0000
7815
7816    nls/Makefile.am: Use LOG_COMPILER
7817    
7818    TESTS_ENVIRONMENT is deprecated
7819    
7820    Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
7821    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7822
7823commit df66d7a98e2bc7f44fb5583b645df87d525f07f1
7824Author: Quentin Glidic <sardemff7+git@sardemff7.net>
7825Date:   Tue Jan 15 21:07:16 2013 +0000
7826
7827    nls/Makefile.am: Remove unneeded $(srcdir)
7828    
7829    Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
7830    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
7831    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7832
7833commit 3facbe5c0df1b5597571b7b00d5f7bdbc92fb278
7834Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7835Date:   Sat Mar 2 12:01:39 2013 -0800
7836
7837    Add <X11/Xresource.h> hint to all Xrm* man pages
7838    
7839    Help users figure out which header file they need to #include
7840    
7841    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7842
7843commit 466404007f2c8f7166e4faddfea1454c5bfe1e9a
7844Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7845Date:   Fri Mar 8 17:13:09 2013 -0800
7846
7847    _xudc_code_to_glyph: check for NULL pointer *before* writing to it, not after
7848    
7849    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7850
7851commit 47d2eff64d167b3245d346d7f38ca703be645e26
7852Author: Ken Moffat <ken@linuxfromscratch.org>
7853Date:   Thu Mar 7 18:27:54 2013 -0500
7854
7855    dead_double_grave and dead_inverted_breve should only have one underscore.
7856    
7857    Correct instances of dead_double_grave and dead_inverted_breve to
7858    dead_doublegrave and dead_invertedbreve.
7859    
7860    Signed-off-by: Ken Moffat <ken@linuxfromscratch.org>
7861    Signed-off-by: James Cloos <cloos@jhcloos.com>
7862
7863commit c23d61d1b84dca3740bf4786978c7908d0065fb9
7864Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7865Date:   Fri Mar 1 18:10:27 2013 -0800
7866
7867    Assume size_t is always available, since it was defined in C89
7868    
7869    Don't provide a fallback definition #ifdef X_NOT_POSIX anymore.
7870    We already use size_t throughout the rest of Xlib, just had this
7871    one instance left in XKBGAlloc.c of a fallback definition.
7872    
7873    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7874
7875commit 9bcfd84aa1410387bc8cf002a5f90f44705aa0d1
7876Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7877Date:   Fri Mar 1 18:09:07 2013 -0800
7878
7879    unifdef XKB_IN_SERVER
7880    
7881    Leftovers from XKB files that were previously shared between the client
7882    and server code, but aren't any more.
7883    
7884    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7885    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7886
7887commit 769a0efa2298040fe8316a89fc9e75fb61e288e5
7888Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7889Date:   Thu Feb 28 20:04:25 2013 -0800
7890
7891    unifdef CRAY & _CRAY
7892    
7893    (mostly performed with unifdef, followed by some manual cleanup of
7894     the remaining code)
7895    
7896    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7897    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7898
7899commit ca106eb03e5f5468df8033300c5caae3d3c6936b
7900Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7901Date:   Thu Feb 28 20:04:25 2013 -0800
7902
7903    unifdef WORD64
7904    
7905    WORD64 seems to have only been defined in <X11/Xmd.h> when building for
7906    CRAY, to handle int being a 64-bit value (ILP64, not LP64) and having
7907    64-bit alignment requirements.
7908    
7909    It hadn't been fully supported even before autotooling, as can be
7910    seen by removed code such as:
7911    
7912     #ifdef WORD64
7913     _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8
7914     #endif
7915    
7916    (mostly performed with unifdef, followed by some manual cleanup of
7917     the remaining code)
7918    
7919    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7920    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7921
7922commit 9399caf2c12cbe1ed56f4f6b368c5811cb5d0458
7923Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7924Date:   Thu Feb 28 20:04:25 2013 -0800
7925
7926    unifdef MUSTCOPY
7927    
7928    MUSTCOPY seems to have only been defined in <X11/Xmd.h> when building for
7929    CRAY, to handle missing some sizes of integer type.
7930    
7931    (mostly performed with unifdef, followed by some manual cleanup of
7932     spacing/indenting in the remaining code)
7933    
7934    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7935    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
7936
7937commit b687440c28c7da6ee0ae44514d20248db5161606
7938Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7939Date:   Sat Feb 16 10:42:23 2013 -0800
7940
7941    Convert more sprintf calls to snprintf
7942    
7943    You could analyze most of these and quickly recognize that there was no
7944    chance of buffer overflow already, but why make everyone spend time doing
7945    that when we can just make it obviously safe?
7946    
7947    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7948
7949commit b092dabbd712d7b656abcf572d253b9b206c0237
7950Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7951Date:   Fri Feb 15 23:43:12 2013 -0800
7952
7953    XKeysymToString: move variable declarations to the scope of their usage
7954    
7955    Makes it easier for readers to understand scope of variable usage, and
7956    clears up gcc warning:
7957    
7958    KeysymStr.c: In function 'XKeysymToString':
7959    KeysymStr.c:128:13: warning: declaration of 'i' shadows a previous local [-Wshadow]
7960    KeysymStr.c:73:18: warning: shadowed declaration is here [-Wshadow]
7961    
7962    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7963
7964commit f0b171c8ea7b055ba520272ea9a2604e18841ac7
7965Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7966Date:   Fri Feb 15 23:34:40 2013 -0800
7967
7968    Preserve constness in casting arguments through the Data*() routines
7969    
7970    Casts were annoying gcc by dropping constness when changing types,
7971    when routines simply either copy data into the request buffer or
7972    send it directly to the X server, and never modify the input.
7973    
7974    Fixes gcc warnings including:
7975    ChProp.c: In function 'XChangeProperty':
7976    ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7977    ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7978    ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7979    ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7980    ChProp.c:83:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7981    SetHints.c: In function 'XSetStandardProperties':
7982    SetHints.c:262:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7983    SetPntMap.c: In function 'XSetPointerMapping':
7984    SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7985    SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7986    StBytes.c: In function 'XStoreBuffer':
7987    StBytes.c:97:33: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7988    StName.c: In function 'XStoreName':
7989    StName.c:40:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7990    StName.c: In function 'XSetIconName':
7991    StName.c:51:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
7992    
7993    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
7994
7995commit 6c558ee357292dd9dfc6d9006f4525f625327c52
7996Author: Alan Coopersmith <alan.coopersmith@oracle.com>
7997Date:   Fri Feb 15 22:58:54 2013 -0800
7998
7999    Fix comment typo & confusing indentation levels in Data() macro definition
8000    
8001    The final } matches the one on the #define line, not one that doesn't
8002    appear after the else statement it was lined up with
8003    
8004    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8005
8006commit afd6593da90e51234d59f8921c411317f91ab48b
8007Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8008Date:   Fri Feb 15 23:25:38 2013 -0800
8009
8010    XStringToKeysym: preserve constness when casting off unsignedness for strcmp
8011    
8012    Fixes gcc warning:
8013    StrKeysym.c:97:17: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8014    
8015    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8016
8017commit 7e3bf4dd83fec22bd568146de75e6d59eff74e21
8018Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8019Date:   Fri Feb 15 23:14:40 2013 -0800
8020
8021    XRebindKeysym: Drop unnecessary const-removing cast
8022    
8023    C89 defines memcpy as taking a const void *, so casting from
8024    const unsigned char * to char * simply angers gcc for no benefit:
8025    
8026    KeyBind.c:1017:24: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8027    
8028    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8029
8030commit 54527eab93d46055cf11eb6c18abb353a03ae544
8031Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8032Date:   Fri Feb 15 22:45:19 2013 -0800
8033
8034    cmsColNm.c: maintain constness of arguments to qsort helper function
8035    
8036    Fixes gcc warning:
8037    
8038    cmsColNm.c: In function 'FirstCmp':
8039    cmsColNm.c:257:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8040    cmsColNm.c:257:45: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
8041    
8042    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8043
8044commit deedeada53676ee529d700bf96fde0b29a3a1def
8045Author: Nickolai Zeldovich <nickolai@csail.mit.edu>
8046Date:   Tue Jan 22 10:03:00 2013 -0500
8047
8048    XListFontsWithInfo: avoid accessing realloc'ed memory
8049    
8050    If exactly one of the two reallocs in XListFontsWithInfo() fails, the
8051    subsequent code accesses memory freed by the other realloc.
8052    
8053    Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
8054    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8055    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8056
8057commit f57fd760cba92ad846917f21e94e73e9c846185f
8058Author: Colin Walters <walters@verbum.org>
8059Date:   Wed Jan 4 17:37:06 2012 -0500
8060
8061    autogen.sh: Implement GNOME Build API
8062    
8063    http://people.gnome.org/~walters/docs/build-api.txt
8064    
8065    Signed-off-by: Adam Jackson <ajax@redhat.com>
8066
8067commit 51c102d39e855cf1704d9eb3afba76a2e73c6b81
8068Author: Adam Jackson <ajax@redhat.com>
8069Date:   Tue Jan 15 14:28:48 2013 -0500
8070
8071    configure: Remove AM_MAINTAINER_MODE
8072    
8073    Signed-off-by: Adam Jackson <ajax@redhat.com>
8074
8075commit 3cd974b1d4d1fa6389d3695fa9fcc0c22a51d50c
8076Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8077Date:   Wed Dec 26 22:57:39 2012 -0800
8078
8079    Remove unused DECnet ("DNETCONN") code from Xlib
8080    
8081    Has never been converted to build in modular builds, so has been unusable
8082    since X11R7.0 release in 2005.  DNETCONN support was removed from xtrans
8083    back in 2008.
8084    
8085    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8086
8087commit a6e5b36a3e6d4a7a9fb4bad905ed127e67b1957e
8088Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8089Date:   Wed Dec 26 22:56:38 2012 -0800
8090
8091    Remove unused TLI ("STREAMSCONN") code from Xlib
8092    
8093    Has never been converted to build in modular builds, so has been unusable
8094    since X11R7.0 release in 2005.  All known platforms with TLI/XTI support
8095    that X11R7 & later releases run on also have (and mostly prefer) BSD
8096    socket support for their networking API.
8097    
8098    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8099
8100commit 59c9ee8cd58857c5769b643611cbe526005a9e45
8101Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8102Date:   Sun Dec 16 17:44:42 2012 -0800
8103
8104    Tell clang not to report -Wpadded warnings on public headers we can't fix
8105    
8106    Better to silence the compiler warning than break ABI.
8107    
8108    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8109
8110commit 0b148750027fd0557c5ed93afda861ddf4b92e0f
8111Author: Jon TURNEY <jon.turney@dronecode.org.uk>
8112Date:   Mon Nov 12 17:27:52 2012 +0000
8113
8114    Fix config check for loadable modules
8115    
8116    The config check of the results of testing for dlfcn.h or dl.h just tests the
8117    value of the ac_cv_ variables, which will be 'yes' or 'no', rather than checking
8118    it is 'yes', so loadable module support would always be detected.
8119    
8120    This is necessary for successful compilation for the MinGW target without the
8121    optional dlfcn-win32 library.
8122    
8123    v2: Also, fixed typoed name of ac_cv_header_dlfcn_h, so check still works
8124    correctly when dlfcn.h is available
8125    
8126    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8127    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
8128
8129commit cd25cab4b5b957641183ce72dd1ae0424aff1663
8130Author: Egbert Eich <eich@freedesktop.org>
8131Date:   Fri May 20 18:27:02 2011 +0200
8132
8133    Install Xcms.txt in $(datadir) rather than $(libdir).
8134    
8135    This file is an architecture independent data and should be where
8136    other databases are.
8137    This is the Xlib provided sample file, applications are free to
8138    use a different one specifying its location in the XCMSDB env
8139    variable.
8140    
8141    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8142    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8143
8144commit 51fef7e273b5c6256e4c9604e3e1afe5dc8f6a1a
8145Author: Egbert Eich <eich@freedesktop.org>
8146Date:   Fri May 20 18:25:24 2011 +0200
8147
8148    Don't hard code path to Xcms.txt file.
8149    
8150    The path to this file is configurable at build time. The source
8151    however contains a hard coded path.
8152    
8153    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8154    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8155
8156commit 9833489e6c3829a1e835bc0a11f028fc180809e4
8157Author: Colin Walters <walters@verbum.org>
8158Date:   Fri Dec 7 08:51:21 2012 -0500
8159
8160    Fix build after dropping locales
8161    
8162    They also needed to be removed from configure.ac
8163    
8164    Signed-off-by: Colin Walters <walters@verbum.org>
8165    Signed-off-by: Julien Cristau <jcristau@debian.org>
8166
8167commit 0a740a574aaf0c0eec78859b773a532cff3b74c3
8168Author: Egbert Eich <eich@freedesktop.org>
8169Date:   Fri May 20 13:04:11 2011 +0200
8170
8171    i18n: Uppercased all occurances if 'iso8859' in the full locale name.
8172    
8173    Making all occurances of iso8859 upper case in the full local name
8174    makes the alias entries more consistent and match the entries on
8175    locale.dir.
8176    
8177    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8178    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8179
8180commit 5e7d589697755a70fb22d85c6a1ae82b39843e53
8181Author: Egbert Eich <eich@freedesktop.org>
8182Date:   Fri May 20 17:55:49 2011 +0200
8183
8184    i18n: Remove ja.S90 and ja.U90 locales.
8185    
8186    Both locales carry a copyright notice and a prorietary statement:
8187    
8188    Copyright 1995 by FUJITSU LIMITED
8189    This is source code modified by FUJITSU LIMITED under the Joint
8190    Development Agreement for the CDEnext PST.
8191    This is unpublished proprietary source code of FUJITSU LIMITED
8192    
8193    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8194    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8195
8196commit 99eae3dcb7bf6764e3b243d2a2934a4d1ecce90e
8197Author: Egbert Eich <eich@freedesktop.org>
8198Date:   Fri May 20 16:46:15 2011 +0200
8199
8200    i18n: Bring locale.dir and compose.dir in sync.
8201    
8202    Some entries for locale/encoding combinations were missing from
8203    either file or just misspelled, some entries were wrong or just
8204    aliases.
8205    
8206    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8207    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8208
8209commit d1e6baa4e290b758e430077cb74e7c03ad850771
8210Author: Egbert Eich <eich@freedesktop.org>
8211Date:   Fri May 20 14:02:39 2011 +0200
8212
8213    i18n: Treat 'a3_AZ' as an alias for 'az_AZ'.
8214    
8215    locale.alias contains a comment:
8216    XCOMM a3 is not an ISO 639 language code, but in Cyrillic, "Z" looks like "3".
8217    Thus lets treat 'a3' as an alias for 'az'.
8218    
8219    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8220    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8221    Reviewed-by: Magnus Kessler <Magnus.Kessler@gmx.net>
8222
8223commit fa2aab0bea18efa26a56977d3166277582ab7b07
8224Author: Egbert Eich <eich@freedesktop.org>
8225Date:   Fri May 20 13:09:38 2011 +0200
8226
8227    i18n: Fixed typos in full locale names.
8228    
8229    Fixing those typos those names actually match entries in
8230    locale.dir.
8231    
8232    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8233    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8234
8235commit b88dd95005dce4c40f9b4d5f938f945e7955fd04
8236Author: Egbert Eich <eich@freedesktop.org>
8237Date:   Sat May 21 07:29:38 2011 +0200
8238
8239    i18n: Remove duplicates and aliases to oneself.
8240    
8241    Aliases to itself don't make much sense.
8242    This changes occurances of:
8243    xy_UV.UTF-8:                    xy_UV.UTF-8
8244    to
8245    xy_UV:                  xy_UV.UTF-8
8246    where applicable.
8247    
8248    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8249    Reviewed-by: James Cloos <cloos@jhcloos.com>
8250
8251commit f198c6aa98f88ff285d903175a3c4c0fd33a4575
8252Author: Jens Herden <jens.herden@email.de>
8253Date:   Fri May 20 17:29:03 2011 +0200
8254
8255    i18n: Add support for Khmer locale and compose table.
8256    
8257    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8258    Reviewed-by: James Cloos <cloos@jhcloos.com>
8259
8260commit 40761898692e5063957bfa2518cca3d35b2e354a
8261Author: Kalman Kemenczy <kkemenczy@novell.com>
8262Date:   Fri May 20 17:03:24 2011 +0200
8263
8264    i18n: Add support for Serbian specific compose table entries.
8265    
8266    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8267    Reviewed-by: James Cloos <cloos@jhcloos.com>
8268
8269commit 7c14aacc9f01d7a975f8d9d033b9b13cbd777a61
8270Author: Egbert Eich <eich@freedesktop.org>
8271Date:   Fri May 20 14:00:35 2011 +0200
8272
8273    i18n: Adding and removing comments.
8274    
8275    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8276    Reviewed-by: James Cloos <cloos@jhcloos.com>
8277
8278commit 7754d68976106183751243c2c35a84134be17b34
8279Author: Egbert Eich <eich@freedesktop.org>
8280Date:   Fri May 20 13:45:31 2011 +0200
8281
8282    i18n: Add missing locales which existed in locale.alias.
8283    
8284    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8285    Reviewed-by: James Cloos <cloos@jhcloos.com>
8286
8287commit 952eccd0d25ba66023acfd31873eee2e71c38c42
8288Author: Egbert Eich <eich@freedesktop.org>
8289Date:   Fri May 20 10:57:57 2011 +0200
8290
8291    i18n: Consolidate compose handling for locales with UTF-8 encoding.
8292    
8293    - add an entry to include the default en_US compose file
8294      for the ja_JP, ko_KR, th_TH, zh_CN, zh_HK and zh_TW locales.
8295    - add missing entries for zh_CN. and zh_HK and am_ET.
8296    - change entries for the UTF-8 encoding for ru_RU, ja_JP,
8297      ko_KR, th_TH and zh_TW to point to their native directory
8298      entries.
8299    
8300    Signed-off-by: Egbert Eich <eich@freedesktop.org>
8301    Reviewed-by: James Cloos <cloos@jhcloos.com>
8302
8303commit d14b6a250f004fa405179db7020f6953001d17b9
8304Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
8305Date:   Mon Oct 22 13:54:11 2012 -0500
8306
8307    XIM: remove Private and Public macros
8308    
8309    Private is a struct member name in mingw-w64 <winioctl.h>, causing this
8310    useless define in a private header to break the build.
8311    
8312    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
8313    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8314
8315commit f2a8def423a46d52e834cf7ea49fa0079427663a
8316Author: Adam Jackson <ajax@redhat.com>
8317Date:   Wed Oct 17 14:40:43 2012 -0400
8318
8319    XErrorDB: Add GLXBadProfileARB
8320    
8321    Signed-off-by: Adam Jackson <ajax@redhat.com>
8322
8323commit d45b3fc19fbe95c41afc4e51d768df6d42332010
8324Author: Ross Burton <ross.burton@intel.com>
8325Date:   Wed Sep 12 14:39:40 2012 +0100
8326
8327    Allow overriding location of keysymdef.h
8328    
8329    Currently keysymdef.h is found by using the includedir of xproto.  This doesn't
8330    work when cross-compiling with a sysroot as that ends up being /usr/include/X11,
8331    not a path into the cross-build environment.
8332    
8333    So, add an option to allow explicitly specifying the location of keysymdef.h,
8334    and verify that the specified or found path exists.
8335    
8336    (original patch by Martin Jansa <martin.jansa@gmail.com>, revised by myself)
8337    
8338    Signed-off-by: Ross Burton <ross.burton@intel.com>
8339    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
8340    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
8341
8342commit 44cdc0dc2c68d67654023ec707b807145d3a38c0
8343Author: Eric S. Raymond <esr@thyrsus.com>
8344Date:   Thu Aug 23 19:15:07 2012 -0400
8345
8346    Renove some unnecessary low-level markup.
8347    
8348    Also, SYNTAX -> SYNOPSIS so function prototypes get parsed by doclifter.
8349    This appears to have been somebody's thinko, it's only in a few of the files.
8350    
8351    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8352
8353commit b686600ab5ca93b5750f827786e79c329ab2db4d
8354Author: Eric S. Raymond <esr@thyrsus.com>
8355Date:   Thu Aug 23 10:53:33 2012 -0400
8356
8357    The .NT/.NE macro pair is no longer used.  Remove it.
8358    
8359    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8360
8361commit b83f2898528c5dd683acd2a2143879a760a8dad5
8362Author: Eric S. Raymond <esr@thyrsus.com>
8363Date:   Thu Aug 23 10:42:46 2012 -0400
8364
8365    Remove the one and only use of the .NT/.NE pair.
8366    
8367    The problem with these macros is that they rely on being able to
8368    center the note label.  That doesn't play well with modern HTML,
8369    not anyway without coomplications like CSS.  This use was just a cute
8370    trick, not adding enough value to be kept.
8371
8372commit e9509fa6745d25eee01ec6f1c34edf8a806d66b3
8373Author: Eric S. Raymond <esr@thyrsus.com>
8374Date:   Thu Aug 23 10:23:45 2012 -0400
8375
8376    The .C{ and .C} macros are never used.  Remove them.
8377    
8378    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8379
8380commit 6b2f7ddeea6c87dbec4ddfc19b1fed33f1bb8575
8381Author: Eric S. Raymond <esr@thyrsus.com>
8382Date:   Thu Aug 23 10:08:08 2012 -0400
8383
8384    The .FN macro, paired with .FD, is also never used. Remove it.
8385    
8386    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8387
8388commit cac6572701c976542da562b2b277b07ffb892a6f
8389Author: Eric S. Raymond <esr@thyrsus.com>
8390Date:   Thu Aug 23 10:03:00 2012 -0400
8391
8392    The ".FD" macro is never used.  Remove it.
8393    
8394    It was a temptation to presentation-level klugery and is best gone.
8395    
8396    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8397
8398commit 6e27a828f39f7028bc7f4a8736e7262fca250632
8399Author: Eric S. Raymond <esr@thyrsus.com>
8400Date:   Thu Aug 23 09:54:25 2012 -0400
8401
8402    Clean up, my last commit missed four cases.
8403    
8404    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8405
8406commit a7fb575957ff4d9dd3671994a005ac3be8bb10fe
8407Author: Eric S. Raymond <esr@thyrsus.com>
8408Date:   Thu Aug 23 08:24:17 2012 -0400
8409
8410    Eliminate all uses of tab stops in the libX11 man pages.
8411    
8412    Two steps: First, expand tabs to 8 spaces in code and structure
8413    listings.  Second, make the .Ds used to wrap code listings switch to
8414    constant-width font (CW) rather than numeric font position 1, which
8415    maps to R on most systems.
8416    
8417    It is possible some archaic systems won't know what CW is, but the
8418    only risk is that code listings won't look quite right on troff
8419    devices; the PostScript and DVI drivers definitely grok it, so those
8420    important cases are OK.
8421    
8422    The purpose of these changes is to get rid of presentation-level
8423    markup so these pages will lift clean to DocBook.
8424    
8425    Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
8426
8427commit 8042f88ace33573f9d0dfaa21ed54ac7cef266d5
8428Author: Will Thompson <will@willthompson.co.uk>
8429Date:   Mon Jul 9 18:00:27 2012 +0100
8430
8431    Add compose sequences for "therefore" and "because".
8432    
8433    These sequences look sensible to me. I added them to the APL-related
8434    section of Mathematical Operators—they're in that section of Unicode
8435    anyway.
8436    
8437    https://bugs.freedesktop.org/show_bug.cgi?id=51922
8438    
8439    Signed-off-by: Will Thompson <will@willthompson.co.uk>
8440    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
8441
8442commit 65358ea5079236b2508f787ac2fb2024a477e36d
8443Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8444Date:   Fri Jun 29 23:08:04 2012 -0700
8445
8446    Convert XCreate{Pix,Bit}map...Data to use C99 designated initializers
8447    
8448    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8449
8450commit 0dc93f7e43deb102b1f8fb7c4c4844cdce7ffd1e
8451Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8452Date:   Fri Jun 29 22:57:13 2012 -0700
8453
8454    XCreate{Pix,Bit}map...Data: Free pixmap in error path if XCreateGC fails
8455    
8456    Fixes leaks in error paths found by Parfait 1.0.0:
8457    
8458    Error: X Resource Leak
8459       Leaked X Resource pix
8460            at line 62 of CrBFData.c in function 'XCreateBitmapFromData'.
8461              pix initialized at line 60 with XCreatePixmap
8462    Error: X Resource Leak
8463       Leaked X Resource pix
8464            at line 70 of CrPFBData.c in function 'XCreatePixmapFromBitmapData'.
8465              pix initialized at line 66 with XCreatePixmap
8466    
8467    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8468    Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
8469
8470commit dce84b8c39ad5a8908c29bb6de25b6c3004c1ab7
8471Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8472Date:   Wed Jun 6 13:31:16 2012 -0700
8473
8474    libX11 spec: Correct prototype for XConvertSelection
8475    
8476    selection & target parameters were accidentally run together
8477    
8478    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8479
8480commit 6c5cb2a90a6479f56855e5167039c37c234cdfe7
8481Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8482Date:   Fri Jun 1 23:37:09 2012 -0700
8483
8484    libX11 1.5.0
8485    
8486    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8487
8488commit 05c587ec3be880721131a17c1dd4366e458fdd8b
8489Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8490Date:   Sat May 26 14:37:28 2012 -0700
8491
8492    libX11 1.4.99.902 (1.5 RC2)
8493    
8494    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8495
8496commit d817834d9772e3500d5102e2eae868b04a33c31f
8497Author: Pander <pander@users.sourceforge.net>
8498Date:   Sat May 5 19:02:10 2012 +0200
8499
8500    Compose: Reassigned squences with minus and a or o
8501    
8502    Reassigned squences with minus and a or o (vice versa and lower and
8503    upper case) to conform existing series and not resulting in tilde.
8504    Also added noe missing underscore sequence.
8505    
8506    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8507
8508commit 6bfd1dca6e7cb2046ee6bf9dbbddc0af5ef7cc00
8509Author: Pander <pander@users.sourceforge.net>
8510Date:   Sat May 5 16:56:05 2012 +0200
8511
8512    Compose: Removed <slash> <U> and vice versa
8513    
8514    Combination with lower case u suffices
8515    
8516    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8517
8518commit 3a8b1637132d1e36eb5e11f52dfb284081772d14
8519Author: Pander <pander@users.sourceforge.net>
8520Date:   Sat May 5 16:44:31 2012 +0200
8521
8522    Compose: Removed <n> <minus> for n with tilde, also vice versa & for upper case
8523    
8524    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8525
8526commit fd514d7a35119dd4413b51c84f2a536f2ca501a6
8527Author: Pander <pander@users.sourceforge.net>
8528Date:   Sat May 5 16:40:20 2012 +0200
8529
8530    Compose: Removed <S> in combination with a numeral for superscript
8531    
8532    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8533
8534commit 1b5cad3ca54410c4edbca79c23c463e9e088bc0d
8535Author: Pander <pander@users.sourceforge.net>
8536Date:   Sat May 5 16:28:01 2012 +0200
8537
8538    Compose: Reassigned <o> <apostrophe> to oacute, also for upper case.
8539    
8540    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8541
8542commit 0bbc0d5e605e2a4a3eb00a229b651d4546e8aef4
8543Author: James Cloos <cloos@jhcloos.com>
8544Date:   Thu May 10 12:27:28 2012 -0400
8545
8546    Remove duplicate compose sequences from commit e51e37c118ae.
8547    
8548    The compose-check script only handles compiled Compose files, not
8549    the Compose.pre files.  One must remember to use:
8550    
8551        ./autogen.sh; make; make check
8552    
8553    when reviewing patches to the Compose.pre files....
8554    
8555    Signed-off-by: James Cloos <cloos@jhcloos.com>
8556
8557commit e51e37c118ae6cb9ced8244ce1c410677e0279ce
8558Author: Geoff Streeter <geoff@dyalog.com>
8559Date:   Thu Mar 22 15:02:00 2012 +0000
8560
8561    Add APL support to compose
8562    
8563    Signed-off-by: Geoff Streeter <geoff@dyalog.com>
8564    Signed-off-by: James Cloos <cloos@jhcloos.com>
8565
8566commit dac90324cee224df977a428afe80d960dceca769
8567Author: Julien Cristau <jcristau@debian.org>
8568Date:   Sat May 5 16:05:07 2012 +0200
8569
8570    configure: make previous change work with older autoconf
8571    
8572    autoconf 2.63 doesn't seem to like the nested AC_CHECK_DECL/FUNC.  So do
8573    the tests separately.
8574    
8575    Reported-by: Dave Airlie
8576    Signed-off-by: Julien Cristau <jcristau@debian.org>
8577
8578commit f5b50af4324186962e258ffe9be78d5ee4681982
8579Author: Julien Cristau <jcristau@debian.org>
8580Date:   Sun Apr 29 16:43:09 2012 +0200
8581
8582    configure: check if issetugid is declared
8583    
8584    GNU/kFreeBSD has issetugid in libc (for legacy apps?), but doesn't
8585    declare it anywhere, causing gcc to error out with
8586    -Werror=implicit-function-declaration.  Use AC_CHECK_DECL in addition to
8587    AC_CHECK_FUNC so we disable this code instead of failing to build it.
8588    
8589    Debian bug#669670 <http://bugs.debian.org/669670>
8590    
8591    Signed-off-by: Julien Cristau <jcristau@debian.org>
8592    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8593    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8594
8595commit 52e1b5cc3b6de76ccf4285b55652474a522ed9a8
8596Author: Peter Hutterer <peter.hutterer@who-t.net>
8597Date:   Mon Apr 30 16:36:47 2012 +1000
8598
8599    Typo fix
8600    
8601    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8602
8603commit b64969f0e510d5d3300cf968741a3726a6409577
8604Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8605Date:   Tue Apr 17 18:12:02 2012 -0700
8606
8607    Add X11R7 sections to the libX11 & XKBlib credits to cover Docbook conversion
8608    
8609    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8610
8611commit 9ea611696f317ac3b3fb67893f1d6d87d49e3b5e
8612Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8613Date:   Tue Apr 17 18:01:36 2012 -0700
8614
8615    Add olinks from libX11 & localedb specs to ICCCM spec
8616    
8617    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8618
8619commit b3c1b8cdab7d14220426c9b997ac362dc16318fc
8620Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8621Date:   Tue Apr 17 17:49:44 2012 -0700
8622
8623    Add olinks from libX11 spec to ICCCM spec
8624    
8625    Also convert ICCCM title mentions from <emphasis> to <citetitle>
8626    
8627    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8628
8629commit ebebb65e753007ad01966dccc90bd6ca9a826488
8630Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8631Date:   Sat Apr 14 23:40:01 2012 -0700
8632
8633    libX11 AppC: Fix section headers that didn't translate from nroff properly
8634    
8635    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8636    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
8637
8638commit d5ab4ae0e74ae1fb30fb72add0751effe2759bf2
8639Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8640Date:   Sat Apr 14 23:13:05 2012 -0700
8641
8642    Add olinks from libX11 spec to x11protocol spec
8643    
8644    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8645    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
8646
8647commit 83878a0e34fffd255597300dd3e6cd43fcd645b0
8648Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8649Date:   Sat Apr 14 22:28:53 2012 -0700
8650
8651    libX11 spec: Remove .br nroff macro left behind in XGetWindowProperty prototype
8652    
8653    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8654
8655commit abc523fce31fcf2687229697a8eb656e343ecb0c
8656Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8657Date:   Thu Mar 15 22:14:45 2012 -0700
8658
8659    libX11 1.4.99.901 (1.5 RC1)
8660    
8661    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8662
8663commit b2cc5905a4a6d519957223e8ba2caef71520040c
8664Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8665Date:   Thu Mar 15 22:03:21 2012 -0700
8666
8667    Remove "register" qualifier that annoys Solaris Studio compiler
8668    
8669    Fixes warning:
8670    "Xrm.c", line 1094: warning: storage class after type is obsolescent
8671    
8672    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8673
8674commit 62d42953893f93a98db0504eaf06d650ceaf5811
8675Author: James Cloos <cloos@jhcloos.com>
8676Date:   Wed Mar 14 17:25:46 2012 -0400
8677
8678    Fix the gtk+ additions
8679    
8680    (Some of) the Dstroke and dstroke entries already were present as U011[01],
8681    even though XK_Dstroke and XK_dstroke are part of the latin2 set in keysymdef.h.
8682    
8683    The addition of <Multi_key> <o> <apostrophe> as a postfix version of
8684    <Multi_key> <apostrophe> <o> blocks the existing entries for ǻ and Ǻ.
8685    That prevents its and <Multi_key> <O> <apostrophe>’s addition.
8686    
8687    Signed-off-by: James Cloos <cloos@jhcloos.com>
8688
8689commit 91bcce48d94792f78333d2aea73961cc2e739d2e
8690Author: Pander <pander@users.sourceforge.net>
8691Date:   Wed Mar 14 12:54:53 2012 -0400
8692
8693    Complete compose key sequences for musical symbols
8694    
8695    Signed-off-by: Pander <pander@users.sourceforge.net>
8696    Signed-off-by: James Cloos <cloos@jhcloos.com>
8697
8698commit cf040016d455bc37f7665d6714337c5eafd8ea94
8699Author: Pander <pander@users.sourceforge.net>
8700Date:   Wed Mar 14 12:46:25 2012 -0400
8701
8702    Add compose sequences from gtk+ to X.Org
8703    
8704    Signed-off-by: Pander <pander@users.sourceforge.net>
8705    Signed-off-by: James Cloos <cloos@jhcloos.com>
8706
8707commit a4c591ced5cac9301b9abfa0e521be2d0b267882
8708Author: Keith Packard <keithp@keithp.com>
8709Date:   Sun Mar 4 02:00:13 2012 -0800
8710
8711    Block for other threads in _XUserLockDisplay
8712    
8713    Wait for all other threads to release the user-level lock when
8714    acquiring it. This ensures that only one thread at a time holds the
8715    user-level lock, necessary as it is a nesting lock and a single
8716    variable is used to determine when the lock is nesting and when it is
8717    contended.
8718    
8719    Signed-off-by: Keith Packard <keithp@keithp.com>
8720    Reviewed-by: Jamey Sharp <jamey@minilop.net>
8721
8722commit ed00b460acb08787b695f27b864e96102dfd4867
8723Author: Jon TURNEY <jon.turney@dronecode.org.uk>
8724Date:   Fri Oct 28 11:09:20 2011 -0500
8725
8726    Don't use caddr_t casts
8727    
8728    (caddr_t) isn't used anywhere else in xcb or libX11.
8729    Cast to (char *) for consistency.
8730    
8731    Removing this cast allows building for MinGW without patching.
8732    
8733    v2: Cast to (char *) rather than just dropping the cast
8734    
8735    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8736    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8737
8738commit 20adca02c2a1d0b7c95ecbe387d68f881fd57754
8739Author: Julien Cristau <jcristau@debian.org>
8740Date:   Sun Feb 19 13:27:38 2012 +0100
8741
8742    Revert "xcb: Add TCP fallback"
8743    
8744    This reverts commit f09c5299a381e2729e800a0ac43f1c0e371f65f6.
8745    
8746    The TCP fallback ended up falling back to UNIX socket connection if
8747    $DISPLAY was set to e.g. some.host:0 and the initial attempt failed.
8748    
8749    Debian bug#659558 <http://bugs.debian.org/659558>
8750    
8751    Signed-off-by: Julien Cristau <jcristau@debian.org>
8752    
8753    Conflicts:
8754    
8755            src/OpenDis.c
8756
8757commit f4378193619baa9bb973c1b5b718721bbcbe92c7
8758Author: James Cloos <cloos@jhcloos.com>
8759Date:   Wed Feb 22 14:13:20 2012 -0500
8760
8761    Make the compose sequence for ẞ work.
8762    
8763    There is no XK_Ssharp symbol for U+1E9E LATIN CAPITAL LETTER SHARP S,
8764    so use the U1e9e symbol in the Compose sequence.
8765    
8766    (Compose sequences do not work when the target symbol is unknown.)
8767    
8768    Signed-off-by: James Cloos <cloos@jhcloos.com>
8769
8770commit 2ca641c3a506dcbee97e279b67990d5387389f36
8771Author: Marko Myllynen <myllynen@redhat.com>
8772Date:   Mon Feb 20 17:04:59 2012 +0200
8773
8774    Use ezh/EZH in compose maps
8775    
8776    Related: https://bugs.freedesktop.org/show_bug.cgi?id=19687
8777    
8778    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
8779    Reviewed-by: Matt Dew <marcoz@osource.org>
8780    Signed-off-by: James Cloos <cloos@jhcloos.com>
8781
8782commit 61725822f20f47684a545c1797183ee7075243ac
8783Author: Marko Myllynen <myllynen@redhat.com>
8784Date:   Sun Feb 20 17:09:43 2011 +0200
8785
8786    Provide translation from XK_permille (ad5) to Unicode (U2030)
8787    
8788    https://bugs.freedesktop.org/show_bug.cgi?id=19687
8789    
8790    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
8791    Reviewed-by: Matt Dew <marcoz@osource.org>
8792    Signed-off-by: James Cloos <cloos@jhcloos.com>
8793
8794commit d2cce0abba0fa0143f49026442c8cab5ed721625
8795Author: Jeremy Huddleston <jeremyhu@apple.com>
8796Date:   Sun Feb 12 19:01:43 2012 -0800
8797
8798    nls: Use LC_CTYPE=C for sed magic
8799    
8800    Stricter versions of sed can trip up if the input does not match
8801    LC_CTYPE
8802    
8803    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
8804
8805commit f180e043f55531933bedfa6e0ff36a00a9ec07f3
8806Author: Frédéric Boiteux < <fboiteux@calistel.com>
8807Date:   Wed Nov 30 12:47:31 2011 +0000
8808
8809    Compositions with the dead greek symbol
8810    
8811    FreeDesktop Bug 21475 <https://bugs.freedesktop.org/show_bug.cgi?id=21475>
8812    
8813    Signed-off-by: Frédéric Boiteux <fboiteux@calistel.com>
8814    Signed-off-by: James Cloos <cloos@jhcloos.com>
8815
8816commit d58e8f8e27790017fcfdeca0843b7318d541c189
8817Author: Ryan Pavlik <rpavlik@iastate.edu>
8818Date:   Sun Jan 1 21:04:52 2012 +0000
8819
8820    Use pthreads on MinGW also
8821    
8822    Use pthreads (provided by the pthreads-win32 compatability library which implements
8823    them using native Win32 threading) on MinGW
8824    
8825    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8826
8827commit bf15ccb6821664746ec23d769d757edf8059007e
8828Author: Ryan Pavlik <rpavlik@iastate.edu>
8829Date:   Mon Jan 23 14:18:02 2012 +0000
8830
8831    Add XWindows.h include to Xxcbint.h
8832    
8833    This avoids some conflicting type re-definition errors which occur if
8834    we attempt to include Windows headers after Xmd.h
8835    
8836    Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
8837
8838commit cadcbd376f0a5d17a71a2fe2f8ced5d93232921a
8839Author: Matt Dew <marcoz@osource.org>
8840Date:   Sat Jan 21 17:59:51 2012 -0700
8841
8842    informaltable & table cleanup
8843    
8844    On certain tables, add top and bottom borders to table header
8845    and a bottom border to the table. This matches what those
8846    tables in the old pdfs looked like.
8847    
8848    the <?dbfo keep-together='always'> prevents tables from
8849    splitting across pages. Useful for tiny tables.
8850    
8851    Converting the colwidth to a floating point, IE, 1* -> 1.0*
8852    cleans up these build errors:
8853    WARNING: table-layout="fixed" and column-width unspecified =>
8854    falling back to proportional-column-width(1)
8855    
8856    Signed-off-by: Matt Dew <marcoz@osource.org>
8857
8858commit bb551f654df8f647c867f79252241964521e689e
8859Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8860Date:   Wed Dec 28 21:22:41 2011 -0800
8861
8862    Add more Xkb man pages to the See Also lists for core keyboard functions
8863    
8864    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8865    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
8866
8867commit b7bb23fe7c8b8a17128e5bc98a05f68595190438
8868Author: Gaetan Nadon <nadon@memsize.(none)>
8869Date:   Fri Dec 30 17:08:14 2011 -0500
8870
8871    docbook.am: embed css styles inside the HTML HEAD element
8872    
8873    Rather than referring to the external xorg.css stylesheet, embed the content
8874    of the file in the html output produced. This is accomplished by using
8875    version 1.10 of xorg-xhtml.xsl.
8876    
8877    This makes the whole html docs tree much more relocatable.
8878    In addition, it eliminates xorg.css as a runtime file which makes
8879    xorg-sgml-doctools a build time only package.
8880    
8881    Signed-off-by: Gaetan Nadon <nadon@memsize.(none)>
8882
8883commit 70505468b7c4a7068cc39be42e421dcee34ec595
8884Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8885Date:   Tue Dec 13 19:58:30 2011 -0800
8886
8887    makekeys: move buf declaration from global to main to silence gcc -Wshadow
8888    
8889    The global was only referenced in the main() function, which passes it
8890    as an argument of the same name to the parse_line() function, leading
8891    to gcc -Wshadow warnings:
8892    
8893    makekeys.c: In function ‘parse_line’:
8894    makekeys.c:58:24: warning: declaration of ‘buf’ shadows a global declaration
8895    makekeys.c:54:13: warning: shadowed declaration is here
8896    
8897    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8898
8899commit 87e10a7b9a97c951ab4d477f61177779ac0a6a66
8900Author: Kusanagi Kouichi <slash@ac.auone-net.jp>
8901Date:   Wed Dec 14 02:17:55 2011 -0500
8902
8903    XQueryColors: Split a request into multiple requests if necessary
8904    
8905    https://bugs.freedesktop.org/show_bug.cgi?id=9160
8906    
8907    Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
8908    Signed-off-by: James Cloos <cloos@jhcloos.com>
8909
8910commit d3b3570592e9b9e57f270a0bd86762fd205a2833
8911Author: Andreas Wettstein <wettstein509@solnet.ch>
8912Date:   Tue Nov 8 20:18:09 2011 +0000
8913
8914    XIM: Allow modifier releases in compose sequences (#26705)
8915    
8916    Currently, only non-modifier keys (actually, keysyms) can be part of a compose
8917    sequence, and they are matched against the defined compose sequences at the
8918    time the key is pressed.  The patch allows to use modifier keys an well, but
8919    matches them on key release, and only if no other key has been pressed after
8920    the modifier.
8921    
8922    Releasing a non-matched modifier during an ongoing compose sequence only aborts
8923    the sequence if any modifier release would have matched.  In particular, if no
8924    compose sequences with modifiers are specified, the compose mechanism works
8925    exactly as without this patch.
8926    
8927    Even if modifiers are part of a compose sequence, they are not filtered.  This
8928    is because modifiers affect the keyboard state no matter what we do here and,
8929    therefore, filtering them only could confuse clients.
8930    
8931    The purpose is this extension to the compose mechanism is to allow to make
8932    better use of keys in convenient reach for touch typing.
8933    
8934    Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
8935    Signed-off-by: James Cloos <cloos@jhcloos.com>
8936
8937commit 56448a626fc90bcf75a1fa2f4c294b0eb1f23bd6
8938Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8939Date:   Tue Nov 29 23:56:57 2011 -0800
8940
8941    Reject negative string counts in copy_string_list
8942    
8943    Silences parfait warning of a potential memory leak:
8944       Memory leak of pointer 'dst' allocated with malloc(length)
8945            at line 160 of FSWrap.c in function 'copy_string_list'.
8946              'dst' allocated at line 145 with malloc(length).
8947              dst leaks when count <= 0 at line 154.
8948    
8949    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8950
8951commit bf2d7c8c6d70539c72560b1921e18df2610acf29
8952Author: Peter Hutterer <peter.hutterer@who-t.net>
8953Date:   Fri Nov 11 14:56:36 2011 +1000
8954
8955    libX11 1.4.99.1
8956    
8957    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8958
8959commit 24283d40b1e4314c6647dda49d2a159833341a8b
8960Author: Derek Buitenhuis <derek.buitenhuis@gmail.com>
8961Date:   Thu Sep 22 18:44:13 2011 -0400
8962
8963    makekeys: Fix build/target word size mismatch when cross-compiling
8964    
8965    Since makekeys is built using build environment's gcc and
8966    runs natively, we have to make sure that the size of the
8967    Signature type is the same on both the native environment
8968    and the target, otherwise we get mismatches upon running X,
8969    and some LSB test failures (xts5).
8970    
8971    Use an unsigned 32-bit integer on all platforms. Also,
8972    eliminate the redundant multiple typedefs for the
8973    Signature type.
8974    
8975    Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
8976    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
8977
8978commit 1c41f3b9b86b5eeedfa3bff92e519d45aa097587
8979Author: Alan Coopersmith <alan.coopersmith@oracle.com>
8980Date:   Wed Nov 9 21:00:36 2011 -0800
8981
8982    XlcSL.c: convert old-style function definitions to ANSI C89 style
8983    
8984    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
8985
8986commit f6dad6aaa384bf836baa28ccb72b476a85c40eff
8987Author: Peter Hutterer <peter.hutterer@who-t.net>
8988Date:   Thu Oct 27 13:53:22 2011 +1000
8989
8990    Use GetReqSized for GetReq and GetReqExtra
8991    
8992    GetEmptyReq and GetResReq cannot do this due to the final typecast -
8993    typically requests that need either of those do not have their own typedef
8994    in the protocol headers.
8995    
8996    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
8997    Reviewed-by: Jamey Sharp <jamey@minilop.net>
8998
8999commit c9c99058b9d98789c0b2d7e78a23443c2b57a047
9000Author: Peter Hutterer <peter.hutterer@who-t.net>
9001Date:   Fri Oct 14 14:51:06 2011 +1000
9002
9003    include: Add GetReqSized() for request buffers of specific size
9004    
9005    Some XI2 requests change in size over different versions and libXi would
9006    need to hack around GetReq and GetReqExtra. Add a new GetReqSized so the
9007    library can explicitly specify the size of the request in 4-byte units.
9008    
9009    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9010    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9011
9012commit ba8a7a19165e30d14bc165f43f67c19b6a115585
9013Author: Peter Hutterer <peter.hutterer@who-t.net>
9014Date:   Thu Oct 27 13:24:10 2011 +1000
9015
9016    Switch GetEmptyReq and GetResReq to call _XGetRequest
9017    
9018    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9019    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9020
9021commit 4a060f993bf676cf21ad9784e010f54134da7b40
9022Author: Peter Hutterer <peter.hutterer@who-t.net>
9023Date:   Mon Oct 17 09:45:15 2011 +1000
9024
9025    Add _XGetRequest as substitute for GetReq/GetReqExtra
9026    
9027    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
9028    Reviewed-by: Jamey Sharp <jamey@minilop.net>
9029
9030commit d8956520deb79c1cbb5e974c175bf8493859b22b
9031Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9032Date:   Fri Oct 14 17:53:00 2011 -0700
9033
9034    Fix "nomal" -> "normal" typo in several comments
9035    
9036    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9037
9038commit 002b36e308a26a152504f9b40aa08a0dce9a7991
9039Author: Xue Wei <Wei.Xue@Sun.COM>
9040Date:   Fri Oct 14 17:39:21 2011 -0700
9041
9042    mbtocs should not truncate input
9043    
9044    Fixes pasting more than 1024 bytes into xterm, as described in
9045    https://bugs.freedesktop.org/show_bug.cgi?id=25209
9046    
9047    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9048
9049commit 9b8d8c9e5b27273e8856a3851ba9b68022bed3cd
9050Author: Marko Myllynen <myllynen@redhat.com>
9051Date:   Mon Oct 10 09:41:13 2011 +0300
9052
9053    Add new compose sequences
9054    
9055    This patch adds few new compose sequences to the en_US map, ligatures
9056    common in typesetting and sequences already in use in the fi_FI map.
9057    
9058    https://bugs.freedesktop.org/show_bug.cgi?id=30621
9059    https://bugs.freedesktop.org/show_bug.cgi?id=34523
9060    
9061    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9062
9063commit 738f7b867341c1da87ee667d48815822715c3e75
9064Author: Choe Hwanjin <choe.hwanjin@gmail.com>
9065Date:   Thu Oct 13 07:58:02 2011 +0900
9066
9067    XIM: Make Xim handle NEED_SYNC_REPLY flag
9068    
9069    NEED_SYNC_REPLY flag should be in Xim not in Xic.
9070    Because the focused Xic can be changed before sending sync reply.
9071    After focused Xic changed, the new Xic doesn't have NEED_SYNC_REPLY
9072    flag enabled, so libX11 doesn't send XIM_SYNC_REPLY packet.
9073    
9074    This patch adds sync reply flag to Xim and removes sync reply
9075    from Xic.
9076    
9077    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=7869
9078    
9079    Signed-off-by: Choe Hwanjin <choe.hwanjin@gmail.com>
9080    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9081
9082commit 50e15379331c436851abb194251cde96999e67b3
9083Author: Bodo Graumann <mail@bodograumann.de>
9084Date:   Mon Oct 10 16:40:52 2011 +0200
9085
9086    libX11: Fixing modifier key range in Xutil.h (Bug #21910)
9087    
9088    IsModifierKey, defined in include/X11/Xutil.h, is a macro determining,
9089    which keys are regarded as modifiers. The constants ISO_Level5_Shift,
9090    ISO_Level5_Latch and ISO_Level5_Lock where excluded previously, leaving
9091    some Neo2 modifiers functionless in combination with compose.
9092    This patch adjusts the range to include the correct, full range of
9093    modifier constants.
9094    
9095    Neo2 Bug 277 <http://wiki.neo-layout.org/ticket/277>
9096    
9097    X.Org Bug 21910 <http://bugs.freedesktop.org/show_bug.cgi?id=21910>
9098    
9099    Signed-off-by: Bodo Graumann <mail@bodograumann.de>
9100    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
9101
9102commit b5a108624331fabf393223c0891914cc54d4caf4
9103Author: Yann Droneaud <yann@droneaud.fr>
9104Date:   Tue Oct 11 17:27:59 2011 +0200
9105
9106    Return name instead of False in XSetICValues()
9107    
9108    In case of error, XSetICValues() must return the first argument
9109    that failed to be set.
9110    
9111    But in some error paths, it returns False, which is converted to NULL,
9112    so the function returns OK in case of error.
9113    
9114    Signed-off-by: Yann Droneaud <yann@droneaud.fr>
9115    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9116
9117commit 10992cb8dc5e4d938e7e5a633b68a81b5875f3ba
9118Author: Yann Droneaud <yann@droneaud.fr>
9119Date:   Sun Oct 9 17:56:45 2011 +0200
9120
9121    Return name instead of value in XGetIMValues() and XSetIMValues()
9122    
9123    As stated in man page (XOpenIM) and Xlib documentation (chapter 13.5.3),
9124    XGetIMValues() and XSetImValues() "returns the name of the first argument
9125    that could not be obtained."
9126    
9127    But currently,
9128    
9129      err = XGetIMValues(im, "invalid", &arg, NULL);
9130    
9131    returns &arg instead of "invalid".
9132    
9133    This patch fixes https://bugs.freedesktop.org/show_bug.cgi?id=12897
9134    
9135    Signed-off-by: Yann Droneaud <yann@droneaud.fr>
9136    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
9137
9138commit f2651e03f3295a453a2965c3749bc8b6e66f1c09
9139Author: Jeremy Huddleston <jeremyhu@apple.com>
9140Date:   Mon Oct 10 14:09:17 2011 -0700
9141
9142    Mark XKeycodeToKeysym as _X_DEPRECATED
9143    
9144    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9145
9146commit 25d7bb9c705623ebbb3afba3b86d0ded5b6bb4fb
9147Author: Jeremy Huddleston <jeremyhu@apple.com>
9148Date:   Sun Oct 9 02:38:28 2011 -0700
9149
9150    Fix nobreakspace for pt_BR.UTF-8
9151    
9152    https://bugs.freedesktop.org/show_bug.cgi?id=31334
9153    
9154    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9155
9156commit 913603660c17ec59a4d39d3e3b9d60469ee5c060
9157Author: Jeremy Huddleston <jeremyhu@apple.com>
9158Date:   Sun Oct 9 02:26:45 2011 -0700
9159
9160    Fix potential uninitialized variable access in _XimMakeICAttrIDList
9161    
9162    Found by clang static analysis
9163    
9164    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9165
9166commit 566ceaf5a92c721ac7155528e4d0d2e5cbef023f
9167Author: Jeremy Huddleston <jeremyhu@apple.com>
9168Date:   Sun Oct 9 02:25:50 2011 -0700
9169
9170    Remove self-resolving aliases
9171    
9172    https://bugs.freedesktop.org/show_bug.cgi?id=30112
9173    
9174    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9175
9176commit 22ba43d198dcca86c88eb15a56fc7d8fc47c422e
9177Author: Matt Dew <marcoz@osource.org>
9178Date:   Fri Oct 7 22:52:30 2011 -0600
9179
9180    Cleanup IDs and links in doc
9181    
9182    1 - fix the capitalization of the ID attributes to match either the
9183         <title> or <funcdef> string it goes with.
9184    2 - fix any <linkend>'s that were affected by 1.
9185    3 - any <function> in the docs that has an actual funcdef,
9186    will become an olink.
9187    
9188    Signed-off-by: Matt Dew <marcoz@osource.org>
9189
9190commit f858f3326adbc0c5711669b92a64a84a9083a055
9191Author: James Cloos <cloos@jhcloos.com>
9192Date:   Tue Oct 4 17:11:11 2011 -0400
9193
9194    [nls] Fix typo/synco.
9195    
9196    The iso8859-1 Compose table includes an optional section which uses
9197    Ctrl<T> as a substitute for <Multi_key>.  In that section the sequence
9198    to generate an @ (at) either was incorrectly copied from the Multi_key
9199    section or was not kept in sync with the Multi_key section.
9200    
9201    Fixing this eliminates the warning from compose-check.pl:
9202    
9203    ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
9204    ┃ Clash with existing sequence in iso8859-1/Compose.pre
9205    ┃  on line 661: Ctrl<T> <A> <A>
9206    ┃    line #661: Ctrl<T> <A> <A>         : "\305"        Aring
9207    ┃    line #480: Ctrl<T> <A> <A>         : "@"   at
9208    ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
9209    
9210    Signed-off-by: James Cloos <cloos@jhcloos.com>
9211
9212commit 7f35f7efc23234b3f8529fee31a802c21846d9ab
9213Author: Jeremy Huddleston <jeremyhu@apple.com>
9214Date:   Mon Sep 26 11:32:56 2011 -0700
9215
9216    Remove conflicting compose sequences for cent and colon
9217    
9218    Regression from 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4
9219    Found by tinderbox
9220    
9221    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9222
9223commit 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4
9224Author: Jeremy Huddleston <jeremyhu@apple.com>
9225Date:   Sun Sep 25 16:29:17 2011 -0700
9226
9227    Add additional compose sequences for pound sterling, yen, and cent (mixed case)
9228    
9229    https://bugs.freedesktop.org/show_bug.cgi?id=1013
9230    
9231    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9232    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
9233
9234commit 0ca4153a373e2af7a9eb3cef003393ab332bc79e
9235Author: Jeremy Huddleston <jeremyhu@apple.com>
9236Date:   Sat Sep 24 16:44:04 2011 -0700
9237
9238    Use a configure check for seteuid
9239    
9240    HP-UX doesn't have seteuid
9241    
9242    https://bugs.freedesktop.org/show_bug.cgi?id=1497
9243    
9244    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9245
9246commit 79594b4d66344f248eb4314ecb78eb81f632c3ab
9247Author: Gaetan Nadon <memsize@videotron.ca>
9248Date:   Tue Sep 20 16:06:10 2011 -0400
9249
9250    localedb specs: use <copyright> for first holder of multi license
9251    
9252    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9253
9254commit 154430268cf004b06920edde2f1812dc16d71fae
9255Author: Gaetan Nadon <memsize@videotron.ca>
9256Date:   Tue Sep 20 16:01:52 2011 -0400
9257
9258    libX11 specs: use <copyright> for first holder of multi license
9259    
9260    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9261
9262commit 66e621f58dee804acb795ff2c79ce6dd38b34a3f
9263Author: Tollef Fog Heen <tfheen@err.no>
9264Date:   Tue Sep 20 13:07:37 2011 -0400
9265
9266    NLS: Add more vulgar fractions
9267    
9268    Add 1/7, 1/9, 1/10 and 0/3 vulgar fractions.
9269    
9270    Signed-off-by: Tollef Fog Heen <tfheen@err.no>
9271    Signed-off-by: James Cloos <cloos@jhcloos.com>
9272
9273commit e37a6da814b5653be46000a9a76902729660a2e6
9274Author: Alexander Polakov <polachok@gmail.com>
9275Date:   Sat Sep 17 20:54:58 2011 +0400
9276
9277    XGrabKey manual page: change XAllowAccess to XAllowEvents in See Also
9278    
9279    There is no XAllowAccess man page to see.
9280    
9281    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9282    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9283
9284commit a23f3323f269b33a43d16dce01395cd28bc41b45
9285Author: Gaetan Nadon <memsize@videotron.ca>
9286Date:   Mon Sep 19 15:27:44 2011 -0400
9287
9288    XKB: provide adequate quotes for the license text
9289    
9290    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9291
9292commit a9c7a5cad91d968f5536ef8fc735036e921832a3
9293Author: Gaetan Nadon <memsize@videotron.ca>
9294Date:   Mon Sep 19 15:18:53 2011 -0400
9295
9296    XIM: refactor the multi licensing legal text
9297    
9298    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9299
9300commit c46f934ed89e7de79746a0387c9f998d91994ea6
9301Author: Gaetan Nadon <memsize@videotron.ca>
9302Date:   Mon Sep 19 10:33:30 2011 -0400
9303
9304    xim trans: restore Fujitsu copyright legal text
9305    
9306    Somehow lost during docbook conversion. text from x.org ftp R7.5.
9307    
9308    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9309
9310commit 3d75f993382bfdc89d31668d7dfc71c91222e0d7
9311Author: Gaetan Nadon <memsize@videotron.ca>
9312Date:   Mon Sep 19 10:11:46 2011 -0400
9313
9314    xtrans: restore X Consortium original legal text
9315    
9316    Asking X Consortium permission to use The Open Group name makes no sense.
9317    Even more so in 1994 before X Window System was passed on to the Open Group.
9318    
9319    Using original text from xorg-docs/general/License
9320    
9321    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9322
9323commit 136a381585a4a9686c11bad1a6130837978e677f
9324Author: Gaetan Nadon <memsize@videotron.ca>
9325Date:   Mon Sep 19 10:03:21 2011 -0400
9326
9327    Framework: restore X Consortium copyright
9328    
9329    Somewhat dammaged during docbook conversion.
9330    Also restore pasrt of the original license text
9331    
9332    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9333
9334commit 33f346878406762704a0c13cdc018111c666f3c1
9335Author: Gaetan Nadon <memsize@videotron.ca>
9336Date:   Mon Sep 19 09:50:00 2011 -0400
9337
9338    localedb: restore X Consortium original legal text
9339    
9340    Asking X Consortium permission to use The Open Group name makes no sense.
9341    Even more so in 1994 before X Window System was passed on to the Open Group.
9342    
9343    Using original text from xorg-docs/general/License
9344    
9345    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9346
9347commit e99c9338e413efca026210b9d830efb5c74d34e5
9348Author: Gaetan Nadon <memsize@videotron.ca>
9349Date:   Mon Sep 19 08:23:18 2011 -0400
9350
9351    specs: support multi licensed copyright notice and license text
9352    
9353    For documentation having multiple licenses, the copyright and legalnotice
9354    elements sequence cannot instantiated multiple times.
9355    The copyright notice and license text are therefore coded inside a
9356    legalnotice element. The role attribute on the paragraph is used to allow
9357    styling of the copyright notice text which should not be italicized.
9358    
9359    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9360
9361commit b9dedc757e667333ce899115d618f25cdaa4dd5e
9362Author: Gaetan Nadon <memsize@videotron.ca>
9363Date:   Mon Sep 19 08:17:50 2011 -0400
9364
9365    localedb: add release info to spec
9366    
9367    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9368
9369commit 4519c89a87ddce01e599542737a27c26030fe4ff
9370Author: Gaetan Nadon <memsize@videotron.ca>
9371Date:   Mon Sep 19 08:05:32 2011 -0400
9372
9373    specs: fix The Open Group license text
9374    
9375    The warranty referred to the X Consortium
9376    
9377    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9378
9379commit 08ac378423e7be72e340197ba5559a1a36f1783c
9380Author: Gaetan Nadon <memsize@videotron.ca>
9381Date:   Mon Sep 19 08:03:39 2011 -0400
9382
9383    specs: The strandard name is still "X Consortium Standard"
9384    
9385    This spec, and fsproto spec, are the only two docs with a different
9386    standard name.
9387    
9388    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9389
9390commit 8dfbeb1b1b1e8aa17f69d0a373155419a5f6a172
9391Author: Gaetan Nadon <memsize@videotron.ca>
9392Date:   Sun Sep 18 13:22:34 2011 -0400
9393
9394    specs: support multi licensed copyright notice and license text
9395    
9396    For documentation having multiple licenses, the copyright and legalnotice
9397    elements sequence cannot instantiated multiple times.
9398    The copyright notice and license text are therefore coded inside a legalnotice
9399    element. The role attribute on the paragraph is used to allow styling of the
9400    copyright notice text which should not be italicized.
9401    
9402    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9403
9404commit 278ca8947c1dabb2b819527dca0fa6190c034f67
9405Author: Gaetan Nadon <memsize@videotron.ca>
9406Date:   Fri Sep 16 11:45:56 2011 -0400
9407
9408    docs: merge copyright holder under the same copyright notice
9409    
9410    As per the docbook markup dtd.
9411    
9412    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9413
9414commit 7ff012bb433109a622ac122ade71669e842ebdcf
9415Author: Gaetan Nadon <memsize@videotron.ca>
9416Date:   Fri Sep 16 10:09:57 2011 -0400
9417
9418    specs: handle multiple sets of copyright notice/license/warranty
9419    
9420    Docbook groups all the <copyright> elements together and all the
9421    <legalnotice> elements together.
9422    
9423    We cannot have a sequence:
9424    <copyright> <legalnotice> <copyright> <legalnotice> [...]
9425    
9426    A workaround, which was done in some documents, is to put the copyright
9427    notice inside the legalnotice in plain text without the <copyright> element.
9428    A formal paragraph title is added here which makes the copyright notice bold,
9429    and makes it much easier to locate.
9430    
9431    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9432    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9433
9434commit 4a550c71b8221c37b1a9378d5a170da9eaa03405
9435Author: Gaetan Nadon <memsize@videotron.ca>
9436Date:   Wed Sep 14 15:55:42 2011 -0400
9437
9438    specs: remove orphan affiliation.
9439    
9440    Authors affiliation are correct.
9441    
9442    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9443
9444commit 0cc02a6df65214560b7575e89cebee741d6d2469
9445Author: Gaetan Nadon <memsize@videotron.ca>
9446Date:   Tue Sep 13 17:39:02 2011 -0400
9447
9448    specs: use appropriate markup for Copyright statements
9449    
9450    Also move <releaseinfo> to match order of appearance
9451    
9452    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9453
9454commit afe13e19ebf0fc4e3460644164433af016f0add7
9455Author: Gaetan Nadon <memsize@videotron.ca>
9456Date:   Mon Sep 12 16:54:45 2011 -0400
9457
9458    docs: use the &fullrelvers; entity to set X11 release information
9459    
9460    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9461
9462commit 22a2153282db13e9d94881022b8c979539841a2c
9463Author: Gaetan Nadon <memsize@videotron.ca>
9464Date:   Sun Sep 11 19:49:53 2011 -0400
9465
9466    docs: remove <productnumber> which is not used by default
9467    
9468    This element is not rendered by default on the title. A template
9469    customization is required to display it.
9470    X Window System does not have a product number.
9471    
9472    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9473
9474commit 719f16570d9fcfd15247813ee51fa51ac8a6ff4c
9475Author: Gaetan Nadon <memsize@videotron.ca>
9476Date:   Sun Sep 11 17:40:21 2011 -0400
9477
9478    docs: use the &fullrelvers; entity to set X11 release information
9479    
9480    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9481
9482commit a6b2992f50b571d612ea9ade631c432c3099bc12
9483Author: Gaetan Nadon <memsize@videotron.ca>
9484Date:   Sun Sep 11 17:38:42 2011 -0400
9485
9486    docs: remove orphan <affiliation>
9487    
9488    Somehow created during the conversion from roff. Unable to locate
9489    the author to which it belongs.
9490    
9491    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9492
9493commit c7420060b6e47b8467ba50c796ec3c1bab090bc7
9494Author: Gaetan Nadon <memsize@videotron.ca>
9495Date:   Sun Sep 11 17:38:05 2011 -0400
9496
9497    docs: remove <productnumber> which is not used by default
9498    
9499    This element is not rendered by default on the title. A template
9500    customization is required to display it.
9501    X Window System does not have a product number.
9502    
9503    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9504
9505commit 7d5b718c1edbc43c54b7001be548e515a49540ea
9506Author: Gaetan Nadon <memsize@videotron.ca>
9507Date:   Fri Sep 9 14:46:39 2011 -0400
9508
9509    docbook.am: embed css styles inside the HTML HEAD element
9510    
9511    Rather than referring to the external xorg.css stylesheet, embed the content
9512    of the file in the html output produced. This is accomplished by using
9513    version 1.10 of xorg-xhtml.xsl.
9514    
9515    This makes the whole html docs tree much more relocatable.
9516    In addition, it eliminates xorg.css as a runtime file which makes
9517    xorg-sgml-doctools a build time only package.
9518    
9519    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9520
9521commit 24632d280491c465dfe208ef788815890892e326
9522Author: Gaetan Nadon <memsize@videotron.ca>
9523Date:   Mon Sep 5 12:53:12 2011 -0400
9524
9525    compose: upgrade makefile to support olinking on chunked html
9526    
9527    The essential differences over the regular docbook.am are:
9528    Adding root.filename parameter for naming of chapters html files.
9529    Using xhtml xmlto format and xorg-chunk.xsl stylesheet
9530    Set olink.base.uri for pdf but not for chunked html
9531    Olink is not applicable to ps and txt formats.
9532    
9533    Html chapters are added to shelf_DATA as they are also installed.
9534    The xml is generated from a perl script and not distributed.
9535    
9536    Requires version 1.10 of xorg-sgml-doctools.
9537    
9538    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9539
9540commit 93ab0a1c9a4f9094cf0c8e1c686130673e681798
9541Author: Gaetan Nadon <memsize@videotron.ca>
9542Date:   Wed Aug 31 19:46:20 2011 -0400
9543
9544    docbook.am: refactor common flags for xmlto and xsltproc
9545    
9546    Maximize reuse and reduce risk of setting the wrong flag
9547    at the wrong place.
9548    
9549    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9550
9551commit 85b725aa9ef76aeb5e94d42e0cd60f896da7ab97
9552Author: Gaetan Nadon <memsize@videotron.ca>
9553Date:   Wed Aug 31 10:46:59 2011 -0400
9554
9555    docbook.am: add search path for local entities
9556    
9557    Currently, only $(XORG_SGML_PATH)/X11 is searched for xml entities.
9558    A module may want to add entities that apply only to itself and
9559    not to all modules, like the xserver does.
9560    
9561    This feature may or may not be used in this module, but all modules
9562    share a copy of docbook.am.
9563    
9564    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9565
9566commit b5ed2d368d47f244d43d2513329fa2758bc4ef54
9567Author: Gaetan Nadon <memsize@videotron.ca>
9568Date:   Thu Aug 25 14:39:20 2011 -0400
9569
9570    docbook.am: explicitly list xmlto flags for each target
9571    
9572    Normal evolution of make targets make it impractical to factor out
9573    common command flags for xmlto.
9574    
9575    The targets now list each command option as its presence or absence
9576    needs to be justified.
9577    
9578    xorg.ss is only needed by xmlto for html.
9579    masterdb is only needed by xmlto.
9580    img.src.path must not be used by html.
9581    xsltproc need to use customization layer xorg-*.xsl.
9582    txt format is not required to search masterdb.
9583    
9584    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9585
9586commit d5b3605f3b0639452e892e9ff3afb18bb7442069
9587Author: Gaetan Nadon <memsize@videotron.ca>
9588Date:   Thu Aug 25 13:12:49 2011 -0400
9589
9590    docbook.am: do not generate docs if docbook customization layer is missing
9591    
9592    The stylesheets used to be only about style, fonts or colors.
9593    Complex features are now used like olink and chunked html which
9594    may cause a build break when stylesheets are missing or lead to hard
9595    to find problem.
9596    
9597    Some modules may be built when stylesheets are present while others
9598    are built without. There is no requirement to build crippled docs.
9599    
9600    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9601
9602commit 7f23c72c94d9d14122426b0d8c66054f1402769a
9603Author: Gaetan Nadon <memsize@videotron.ca>
9604Date:   Fri Aug 5 14:33:32 2011 -0400
9605
9606    libX11 specs: review doclifter generated tables
9607    
9608    Many tables had a questionnable layout and some had information dropped.
9609    Each table was cross-referenced with a pre-docbook version
9610    to ensure semantic integrity.
9611    
9612    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9613
9614commit 1efdbeb8cd08f443188ac783cbab6ba0654a66d6
9615Author: Gaetan Nadon <memsize@videotron.ca>
9616Date:   Wed Jul 20 15:28:47 2011 -0400
9617
9618    credits.xml: remove toc from Acknowledgments
9619    
9620    There should be no toc for a simple preface with only one
9621    Acknowledgments section.
9622    
9623    Use <simplesect> markup rather than sect1.
9624    
9625    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9626
9627commit 77cc17ed0817c1df3ac5feb9dc83a5ba5d021a4b
9628Author: Gaetan Nadon <memsize@videotron.ca>
9629Date:   Thu Aug 4 19:59:37 2011 -0400
9630
9631    libX11 specs: move </para> above <varaiablelist>
9632    
9633    Many, but not all, function synopsis have a paragraph that
9634    nests the variable list. The code was generated by doclifter
9635    and there is no apprent reasons as to why it was done this way.
9636    Found while investigating a spacing issue.
9637    
9638    Before the patch:
9639    <para>Some blurb about the function
9640    <funcsynopsis>
9641    </funcsynopsis>
9642    <variablelist>
9643    </variablelist>
9644    </para>
9645    
9646    After the patch:
9647    <para>Some blurb about the function</para>
9648    <funcsynopsis>
9649    </funcsynopsis>
9650    <variablelist>
9651    </variablelist>
9652    
9653    There are no noticable differences when reading the doc
9654    other than than the removed few pixels of spacing between
9655    the function sysnopsis and the variable list block.
9656    
9657    In some cases, there are no "blurb about the fucntion"
9658    and the empty paragraph is removed.
9659    
9660    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9661
9662commit 18595f0d8f95ff1c12e7b9f73bd4b30dda5c7168
9663Author: Gaetan Nadon <memsize@videotron.ca>
9664Date:   Thu Jul 28 19:46:14 2011 -0400
9665
9666    compose specs: generate chunked html
9667    
9668    For large 600 page documents such as this one, chunked html
9669    provide faster browser load time and better navigation.
9670    
9671    Simply click on the locale of your choice in the toc and
9672    the browser loads just that one file. Being a DocBook,
9673    it benefits from all of the usual features and can be
9674    integrated with the rest of the documentation.
9675    
9676    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9677    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9678
9679commit 5cc8815b8a117cf0b6af37e7d55c5a420d9fff57
9680Author: Gaetan Nadon <memsize@videotron.ca>
9681Date:   Thu Jun 23 15:16:47 2011 -0400
9682
9683    specs: build compose keys tables in specs/i18n/compose
9684    
9685    Build the tables article together with the rest of the specs.
9686    Make it transparent that the source in generated in nls.
9687    Reuse docbook.am and get all the features such as olink.
9688    
9689    The docbook article file stem and id must be the same. The new name
9690    for the main article is libX11-keys. The new installation location
9691    is $docdir/libX11/i18n/compose.
9692    
9693    The nls dir retains the role of generating the DocBook/XML source
9694    but does not build neither installs output formats or generated source.
9695    
9696    The tables article now has to specify that each included locale
9697    section is a dependency. It did not matter before as they were web links.
9698    
9699    The xorg-sgml-doctools masterdb shall be updated to include this new doc.
9700    Install location moved from doc/libX11/Compose to doc/libX11/i18n/compose.
9701    
9702    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9703    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9704
9705commit adebbe3856b4aee39c0c7af902c1dea847dd6d7c
9706Author: Gaetan Nadon <memsize@videotron.ca>
9707Date:   Thu Jun 23 14:43:31 2011 -0400
9708
9709    nls: restructure charts as a single article with sections
9710    
9711    Looks more like a real article with a toc rather than individual
9712    web pages. Looks nicer in pdf.
9713    
9714    Each locale is a "section" rather than an "article".
9715    Using XInclude to aggregate xml source files gets you the toc for free.
9716    
9717    The single document is over 600 pages while there were 62 separate
9718    documents previously. FOP version 1.0 is required to handle missing
9719    character like capital sharp s.
9720    
9721    Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9722    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9723
9724commit fc74dc12b1ff3c43e240e1a713316ce1bf525d61
9725Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9726Date:   Thu Jul 28 19:32:40 2011 -0700
9727
9728    libX11 1.4.4
9729    
9730    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9731
9732commit ab1f30231755d99ed123af6873b981834640f01d
9733Author: Matt Dew <marcoz@osource.org>
9734Date:   Wed Jul 13 12:33:40 2011 -0600
9735
9736    Add id attributes to error codes to allow linking from other docs.
9737    
9738    Signed-off-by: Matt Dew <marcoz@osource.org>
9739    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9740    Acked-by: Gaetan Nadon <memsize@videotron.ca>
9741
9742commit 5c831fef402914ccf2ec14005c25be48852f119b
9743Author: David Coppa <dcoppa@gmail.com>
9744Date:   Tue Jul 12 10:05:47 2011 +0200
9745
9746    Fix libpthread linkage on OpenBSD.
9747    
9748    OpenBSD prefers to use the -pthread to fetch pthread libs when needed.
9749    
9750    Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr>
9751    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9752
9753commit 445b0b3efc04a74fc77cd3e483b25cfb5da82f12
9754Author: Gaetan Nadon <memsize@videotron.ca>
9755Date:   Wed Jun 22 20:46:26 2011 -0400
9756
9757    config: use XORG_WITH_PERL macro to replace custom program check
9758    
9759    New in version util-macros 1.15, the macro does the usual checking
9760    for program path and provides an interface for makefile variables.
9761    
9762    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9763    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9764
9765commit 47d99a324db9232029a22dd523e3bc3ed0936aca
9766Author: Gaetan Nadon <memsize@videotron.ca>
9767Date:   Wed Jun 22 20:41:18 2011 -0400
9768
9769    config: update XORG_WITH_FOP macro usage
9770    
9771    A new feature to test for a minimum version has been added.
9772    
9773    The parameter position changed. Due to limited usage
9774    and requiring 1.15, it turns out to be backward compatible.
9775    
9776    There is no functional change to how docs are build.
9777    
9778    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9779
9780commit e659683ad51ecc26c989fbdefdfe20f83af2c5a5
9781Author: Gaetan Nadon <memsize@videotron.ca>
9782Date:   Wed Jun 22 11:48:49 2011 -0400
9783
9784    Remove unused xmlrules.in in specs, now in /nls
9785    
9786    The specs dir has been converted to docbook.am
9787    
9788    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9789
9790commit 9992092609bd83a0db778e9a9ac1e879d445e637
9791Author: Gaetan Nadon <memsize@videotron.ca>
9792Date:   Thu Mar 31 10:23:22 2011 -0400
9793
9794    nls: move xmlrules.in from specs to nls
9795    
9796    It is no longer used in the specs dir.
9797    Remove html from CLEANFILES as it breaks. Make do not clean directories
9798    Remove SUFFIXES as pattern rules are used rather than suffix rules.
9799    
9800    Other improvements are possible, the first priority is to move it out
9801    of the specs dir.
9802    
9803    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
9804    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9805
9806commit b370eeb0191b0e2a8bde2f03a442a8ca62e35d58
9807Author: Gaetan Nadon <memsize@videotron.ca>
9808Date:   Sun Jun 5 16:27:36 2011 -0400
9809
9810    Install target dbs alongside generated documents
9811    
9812    This matches a change in xorg-sgml-docs whereby the masterdb will look for
9813    the target dbs into the same location as the generated documents.
9814    
9815    The target dbs are now installed alongside the generated documents.
9816    Previously they are installed in $prefix/sgml/X11/dbs alongside masterdb which
9817    has the potential of installing outside the package prefix and cause
9818    distcheck to fail when user does not have write permission in this package.
9819    
9820    Requires XORG_CHECK_SGML_DOCTOOLS(1.8) which was released 2011-06-11
9821
9822commit 4e741654b7e0a283779b8b4af5bb32e05b7fc678
9823Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9824Date:   Tue Jun 7 18:41:30 2011 -0700
9825
9826    specs/libX11: Turn appendix references into links
9827    
9828    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9829
9830commit a3d1152bc4b300f59ea2f08925739aab1ebfbc4e
9831Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9832Date:   Tue Jun 7 18:37:41 2011 -0700
9833
9834    specs/libX11: Turn chapter references into links
9835    
9836    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9837
9838commit d0cc949dfe44a41a71f19e4fe6c7547bb3d9bdbe
9839Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9840Date:   Tue Jun 7 00:52:32 2011 -0700
9841
9842    specs/libX11: Turn many "see section ..." into links
9843    
9844    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9845
9846commit 02d4c08dead2f266809b3f93ec72377783423ac7
9847Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9848Date:   Sun Jun 5 22:50:37 2011 -0700
9849
9850    specs/libX11: Convert some header filenames to filename tags
9851    
9852    perl -i -p -e 's{^&lt;(.*\.h)&gt;\ *}{<filename class="headerfile">&lt;\1&gt;</filename>}' *.xml
9853    
9854    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9855
9856commit c51a011329afa6e5b9d2b285349c132683ecf9fb
9857Author: Matt Dew <marcoz@osource.org>
9858Date:   Tue May 31 20:03:23 2011 -0600
9859
9860    Add id attributes to funcsynopsis to allow other docs to olink to them.
9861    
9862    Signed-off-by: Matt Dew <marcoz@osource.org>
9863    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9864    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
9865
9866commit 6841260c8bb15404a0b4805bee3b0bdfec7176b3
9867Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9868Date:   Fri May 27 20:14:36 2011 -0700
9869
9870    Bug 37624 - parameter x is missing from synopsis of XDrawString in libX11.html
9871    
9872    https://bugs.freedesktop.org/show_bug.cgi?id=37624
9873    
9874    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9875
9876commit 205af6a4e557c62a6395feadc1c89f4a9fe1713b
9877Author: Matt Dew <marcoz@osource.org>
9878Date:   Sat May 21 00:23:06 2011 -0600
9879
9880    Add id's to functions, to make clickable links.
9881    
9882    Signed-off-by: Matt Dew <marcoz@osource.org>
9883    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
9884
9885commit 5c06bc594473f6ab234724cd90db32e7b57fe811
9886Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9887Date:   Fri May 20 15:38:08 2011 -0700
9888
9889    libX11 spec: fix monospaced column alignment after deligaturization
9890    
9891    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9892
9893commit ef0019c714c273cb9b4ad3eba52d0b905109826b
9894Author: James Cloos <cloos@jhcloos.com>
9895Date:   Fri May 20 15:51:46 2011 -0400
9896
9897    Expand latin ligatures out to their NFKC equivalents.
9898    
9899    Ligatures should be done by the typesetting system.
9900    Manuscripts should not bother with them.
9901    
9902    Signed-off-by: James Cloos <cloos@jhcloos.com>
9903    Reviewed-by: Keith Packard <keithp@keithp.com>
9904
9905commit 761b8aa0c9b3c58c478ac5ea1b3aaafadcfc1325
9906Author: Daniel Stone <daniel@fooishbar.org>
9907Date:   Thu May 12 16:21:50 2011 +0200
9908
9909    XCB: Add more friendly error messages for common asserts
9910    
9911    This patch adds more friendly error messages for three common classes of
9912    assertion:
9913        - missed sequence numbers due to being griefed by another thread
9914        - unknown requests in queue due to being griefed by another thread
9915        - extensions dequeuing too much or too little reply data
9916    
9917    It adds error messages offering advice (e.g. call XInitThreads() first)
9918    on stderr, but still generates actual assertions.  Hopefully this means
9919    it's a little more Googleable and a little less frightening.
9920    
9921    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
9922    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9923
9924commit 159bf292477048b9a2f074735afc516f52c93d80
9925Author: Gaetan Nadon <memsize@videotron.ca>
9926Date:   Thu May 19 20:22:11 2011 -0400
9927
9928    Remove misplaced hyphens in libX11 DocBook/XML #37364
9929    
9930    Reported-by: Christopher Yeleighton <giecrilj@stegny.2a.pl>
9931    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9932    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9933    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
9934
9935commit 0ee6d8247d397500ae183ef180b6ff21bceefa1c
9936Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9937Date:   Tue May 17 20:49:59 2011 -0700
9938
9939    Fix man page and comment references to use XFreeModifiermap (lowercase map)
9940    
9941    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9942
9943commit 3275ec4ca8790d571b6a1902367ca5f68a2dc5fd
9944Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9945Date:   Tue May 17 20:30:54 2011 -0700
9946
9947    XKB man pages: Fix coordinantes typo in multiple pages
9948    
9949    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9950
9951commit ca33a80606cd68e4572fa49ed6aaa1d1fc31a47f
9952Author: Alan Coopersmith <alan.coopersmith@oracle.com>
9953Date:   Tue May 17 16:27:32 2011 -0700
9954
9955    DisplayOfCCC.man: Fix typo "ClientWhitePointOfCC" -> "ClientWhitePointOfCCC"
9956    
9957    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
9958
9959commit ae39d82b01359b07dd1fe092d867f79ea95a5438
9960Author: Jeremy Huddleston <jeremyhu@apple.com>
9961Date:   Sun May 8 09:08:07 2011 -0700
9962
9963    Silence clang static analysis warnings for SetReqLen
9964    
9965    This provides a simplified version of the SetReqLen macro when using clang for
9966    static analysis.  Prior to this change, we would see many Idempotent operation
9967    warnings inside this macro due to the common case of calling with arg2 and
9968    arg3 being the same variable.  This has no effect on code produced during
9969    compilation, but it silences a number of false positives in static analysis.
9970    
9971    XIPassiveGrab.c:170:5: warning: Assigned value is always the same as the existing value
9972        SetReqLen(req, num_modifiers, num_modifiers);
9973        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
9974    In file included from XIPassiveGrab.c:26:
9975    .../include/X11/Xlibint.h:580:8: note: instantiated from:
9976                n = badlen; \
9977                  ^
9978    
9979    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9980
9981commit 130af2bc02090ee2526adb2f9803fc07019b8f1f
9982Author: Jeremy Huddleston <jeremyhu@apple.com>
9983Date:   Wed May 4 15:34:31 2011 -0700
9984
9985    XKB: XkbComputeSectionBounds: Check correct bounds in default switch-case
9986    
9987    XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds')
9988            _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1);
9989                                   ^~~~~~~
9990    
9991    Found-by: clang static analyzer
9992    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
9993    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
9994
9995commit c987ef04bea345d5810e60cf6c53b747ddbe910e
9996Author: Jeremy Huddleston <jeremyhu@apple.com>
9997Date:   Wed May 4 15:31:17 2011 -0700
9998
9999    Revert "XKB: Avoid a possible NULL dereference"
10000    
10001    Sorry for the noise.  I accidentally pushed and didn't mean to. =(
10002    
10003    This reverts commit 4024091678ea07e0d898b798df9b29f3bf68eb08.
10004
10005commit 46d04bc4e7ff6c86385002c929d8eb0310f737a8
10006Author: Jeremy Huddleston <jeremyhu@apple.com>
10007Date:   Wed May 4 11:35:31 2011 -0700
10008
10009    Dead code removal
10010    
10011    XKBGeom.c:118:27: warning: Assigned value is always the same as the existing value
10012        for (key=row->keys,pos=k=0;k<row->num_keys;k++,key++) {
10013                           ~~~^~~~
10014    XKBGeom.c:115:5: warning: Value stored to 'pos' is never read
10015        pos= 0;
10016        ^    ~
10017    
10018    Found-by: clang static analyzer
10019    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10020
10021commit 4024091678ea07e0d898b798df9b29f3bf68eb08
10022Author: Jeremy Huddleston <jeremyhu@apple.com>
10023Date:   Wed May 4 11:34:53 2011 -0700
10024
10025    XKB: Avoid a possible NULL dereference
10026    
10027    XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds')
10028            _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1);
10029                                   ^~~~~~~
10030    
10031    Found-by: clang static analyzer
10032    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10033
10034commit 626e5e34be55b7cf734d745bd40a7ee3359029f8
10035Author: Jeremy Huddleston <jeremyhu@apple.com>
10036Date:   Wed May 4 11:21:41 2011 -0700
10037
10038    Annotate _XIOError as _X_NORETURN
10039    
10040    Found-by: clang static analyzer
10041    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10042
10043commit d3d7896408b435ab42656398839ff4351a37724d
10044Author: Jeremy Huddleston <jeremyhu@apple.com>
10045Date:   Tue May 3 09:32:53 2011 -0700
10046
10047    clang analyzer: Don't warn about Xmalloc(0)
10048    
10049    This will prevent a number of false positives in where clang's
10050    static analysis reports about calls to malloc(0).
10051    
10052    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10053
10054commit 393921cf2188b2b0713cc157effaf17d0abab783
10055Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10056Date:   Thu Apr 21 16:21:04 2011 -0700
10057
10058    Delete special case code to append "/sparcv9" to i18n module path
10059    
10060    Was triggered by defined(__sparcv9) so only built on Solaris SPARC 64-bit.
10061    Inconsistent with all other platforms, and a bit overcomplicated.
10062    
10063    Should anyone need to continue using that path, simply build with
10064    a #define POSTLOCALELIBDIR "sparcv9" to get the same result.
10065    
10066    Fixes Solaris bug 7038737:
10067     sparcv9 Xlib looking in wrong path for i18n loadable modules
10068    
10069    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10070    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
10071
10072commit b9ff9cfd5c56ae6d52c6503f7019756d9de29557
10073Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10074Date:   Thu Apr 14 13:47:12 2011 -0700
10075
10076    specs/libX11: Fix XOpenDisplay() prototype in chapter 2 [bug 36244]
10077    
10078    Was incorrectly showing AllPlanes() instead when describing XOpenDisplay()
10079    
10080    https://bugs.freedesktop.org/show_bug.cgi?id=36244
10081    
10082    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10083
10084commit b850adbdebcf500c659f85285d4d7374e15857f5
10085Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10086Date:   Tue Apr 12 22:30:45 2011 -0700
10087
10088    Convert malloc(strlen()); strcpy() sets to strdup
10089    
10090    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10091    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
10092
10093commit 09194042d3dc44a463add1f7c122a68ffd5ef0bf
10094Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10095Date:   Tue Apr 12 21:27:45 2011 -0700
10096
10097    Replace Xmalloc+bzero pairs with Xcalloc calls
10098    
10099    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10100    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
10101
10102commit ac1e2bff7121987fd768500a11d428d9fb9447c5
10103Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10104Date:   Fri Mar 25 14:47:35 2011 +0200
10105
10106    om: Fix memory leaks on get_font_name error paths.
10107    
10108    While at it, remove unneeded check for NULL before Xfree.
10109    
10110    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10111    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10112
10113commit 7c362e275c93c92b4e68fe862e73ee36665de703
10114Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10115Date:   Wed Mar 30 15:06:10 2011 +0300
10116
10117    xcms: Fix memory leaks on LINEAR_RGB_InitSCCData error path.
10118    
10119    pScreenData is replaced when building per visual intensity tables. If
10120    malloc failed the old value of pScreenData (stored also in
10121    pScreenDefaultData) was being leaked. Also, property_return wasn't
10122    free'd in that case.
10123    
10124    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10125    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10126
10127commit 5c810e2ac233e00f361549bafb9b59e8a9e05eff
10128Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10129Date:   Wed Mar 30 15:19:28 2011 +0300
10130
10131    xcms: Fix error on LINEAR_RGB_InitSCCData error path.
10132    
10133    Due to what looks like a copy & paste error, pScreenData->pBlueTbl would
10134    be accessed after being free'd.
10135    
10136    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10137    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10138
10139commit d749948f9492fd9b61c74655a08e32c595e0e3a5
10140Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10141Date:   Fri Mar 25 16:06:15 2011 +0200
10142
10143    om: Fix potential memory leak in init_om.
10144    
10145    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10146    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10147
10148commit d0749d6abdf0fd4d8b4e59b02dad8ccda3f10995
10149Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10150Date:   Thu Mar 24 20:40:41 2011 +0200
10151
10152    om: Fix memory leak on read_EncodingInfo error path.
10153    
10154    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10155    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10156
10157commit 4b2e8d00f5b6969c14003ee8eb258b9f0e4dd7c3
10158Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10159Date:   Thu Mar 24 19:55:33 2011 +0200
10160
10161    Fix memory leaks on _XimCbDispatch error path.
10162    
10163    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10164    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10165
10166commit 46e6c78b1a89e4774e0f7e0f4d6d0fd060c3897e
10167Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10168Date:   Thu Mar 24 19:52:05 2011 +0200
10169
10170    Fix memory leak on _XimCommitRecv error path.
10171    
10172    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10173    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10174
10175commit 0ace642a2d47265f01450bfa2b2fd48eb6956a83
10176Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10177Date:   Thu Mar 24 19:48:44 2011 +0200
10178
10179    Fix memory leaks on _XimWriteCachedDefaultTree error paths.
10180    
10181    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10182    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10183
10184commit e29be94edbb58b3b8dab545377a710d1f73b61e0
10185Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10186Date:   Thu Mar 24 19:42:36 2011 +0200
10187
10188    Fix memory leaks on _XimGetAttributeID error paths.
10189    
10190    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10191    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10192
10193commit d144a50512466766f55ce61e3884925334b08f0d
10194Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10195Date:   Thu Mar 24 19:36:56 2011 +0200
10196
10197    Fix memory leaks on _XimProtoCreateIC error paths.
10198    
10199    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10200    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10201
10202commit 6a452f7a98499508f753cb8a7c3f08bcbec736b9
10203Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10204Date:   Thu Mar 24 16:24:29 2011 +0200
10205
10206    Fix leaks in _XimEncodingNegotiation error paths.
10207    
10208    name_ptr and detail_ptr weren't free'd in some cases before returning
10209    False.
10210    
10211    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10212    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10213
10214commit 396e5a452a59c1f121220ba72167b720a863b30f
10215Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10216Date:   Thu Mar 24 14:17:44 2011 +0200
10217
10218    FSWrap: fix potential leak in copy_string_list
10219    
10220    If list_count is 0, dst would be allocated and leaked.
10221    
10222    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10223    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10224
10225commit 1a944260182bb552b954d69f6355c2760d4415df
10226Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10227Date:   Wed Mar 23 14:36:22 2011 +0200
10228
10229    Fix memory leak with broken bitmap files in XReadBitmapFileData
10230    
10231    Bitmap file data is read looping through the lines in the input file. If
10232    there is extra data after the bitmap, these lines will be processed and
10233    if this data represents another bitmap it will replace the one read
10234    before causing the memory allocated for bits to leak.
10235    
10236    This changes the code to stop processing the file once a bitmap was
10237    read.
10238    
10239    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10240    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10241
10242commit c1c91e9a221a0e762d96cc12a3f189d6e59f4865
10243Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10244Date:   Tue Apr 12 20:50:42 2011 -0700
10245
10246    Fix "attrinute" typo in comments in ximcp
10247    
10248    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10249
10250commit 7103b83a88edb9c93acfa68f3b556595b087baff
10251Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10252Date:   Fri Apr 8 21:36:11 2011 -0700
10253
10254    Make doc install subdirectories more like the pre-docbook.am paths
10255    
10256    Haven't restored the "specs" path in the install path, but have restored
10257    libX11 & i18n subdirectories in the path for better logical grouping.
10258    
10259    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10260    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10261
10262commit 9a1ae7a90f815a737a942e42d5eb1ffff400977b
10263Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10264Date:   Fri Apr 8 13:24:28 2011 -0700
10265
10266    Install xml versions of specs even if HAVE_XMLTO is false
10267    
10268    Moves HAVE_XMLTO check into docbook.am, more closely matches behaviour
10269    from before docbook.am changes (commit e8c76407d2f6e3)
10270    
10271    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10272    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10273
10274commit a0ad0d5c99023bb9a8ce3944dbc3267f5265721e
10275Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10276Date:   Tue Apr 5 13:29:04 2011 -0700
10277
10278    libX11 1.4.3
10279    
10280    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10281
10282commit e8c76407d2f6e3b11babdb84426e43e780c859c3
10283Author: Gaetan Nadon <memsize@videotron.ca>
10284Date:   Fri Feb 25 09:23:54 2011 -0500
10285
10286    Documentation: add Docbook external references support
10287    
10288    When writing technical documentation, it is often necessary to cross
10289    reference to other information. When that other information is not in the
10290    current document, additional support is needed, namely <olink>.
10291    
10292    A new feature with version 1.7 of xorg-sgml-doctools adds references to
10293    other documents within or outside this package.
10294    
10295    This patch adds technical support for this feature but does not change
10296    the content of the documentation as seen by the end user.
10297    
10298    Each book or article must generate a database containing the href
10299    of sections that can be referred to from another document. This database
10300    is installed in DATAROOTDIR/sgml/X11/dbs. There is a requirement that
10301    the value of DATAROOTDIR for xorg-sgml-doctools and for the package
10302    documentation is the same. This forms a virtual document tree.
10303    
10304    This database is consulted by other documents while they are being generated
10305    in order to fulfill the missing information for linking.
10306    Refer to the xorg-sgml-doctools for further technical information.
10307    
10308    Co-authored-by: Matt Dew <marcoz@osource.org>
10309    
10310    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10311
10312commit 25eb76b3d2726f2f1de759901971ae53b2539dc4
10313Author: Harshula Jayasuriya <harshula@gmail.com>
10314Date:   Mon Mar 21 14:49:37 2011 +0000
10315
10316    Add #define XK_SINHALA
10317    
10318    Add #define XK_SINHALA so that the Sinhala keysyms can be used by
10319    the lk xkb keymap.
10320    
10321    Signed-off-by: Harshula Jayasuriya <harshula@gmail.com>
10322    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
10323
10324commit db8b20b789112717ac0590b40f0b4dc2171797d0
10325Author: Jeremy Huddleston <jeremyhu@apple.com>
10326Date:   Thu Mar 17 16:15:00 2011 -0700
10327
10328    configure.ac: Bump version to 1.4.2
10329    
10330    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
10331
10332commit 83e1ba59c48c79f8b0a7e7aa0b9c9cfd84fa403d
10333Author: Jamey Sharp <jamey@minilop.net>
10334Date:   Tue Mar 15 16:48:07 2011 -0700
10335
10336    Call _XErrorFunction without holding the Display lock.
10337    
10338    Historically, Xlib dropped the Display lock around the upcall to any
10339    user-supplied _XErrorFunction, but somewhere along the way I quit doing
10340    that if you built with XCB. The reasons are lost somewhere in the
10341    pre-git history of Xlib/XCB, and I can't now see any reason to hold the
10342    lock.
10343    
10344    The documentation for XSetErrorHandler still applies though:
10345    
10346        Because this condition is not assumed to be fatal, it is acceptable
10347        for your error handler to return; the returned value is ignored.
10348        However, the error handler should not call any functions (directly
10349        or indirectly) on the display that will generate protocol requests
10350        or that will look for input events.
10351    
10352    So while you are now once again permitted to re-enter Xlib from the
10353    error handler, you're only allowed to call non-protocol functions.
10354    
10355    Signed-off-by: Jamey Sharp <jamey@minilop.net>
10356
10357commit fd85aca7a616c595fc17b2520f84316a11e8906f
10358Author: Jamey Sharp <jamey@minilop.net>
10359Date:   Mon Mar 14 14:45:35 2011 -0700
10360
10361    Ignore user locks after sleeping in _XReply and _XReadEvents.
10362    
10363    This bug appears as a hang in applications that wait for replies from
10364    multiple threads, where one such thread has taken a user lock using
10365    XLockDisplay.
10366    
10367    Prior to this fix, the code could deadlock in this way: If thread 1 goes
10368    to sleep waiting for a reply, and then thread 2 takes a user lock and
10369    waits for a reply, then thread 2 will wait for thread 1 to process its
10370    reply (because responses must be processed in order), but thread 1 will
10371    wait for thread 2 to drop its user lock.
10372    
10373    Fixed by making thread 1 not wait for thread 2 to drop its user lock.
10374    This makes the semantics of user locks hard to define, but they were
10375    already hard to define. The new behavior appears to be consistent with
10376    the way Xlib worked historically, anyway.
10377    
10378    Fixes: http://lists.freedesktop.org/archives/xcb/2011-March/006802.html
10379    
10380    There was a similar potential for deadlock in _XReadEvents, fixed the
10381    same way, with the same caveats about user-lock semantics.
10382    
10383    Signed-off-by: Jamey Sharp <jamey@minilop.net>
10384
10385commit 690f8bffd48a4e7e74298360ddd0431dc95dcd3f
10386Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10387Date:   Tue Jan 18 15:58:20 2011 +0200
10388
10389    xkb: XkbPropertyPtr determined allocation success from wrong variables
10390    
10391    Cannot reach dead statement "return NULL;"
10392    
10393    Check for the NULLness of prop->name and prop->value instead of
10394    name and value, which was checked earlier anyway. Decided against
10395    using strdup due to curious memory allocation functions and the
10396    rest of the xkb not using it either.
10397    
10398    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10399    Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>
10400
10401commit 6a4d027284e7bb5dd458157947bbb1ff580ad071
10402Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10403Date:   Mon Jan 10 16:37:22 2011 +0200
10404
10405    keyBind: Use Xcalloc to initialize allocated _XKeytrans
10406    
10407    Using uninitialized value "p->modifiers"
10408    
10409    Small fix by using Xcalloc instead of Xmalloc
10410    
10411    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10412    Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>
10413
10414commit b993d73bb3214ecc24646f5427c8003b816c6921
10415Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10416Date:   Mon Jan 10 16:22:45 2011 +0200
10417
10418    im/ximcp: release modifiermap before returning
10419    
10420    Variable "map" goes out of scope
10421    
10422    Release modifiermap before returning. Reordered code to call
10423    XGetModifierMapping after the first return from the function.
10424    
10425    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10426    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
10427
10428commit 807a7fc0354f2212dfa5ff1f9e4ede56d8e69ef4
10429Author: Gaetan Nadon <memsize@videotron.ca>
10430Date:   Fri Feb 25 09:23:54 2011 -0500
10431
10432    Docbook: change the book id to match the xml file basename
10433    
10434    This is required for the up-coming external references support.
10435    
10436    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10437
10438commit 40812b53ff5fe548f6eaf43ba4c8781cb43dab43
10439Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
10440Date:   Mon Feb 21 21:54:17 2011 +0100
10441
10442    Make the Local XIM request key releases for braille
10443    
10444    Braille chords management needs key release events. We need to explicitly
10445    request then, else GTK would not pass them throught XFilterEvent and braille
10446    wouldn't work.
10447    
10448    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
10449
10450commit c97c42c49cd5095462abecdf908b416fb0b540b6
10451Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
10452Date:   Mon Feb 21 17:27:38 2011 +0100
10453
10454    Match braille patterns with compose tree
10455    
10456    Braille patterns should also be usable in Compose.  This combines the
10457    implementation of braille chords and compose tree: only emit the braille
10458    pattern if it can not be found in the compose tree, if any.
10459    
10460    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
10461
10462commit 0c6ca565d7c8a47ef3ea823569a9ca5298a5307d
10463Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
10464Date:   Mon Feb 21 15:56:54 2011 +0100
10465
10466    Fix status reporting for braille patterns
10467    
10468    _XimLocalMbLookupString can return a braille keysym even if _Xlcwctomb can't
10469    convert to the current MB charset.
10470    _XimLocalUtf8LookupString needs to set the braille keysym and status too.
10471    
10472    Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
10473
10474commit 993abe751f4141f54d8d28b8b73588a1c9085970
10475Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10476Date:   Fri Feb 11 14:49:17 2011 -0800
10477
10478    Clean up memory when first XCreateRegion succeeds and second fails
10479    
10480    Error: Memory leak (CWE 401)
10481       Memory leak of pointer 's' allocated with XCreateRegion()
10482            at line 387 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XShrinkRegion'.
10483              's' allocated at line 387 with XCreateRegion().
10484              s leaks when s != 0 at line 387.
10485    Error: Memory leak (CWE 401)
10486       Memory leak of pointer 'tra' allocated with XCreateRegion()
10487            at line 1452 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XXorRegion'.
10488              'tra' allocated at line 1451 with XCreateRegion().
10489              tra leaks when tra != 0 at line 1451.
10490    
10491    [ This bug was found by the Parfait 0.3.6 bug checking tool.
10492      For more information see http://labs.oracle.com/projects/parfait/ ]
10493    
10494    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10495
10496commit 6ac417cea1136a3617f5e40f4b106aaa3f48d6c2
10497Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10498Date:   Fri Feb 11 14:20:24 2011 -0800
10499
10500    ximcp: Prevent memory leak & double free if multiple %L in string
10501    
10502    In the highly unlikely event that TransFileName was passed a path
10503    containing multiple %L entries, for each entry it would call
10504    _XlcFileName, leaking the previous results, and then for each entry it
10505    would copy from that pointer and free it, resulting in invalid pointers
10506    & possible double frees for each use after the first one freed it.
10507    
10508    Error: Use after free (CWE 416)
10509       Use after free of pointer 'lcCompose'
10510            at line 358 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
10511              Previously freed at line 360 with free.
10512    Error: Use after free (CWE 416)
10513       Use after free of pointer 'lcCompose'
10514            at line 359 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
10515              Previously freed at line 360 with free.
10516    Error: Double free (CWE 415)
10517       Double free of pointer 'lcCompose'
10518            at line 360 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'.
10519              Previously freed at line 360 with free.
10520    
10521    [ This bug was found by the Parfait 0.3.6 bug checking tool.
10522      For more information see http://labs.oracle.com/projects/parfait/ ]
10523    
10524    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10525
10526commit 502d414118c97d35a44f8e295709682022876331
10527Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10528Date:   Thu Feb 3 17:08:57 2011 +0200
10529
10530    xcms/cmsProp: don't deal with uninitialized values, fail instead
10531    
10532    Properly handle the return value of XGetWindowProperty by considering
10533    if after the loop as well.
10534    
10535    Using freed pointer "prop_ret"
10536    
10537    There were numerous things wrong in how this function interacted with
10538    XGetWindowProperty.
10539    
10540    None of the local variables were initialized and remained that way if
10541    the call to XGetWindowProperty returned 1 (not Succeed). That doesn't
10542    result in after_ret being initialized in which case if it happens to
10543    be 0, the loop was exited. In that case format_ret and nitems_ret were
10544    uninitialized and the function might return with success (but with
10545    uninitialized pointer in prop_ret) or XcmsFailure.
10546    
10547    As the buffer enlarging code was called only when XGetWindowProperty
10548    failed (returned not Success), after_ret would not have been
10549    initialized. It would have been initialized only if the
10550    XGetWindowProperty has returned Success earlier, but in that case the
10551    code fragment would not have been reached.
10552    
10553    This patch alters the function to return XcmsFailure if the call to
10554    XGetWindowProperty fails.
10555    
10556    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10557    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10558    Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
10559    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10560
10561commit c37e278993b9e5b3d7025ef4c434373a011996ec
10562Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10563Date:   Mon Jan 31 14:02:07 2011 +0200
10564
10565    xcms/LRGB: don't double-free property_return
10566    
10567    property_return was free'd before and in the case the conditional is true,
10568    the call to XcmsGetProperty failed which means that property_return wasn't
10569    set so there is no need to free it again.
10570    
10571    Double free of pointer "property_return" in call to "free"
10572    
10573    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10574    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10575    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10576
10577commit 50f4107811249806718a100f9d34f996c58e5e25
10578Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10579Date:   Wed Feb 2 08:58:45 2011 -0800
10580
10581    Xrm.c: ReadInFile: refactor fstat error handling
10582    
10583    We can simplify the fstat failure case now that the GetFileSize macro
10584    has been expanded inline.
10585    
10586    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10587    Reviewed-by: Julien Cristau <jcristau@debian.org>
10588
10589commit 5e9c40fcb5da43c9fdacf12967d090bf202daf2a
10590Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10591Date:   Wed Feb 2 08:56:00 2011 -0800
10592
10593    Expand GetSizeOfFile() macro at the one place it's called
10594    
10595    Removes XrmI.h header that only contained this single macro
10596    
10597    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10598    Reviewed-by: Julien Cristau <jcristau@debian.org>
10599
10600commit 450e17422c0e374d25c643f343ea268cec68da38
10601Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10602Date:   Mon Jan 31 14:01:57 2011 +0200
10603
10604    XlibInt: Use strncpy+zero termination instead of strcpy to enforce buffer size
10605    
10606    Possible overrun of 8192 byte fixed size buffer "buffer" by copying
10607    "ext->name" without length checking
10608    
10609    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10610    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10611    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10612    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10613
10614commit e2566e43b02d2d7b7c1c3bb7db7c5ae81c1245fa
10615Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10616Date:   Mon Jan 31 14:02:13 2011 +0200
10617
10618    lc/def/lcDefConv: Use Xcalloc to avoid use of uninitialized memory
10619    
10620    Fixed by zero'ing conv on allocation with Xcalloc. Then
10621    close_converter works properly.
10622    
10623    Using uninitialized value "conv->state" in call to function "close_converter"
10624    
10625    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10626    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10627    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10628    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10629
10630commit 03f81ad8fb9783986cf9b17661dd31e95c396681
10631Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10632Date:   Mon Jan 31 14:02:16 2011 +0200
10633
10634    xlibi18n/lcFile: Removed superfluous check for NULL target_dir
10635    
10636    The situation is already handled before this code.
10637    
10638    Cannot reach dead expression "0U" inside statement "if (1U + (target_dir ? strl..."
10639    
10640    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10641    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10642    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10643    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10644
10645commit 1346b9ea3b3882201ff8c3ee462ff4b0d4edf639
10646Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10647Date:   Mon Jan 31 14:02:15 2011 +0200
10648
10649    ximcp/imLckup: Handle negative return value from _Xlcwctomb
10650    
10651    Fixed by negative value to memcpy by checking for the negative return
10652    value of _Xlcwctomb and returning 0/XLookupNone in that case.
10653    
10654    a negative value was passed to memcpy
10655    
10656    Unfortunately the other return values for *status don't fit into the
10657    error (which appears to indicate some internal error or running out of
10658    memory). The other valid status codes are XBufferOverflow,
10659    XLookupNone, XLookupChars, XLookupKeySym, and XLookupBoth. Each of
10660    these has a specific meaning attached.
10661    
10662    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10663    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10664    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10665
10666commit 79a5c86e020f08cc108184298a72e2777036de39
10667Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10668Date:   Mon Jan 31 14:02:14 2011 +0200
10669
10670    ximcp/imTrX: Handle failing XGetWindowProperty
10671    
10672    Checked return value of XGetWindowProperty and return false if it fails.
10673    
10674    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
10675    
10676    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10677    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10678    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10679
10680commit 2ace8d5c89c8f6d9f42b4068f4b508ca28f0ced1
10681Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10682Date:   Mon Jan 31 14:02:12 2011 +0200
10683
10684    XlibInt: info_list->watch_data was reallocated, but result was discarded
10685    
10686    info_list->watch_data was being reallocated, but the return value of
10687    the reallocation was stored only into a local variable. This might
10688    cause some funky behavior and crashes.
10689    
10690    Variable "wd_array" goes out of scope
10691    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)))"
10692    
10693    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10694    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10695    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10696    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10697
10698commit 2b16a7e683e355c9746290b2cee2fd0dd2bf342a
10699Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10700Date:   Mon Jan 31 14:02:10 2011 +0200
10701
10702    GetProp: Zero-initialized error so its resourceID field is initialized
10703    
10704    Using uninitialized value "error.resourceID" in call to function "_XError"
10705    
10706    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10707    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10708    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10709    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10710
10711commit 85e9f38e016137f0ff2791eb0d092ab027382d2c
10712Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10713Date:   Tue Feb 1 11:07:25 2011 +0200
10714
10715    xcms/LRGB: Add a label for freeing property_return.
10716    
10717    The rest of the code uses goto's to free memory allocated later
10718    and prevent memory leaks, but there were several paths were
10719    property_return was free'd just before a goto.
10720    
10721    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10722    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10723    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10724    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10725
10726commit 3161dc57d4e9b70f852f05e5e474455e121b06ab
10727Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10728Date:   Mon Jan 31 14:02:08 2011 +0200
10729
10730    xcms/LRGB: Fix potential resource leak.
10731    
10732    property_return was not free'd if the allocation of pRedTbl failed.
10733    
10734    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10735    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10736    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10737
10738commit be3e6c205d94dedc1cdebf5d17b987f0f828377a
10739Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10740Date:   Mon Jan 31 14:02:06 2011 +0200
10741
10742    Xrm: Handle the extremely unlikely situation of fstat failing
10743    
10744    Tracked variable "size" was passed to a negative sink.
10745    
10746    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10747    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10748    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10749    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10750
10751commit 3fae16c64d6ef76fd4a25a54c7f7de76596457db
10752Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10753Date:   Mon Jan 31 14:02:05 2011 +0200
10754
10755    Xrm: NEWTABLE had a memory leak after a memory allocation error
10756    
10757    The NEWTABLE macro missed freeing its allocated memory on subsequent
10758    memory allocation errors. Added call to Xfree.
10759    
10760    Variable "table" goes out of scope
10761    
10762    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10763    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10764    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10765    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10766
10767commit 7110ac653349a23c80c365f11f6270dc27c8975a
10768Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10769Date:   Mon Jan 31 14:02:03 2011 +0200
10770
10771    ImUtil: Handle a memory leak in one early return branch
10772    
10773    Fixed memory leak by adding Xfree for image
10774    
10775    Variable "image" goes out of scope
10776    
10777    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10778    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10779    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10780    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10781
10782commit d695f5da9f5b778e54b6987d9177048e32818c4d
10783Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10784Date:   Mon Jan 31 14:02:01 2011 +0200
10785
10786    ximcp/imRm: Handle leaking colormap_ret
10787    
10788    Fixed memory leak by adding Xfree for colormap_ret
10789    
10790    Variable "colormap_ret" goes out of scope
10791    
10792    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10793    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10794    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10795    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10796
10797commit f9eda19d08c1ea0973dfe0bc10a2519d6fd26cc7
10798Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10799Date:   Mon Jan 31 14:01:59 2011 +0200
10800
10801    ximcp/imRmAttr: Handle leaking missing_list
10802    
10803    Fixed memory leak by adding Xfree and initializing missing_list with NULL
10804    
10805    Variable "missing_list" goes out of scope
10806    
10807    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10808    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10809    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10810    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10811
10812commit 3183269e0979c9dbce9d55d9e03937897dc9fb3b
10813Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10814Date:   Mon Jan 31 14:01:58 2011 +0200
10815
10816    ximcp/imRmAttr: Handle leaking colormap_ret
10817    
10818    XFree colormap_ret and initialize it when appropriate.
10819    
10820    Variable "colormap_ret" goes out of scope
10821    
10822    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10823    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10824    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10825
10826commit f66a032a937030f2d9baa81744d36dc585bb085c
10827Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10828Date:   Mon Jan 31 14:01:56 2011 +0200
10829
10830    xlibi18n/lcGeneric: Initialize uninitialized local variable
10831    
10832    Using uninitialized value "new"
10833    
10834    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10835    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10836    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10837    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10838
10839commit d0266e06d38110ec908ca28379014eff743630b7
10840Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10841Date:   Mon Jan 31 14:01:54 2011 +0200
10842
10843    xcmx/cmxColNm: Removed unused assignments to pBuf (in two functions)
10844    
10845    Pointer "pBuf" returned from "fgets(buf, 256, stream)" is never used
10846    
10847    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10848    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10849    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10850    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10851
10852commit cc686655d7bfdeab8b67e01a24bd452a2e9e3fcf
10853Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10854Date:   Tue Feb 1 12:30:56 2011 +0200
10855
10856    XDefaultOMIF.c: Fix memory leaks in get_font_name
10857    
10858    Instead of copying the value returned by get_prop_name and then releasing it,
10859    directly use the return value of get_prop_name, which allocates memory for the
10860    name.
10861    
10862    If get_prop_name returns NULL, continue on to XFreeFont to release the font
10863    before returning the NULL via the normal function return.
10864    
10865    Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10866    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10867
10868commit 95796351435d81eaef8166d49ba3a6e7b633d70a
10869Author: Gaetan Nadon <memsize@videotron.ca>
10870Date:   Mon Jan 31 14:50:19 2011 -0500
10871
10872    config: comment, minor upgrade, quote and layout configure.ac
10873    
10874    Group statements per section as per Autoconf standard layout
10875    Quote statements where appropriate.
10876    Autoconf recommends not using dnl instead of # for comments
10877    
10878    Use AC_CONFIG_FILES to replace the deprecated AC_OUTPUT with parameters.
10879    Remove redundant AC_CANONICAL_HOST included in XORG_DEFAULT_OPTIONS
10880    
10881    This helps automated maintenance and release activities.
10882    Details can be found in http://wiki.x.org/wiki/NewModuleGuidelines
10883    
10884    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10885
10886commit e994ab227ed28655565c3193e63733630105e7f9
10887Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10888Date:   Tue Jan 18 12:49:48 2011 +0200
10889
10890    Comparing array against NULL is not useful "&xkb->server->vmods != NULL"
10891    
10892    Removed superfluous comparison.
10893    
10894    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
10895    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10896
10897commit 4115f051e62f9b098efce691e070d44f09f30f1c
10898Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10899Date:   Wed Jan 12 15:51:11 2011 +0200
10900
10901    Variable "entry" tracked as NULL was dereferenced.
10902    
10903    Check entry for non-nullness before dereferencing it
10904    
10905    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
10906    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10907    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10908
10909commit 0edb76d1d5fdca5a2543332699be2e72386dab24
10910Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10911Date:   Thu Jan 27 09:54:00 2011 +0200
10912
10913    Dereferencing possibly NULL "str" in call to function "memcpy" (Deref assumed on the basis of 'nonnull' parameter attribute.)
10914    
10915    If _XkbGetReadBufferPtr returns NULL, goto BAILOUT
10916    
10917    Reviewed-by: Dirk Wallenstein <halsmit@t-online.de>
10918    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10919    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10920
10921commit 59da8a211ef723909d0530c0331d541db8e63378
10922Author: Erkki Seppälä < <erkki.seppala@vincit.fi>
10923Date:   Mon Jan 10 16:17:47 2011 +0200
10924
10925    Using freed pointer "e"
10926    
10927    Reordered code to first to do the comparison and then to release data
10928    
10929    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10930    Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com>
10931    Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
10932
10933commit 0f11c229f7099f7c5aeed4691b358dca151dac7d
10934Author: Matt Dew <marcoz@osource.org>
10935Date:   Thu Jan 27 00:31:23 2011 -0700
10936
10937    Remove <literal> tags from compose-chart.pl so pdf building doesn't die on soft-hyphens.
10938    
10939    Signed-off-by: Matt Dew <marcoz@osource.org>
10940    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10941
10942commit ca00bb202f3afcdbbcb8c4cb50308c5dd03f2322
10943Author: Gaetan Nadon <memsize@videotron.ca>
10944Date:   Fri Jan 28 16:07:07 2011 -0500
10945
10946    config: replace deprecated AC_HELP_STRING with AS_HELP_STRING
10947    
10948    This silences an Automake warning.
10949    
10950    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
10951
10952commit 8dac08c7515cb6f8a8f4d0a721713697eae755d8
10953Author: Gaetan Nadon <memsize@videotron.ca>
10954Date:   Fri Jan 28 14:59:04 2011 -0500
10955
10956    config: remove unrequired AC_HEADER_STDC
10957    
10958    Autoconf says:
10959    "This macro is obsolescent, as current systems have conforming
10960    header files. New programs need not use this macro".
10961
10962commit d4483375e37a72631d5821413cb75a423e990ffe
10963Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10964Date:   Tue Jan 18 19:37:02 2011 -0800
10965
10966    config: Use correct AC_CONFIG_HEADERS macro
10967    
10968    Replaces obsolete AM_CONFIG_HEADER and undocumented AC_CONFIG_HEADER
10969    
10970    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10971    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10972    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
10973
10974commit c6405fe931f313a0d159bdad00244fc8aae12468
10975Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10976Date:   Sun Jan 16 11:55:23 2011 -0800
10977
10978    config: remove AC_PROG_CC as it overrides AC_PROG_C_C99
10979    
10980    XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls
10981    AC_PROG_C_C99. This sets gcc with -std=gnu99.
10982    If AC_PROG_CC macro is called afterwards, it resets CC to gcc.
10983    
10984    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
10985    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
10986
10987commit 17aa1f2305da0952ad594c7329d08aba21f13ac8
10988Author: Alan Coopersmith <alan.coopersmith@oracle.com>
10989Date:   Sun Jan 16 11:40:59 2011 -0800
10990
10991    Update to xf86bigfontproto >= 1.2.0 header name
10992    
10993    Clears compile-time warning of:
10994    "X11/extensions/xf86bigfstr.h", line 1: #warning: "xf86bigfstr.h is obsolete and may be removed in the future."
10995    "X11/extensions/xf86bigfstr.h", line 2: #warning: "include <X11/extensions/xf86bigfproto.h> for the protocol defines."
10996    
10997    Requires xf86bigfontproto >= 1.2.0 if --disable-xf86bigfont is not
10998    passed to configure.
10999    
11000    Also removes unnecessary AC_SUBST of BIGFONT_CFLAGS & BIGFONT_LIBS
11001    that PKG_CHECK_MODULES does automatically
11002    
11003    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11004    Reviewed-by: Julien Cristau <jcristau@debian.org>
11005
11006commit a9228fcc676aacf9a760dd94891c89f9bc82b20d
11007Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11008Date:   Tue Jan 11 17:55:22 2011 -0800
11009
11010    libX11 1.4.1
11011    
11012    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11013
11014commit 92fa96451af37a7a0f2592de07643fdaed7a9efd
11015Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
11016Date:   Mon Jan 3 12:25:28 2011 -0500
11017
11018    Initialize event type
11019    
11020    If we receive unsupported event closing connection triggers valgrind
11021    error.
11022    
11023    ==12017== Conditional jump or move depends on uninitialised value(s)
11024    ==12017==    at 0x487D454: _XFreeDisplayStructure (OpenDis.c:607)
11025    ==12017==    by 0x486857B: XCloseDisplay (ClDisplay.c:72)
11026    *snip*
11027    ==12017==  Uninitialised value was created by a heap allocation
11028    ==12017==    at 0x4834C48: malloc (vg_replace_malloc.c:236)
11029    ==12017==    by 0x4894147: _XEnq (XlibInt.c:877)
11030    ==12017==    by 0x4891BF3: handle_response (xcb_io.c:335)
11031    ==12017==    by 0x4892263: _XReply (xcb_io.c:626)
11032    *snip*
11033    
11034    Problem is that XFreeDisplaySturture is checking for qelt->event.type ==
11035    GenericEvent while _XUnknownWireEvent doesn't store the type.
11036    
11037    Reviewed-by: Adam Jackson <ajax@redhat.com>
11038    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11039    Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
11040
11041commit fa9747b9d4443eef9b1687b8b1208801f5c585da
11042Author: Fernando Carrijo <fcarrijo.lists@gmail.com>
11043Date:   Tue Dec 28 16:18:57 2010 -0500
11044
11045    specs: convert images from gif to svg format.
11046    
11047    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11048
11049commit 0012e2a4eaddc0840617b4758931ad976ca2eb7c
11050Author: Philipp Reh <sefi@s-e-f-i.de>
11051Date:   Wed Dec 22 15:14:05 2010 +0000
11052
11053    Events: Store event cookie when dequeuing event
11054    
11055    When we dequeue an event in XCheckTypedEvent or XCheckTypedWindowEvent,
11056    make sure to store the corresponding cookie too.
11057    
11058    Signed-off-by: Philipp Reh <sefi@s-e-f-i.de>
11059    Reviewed-by: Daniel Stone <daniel@fooishbar.org>
11060
11061commit d2714d65e85b44abedf5f82e1a31506dba397ef2
11062Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11063Date:   Tue Dec 21 18:47:16 2010 -0800
11064
11065    Revert "Mark atom names argument to XInternAtoms as const"
11066    
11067    This reverts commit c8701115462b482d99ecff24d9de0f2806084ba5.
11068    
11069    The constification of a pointer to a pointer caused unexpected issues,
11070    and xorg-devel was unable to come up with a clean, safe, reasonable way
11071    to handle them, so we're chalking this up for now as yet another mistake
11072    in the Xlib API definition we'll be living with.
11073    
11074    See https://bugs.freedesktop.org/show_bug.cgi?id=32098 for details.
11075    
11076    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11077
11078commit 6459f9a48a82de3bc0235bd36acab2df11e436d5
11079Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11080Date:   Mon Dec 20 20:54:07 2010 -0800
11081
11082    specs/XKB: make acknowlegement section <preface> instead of <chapter>
11083    
11084    Fixes numbering of actual chapters to match their filenames and to
11085    make the table/figure numbering match the references in the text.
11086    
11087    Reported-by: jelmd on freenode irc
11088    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11089
11090commit 42c653d5413fc5870486a9cf51dd28d1e88f7c69
11091Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11092Date:   Sun Dec 19 23:53:02 2010 -0800
11093
11094    Add XKBlib spec images to dist_spec_DATA for distribution & installation
11095    
11096    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11097    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11098
11099commit 8f8a3c8c42af82976d261ac2447cfbb0f90fa77d
11100Author: Matt Dew <matt@osource.org>
11101Date:   Tue Nov 30 10:52:07 2010 -0500
11102
11103    specs: convert XKBlib spec from Framemaker to DocBook.xml
11104    
11105    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11106
11107commit c8701115462b482d99ecff24d9de0f2806084ba5
11108Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11109Date:   Sun Nov 21 20:49:05 2010 -0800
11110
11111    Mark atom names argument to XInternAtoms as const
11112    
11113    Updates code & docs for XInternAtoms.
11114    
11115    The single atom name argument to XInternAtom was already const char *
11116    in the code, but not the docs, so updated it in the docs too.
11117    
11118    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11119    Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
11120
11121commit 24db5b5ff0d51b0ed9d9701a792c5f14f23b638d
11122Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11123Date:   Sun Nov 21 14:04:32 2010 -0800
11124
11125    libX11 1.4.0
11126    
11127    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11128
11129commit 18d3c3e9de19e936281b111d5a82acfb3a3f552e
11130Author: Dan Nicholson <dbn.lists@gmail.com>
11131Date:   Sat Nov 6 21:58:10 2010 +0000
11132
11133    config: Check host string when deciding architecture to build for
11134    
11135    When checking for the OS/2 platform, $target_os is used. However, unless
11136    building a cross compiler, the $host* strings contain the platform
11137    details for the build system. See:
11138    
11139    http://www.gnu.org/software/automake/manual/automake.html#Cross_002dCompilation
11140    
11141    $host_os is already being used to determine the transport and threading
11142    options.
11143    
11144    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11145    Reviewed-by: Julien Cristau <jcristau@debian.org>
11146    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11147
11148commit affc2488a7f2660a74dc8354fc3e0bff2c4f879c
11149Author: Dan Nicholson <dbn.lists@gmail.com>
11150Date:   Sat Nov 6 21:58:09 2010 +0000
11151
11152    config: Fix architecture check for OS/2 to skip nios2 cpu
11153    
11154    The OS/2 platform requires some utility functions as well as having a
11155    non-32 bit wchar_t. Fix the configure check so that it doesn't also
11156    affect the nios2 cpu, which wouldn't influence these operating system
11157    issues.
11158    
11159    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11160    Tested-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
11161    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11162    Reviewed-by: Julien Cristau <jcristau@debian.org>
11163    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11164
11165commit 8cbca8a10761d1ea75a75bafa647632d6c0dac71
11166Author: Gaetan Nadon <memsize@videotron.ca>
11167Date:   Tue Nov 9 13:04:44 2010 -0500
11168
11169    config: HTML file generation: use the installed copy of xorg.css
11170    
11171    Currenlty the xorg.css file is copied in each location
11172    where a DocBook/XML file resides. This produces about
11173    70 copies in the $(docdir) install tree.
11174    
11175    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11176
11177commit 3e8907305e1818369aef5a5c0da61f09e20de4f8
11178Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11179Date:   Fri Oct 29 22:02:10 2010 -0700
11180
11181    libX11 1.3.99.903 (1.4.0 RC3)
11182    
11183    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11184
11185commit 5d245d8013289b13e0c42100951b26166c7fada4
11186Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11187Date:   Fri Oct 29 22:01:39 2010 -0700
11188
11189    Require xorg-macros 1.11 now that it is released
11190    
11191    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11192
11193commit 1ec89689fc771f116a6165226b9e076f54254a40
11194Author: Dan Nicholson <dbn.lists@gmail.com>
11195Date:   Fri Oct 22 19:09:07 2010 -0700
11196
11197    docs: Disable fop documentation by default
11198    
11199    fop is used to generate the pdf and ps formats of the documentation.
11200    This can significantly slow down the build, especially when creating all
11201    the compose key charts. Since few people probably want the full set of
11202    doc formats, set the default to 'no'.
11203    
11204    The default parameter for XORG_WITH_FOP is only available in recent
11205    macros. Users generating configure from older macros will just get
11206    'auto' as the default.
11207    
11208    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11209    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11210    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11211    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11212
11213commit 03877bda911b72ac60a7144d6eced7d5be4b4d72
11214Author: Jon TURNEY <jon.turney@dronecode.org.uk>
11215Date:   Wed Apr 14 13:38:18 2010 +0100
11216
11217    Add C.UTF-8 and C.ASCII locale aliases
11218    
11219    Add C.UTF-8 locale as an alias for en_US.UTF-8
11220    Add C.ASCII locale as an alias for C
11221    
11222    (C.UTF-8 is the default locale for cygwin.  It also exists in debian,
11223    although I don't think it's intended for use by humans.)
11224    
11225    Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
11226    Acked-by: Julien Cristau <jcristau@debian.org>
11227    Reviewed-by: James Cloos <cloos@jhcloos.com>
11228
11229commit 81d1d9a6bdca779a44f931730b866eea75777c00
11230Author: Dan Nicholson <dbn.lists@gmail.com>
11231Date:   Sat Oct 9 16:00:47 2010 -0700
11232
11233    docs: Remove directory prerequisites from make rules
11234    
11235    Make expects prerequisites to be files with valid timestamps, and
11236    directories are treated as always being out of date. Thus, any targets
11237    depending on directories will always be rebuilt.
11238    
11239    Instead, the doc rules are changed to always create the target's leading
11240    directory. This should prevent the documentation from being rebuilt when
11241    "make install" is run.
11242    
11243    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
11244    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
11245    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11246
11247commit d0cbf388919364fe6b5b9127f36426eb744090a0
11248Author: Jeremy Huddleston <jeremyhu@apple.com>
11249Date:   Sun Sep 26 21:21:16 2010 -0700
11250
11251    Add an "X11_" string to header guards to avoid possible collision
11252    
11253    This addresses a build failure which can result from <X11/Xlocale.h> and
11254    <xlocale.h> being included in the same code since they both used the same
11255    _XLOCALE_H_ protection.
11256    
11257    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
11258
11259commit b51ac675f44f5ce1a28c5734bf3c26983f8b4192
11260Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11261Date:   Tue Sep 21 19:53:03 2010 -0700
11262
11263    libX11 1.3.99.902 (1.4.0 RC2)
11264    
11265    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11266
11267commit 986bb6d1d54368fe91e3ea24f518d43ce6179782
11268Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11269Date:   Tue Sep 14 00:10:31 2010 -0700
11270
11271    Bug 19379 - Provide docs with overview of all compose key combinations
11272    
11273    Adds compose-chart.pl to generate DocBook/XML documents listing compose
11274    keys, and Makefile rules to generate HTML & PDF output from them if xmlto
11275    is present.
11276    
11277    https://bugs.freedesktop.org/show_bug.cgi?id=19379
11278    
11279    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11280    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
11281    Reviewed-by: James Cloos <cloos@jhcloos.com>
11282    Tested-by: Gaetan Nadon <memsize@videotron.ca>
11283
11284commit 3eb064071695ebf0f371163ed818a428dfeba8e6
11285Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11286Date:   Sat Sep 11 00:49:21 2010 -0700
11287
11288    Make locale data build non-recursive / parallelizable
11289    
11290    On a 4 core CPU with gmake -j 16 the nls subdir builds in half the time,
11291    plus this simplifies the next set of changes.
11292    
11293    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11294    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
11295    Reviewed-by: James Cloos <cloos@jhcloos.com>
11296    Tested-by: Gaetan Nadon <memsize@videotron.ca>
11297
11298commit a3fc78ac352e5a70a958996ef6aec50f653974d1
11299Author: Gaetan Nadon <memsize@videotron.ca>
11300Date:   Fri Sep 17 17:38:11 2010 -0400
11301
11302    config: remove man page suffix from bottom summary
11303    
11304    The man page suffix is the same for all libraries on a given
11305    platform and is not configurable. It should have been removed
11306    in commit 09edc6de6.
11307    
11308    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11309
11310commit a52b0068f36c739eb7e426214cd72a8d9bb776ca
11311Author: Gaetan Nadon <memsize@videotron.ca>
11312Date:   Fri Sep 17 17:29:30 2010 -0400
11313
11314    config: remove unhelpful comment in .gitignore
11315    
11316    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11317
11318commit 10dd881e7818be9074da4ab3a68577adcdaf71d2
11319Author: Adam Jackson <ajax@redhat.com>
11320Date:   Wed Sep 8 10:44:23 2010 -0400
11321
11322    Zero buffer data in BufAlloc()
11323    
11324    Inspired by a pattern in NoMachine's NX.  Consistently zeroed buffers
11325    compress better with ssh and friends.  Note that you'll need to rebuild
11326    all your protocol libraries to take advantage of this.
11327    
11328    Signed-off-by: Adam Jackson <ajax@redhat.com>
11329    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
11330    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11331
11332commit da099f0d2015831ea5f4a5d0740ed962aa4d5d8b
11333Author: James Cloos <cloos@jhcloos.com>
11334Date:   Sun Sep 12 17:00:54 2010 -0400
11335
11336    Revert “Dolt-ify"
11337    
11338    Libtool’s is now sufficiently fast that DOLT is no longer
11339    worth the bother, even on those few systems where is works.
11340    
11341    This reverts commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60
11342    and part of commit d31e644c65c52828ea3e7abd94a8cf9aee12265c.
11343    
11344    It conflicted with commit f6a4fd0c7615684d08e848245849dea4017a5214
11345    which moved dolt from configure.ac to m4/dolt.m4.
11346    
11347    And it addresses: http://bugs.freedesktop.org/show_bug.cgi?id=28188
11348    
11349    Signed-off-by: James Cloos <cloos@jhcloos.com>
11350
11351commit 2661fbe6b809c937a60c2c1ca5ddb8280e399bd8
11352Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11353Date:   Thu Sep 9 19:24:29 2010 -0700
11354
11355    XIM spec title page minor formatting cleanup
11356    
11357    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11358
11359commit f99f1380bbf7bb2a0c491acad4a3d8db44bfd752
11360Author: Jens Petersen <petersen@redhat.com>
11361Date:   Mon Aug 23 18:08:10 2010 -0700
11362
11363    Bug 29773: aliases for nb_NO.utf8 and nn_NO.utf8
11364    
11365    <https://bugs.freedesktop.org/show_bug.cgi?id=29773>
11366    
11367    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11368
11369commit b15b8a558ec64c834cc8f6a52d7f3c4f530c8c4b
11370Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11371Date:   Fri Sep 3 23:11:53 2010 -0700
11372
11373    Sun's copyrights are now owned by Oracle
11374    
11375    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11376
11377commit 7742bf62b1fa652da4270587e280249945367cae
11378Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11379Date:   Fri Sep 3 18:31:44 2010 -0700
11380
11381    libX11 1.3.99.901 (1.4.0 RC1)
11382    
11383    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11384
11385commit 5a02d6d74b5451d48d8f55709dfd4ecd3c068586
11386Author: Jeroen Hoek <mail@jeroenhoek.nl>
11387Date:   Sun Aug 29 09:47:06 2010 -0400
11388
11389    libX11/nls compose tables, add LATIN CAPITAL LETTER SHARP S
11390    
11391    Add LATIN CAPITAL LETTER SHARP S
11392    (See https://bugs.freedesktop.org/show_bug.cgi?id=29448 for rationale)
11393    
11394    Signed-off-by: James Cloos <cloos@jhcloos.com>
11395
11396commit 09edc6de6619a2eabda3b808ebff4165550664f1
11397Author: Gaetan Nadon <memsize@videotron.ca>
11398Date:   Fri Aug 13 14:21:08 2010 -0400
11399
11400    config: remove man-pages configuration option
11401    
11402    This option was added in commit 6e752ea120 with no explanation.
11403    The section number is provoded by XORG_MANPAGE_SECTIONS
11404    There is no case where libX11 should be different than other libs
11405    The option was also used to disable building of the man pages,
11406    which build in 14 secs. No indication this is required.
11407    
11408    If there is a requirement from system builders to disable building
11409    of man pages, it could be done consistently for all modules.
11410    
11411    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11412
11413commit f92e754297ec5fdb81068b56a4435026666224fa
11414Author: Aaron Plattner <aplattner@nvidia.com>
11415Date:   Sun Aug 15 21:51:38 2010 -0700
11416
11417    XOpenDisplay: save the correct display_name value
11418    
11419    The X Test Suite's XDisplayString test checks the invariant
11420    XDisplayString(XOpenDisplay(str)) == str.  The Xlib XOpenDisplay violates this
11421    invariant by expanding str to the canonical form "host:display.scrn" (unless
11422    HAVE_LAUNCHD is set and it starts with "/tmp/launch").  E.g., this expands ":1"
11423    to ":1.0":
11424    
11425      400|26 1 1 19:26:41|IC Start
11426      200|26 1 19:26:41|TP Start
11427      520|26 1 00032625 1 1|VSW5TESTSUITE PURPOSE 1
11428      520|26 1 00032625 1 2|Assertion XDisplayString-1.(A)
11429      520|26 1 00032625 1 3|A call to XDisplayString returns the string that was used
11430      520|26 1 00032625 1 4|as the argument to the XOpenDisplay call that returned the
11431      520|26 1 00032625 1 5|value used as the display argument.
11432      520|26 1 00032625 1 6|METH: Open a connection using XOpenDisplay.
11433      520|26 1 00032625 1 7|METH: Obtain the display string using XDisplayString.
11434      520|26 1 00032625 1 8|METH: Verify that the value of the string is the parameter used in XOpenDisplay.
11435      520|26 1 00032625 1 9|METH: Close the display using XCloseDisplay.
11436      520|26 1 00032625 1 10|REPORT: XDisplayString() returned ":1.0" instead of ":1".
11437      220|26 1 1 19:26:41|FAIL
11438      410|26 1 1 19:26:41|IC End
11439    
11440    Fix this by deleting all of the code to construct the canonical path and just
11441    stashing a copy of the original display_name in dpy->display_name.
11442    
11443    Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
11444    Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
11445
11446commit 1dc401f678469d0235e6d0b28eb4356f03327c9f
11447Author: Gaetan Nadon <memsize@videotron.ca>
11448Date:   Fri Aug 13 13:06:51 2010 -0400
11449
11450    man: xkb: remove unused variable LIB_MAN_DIR_SUFFIX
11451    
11452    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11453
11454commit cbe9eebd11e5ebab4280d25b57b4cd4001241619
11455Author: Gaetan Nadon <memsize@videotron.ca>
11456Date:   Fri Aug 13 11:42:29 2010 -0400
11457
11458    man: simplify building of shadow man pages
11459    
11460    Store the shadow files in git as any other man page.
11461    Move man pages to man dir and use the common makefile
11462    
11463    Local fix in CVS for bug 5628 is not required
11464    as the problem has been fixed in
11465    util-macros d9062e4077ebfd0985baf8418f3d0f111b9ddbba
11466    
11467    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11468
11469commit bfd899c9af2318ecbe84cb69007dba4223fdb502
11470Author: Gaetan Nadon <memsize@videotron.ca>
11471Date:   Mon Aug 9 09:39:26 2010 -0400
11472
11473    config: reinstate XORG_PROG_RAWCPP erroneoulsy removed
11474    
11475    in commit 76e07ef6911734eac418e399b114f1b544512736.
11476    
11477    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11478
11479commit 76e07ef6911734eac418e399b114f1b544512736
11480Author: Gaetan Nadon <memsize@videotron.ca>
11481Date:   Fri Aug 6 14:01:51 2010 -0400
11482
11483     man: using the C preprocessor is not required for man pages.
11484    
11485    There were no special symbols needing cpp.
11486    Everything can be handled by the default MAN_SUBSTS in util-macros.
11487    
11488    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11489
11490commit 4b8ff7db39f2fe7ef12968d462aaf3f9054b6c18
11491Author: Jamey Sharp <jamey@minilop.net>
11492Date:   Fri Aug 6 15:51:56 2010 -0700
11493
11494    Fix use-after-free in _XReply on X errors.
11495    
11496    _XReply would always call dequeue_pending_request on errors.  When it
11497    got an error for the current request, it would call dequeue, then break
11498    out of the loop; then, if it had an error in the event queue, it would
11499    compare it with the sequence number of the now-freed pending request.
11500    _XReply already stored that sequence number in dpy->last_request_read
11501    before freeing it, so look at that instead.
11502    
11503    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29412
11504    
11505    Signed-off-by: Jamey Sharp <jamey@minilop.net>
11506    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
11507
11508commit 9fa146b30046396b70d64986e50d6617b3a8ac48
11509Author: Gaetan Nadon <memsize@videotron.ca>
11510Date:   Sat Jul 31 16:15:35 2010 -0400
11511
11512    specs: xsl stylesheet requires abs path to images
11513    
11514    Using abs_builddir requires automake 1.10 or later.
11515    
11516    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11517
11518commit a8727d4bc39015e303b4128d0ad8aaf6d5fc9f0a
11519Author: Gaetan Nadon <memsize@videotron.ca>
11520Date:   Sun Aug 1 14:13:07 2010 -0400
11521
11522    specs: update .gitignore now that all groff generated files are gone
11523    
11524    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11525
11526commit f70527f05677aaf6a493ba42d2d95f738aaebafa
11527Author: Adam Jackson <ajax@redhat.com>
11528Date:   Thu Jul 29 11:11:21 2010 -0400
11529
11530    nls: Switch one of the interrobang sequences to gnaborretni
11531    
11532    Since gnaborretni is primarily used in LTR locales, the ? part of the
11533    sequence reflects the position of the bulge in the ?.  When scanning an
11534    inverted interrobang left-to-right, you see the bulge first, so the
11535    compose sequence ?! is used; upright interrobang shows the bulge
11536    last, so it's !?.
11537    
11538    This is a change in behaviour, but I was unable to come up with anything
11539    better.  ??!! or !!?? would match the patterns for ¿ and ¡, but they
11540    would delay evaluation of them; "¿Qué?" wouldn't show the ¿ until you hit
11541    Q.  Likewise ?!?! and !?!? would delay showing the interrobang itself.
11542    ~!? and ~?! were considered but are arguably less intuitive.
11543    
11544    Reviewed-by: James Cloos <cloos@jhcloos.com>
11545    Signed-off-by: Adam Jackson <ajax@redhat.com>
11546
11547commit 511c4f6d29b2da4f71093feabcbb3913cb5d12a7
11548Author: Gaetan Nadon <memsize@videotron.ca>
11549Date:   Wed Jul 28 10:27:39 2010 -0400
11550
11551    specs: move indexterm from glossdef to glossterm
11552    
11553    This move fixes a Java class cast exception in the glossary.
11554    The problem was introduced in commit
11555    26f4f0d50840fe5ba4c46aae0a8e68db0059434b
11556    
11557    It may not happen on all versions of the doc toolchain.
11558    There is no reason why indexterm cannot appear in glossdef,
11559    this is a workaround to an implementation problem found by
11560    trial and error.
11561    
11562    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11563    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11564    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11565
11566commit 87a108f3aa565aa803dacb3a53b811f45025b609
11567Author: Matt Dew <matt@osource.org>
11568Date:   Sat Jul 24 14:46:57 2010 -0400
11569
11570    specs: replace troff source with docbook-xml source
11571    
11572    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11573
11574commit 8e603413586ff42b1351688f6f99318cbe243bca
11575Author: Julien Cristau <jcristau@debian.org>
11576Date:   Sat Jul 24 22:13:07 2010 +0200
11577
11578    Compose.man: default user compose file is .XCompose, not .Xcompose
11579    
11580    The path was correct in the DESCRIPTION section, but not in FILES.
11581    
11582    Signed-off-by: Julien Cristau <jcristau@debian.org>
11583
11584commit 554da76ece85d0fc0cada45a86860e69c2107e9a
11585Author: Daniel Stone <daniel@fooishbar.org>
11586Date:   Tue Jul 20 12:34:48 2010 +0100
11587
11588    NLS: Add \o/ Compose sequence
11589    
11590    Unicode is even more pointlessly awesome! Add a binding for Compose-\o/
11591    to U+1F64C PERSON RAISING BOTH HANDS IN CELEBRATION.
11592    
11593    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11594
11595commit 2e7a18b6a617b9b4bfcea2d36f2bd2d7e0c4a3dd
11596Author: Daniel Stone <daniel@fooishbar.org>
11597Date:   Fri Jul 9 18:13:13 2010 +0100
11598
11599    XStringToKeysym: Cope with 0x1234cafe-style input
11600    
11601    If we get input in the style of 0xdeadbeef, just return that exact
11602    keysym.  Introduces a dependency on strtoul, which I'm told is OK on all
11603    the systems we care about.
11604    
11605    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11606
11607commit 3df45ed0c29b98ff468a0ff0ba24830bb664fd5a
11608Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11609Date:   Mon Jul 19 17:58:27 2010 -0700
11610
11611    specs/libX11: Fix column count of Gravity Attributes table
11612    
11613    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11614
11615commit 53bcba0d1d2dc3fd5e0de4bae3da30a1aa31a0c4
11616Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11617Date:   Mon Jul 19 14:43:38 2010 -0700
11618
11619    specs/libX11: Convert \- to &minus; and \^ to either &hairsp; or removed
11620    
11621    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11622
11623commit 5a0b45275638281e1bb2ae7d3e16b98e6470dae9
11624Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11625Date:   Mon Jul 19 11:28:50 2010 -0700
11626
11627    specs/libX11: Manual cleanup pass over Ch. 1
11628    
11629    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11630
11631commit 21567992958a6f12b3d0186f2f27a5b2a41b8448
11632Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11633Date:   Mon Jul 19 14:51:10 2010 -0700
11634
11635    specs/libX11: make sure all files have DOCTYPEs so standard entities work
11636    
11637    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11638
11639commit c944a8521f5760b485192658b921145159cdb439
11640Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11641Date:   Sun Jul 18 13:12:20 2010 -0700
11642
11643    specs/libX11: Fix up list of header files in Ch.1 & add index entries
11644    
11645    Combination of manual editing and automatic substitution via:
11646    
11647    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>$4<para>}g' CH01.xml
11648    
11649    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11650
11651commit bb66e8f1ce330043278d3e8cbc7d5d3bc56ee030
11652Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11653Date:   Sun Jul 18 12:47:27 2010 -0700
11654
11655    specs/libX11: Add indexterms for headerfiles that .hN used to provide
11656    
11657    Automatic substitution performed via:
11658    perl -i -p -e 's{^&lt;<filename class="headerfile">([^<]+)</filename>&gt;(.*)$}{<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;$1&gt;</filename></secondary></indexterm>}' *.xml
11659    
11660    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11661
11662commit d0b0d215cb85da48b323b1392149ce181e7ef5ec
11663Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11664Date:   Fri Jul 16 21:15:13 2010 -0700
11665
11666    specs/libX11: convert multicolumn lists from tables to <simplelist>
11667    
11668    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11669
11670commit e679f80c6210c0aefe3669f8f536353619097a4b
11671Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11672Date:   Fri Jul 16 20:51:52 2010 -0700
11673
11674    specs/libX11: Tag WM_* as <property>
11675    
11676    Mass substitution done by:
11677     perl -i -p -e 'if ($_ !~ m{^\<}) { $_ =~ s{(WM_\w+)}{<property>$1</property>}g; }' *.xml
11678    
11679    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11680
11681commit c7b7e59b3b22221d0be6286c540001c360308f69
11682Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11683Date:   Fri Jul 16 20:41:33 2010 -0700
11684
11685    specs/libX11: Convert simpler eqn markup to docbook tags
11686    
11687    Mostly "sup" to <superscript>
11688    
11689    There's several more complicated equations that will probably need
11690    MathML or SVG to solve.
11691    
11692    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11693
11694commit 897486c54c6a54771867d667441aaf9a4b9c35ca
11695Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11696Date:   Fri Jul 16 00:50:39 2010 -0700
11697
11698    specs/libX11: App. D: convert literallayouts to synopsis or programlisting
11699    
11700    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11701
11702commit 692906c3b3d8aa8b8927fbc230c3050d633785d2
11703Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11704Date:   Fri Jul 16 00:36:43 2010 -0700
11705
11706    specs/libX11: Fix section headers in Appendix D
11707    
11708    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11709
11710commit bcc41baa02e137884d847aec0f3ff8ca5c85c32f
11711Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11712Date:   Wed Jul 14 07:22:26 2010 -0700
11713
11714    specs/libX11: Explicitly tag document as English
11715    
11716    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11717
11718commit 6783544706ff370e900c137f951e90230586d6dc
11719Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11720Date:   Mon Jul 12 22:03:18 2010 -0700
11721
11722    specs/libX11: Appendix C: convert literallayouts to synopsis or programlisting
11723    
11724    Fixes display of a bunch of function prototypes and sample code
11725    
11726    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11727
11728commit 8834cd9285e6aaf437aee56292c508d99a090fa9
11729Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11730Date:   Mon Jul 12 18:37:16 2010 -0700
11731
11732    xmlrules.in: Add chunked-html rule to manually generate smaller files
11733    
11734    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11735
11736commit 53f78680c34ebc94ce26b5e14c2a0003435cf10d
11737Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11738Date:   Mon Jul 12 18:35:59 2010 -0700
11739
11740    specs/libX11: Convert Appendix A tables to real tables
11741    
11742    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11743
11744commit 705a1257a8a47f4a1f94979aee09a837500ede7d
11745Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11746Date:   Mon Jul 12 18:38:09 2010 -0700
11747
11748    specs/libX11: Fix out-of-place text in Chapter 1
11749    
11750    DTD violation error introduced in 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4
11751    
11752    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11753
11754commit a151346a2b7810e988f1de4b1e00b37672dc587a
11755Author: Jeremy Huddleston <jeremyhu@apple.com>
11756Date:   Mon Jul 12 16:52:12 2010 -0700
11757
11758    launchd: Explicitly search /sbin
11759    
11760    Previously, launchd wasn't found if /sbin wasn't in the user's PATH.
11761    https://bugs.freedesktop.org/show_bug.cgi?id=29028
11762    
11763    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
11764
11765commit 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4
11766Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11767Date:   Mon Jul 12 13:38:04 2010 -0700
11768
11769    Replace untranslated .hN macros with <filename> tags to show header names
11770    
11771    Translation performed by:
11772     perl -i -p -e 's{\<\!-- .hN (\S+) (\S+)?\s*-->}
11773                     {&lt;<filename class="headerfile">$1</filename>&gt;$2}' *.xml
11774    
11775    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11776
11777commit 75480440946603b8efdbbf78b88d59c641b6d2c8
11778Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11779Date:   Sat Jul 10 00:06:13 2010 -0700
11780
11781    specs/libX11: Mass substitution of <function> tags with more specific tags
11782    
11783    Matched names from X headers & "nm libX11.so" output to names in spec to
11784    map to more specific tags.   Tags used:
11785    
11786                        <code>   code fragments
11787                    <constant>   enum values
11788                   <errorname>   X protocol errors
11789                    <filename>   filenames
11790                    <function>   functions
11791                    <function>   function-like macros (#define foo(a,b)...)
11792                 <returnvalue>   function return codes
11793                 <structfield>   struct members
11794                  <structname>   struct names (even when typedefed)
11795                      <symbol>   simple value #defines (#define NAME value)
11796                  <systemitem>   X protocol requests
11797    <systemitem class="event">   X protocol events
11798                        <type>   non-struct typedefs
11799                     <varname>   global variables
11800    
11801    (Also fixed a couple typos detected by failures of this matching, such as
11802     XESSetPrintErrorValues -> XESetPrintErrorValues.)
11803    
11804    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11805
11806commit 26f4f0d50840fe5ba4c46aae0a8e68db0059434b
11807Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11808Date:   Fri Jul 9 00:37:23 2010 -0700
11809
11810    specs/libX11: Glossary cross-reference links
11811    
11812    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11813
11814commit 930b52d84031de0e15e0a11cf4ecfd3b2f59e073
11815Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11816Date:   Thu Jul 8 21:19:08 2010 -0700
11817
11818    specs/libX11: Make whitespace around <function> tags more uniform
11819    
11820    Simplifies regular expressions for further mass substitutions.
11821    
11822    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11823
11824commit 730ce6b8aca77b77385aa894d234cbde5fcbce37
11825Author: Gaetan Nadon <memsize@videotron.ca>
11826Date:   Fri Jul 9 20:10:37 2010 -0400
11827
11828    specs: use pattern rules rather than suffix rules
11829    
11830    This allows target to rebuild when included .xml files are changed.
11831    
11832    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11833
11834commit 5527b4bc8ce7de60123eb28789ddef54aa48a378
11835Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11836Date:   Thu Jul 8 20:56:41 2010 -0700
11837
11838    specs/libX11: Move punctuation outside of the <function> tags
11839    
11840    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11841
11842commit d66d2134dfc38cd866c7c2d9ea45ed4b4dad23f0
11843Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11844Date:   Thu Jul 8 20:42:50 2010 -0700
11845
11846    specs/libX11: Mark a number of <acronym>s
11847    
11848    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11849
11850commit d5bbb12f55ac9f691eab6242cedc53207275131b
11851Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11852Date:   Thu Jul 8 19:31:36 2010 -0700
11853
11854    specs/libX11: Glossary terms should not be marked as functions
11855    
11856    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11857
11858commit 5decf7bc5dae0ae0c45c47eaaa9c4fdf9515ca1c
11859Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11860Date:   Thu Jul 8 19:28:18 2010 -0700
11861
11862    specs/libX11: Clean up author list & acknowledgements a little
11863    
11864    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11865
11866commit 0cd29bdb332c333e5123dce65b25ad1d97fbdae3
11867Author: Alan Coopersmith <alan.coopersmith@oracle.com>
11868Date:   Thu Jul 8 17:30:34 2010 -0700
11869
11870    specs/libX11: Convert troff .IN macros to docbook <indexterm> tags
11871    
11872    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11873
11874commit a67d99ccbe22c8ce5f7a12f13b8991d9e0cf4491
11875Author: Gaetan Nadon <memsize@videotron.ca>
11876Date:   Thu Jul 8 17:29:21 2010 -0400
11877
11878    specs: specdir is required to install xml files
11879    
11880    The source is installed as well as the targets.
11881    This failed when configuring with --without-xmlto
11882    
11883    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11884
11885commit e14273f44c1501ab51db4adcb83b18a1073787d8
11886Author: Matt Dew <matt@osource.org>
11887Date:   Thu Jul 8 14:42:32 2010 -0400
11888
11889    specs: replace troff source with docbook-xml source
11890    
11891    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
11892
11893commit bea0873caf50e9ed1b89255775d9ab912cbecd45
11894Author: Daniel Stone <daniel@fooishbar.org>
11895Date:   Thu Jul 8 16:49:51 2010 +0100
11896
11897    XStringToKeysym: Check strdup() return value
11898    
11899    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11900    Reviewed-by: Keith Packard <keithp@keithp.com>
11901
11902commit cffa71b4a5743e2b6675b9a917d15aef4177d513
11903Author: Marko Myllynen <myllynen@redhat.com>
11904Date:   Mon Jun 28 15:08:05 2010 +0300
11905
11906    Fix two typos in SFS 5966 Annex 3
11907    
11908    X.Org bug#28792 <https://bugs.freedesktop.org/show_bug.cgi?id=28792>
11909    
11910    Signed-off-by: Julien Cristau <jcristau@debian.org>
11911
11912commit eb023c0f8919e809b8b609e1467b14d20a290aa7
11913Author: Daniel Stone <daniel@fooishbar.org>
11914Date:   Tue Jun 15 18:49:43 2010 +0100
11915
11916    Delete now-redundant XKeysymDB
11917    
11918    Since XStringToKeysym now supports all the vendor keysyms, just delete
11919    our XKeysymDB, which was incomplete at best, misleading at worst, and
11920    always an annoyance.
11921    
11922    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11923    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11924
11925commit 00175397480b76d32bf82b0c7c94c91a2a95954e
11926Author: Daniel Stone <daniel@fooishbar.org>
11927Date:   Tue Jun 15 18:48:48 2010 +0100
11928
11929    makekeys: Scan vendor keysyms as well as core
11930    
11931    Since we can't really live without vendor keysyms, scan them all in to
11932    generate ks_tables.h, rather than only doing the core ones, and leaving
11933    the vendor syms to be manually synchronised with XKeysymDB.
11934    
11935    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11936    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11937
11938commit ebd6ef0a4db0ddef0ae17ad14571518ccdeea5ba
11939Author: Daniel Stone <daniel@fooishbar.org>
11940Date:   Tue Jun 15 18:47:37 2010 +0100
11941
11942    XStringToKeysym: Special case for XF86 keysyms
11943    
11944    Some XFree86 keysyms were in XKeysymDB as XF86_foo, despite really being
11945    XF86foo.  So, if we get to the bottom of XStringToKeysym and haven't
11946    found our XF86_foo, try it again as XF86foo.
11947    
11948    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11949    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11950
11951commit 8c2ffce9e5c6eef8a04f7e2732db46b3b62e13bb
11952Author: Daniel Stone <daniel@fooishbar.org>
11953Date:   Tue Jun 15 17:20:48 2010 +0100
11954
11955    configure.ac: Change from deprecated AC_DEFINE_DIR to AX_
11956    
11957    AC_DEFINE_DIR is deprecated as it's squatting on the autoconf-builtin
11958    namespace, so start using the more proper AX_DEFINE_DIR instead.
11959    
11960    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
11961    Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
11962    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11963
11964commit 6de368c9aa7ccd2fcd62fca5a2b278913db4d03d
11965Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
11966Date:   Thu Jul 1 06:50:47 2010 -0300
11967
11968    Purge macros NEED_EVENTS and NEED_REPLIES
11969    
11970    Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
11971    Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com>
11972    Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
11973
11974commit 0b724231be17f19538ee115a03b82b221b1f83c0
11975Author: Adam Jackson <ajax@redhat.com>
11976Date:   Wed Jun 30 16:30:49 2010 -0400
11977
11978    Use -version-number consistently instead of -version-info
11979    
11980    The latter is libtool braindamage.
11981    
11982    Signed-off-by: Adam Jackson <ajax@redhat.com>
11983
11984commit 241a990afcf9c967c587ad6fd245df21b68b1ad9
11985Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
11986Date:   Tue Jun 29 18:32:42 2010 +0200
11987
11988    Make Compose-comma map to ogonek for I and U in UTF-8 locales.
11989    
11990    With the preceding patch, it makes the UTF-8 compose map consistent with
11991    the ISO 8859-4 and -13 maps.
11992
11993commit af55e582f485a668c2bf43129be972bc65f03c60
11994Author: Juliusz Chroboczek <jch@pps.jussieu.fr>
11995Date:   Tue Jun 29 18:20:18 2010 +0200
11996
11997    Make Compose-comma map to Ogonek for A and E in UTF-8 locales.
11998    
11999    This makes the UTF-8 compose file consistent with the ISO 8859-2 compose file
12000    as far as Polish is concerned.  We only sacrifice one pair of characters,
12001    e-cedilla, which is not used in any language.
12002
12003commit d6ba13009fc9ef876a104b907ffef73c6e405d4b
12004Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12005Date:   Thu Jun 24 13:13:11 2010 -0700
12006
12007    Define FILE_MAN_DIR_SUFFIX so XCompose shadow page has correct path
12008    
12009    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12010
12011commit 978c7238789037de917b9423ea9adfb006da1260
12012Author: Jamey Sharp <jamey@minilop.net>
12013Date:   Sat Jun 19 10:44:55 2010 -0700
12014
12015    poll_for_response: Really handle xcb_poll_for_reply getting a reply.
12016    
12017    Don't lose async replies. That's bad.
12018    
12019    `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the
12020    _XReply path worked; that path waited for replies, rather than polling.
12021    
12022    However, XRecordProcessReplies, which does nothing but call XPending,
12023    relied on the event-handling path to process async replies, and that was
12024    busted.
12025    
12026    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595
12027    
12028    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12029
12030commit ef67486c5a8aada24fd95b2991a3c4979f53805f
12031Author: Jamey Sharp <jamey@minilop.net>
12032Date:   Mon Jun 21 13:34:15 2010 -0700
12033
12034    Revert "xcb_io.c: poll_for_response doesn't guarantee there's a pending request."
12035    
12036    This reverts commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e, because as
12037    a matter of fact, if poll_for_response returns NULL when we know
12038    dpy->xcb->next_event is non-NULL, there *is* guaranteed to be a pending
12039    request.
12040
12041commit 301ec5b41e9d253a446db656e1789ac7345dc433
12042Author: Jamey Sharp <jamey@minilop.net>
12043Date:   Mon Jun 21 13:24:24 2010 -0700
12044
12045    Revert "poll_for_response: Really handle xcb_poll_for_reply getting a reply."
12046    
12047    This reverts commit c115095d7f2bc4f5a4fb26380e3698fefdad7611. We *do*
12048    need to check poll_for_event every time through the loop in
12049    poll_for_response, so the commit did too much.
12050
12051commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e
12052Author: Jamey Sharp <jamey@minilop.net>
12053Date:   Mon Jun 21 10:54:02 2010 -0700
12054
12055    xcb_io.c: poll_for_response doesn't guarantee there's a pending request.
12056    
12057    Fixes the second bug reported in:
12058    https://bugs.freedesktop.org/show_bug.cgi?id=28595
12059    
12060    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12061
12062commit c115095d7f2bc4f5a4fb26380e3698fefdad7611
12063Author: Jamey Sharp <jamey@minilop.net>
12064Date:   Sat Jun 19 10:44:55 2010 -0700
12065
12066    poll_for_response: Really handle xcb_poll_for_reply getting a reply.
12067    
12068    Don't lose async replies. That's bad.
12069    
12070    `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the
12071    _XReply path worked; that path waited for replies, rather than polling.
12072    
12073    However, XRecordProcessReplies, which does nothing but call XPending,
12074    relied on the event-handling path to process async replies, and that was
12075    busted.
12076    
12077    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595
12078    
12079    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12080    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12081
12082commit a25ae169862ab9b76daf259613b37c6b07bc2ef2
12083Author: Marko Myllynen <myllynen@redhat.com>
12084Date:   Sun Jun 13 19:23:05 2010 +0300
12085
12086    Implement SFS 5966 Annex 4 for Finland
12087    
12088    This patch adds Annex 4 of SFS 5966 for Finland and fixes two typos in
12089    Annex 3 (which were actually copied verbatim from the standard text).
12090    
12091    The implementation of the standard is now complete.
12092    
12093    X.Org bug#28498 <https://bugs.freedesktop.org/show_bug.cgi?id=28498>
12094    
12095    Signed-off-by: Julien Cristau <jcristau@debian.org>
12096
12097commit 47b04195d8a31c8f9e6dd804196162c6cfca3ac6
12098Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12099Date:   Fri Jun 4 13:20:17 2010 -0700
12100
12101    LINEAR_RGB_InitSCCData: When malloc fails, don't try to free unallocated bits
12102    
12103    One of the malloc failure checks had a goto to the wrong spot in the
12104    list of cleanup free() calls to unwind at the end, and was freeing
12105    bits that hadn't been initialized/allocated yet, since they would be
12106    stored in the struct that just failed to be allocated.
12107    
12108    Error: Null pointer dereference (CWE 476)
12109       Read from pointer that could be constant 'NULL'
12110            at line 805 of /export/alanc/X.Org/sx86/lib/libX11/src/xcms/LRGB.c in function 'LINEAR_RGB_InitSCCData'.
12111              Pointer checked against constant 'NULL' at line 754 but does not protect the dereference.
12112    
12113    [ This bug was found by the Parfait bug checking tool.
12114      For more information see http://research.sun.com/projects/parfait ]
12115    
12116    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12117
12118commit aebbf3623888119b43893b253195d93b613a6e1e
12119Author: Josh Triplett <josh@joshtriplett.org>
12120Date:   Sun Jun 6 19:11:55 2010 -0700
12121
12122    Mark the rest of Xprivate.h as _X_HIDDEN.
12123    
12124    None of the functions in Xprivate.h should have any callers outside of
12125    Xlib, by definition.
12126    
12127    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12128    Suggested-by: Jamey Sharp <jamey@minilop.net>
12129    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12130
12131commit abcd1b67c8135ad652833e5a60d818837c39fce1
12132Author: Julien Cristau <jcristau@debian.org>
12133Date:   Fri Jun 4 13:57:45 2010 +0200
12134
12135    Hide _XSeqSyncFunction
12136    
12137    Commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43 made _XSeqSyncFunction
12138    non-static, but we don't need to export it.
12139    
12140    Signed-off-by: Julien Cristau <jcristau@debian.org>
12141    Reviewed-by: Adam Jackson <ajax@redhat.com>
12142    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12143
12144commit 15e5eaf62897b3179d1fbe457cb19f886f0449f8
12145Author: Josh Triplett <josh@joshtriplett.org>
12146Date:   Thu Jun 3 09:41:01 2010 -0700
12147
12148    Remove support for building without XCB
12149    
12150    And there was much rejoicing.
12151    
12152    Signed-off-by: Josh Triplett <josh@joshtriplett.org>
12153    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12154    
12155    Consensus on #xorg-devel agrees with removing --without-xcb; in
12156    particular, acks from Adam Jackson, Daniel Stone, Kristian Høgsberg,
12157    Julien Cristau, and Rémi Cardona.
12158
12159commit fb19eb767a32fd4ff74021c51bc6f60eb8bdff74
12160Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12161Date:   Thu Jun 3 15:21:40 2010 -0700
12162
12163    libX11 1.3.4
12164    
12165    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12166
12167commit 8f3db40ca5108a919244f3fff7466d01a14b3ce2
12168Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12169Date:   Thu Jun 3 13:08:17 2010 -0700
12170
12171    Workaround bug in groff flag processing that breaks distcheck
12172    
12173    At least with the groff 1.19.2 package I have installed, groff passes
12174    on the -I flags for the include path to grohtml, which if they come
12175    after the -P-I... flag we pass to grohtml to specify the image file
12176    name pattern cause it to override that flag and put the images in
12177    the wrong place, breaking "make distcheck" - changing the flag order
12178    works around this.
12179    
12180    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12181
12182commit bdb31a1fb707cebccc9efbbf68cb55c10fa4ea3e
12183Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12184Date:   Thu Jun 3 11:25:33 2010 -0700
12185
12186    Fix typo that made configure always report "none" for man page suffix
12187    
12188    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12189
12190commit 6bb7d2d21d6b58019fb37b724b59c9744f30b9ce
12191Author: Paul Bender <pebender@gmail.com>
12192Date:   Thu Jun 3 11:20:26 2010 -0700
12193
12194    Bug 22591 - configure does not obey the --enable-*-transport options
12195    
12196    https://bugs.freedesktop.org/show_bug.cgi?id=22591
12197    
12198    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12199
12200commit d7f35946d673bb00f2768e025b6e1017422b2174
12201Author: Paul Bender <pebender@gmail.com>
12202Date:   Thu Jun 3 11:17:45 2010 -0700
12203
12204    Bug 22590 - libX11 1.2.1 has broken abstract namespace support
12205    
12206    https://bugs.freedesktop.org/show_bug.cgi?id=22590
12207    
12208    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12209
12210commit b2487d07f7b355f230a56e32c763acd08862735c
12211Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12212Date:   Thu Jun 3 11:06:15 2010 -0700
12213
12214    Bug 22584 - libX11 does not cross compile
12215    
12216    Adaptation of patch submitted by Paul Bender in attachment 27301 to
12217    https://bugs.freedesktop.org/show_bug.cgi?id=22584
12218    
12219    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12220    Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
12221    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
12222
12223commit 4378219ff8c1418418c70086085358b69d574e74
12224Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12225Date:   Tue Jun 1 18:55:48 2010 -0700
12226
12227    Clarify requirements in XRestackWindows man page
12228    
12229    The required common parent window is not specified in the arguments,
12230    just implied as the parent of the first window in the list.
12231    
12232    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12233
12234commit f09c5299a381e2729e800a0ac43f1c0e371f65f6
12235Author: Jeremy Huddleston <jeremyhu@apple.com>
12236Date:   Wed May 12 16:42:18 2010 -0700
12237
12238    xcb: Add TCP fallback
12239    
12240    If we fail to connect to a UNIX socket and the transport isn't specified,
12241    fallback on TCP.  This matches behavior with the xtrans codepath and the
12242    Xlib spec.
12243    
12244    http://lists.x.org/archives/xorg-devel/2010-April/007915.html
12245    
12246    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12247    Reviewed-by: Jamey Sharp <jamey@minilop.net>
12248
12249commit fd82552d5c0ce1931f29006a0c36f5e03cf8577e
12250Merge: aae2a4a7 933aee1d
12251Author: Jamey Sharp <jamey@minilop.net>
12252Date:   Mon May 10 16:51:24 2010 -0700
12253
12254    Merge branch 'xlib-xcb-thread-fixes'
12255
12256commit aae2a4a7aab26de3fa715d6ecd0a0e0926b37fc9
12257Author: Jeremy Huddleston <jeremyhu@apple.com>
12258Date:   Fri Apr 23 21:50:29 2010 -0700
12259
12260    Don't append the screen number when using a launchd socket
12261    
12262    ssh gets confused by this.  XQuartz is the only DDX using this
12263    functionality, and it doesn't support different screens, so
12264    let's just not include this until most ssh know how to handle
12265    this.
12266    
12267    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12268
12269commit adcd0ec209abf832a769d52db660fb37eaad6e0c
12270Author: Jeremy Huddleston <jeremyhu@apple.com>
12271Date:   Fri Apr 23 16:33:44 2010 -0700
12272
12273    Remove launchd logic from _XConnectXCB as it's handled in XCB
12274    
12275    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12276
12277commit d232b259c36fdde1f4179822809fec1480867dc5
12278Author: Jeremy Huddleston <jeremyhu@apple.com>
12279Date:   Tue Feb 2 17:01:28 2010 -0800
12280
12281    Fix various build warnings
12282    
12283    imLcIm.c: In function '_XimCachedFileName':
12284    imLcIm.c:361: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int'
12285    imLcIm.c:364: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int'
12286    
12287    imRm.c: In function '_XimDefaultArea':
12288    imRm.c:597: warning: cast from pointer to integer of different size
12289    imRm.c: In function '_XimDefaultColormap':
12290    imRm.c:626: warning: cast from pointer to integer of different size
12291    
12292    lcFile.c:224: warning: no previous prototype for 'xlocaledir'
12293    
12294    lcUTF8.c: In function 'iconv_cstombs':
12295    lcUTF8.c:1841: warning: assignment discards qualifiers from pointer target type
12296    lcUTF8.c:1869: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
12297    lcUTF8.c:1873: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
12298    lcUTF8.c: In function 'iconv_mbstocs':
12299    lcUTF8.c:1935: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
12300    lcUTF8.c: In function 'iconv_mbtocs':
12301    lcUTF8.c:2031: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
12302    lcUTF8.c: In function 'iconv_mbstostr':
12303    lcUTF8.c:2121: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness
12304    lcUTF8.c: In function 'iconv_strtombs':
12305    lcUTF8.c:2180: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness
12306    lcUTF8.c: In function '_XlcAddGB18030LocaleConverters':
12307    lcUTF8.c:2367: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12308    lcUTF8.c:2368: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12309    lcUTF8.c:2373: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12310    lcUTF8.c:2374: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12311    lcUTF8.c:2375: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12312    lcUTF8.c:2376: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12313    lcUTF8.c:2377: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type
12314    
12315    XlibInt.c: In function '_XGetHostname':
12316    XlibInt.c:3441: warning: implicit declaration of function 'gethostname'
12317    XlibInt.c:3441: warning: nested extern declaration of 'gethostname'
12318    
12319    ConnDis.c: In function '_XDisconnectDisplay':
12320    ConnDis.c:540: warning: old-style function definition
12321    ConnDis.c: In function '_XSendClientPrefix':
12322    ConnDis.c:554: warning: old-style function definition
12323    ConnDis.c: In function 'XSetAuthorization':
12324    ConnDis.c:677: warning: old-style function definition
12325    
12326    Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12327
12328commit 3e11c73187acb2e2be9a812840bbbea947527ccb
12329Author: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
12330Date:   Fri Apr 23 00:01:54 2010 +0200
12331
12332    Allow X11 users to compose anarchism
12333    
12334    I can easily compose the hammer and sickle (the symbol representative
12335    of communism) by pressing the key sequence:
12336    
12337      <Compose> <C> <C> <C> <P> -> ☭
12338    
12339    But i can't easily make the circled A (the symbol representative of
12340    anarchism).
12341    
12342    I'd like to be able to use <Compose> <O> <A> (this is a
12343    currently unused key sequence) to generate the symbol:  Ⓐ
12344    
12345    Debian bug#555938 <http://bugs.debian.org/555938>
12346    
12347    Signed-off-by: Julien Cristau <jcristau@debian.org>
12348
12349commit 933aee1d5c53b0cc7d608011a29188b594c8d70b
12350Author: Jamey Sharp <jamey@minilop.net>
12351Date:   Fri Apr 16 20:18:28 2010 -0700
12352
12353    Fix Xlib/XCB for multi-threaded applications (with caveats).
12354    
12355    Rather than trying to group all response processing in one monolithic
12356    process_responses function, let _XEventsQueued, _XReadEvents, and
12357    _XReply each do their own thing with a minimum of code that can all be
12358    reasoned about independently.
12359    
12360    Tested with `ico -threads 20`, which seems to be able to make many
12361    icosahedrons dance at once quite nicely now.
12362    
12363    Caveats:
12364    
12365    - Anything that was not thread-safe in Xlib before XCB probably still
12366      isn't. XListFontsWithInfo, for instance.
12367    
12368    - If one thread is waiting for events and another thread tries to read a
12369      reply, both will hang until an event arrives. Previously, if this
12370      happened it might work sometimes, but otherwise would trigger either
12371      an assertion failure or a permanent hang.
12372    
12373    - Versions of libxcb up to and including 1.6 have a bug that can cause
12374      xcb_wait_for_event or xcb_wait_for_reply to hang if they run
12375      concurrently with xcb_writev or other writers. So you'll want that fix
12376      as well.
12377    
12378    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12379    Reviewed-by: Josh Triplett <josh@freedesktop.org>
12380
12381commit aab43278ae619eb57d2dd9c7396f460f078588fc
12382Author: Jamey Sharp <jamey@minilop.net>
12383Date:   Fri Apr 16 22:24:16 2010 -0700
12384
12385    Use InternalLockDisplay on code paths called from LockDisplay.
12386    
12387    It's easier to reason about the code when we can't re-enter the
12388    Xlib-private sync-handlers while they're already running.
12389    
12390    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12391    Reviewed-by: Josh Triplett <josh@freedesktop.org>
12392
12393commit 660b7d05f4ca4ab4661c9fe7ce655a4909b4e556
12394Author: Jamey Sharp <jamey@minilop.net>
12395Date:   Fri Apr 16 19:45:11 2010 -0700
12396
12397    Fix _XSend to enqueue the right range of pending requests.
12398    
12399    _XSend was off-by-one on both ends. It should not re-enqueue the last
12400    request that was already flushed, but it should enqueue the last request
12401    currently being flushed.
12402    
12403    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12404    Reviewed-by: Josh Triplett <josh@freedesktop.org>
12405
12406commit f2735889908d6e5a7f8dbee42f00c54a52665191
12407Author: Jamey Sharp <jamey@minilop.net>
12408Date:   Fri Apr 16 19:45:11 2010 -0700
12409
12410    Pending requests are always added in-order.
12411    
12412    Replace insert_pending_request, which did an in-order search of the
12413    queue to find the right insertion point, with a simpler
12414    append_pending_request, and use that in _XSend as well.
12415    
12416    Includes assertions to check that the list of pending requests is in
12417    order by sequence number and does not have duplicates.
12418    
12419    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12420    Reviewed-by: Josh Triplett <josh@freedesktop.org>
12421
12422commit d9cf5885b0f97942fbbd2a7cc50118132ece50f6
12423Author: Markus Duft <markus.duft@salomon.at>
12424Date:   Fri Apr 16 08:39:34 2010 -0700
12425
12426    Bug 26839: Fix build problem on Interix (POSIX layer on Windows)
12427    
12428    https://bugs.freedesktop.org/show_bug.cgi?id=26839
12429    
12430    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12431
12432commit a3f5f1b90936d23e9894e3261b2d77fb7b32a51a
12433Author: Josh Triplett <josh@freedesktop.org>
12434Date:   Thu Apr 15 14:24:21 2010 -0700
12435
12436    Stop returning an int from _XIDHandler and _XSeqSyncFunction
12437    
12438    _XIDHandler and _XSeqSyncFunction originally ran from dpy->synchandler, and
12439    thus had to return an int.  Now, they only run from _XPrivSyncHandler or
12440    LockDisplay, neither of which needs to check their return value since they
12441    always returned 0.  Make them both void.
12442    
12443    Signed-off-by: Josh Triplett <josh@freedesktop.org>
12444    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12445
12446commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43
12447Author: Jamey Sharp <jamey@minilop.net>
12448Date:   Thu Apr 15 13:05:08 2010 -0700
12449
12450    Move XID and sync handling from SyncHandle to LockDisplay to fix races.
12451    
12452    XID and sync handling happened via _XPrivSyncHandler, assigned to
12453    dpy->synchandler and called from SyncHandle.  _XPrivSyncHandler thus ran
12454    without the Display lock, so manipulating the Display caused races, and
12455    these races led to assertions in multithreaded code (demonstrated via
12456    ico).
12457    
12458    In the XTHREADS case, after you've called XInitThreads, we can hook
12459    LockDisplay and UnlockDisplay.  Use that to run _XIDHandler and
12460    _XSeqSyncHandler from LockDisplay rather than SyncHandle; we then know
12461    that we hold the lock, and thus we can avoid races.  We think it makes
12462    sense to do these both from LockDisplay rather than UnlockDisplay, so
12463    that you know you have valid sync and a valid XID before you start
12464    setting up the request you locked to prepare.
12465    
12466    In the !XTHREADS case, or if you haven't called XInitThreads, you don't
12467    get to use Xlib from multiple threads, so we can use the logic we have
12468    now (with synchandler and savedsynchandler) without any concern about
12469    races.
12470    
12471    This approach gets a bit exciting when the XID and sequence sync
12472    handlers drop and re-acquire the Display lock. Reacquisition will re-run
12473    the handlers, but they return immediately unless they have work to do,
12474    so they can't recurse more than once.  In the worst case, if both of
12475    them have work to do, we can nest the Display lock three deep.  In the
12476    case of the _XIDHandler, we drop the lock to call xcb_generate_id, which
12477    takes the socket back if it needs to request more XIDs, and taking the
12478    socket back will reacquire the lock; we take care to avoid letting
12479    _XIDHandler run again and re-enter XCB from the return_socket callback
12480    (which causes Very Bad Things, and is Not Allowed).
12481    
12482    Tested with ico (with 1 and 20 threads), and with several test programs
12483    for XID and sequence sync.  Tested with and without XInitThreads(), and
12484    with and without XCB.
12485    
12486    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=23192
12487    
12488    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12489    Signed-off-by: Josh Triplett <josh@freedesktop.org>
12490
12491commit b089b53b697c2851db2985d32af3b29f1da5e31e
12492Author: Jamey Sharp <jamey@minilop.net>
12493Date:   Wed Apr 14 12:59:11 2010 -0700
12494
12495    Honest. Extensions get to filter async errors too.
12496    
12497    Under some circumstances, errors are picked up by calling
12498    xcb_poll_for_reply, rather than xcb_poll_for_event, because Xlib issued
12499    the requests with the XCB_REQUEST_CHECKED flag. That happens when either
12500    an async handler is queued at the time the requests are flushed, or when
12501    XSetEventQueueOwner has been used to prevent Xlib from processing XCB's
12502    event queue.
12503    
12504    This bugfix extends 405132dab64bf2375f8e57d02b1b53da2311933a to cover
12505    those cases.
12506    
12507    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545
12508    
12509    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12510
12511commit a15c31274650e391bc6de5d0951eb4464c228139
12512Author: Jamey Sharp <jamey@minilop.net>
12513Date:   Wed Apr 14 12:10:34 2010 -0700
12514
12515    _XError already runs async handlers; only call them directly for replies.
12516    
12517    The previous behavior probably would have triggered bug reports someday.
12518    
12519    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12520
12521commit 405132dab64bf2375f8e57d02b1b53da2311933a
12522Author: Jamey Sharp <jamey@minilop.net>
12523Date:   Tue Apr 13 12:12:36 2010 -0700
12524
12525    Extensions get to filter async errors too.
12526    
12527    Apparently I misread XlibInt.c:_XReply and thought that handlers set
12528    with XESetError should be consulted only for the sequence number that
12529    _XReply is currently looking for. In fact, the error handlers are also
12530    consulted when an error arrives for a request that was not expected to
12531    have a reply.
12532    
12533    However, in an odd twist, the error handlers are *not* consulted outside
12534    of _XReply--that is, when looking for events, or waiting to be able to
12535    flush the output queue. So this patch takes some small pains to preserve
12536    that behavior, though it seems likely to have been unintentional.
12537    
12538    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545
12539    
12540    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12541
12542commit d3eab4a38f3e53ef21eb3b6fa66ead7afecf6227
12543Author: Jamey Sharp <jamey@minilop.net>
12544Date:   Tue Apr 13 12:49:59 2010 -0700
12545
12546    Prefer the xcb_generic_error_t we already have over casting to xEvent.
12547    
12548    Just a minor cleanup.
12549    
12550    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12551
12552commit 75ea8c37935ccc911557d16a303ba595b8ab106b
12553Author: Jamey Sharp <jamey@minilop.net>
12554Date:   Mon Apr 12 11:30:20 2010 -0700
12555
12556    Run the user's synchandler as well as any internal synchandlers.
12557    
12558    Fixes https://bugs.freedesktop.org/show_bug.cgi?id=27595
12559    
12560    Signed-off-by: Jamey Sharp <jamey@minilop.net>
12561
12562commit c3f3e4a9e531d010312c97e753d6e543e607094d
12563Author: Kusanagi Kouichi <slash@ac.auone-net.jp>
12564Date:   Fri Feb 12 17:16:10 2010 +0900
12565
12566    Fix manual typos.
12567    
12568    Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
12569    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12570
12571commit 75c6c5595408caba9ef8233839d77021d02f86f2
12572Author: Dirk Wallenstein <dirkwallenstein@t-online.de>
12573Date:   Fri Feb 5 20:42:46 2010 +0100
12574
12575    man: Correct the XkbAllAccessXEventsMask mask name
12576    
12577    This mask has probably been renamed but not been updated in the manuals.
12578    
12579    Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de>
12580    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12581
12582commit 9262aceaff24c954cab8c252690032bc2ec25e7d
12583Author: Dirk Wallenstein <dirkwallenstein@t-online.de>
12584Date:   Fri Feb 5 20:42:45 2010 +0100
12585
12586    man: Add missing geometry component flag
12587    
12588    Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de>
12589    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12590
12591commit 50cf134465f6fd263f37acecac94518468ef1c6d
12592Author: Dirk Wallenstein <halsmit@t-online.de>
12593Date:   Tue Feb 2 21:24:32 2010 +0100
12594
12595    man: Return value of XkbGetState is Status and not Bool
12596    
12597    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
12598    Reviewed-by: Julien Cristau <jcristau@debian.org>
12599    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12600
12601commit 7a93ae2bb5292a97fdd27ed818bc81248b37f641
12602Author: Dirk Wallenstein <halsmit@t-online.de>
12603Date:   Tue Feb 2 21:24:30 2010 +0100
12604
12605    man: Fix return value specification of XkbKeyActionEntry
12606    
12607    The XkbKeyActionEntry macro expands to a pointer.
12608    
12609    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
12610    Reviewed-by: Julien Cristau <jcristau@debian.org>
12611    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12612
12613commit e9884d4a05a5661ec343ea8a2aa0562b6419e086
12614Author: Dirk Wallenstein <halsmit@t-online.de>
12615Date:   Tue Feb 2 21:24:29 2010 +0100
12616
12617    man: Redirect users from XKeycodeToKeysym to XkbKeycodeToKeysym #25732
12618    
12619    XKeycodeToKeysym keeps compatibility with pre-XKB and thus only sees 2
12620    groups with 2 levels each. It wraps the index into the next group.
12621    This behavior confuses the unaware user, and therefore this will add a
12622    reference to XkbKeycodeToKeysym in the corresponding man paragraph.
12623    
12624    Another bug had that issue, too. #5349
12625    
12626    Signed-off-by: Dirk Wallenstein <halsmit@t-online.de>
12627    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12628
12629commit 6474cdf97cd648afaeaa872155f3fa21cfb43f80
12630Author: Alan Coopersmith <alan.coopersmith@oracle.com>
12631Date:   Thu Apr 8 19:59:55 2010 -0700
12632
12633    Fix typo in new fi_FI.UTF-8 that was reported by "make check"
12634    
12635    Unrecognized pattern in Compose on line #154:
12636      <dead_diaeresis> <space>                " "¨"
12637    
12638    Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
12639
12640commit 91bd5e7af2d3a9526aefb73dc10f417896f0e4dd
12641Author: Marko Myllynen <myllynen@redhat.com>
12642Date:   Thu Apr 8 19:52:42 2010 -0700
12643
12644    Bug 27465 - Rewritten fi_FI.UTF-8 Compose file
12645    
12646    I've rewritten the fi_FI.UTF-8 Compose file so that it
12647    
12648    - includes en_US.UTF-8/Compose for base compose definitions (and thus gets any
12649    possible additions to en_US.UTF-8/Compose automatically included)
12650    - overrides any en_US.UTF-8/Compose definitions with fi_FI.UTF-8/Compose
12651    definitions (thus making it safe to include en_US.UTF-8/Compose)
12652    - lists all the definitions specified in the SFS 5966 standard (thus making it
12653    easy to verify that the implementation is valid and complete)
12654    - adds a header about the file itself as required by the standard
12655    
12656    The rewritten version is ~320 lines compared to over 5000 lines of the previous
12657    version.
12658    
12659    https://bugs.freedesktop.org/show_bug.cgi?id=27465
12660
12661commit f6a4fd0c7615684d08e848245849dea4017a5214
12662Author: Gaetan Nadon <memsize@videotron.ca>
12663Date:   Thu Apr 1 21:46:12 2010 -0400
12664
12665    config: update and relocate AC_DEFINE_DIR macro
12666    
12667    Remove deprecated acinclude.m4 macro container file
12668    Use separate macro files as per autoconf recommendation
12669    Use the latest macro from GNU (ax) which replaces
12670    the non-gnu version (ac)
12671    This preserves the Autoconf macro AC namespace.
12672    
12673    Also moved out of acinclude.m4 is the DOLT macro
12674    
12675    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12676
12677commit 61d5231db418cec51bd718633c3bba39b18689d6
12678Author: Gaetan Nadon <memsize@videotron.ca>
12679Date:   Mon Mar 29 14:53:48 2010 -0400
12680
12681    config: remove the pkgconfig pc.in file from EXTRA_DIST
12682    
12683    Automake always includes it in the tarball.
12684    
12685    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12686
12687commit 69fc76a0321c6b25cc4286173c53435c69ab8e0e
12688Author: Fernando Carrijo <fcarrijo@yahoo.com.br>
12689Date:   Wed Mar 17 23:53:16 2010 -0300
12690
12691    No need for req->firstKeyCode to be set twice.
12692    
12693    There is no reason to set req->firstKeyCode twice when a client, wishful
12694    for changing keyboard mappings, calls XChangeKeyboardMapping. This patch
12695    fixes the mistake by making no functional changes to libX11 whatsoever.
12696    
12697    Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br>
12698    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
12699
12700commit b5797597f83ee62228abfb88f026ef92163e07a5
12701Author: Gaetan Nadon <memsize@videotron.ca>
12702Date:   Tue Feb 16 10:37:21 2010 -0500
12703
12704    config: move CWARNFLAGS from configure.ac to Makefile.am
12705    
12706    Compiler warning flags should be explicitly set in the makefile
12707    rather than being merged with other packages compiler flags.
12708    
12709    Reviewed-by: Dan Nicholson <dbn.lists@gmail.com>
12710    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12711
12712commit 052c4d7995b0e52a038933beb85cd544571c6b7e
12713Author: Gaetan Nadon <memsize@videotron.ca>
12714Date:   Tue Feb 16 14:28:21 2010 -0500
12715
12716    specs: change install cmd due to automake 1.11
12717    
12718    specData_INSTALL is defined in 1.9 and 1.10 but not 1.11
12719    
12720    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12721
12722commit 7277a18cda9f5a6807d11cf4ceb9449b1c302c18
12723Author: Gaetan Nadon <memsize@videotron.ca>
12724Date:   Tue Feb 9 17:27:08 2010 -0500
12725
12726    doc: use $(mkdir_p) rather than $(MKDIR_P) due to automake 1.9.6
12727    
12728    $(MKDIR_P) is not defined in automake 1.9.
12729    
12730    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12731
12732commit e66eda95b520727969d721416d4ea49b54ef2b03
12733Author: Gaetan Nadon <memsize@videotron.ca>
12734Date:   Tue Feb 2 16:29:26 2010 -0500
12735
12736    specs: install html images in $docdir with html files
12737    
12738    The images required by the html files have been omitted.
12739    
12740    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12741
12742commit 137d1ff6be99d040e77c8c4c52029c5bdf1f4808
12743Author: Gaetan Nadon <memsize@videotron.ca>
12744Date:   Sun Jan 31 14:16:20 2010 -0500
12745
12746    doc: use new macros to control doc generation
12747    
12748    Namely XORG_WITH_GROFF for the groff generation tool
12749    XORG_WITH_PS2PDF for the conversion of PS docs to PDF
12750    XORG_ENABLE_SPECS for the generation of specs
12751    
12752    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12753
12754commit 9ca583c5d6fcf9cd8151ee7b019630b141d32fab
12755Author: Gaetan Nadon <memsize@videotron.ca>
12756Date:   Tue Jan 26 19:41:08 2010 -0500
12757
12758    doc: clean-up generated html images
12759    
12760    Generate images in /images as is the convention
12761    Provide a base file name for images rather than process ID
12762    Remove images directory when running make clean
12763    
12764    Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
12765
12766commit eb289b34bfc9978eef724e0f268975938334c06c
12767Author: Kristian Høgsberg  <krh@bitplanet.net>
12768Date:   Thu Jan 21 13:39:17 2010 -0500
12769
12770    XErrorDB: Add new DRI2 request names
12771
12772commit 6babf2123d05adb9349394c49cc2d81d1f66cf9e
12773Author: Julien Cristau <jcristau@debian.org>
12774Date:   Sat Jan 16 22:47:03 2010 +0000
12775
12776    man: Fix typo in Makefile
12777    
12778    Missing line continuation was preventing the XCompose alias from being
12779    generated.
12780    
12781    Signed-off-by: Julien Cristau <jcristau@debian.org>
12782    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12783
12784commit 68530599e61cf9fef3fcc91650e7cef21837fce7
12785Author: Alan Coopersmith <alan.coopersmith@sun.com>
12786Date:   Sat Jan 16 11:41:32 2010 -0800
12787
12788    libX11 1.3.3
12789    
12790    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12791
12792commit 1e5af224b9fbdb154620f898558704f2bc4bf4e9
12793Author: Alan Coopersmith <alan.coopersmith@sun.com>
12794Date:   Thu Jan 14 18:33:39 2010 -0800
12795
12796    Update COPYING file with actual licenses from libX11 code & docs
12797    
12798    COPYING was previously a years out-of-date copy of the LICENSES
12799    doc summarizing all licenses for all XFree86 components, many of
12800    which did not apply to libX11.
12801    
12802    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12803
12804commit ddb1786720d6136b1b28be599c75c10ad1b76a2a
12805Author: Alan Coopersmith <alan.coopersmith@sun.com>
12806Date:   Thu Jan 14 17:38:26 2010 -0800
12807
12808    Purge CVS/RCS id tags
12809    
12810    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12811
12812commit c1db9ddcfa7091f173478dca45bf720badedfca4
12813Author: Alan Coopersmith <alan.coopersmith@sun.com>
12814Date:   Thu Jan 14 15:39:14 2010 -0800
12815
12816    Update Sun license notices to current X.Org standard form
12817    
12818    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12819
12820commit 1aaa7c0358c5c035b99625f5715cb722bc2d2a2f
12821Author: Alan Coopersmith <alan.coopersmith@sun.com>
12822Date:   Thu Jan 14 08:28:22 2010 -0800
12823
12824    Add %S substitutions to Compose man page
12825    
12826    Commit 9df349a7894725f9469b106af645f57f7f3f9af3 added support for %S
12827    in Compose file include statements - this documents it in the Compose
12828    file man page.
12829    
12830    Also changes the existing substitution documentation to list format
12831    instead of a paragraph format to allow easier adding of %S.
12832    
12833    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12834
12835commit 94b45db93058cffa25598fe27dd3e385ace9d305
12836Author: Richard Purdie <rpurdie@linux.intel.com>
12837Date:   Thu Jan 14 08:04:28 2010 -0800
12838
12839    Fix configure.ac PKG_ macro calls
12840    
12841    If the first call to PKG_CHECK_MODULES() may not be called during execution
12842    which is the case here since its inside a case statement, call
12843    PKG_PROG_PKG_CONFIG() to ensure things worked correctly.
12844    
12845    Fixes a configure failure introduced by commit
12846    87529c039050ce3336ff9ce00f1b5a21d15690da when xcb is enabled.
12847    
12848    http://bugs.freedesktop.org/show_bug.cgi?id=26041
12849    
12850    Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
12851    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12852
12853commit df559d8bc2238a2c369152fcb06e5b5ab132a5dc
12854Author: Alan Coopersmith <alan.coopersmith@sun.com>
12855Date:   Tue Jan 12 09:18:15 2010 -0800
12856
12857    Raise xorg-macros requirement to 1.4
12858    
12859    Needed since the changes in 464390f16d7ed4aa5bf80f89863ba92273075ec2
12860    depend on the INSTALL file delivered in xorg-macros 1.4
12861    
12862    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12863
12864commit 9df349a7894725f9469b106af645f57f7f3f9af3
12865Author: James Cloos <cloos@jhcloos.com>
12866Date:   Wed Jan 13 11:04:38 2010 -0500
12867
12868    Allow inclusion of system level compose files.
12869    
12870    With the release of XFree86 4.4 an inclusion system of compose files was
12871    added to Xlib to allow inclusion of the default compose file (with %L),
12872    any compose files from user’s home directory (with %H), or a compose
12873    file with a hard coded path¹. However, even today including system level
12874    compose files is not possible in a platform independent manner although
12875    the machinery for including compose files and overriding previously
12876    defined compositions is already in place.
12877    
12878    With the ability to include system level compose files one could greatly
12879    reduce the need for compose file rule duplication and the work needed to
12880    propagate changes in one compose file to others. For example, currently
12881    the Finnish compose file fi_FI.UTF-8 weights over 5000 lines² but it
12882    is almost identical with en_US.UTF-8 except for perhaps half a dozen
12883    compositions.
12884    
12885    This commit allows one to include system level compose files with the
12886    following kind of syntax:
12887    
12888    include "%S/en_US.UTF-8/Compose"
12889    
12890    1] http://www.xfree86.org/4.4.0/RELNOTES5.html#42
12891    2] http://cgit.freedesktop.org/xorg/lib/libX11/tree/nls/fi_FI.UTF-8/Compose.pre
12892    
12893    Signed-off-by: Marko Myllynen <myllynen@redhat.com>
12894    Signed-off-by: James Cloos <cloos@jhcloos.com>
12895
12896commit ccf21a8877f40136e25f2f62d0668b0c76b3cfbd
12897Author: Thien-Thi Nguyen <ttn@gnuvola.org>
12898Date:   Tue Dec 1 10:31:47 2009 +0100
12899
12900    libX11: Fix comment: Invert polarity (direction) of reformat description.
12901    
12902    Signed-off-by: Thien-Thi Nguyen <ttn@gnuvola.org>
12903    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12904
12905commit 4d4676c0d6324d2ff9d00d15ab485fa5d848a369
12906Author: Osamu Sayama <osamu.sayama@sun.com>
12907Date:   Tue Jan 5 17:26:40 2010 -0800
12908
12909    set_fontset_extents crash after get_rotate_fontname fix in 2bef065b70f7
12910    
12911    - In set_fontset_extents, check font_data is not NULL before running
12912      loop that may increment it to a non-NULL invalid pointer.
12913    - Make sure get_rotate_fontname counts the final field
12914    
12915    Fixes OpenSolaris bug 6913809: X application dumps core in ja_JP.UTF-8 locale
12916    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6913809>
12917    
12918    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12919    Reviewed-by: Adam Jackson <ajax@redhat.com>
12920
12921commit 87529c039050ce3336ff9ce00f1b5a21d15690da
12922Author: Alan Coopersmith <alan.coopersmith@sun.com>
12923Date:   Tue Jan 5 18:02:37 2010 -0800
12924
12925    Merge X11, XKBPROTO, & XPROTO pkg-config lists
12926    
12927    Since the XPROTO_CFLAGS & XKBPROTO_CFLAGS are just merged into X11_CFLAGS
12928    in configure.ac anyway, might as well combine the lists passed to
12929    PKG_CHECK_MODULES to reduce duplication in the flags.
12930    
12931    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12932    Reviewed-by: Rémi Cardona <remi@gentoo.org>
12933    Acked-by: Peter Hutterer <peter.hutterer@who-t.net>
12934
12935commit 2dd053667b44c55e2bf601aec7b75fc6fee1cf44
12936Author: Alan Coopersmith <alan.coopersmith@sun.com>
12937Date:   Tue Jan 5 17:56:58 2010 -0800
12938
12939    Add XCompose man page shadow for Compose man page
12940    
12941    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12942
12943commit 02543bba816e065c02cd36e0c9e21519a9ee9c96
12944Author: Alan Coopersmith <alan.coopersmith@sun.com>
12945Date:   Tue Jan 5 17:47:44 2010 -0800
12946
12947    Add $(AM_V_GEN) to silence lint rules too
12948    
12949    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12950
12951commit db7c6fdeeaef9475458498e4cf09d6b1329e9aa3
12952Author: Alan Coopersmith <alan.coopersmith@sun.com>
12953Date:   Tue Jan 5 17:42:42 2010 -0800
12954
12955    Remove GCC_WARNINGS now that XORG_CWARNFLAGS sets them for us
12956    
12957    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
12958
12959commit 6b4e526e7ba105622f6b143e8e2639b44cb6cc19
12960Author: Gaetan Nadon <memsize@videotron.ca>
12961Date:   Sun Nov 29 10:44:24 2009 -0500
12962
12963    configure.ac: restore correct order for XORG_DEFAULT_OPTIONS
12964    
12965    It must be following AC_USE_SYSTEM_EXTENSION
12966
12967commit 3dbaa11d1c8705e1b309c6686f354766b651ad20
12968Author: Gaetan Nadon <memsize@videotron.ca>
12969Date:   Sun Nov 29 10:01:44 2009 -0500
12970
12971    Add .gitignore in /specs for generated files
12972    
12973    Being in /specs, it will apply for all 3 subdirs
12974
12975commit f45d39d37aab04742e44cd8c3c993aad7587e40c
12976Author: Gaetan Nadon <memsize@videotron.ca>
12977Date:   Fri Nov 27 20:56:03 2009 -0500
12978
12979    Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES
12980    
12981    Now that the INSTALL file is generated.
12982    Allows running make maintainer-clean.
12983
12984commit 464390f16d7ed4aa5bf80f89863ba92273075ec2
12985Author: Gaetan Nadon <memsize@videotron.ca>
12986Date:   Wed Oct 28 14:09:10 2009 -0400
12987
12988    INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206
12989    
12990    Add missing INSTALL file. Use standard GNU file on building tarball
12991    README may have been updated
12992    Remove AUTHORS file as it is empty and no content available yet.
12993    Remove NEWS file as it is empty and no content available yet.
12994
12995commit 6360e7f0ce6a75da2bed33ede7fea783a1fb80e1
12996Author: Gaetan Nadon <memsize@videotron.ca>
12997Date:   Tue Oct 27 15:07:25 2009 -0400
12998
12999    Deploy the new XORG_DEFAULT_OPTIONS #24242
13000    
13001    This macro aggregate a number of existing macros that sets commmon
13002    X.Org components configuration options. It shields the configuration file from
13003    future changes.
13004
13005commit d02f943c98fde2f14319bc57fd9ad77eb9a2a572
13006Author: Gaetan Nadon <memsize@videotron.ca>
13007Date:   Mon Oct 26 22:08:42 2009 -0400
13008
13009    Makefile.am: ChangeLog not required: EXTRA_DIST or *CLEANFILES #24432
13010    
13011    ChangeLog filename is known to Automake and requires no further
13012    coding in the makefile.
13013
13014commit f77c89c751a6a63c8ef11ecdddca2aed11ff6e29
13015Author: Gaetan Nadon <memsize@videotron.ca>
13016Date:   Thu Oct 22 12:34:19 2009 -0400
13017
13018    .gitignore: use common defaults with custom section # 24239
13019    
13020    Using common defaults will reduce errors and maintenance.
13021    Only the very small or inexistent custom section need periodic maintenance
13022    when the structure of the component changes. Do not edit defaults.
13023
13024commit 46c7b0e9d0d85cf5ccc0d40d19821fcc3335503b
13025Author: Alan Coopersmith <alan.coopersmith@sun.com>
13026Date:   Thu Nov 12 23:05:40 2009 -0800
13027
13028    Add compose-check.pl to EXTRA_DIST
13029    
13030    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13031
13032commit 54c64267cc8bc98641cc39a22cb7bd71673e89e0
13033Author: Julien Cristau <jcristau@debian.org>
13034Date:   Thu Oct 29 17:17:24 2009 +0100
13035
13036    man: fix XCopyGC argument order
13037    
13038    Ubuntu bug#408337
13039
13040commit aad10032651cdc2a53b359035954454a28d6db67
13041Author: Alan Coopersmith <alan.coopersmith@sun.com>
13042Date:   Fri Oct 23 13:55:14 2009 -0700
13043
13044    libX11 1.3.2
13045    
13046    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13047
13048commit 6303ada89cb228c0f9656036f798703afb42fc29
13049Author: Peter Hutterer <peter.hutterer@who-t.net>
13050Date:   Wed Oct 21 12:42:07 2009 +1000
13051
13052    Add smiley faces to compose sequences.
13053    
13054    I wonder how we could have lasted that long without them.
13055    
13056    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13057    Acked-by: Daniel Stone <daniel@fooishbar.org>
13058    Acked-By: James Cloos <cloos@jhcloos.com>
13059
13060commit 0e104ebd8628803c27e36b16922ad1edd891325a
13061Author: Alan Coopersmith <alan.coopersmith@sun.com>
13062Date:   Thu Oct 22 23:12:30 2009 -0700
13063
13064    Add man page for Compose file format
13065    
13066    Based on grammar description in modules/im/ximcp/imLcPrs.c and
13067    note on XFree86 changes formerly found in xorg-docs RELNOTES.sgml
13068    
13069    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13070
13071commit 9c95f2af7c442b3a59b1a30cf804f1ef4e7fc5b5
13072Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
13073Date:   Tue Oct 20 12:46:03 2009 -0700
13074
13075    Add extra configuration and sanity checks for groff and ps2pdf
13076    
13077    1) Add AC_ARG_VAR for GROFF and PS2PDF to inform users of these
13078       environment variables.
13079    2) Check that groff -ms works
13080    
13081    Some distributions ship the ms macros as a separate package which may
13082    not be installed together with groff, so we need to make sure that groff
13083    works and the required macros are actually installed before attempting
13084    to build the specs.
13085    
13086    Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org>
13087    Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
13088    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13089
13090commit d3f801fd2f9198eaad6797414dba652f9c006c6d
13091Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13092Date:   Sun Oct 18 17:34:53 2009 -0500
13093
13094    Fix VPATH build of libX11 specs
13095    
13096    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13097
13098commit ad15e1a89d30ccc11d80745897b83def1448e8c0
13099Author: Alan Coopersmith <alan.coopersmith@sun.com>
13100Date:   Sat Oct 17 16:14:34 2009 -0700
13101
13102    libX11 1.3.1
13103    
13104    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13105
13106commit 0cbf98c17a7484bb58b2464b98d63bb3b4ea2594
13107Author: Alan Coopersmith <alan.coopersmith@sun.com>
13108Date:   Sat Oct 17 16:28:25 2009 -0700
13109
13110    Fix make distcheck
13111    
13112    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13113
13114commit 082e62ad268ef16a3bebc5a3c9fa008dbdc483ed
13115Author: Alan Coopersmith <alan.coopersmith@sun.com>
13116Date:   Fri Oct 16 19:25:35 2009 -0700
13117
13118    Use $(AM_V_GEN) to silence echo commands for generating shadow man pages
13119    
13120    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13121
13122commit 4e66da0783b2e5e3b288aaecd3c89396ed425c20
13123Author: Alan Coopersmith <alan.coopersmith@sun.com>
13124Date:   Wed Oct 14 16:18:24 2009 -0700
13125
13126    Move libX11 & XIM/locale specs from xorg-docs
13127    
13128    If groff is found, and --disable-specs is not passed to configure,
13129    specs will be converted to text, html and ps (or pdf if ps2pdf is
13130    found) and installed to $(docdir)
13131    
13132    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13133
13134commit 5d3d817a42ddcc8d0c6efd33efd1442fe14f5c6b
13135Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13136Date:   Tue Oct 13 19:15:51 2009 -0500
13137
13138    Provide _Xsetlocale compat wrappers on Cygwin
13139    
13140    Previous versions of Cygwin did not have proper locale support, so Cygwin/X
13141    defined X_LOCALE, using _Xsetlocale instead.  Cygwin 1.7 has added locale
13142    support, but we can't remove the _Xsetlocale entry point without breaking
13143    ABI.
13144    
13145    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13146
13147commit 2c8b3a877a713bb66a6316a7051b43c46af6e1a0
13148Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13149Date:   Tue Oct 13 19:15:49 2009 -0500
13150
13151    dolt: add Cygwin to supported platforms
13152    
13153    Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
13154
13155commit a2c8e3e34b15b57ff881a52101fc961a602f35e4
13156Author: Alan Coopersmith <alan.coopersmith@sun.com>
13157Date:   Wed Oct 14 13:23:30 2009 -0700
13158
13159    Recognize XSUNBUFFERSIZE alias for XLIBBUFFERSIZE on Solaris
13160    
13161    Also fix indenting of the XLIBBUFFERSIZE code to match surrounding code
13162    
13163    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13164
13165commit 34ddfca7b7d58240b0fe74bec6b2e0652d15c051
13166Author: Jon TURNEY <jon.turney@dronecode.org.uk>
13167Date:   Tue Feb 10 17:47:25 2009 +0000
13168
13169    Include sys/select.h for select() and struct timeval, if it exists
13170    
13171    This is a cygwin build fix
13172
13173commit 383165916ddac91740d4c780174d4c0d07cdb994
13174Author: Xake <xake@rymdraket.net>
13175Date:   Sun Sep 27 11:16:36 2009 +0200
13176
13177    Use AM_V_GEN instead of customized macros for AM_SILENT_RULES
13178    
13179    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13180
13181commit 854269d55cdda8caf425515bfed2855e211a5ada
13182Author: Julien Cristau <jcristau@debian.org>
13183Date:   Tue Oct 6 16:11:24 2009 +0200
13184
13185    configure: quote argument to m4_pattern_forbid
13186    
13187    Without this, configure spits out
13188    ../configure: line 12364: ac_fn_c_check_member: command not found
13189    ../configure: line 12378: ac_fn_c_check_type: command not found
13190    
13191    Also anchor the pattern to make it stricter.
13192    
13193    Signed-off-by: Julien Cristau <jcristau@debian.org>
13194
13195commit 3bb020587ce74e0737ec7aceb20041f1e77d3b87
13196Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
13197Date:   Thu Oct 1 22:20:38 2009 -0700
13198
13199    Split CFLAGS into CPPFLAGS and CFLAGS
13200    
13201    On some build systems, CPPFLAGS is set to "-I/some/prefix/include".  If older
13202    X11 headers are in /some/prefix/include, they will be preferred over the
13203    shipped headers.  This corrects that problem.
13204
13205commit d54caf1c9c55af8247621b7ba6afb20b23699839
13206Author: Peter Hutterer <peter.hutterer@who-t.net>
13207Date:   Fri Oct 2 10:59:08 2009 +1000
13208
13209    libX11 1.3
13210    
13211    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13212
13213commit dd201bcf9e7f8863b7ef32273a5ef021678133c1
13214Author: Peter Hutterer <peter.hutterer@who-t.net>
13215Date:   Fri Oct 2 12:10:29 2009 +1000
13216
13217    nls: remove duplicate Compose sequences from pt_BR.UTF-8
13218    
13219    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13220
13221commit a293ae9e83739067fc92db1a39c262857bc2283e
13222Author: Peter Hutterer <peter.hutterer@who-t.net>
13223Date:   Fri Sep 25 11:19:41 2009 +1000
13224
13225    Add XF86TouchpadToggle to XKeysymDB
13226    
13227    Lenovo laptops provide a key to enable or disable the touchpad and the
13228    trackstick. This key is usually located on Fn + F8.
13229    
13230    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13231    Acked-by: Adam Jackson <ajax@redhat.com>
13232
13233commit 69839f8903a24eab08f17a781b3797fb64dce9cf
13234Author: Alan Coopersmith <alan.coopersmith@sun.com>
13235Date:   Sun Sep 27 10:34:16 2009 -0700
13236
13237    Bug 24173: libX11 from git fails to build with automake older then 1.11
13238    
13239    AM_CONDITIONAL must come *before* the AC_OUTPUT that creates the
13240    Makefiles, instead of after.
13241    <http://bugs.freedesktop.org/show_bug.cgi?id=24173>
13242    
13243    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13244
13245commit f5effd041f58ef07703cca2b4f396758811e1eec
13246Author: Alan Coopersmith <alan.coopersmith@sun.com>
13247Date:   Wed Apr 15 10:59:23 2009 -0700
13248
13249    Resolve conflicting Compose sequences in iso8859-2, el_GR.UTF-8 & pt_BR.UTF-8
13250    
13251    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13252
13253commit 3843778358d3a0cd6a2d07dba5dd061248053ac9
13254Author: Alan Coopersmith <alan.coopersmith@sun.com>
13255Date:   Wed Apr 15 10:56:09 2009 -0700
13256
13257    Add perl script to check for duplicate or conflicting compose file entries
13258    
13259    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13260
13261commit 19cc5e1fa17a285045662820a8b4de2a0f9a194d
13262Author: Alan Coopersmith <alan.coopersmith@sun.com>
13263Date:   Fri Sep 18 17:10:04 2009 -0700
13264
13265    Use make rules instead of shell for loops to generate shadow man pages
13266    
13267    Allows parallel make and simpler build logs/error reporting
13268    
13269    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13270
13271commit 7dabcac973d0b6692a3cd62bd6d8e0467b24200b
13272Author: Alan Coopersmith <alan.coopersmith@sun.com>
13273Date:   Fri Sep 18 16:58:53 2009 -0700
13274
13275    Add AM_SILENT_RULES support for cpp rules for man & nls files
13276    
13277    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13278
13279commit bfa19cddd8546b4930a773f3bbd81137c362d82b
13280Author: Alan Coopersmith <alan.coopersmith@sun.com>
13281Date:   Fri Sep 18 16:58:16 2009 -0700
13282
13283    Update to using xorg-macros 1.3 & XORG_DEFAULT_OPTIONS
13284    
13285    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13286
13287commit 51396066c8003a96a6399c9a4bed2a81e512b582
13288Author: Mikko Niskanen <mikko.niskanen@iki.fi>
13289Date:   Fri Sep 4 10:11:15 2009 +1000
13290
13291    Fix wrong typedef on HP-UX (#18998)
13292    
13293    shl_dt doesn't exist, the type is shl_t.
13294    
13295    X.Org Bug 18998 <http://bugs.freedesktop.org/show_bug.cgi?id=18998>
13296    
13297    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13298
13299commit 615220a312b9430a580fe6dcf51703c6ef244f66
13300Author: Paul Bender <pebender@gmail.com>
13301Date:   Fri Sep 4 09:57:10 2009 +1000
13302
13303    Don't require xdmcp in configure.ac (#22583)
13304    
13305    X.Org Bug 22583 <http://bugs.freedesktop.org/show_bug.cgi?id=22583>
13306    
13307    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13308
13309commit 20f9ecd86ad2a9ac6966f7eee32234cb5ef77c29
13310Author: Peter Hutterer <peter.hutterer@who-t.net>
13311Date:   Thu Sep 3 14:02:44 2009 +1000
13312
13313    man: fix parameters to XkbAllocGeomOverlay{Rows|Keys} (#23499)
13314    
13315    X.Org Bug 23499 <http://bugs.freedesktop.org/show_bug.cgi?id=23499>
13316    
13317    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13318
13319commit c2814a614dc4d9409bfa0f44c18bfd69ad7f7b85
13320Author: Peter Hutterer <peter.hutterer@who-t.net>
13321Date:   Thu Sep 3 12:05:25 2009 +1000
13322
13323    man: XQueryTree may return BadWindow. (#23416)
13324    
13325    X.Org Bug 23416 <http://bugs.freedesktop.org/show_bug.cgi?id=XXX>
13326    
13327    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13328
13329commit dbe98d456ccc6eeca9fa3e241a3db0a4d83c5a65
13330Author: Alan Coopersmith <alan.coopersmith@sun.com>
13331Date:   Fri Aug 28 23:07:58 2009 +0800
13332
13333    Fix version tag in .TH line of several XKB man pages
13334    
13335    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13336
13337commit bf24400936c10af6f5aa0c75cfe2207ab9b680b4
13338Author: Alan Coopersmith <alan.coopersmith@sun.com>
13339Date:   Fri Aug 28 23:04:38 2009 +0800
13340
13341    XkbSetDeviceButtonActions.man: remove non-existent actions argument
13342    
13343    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13344
13345commit 53affa9335116f2d367f041e6502a411d4619e47
13346Author: Alan Coopersmith <alan.coopersmith@sun.com>
13347Date:   Fri Aug 28 23:00:17 2009 +0800
13348
13349    XkbQueryExtension.man: Arguments should be pointers
13350    
13351    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13352
13353commit 6233948885acc5873a5abddfff235afec555f3c2
13354Author: Alan Coopersmith <alan.coopersmith@sun.com>
13355Date:   Fri Aug 28 22:53:03 2009 +0800
13356
13357    XkbSAActionSetCtrls.man: Fix typo in formatting macro
13358    
13359    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13360
13361commit 28a9ca57cdec4aa9ca92322f963b01f0c2daf47a
13362Author: Alan Coopersmith <alan.coopersmith@sun.com>
13363Date:   Fri Aug 28 22:49:31 2009 +0800
13364
13365    Convert Xkb API man pages to ANSI prototypes
13366    
13367    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13368
13369commit fa2eecca85baab9616f7143cc1a853a594b4a90c
13370Author: James Cloos <cloos@jhcloos.com>
13371Date:   Mon Aug 24 06:35:17 2009 -0400
13372
13373    Add some (Serbian) Cyrillic NFD sequences.
13374    
13375    A number of characters in use in the various countries which use the
13376    Cyrillic script do not appear as pre-composed characters in The UCS
13377    or Unicode; they are only available as combining-character sequences.
13378    
13379    This commit adds support for using (prefix) dead keys and Multi_key-
13380    initiated sequences to enter a number of these combining-character
13381    sequences.  This ensures that users can enter these scripts even
13382    when using the current Cyrillic keymaps, which lack support for
13383    the combining characters.
13384    
13385    Please see the discussions on the xkb mailing list.
13386    
13387    Signed-off-by: James Cloos <cloos@jhcloos.com>
13388
13389commit d1bdc909f9246119696c8b0d9afb7bd8afb71b60
13390Author: Julien Cristau <jcristau@debian.org>
13391Date:   Wed Aug 5 18:14:23 2009 +0200
13392
13393    man/xkb: delete spurious newline in .TH headers
13394
13395commit ee723b83b24682db833a2f0abd96cd319b8a62af
13396Author: Julien Cristau <jcristau@debian.org>
13397Date:   Wed Aug 5 16:45:19 2009 +0200
13398
13399    man: use __libmansuffix__ instead of 3X11 for references to other pages
13400
13401commit 595e204feb82c798a92eea41fea03be6476ac181
13402Author: Julien Cristau <jcristau@debian.org>
13403Date:   Wed Aug 5 16:43:36 2009 +0200
13404
13405    man/xkb: use __libmansuffix__ instead of hardcoding 3Xkb for manpage sections
13406
13407commit 9da7e230d5320e1556ad2084fcd06ee7994385ea
13408Author: Peter Hutterer <peter.hutterer@who-t.net>
13409Date:   Wed Aug 5 14:15:02 2009 +1000
13410
13411    Bump to 1.2.99.901 (1.3 RC1)
13412    
13413    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13414
13415commit 8f78c7b4e3570cd46c5a220982963c17fe2157b8
13416Author: Filippo Giunchedi <filippo@debian.org>
13417Date:   Sat Jun 6 16:56:54 2009 +0200
13418
13419    nls: add {left,right}wards arrow to compose table
13420    
13421    Debian bug#532117 <http://bugs.debian.org/532117>
13422    
13423    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13424
13425commit 7949bfa00390241d994f32463e50d4bd78920568
13426Author: Julien Cristau <jcristau@debian.org>
13427Date:   Fri Jul 31 13:33:52 2009 +0200
13428
13429    Update library version for new symbols
13430    
13431    Commit 554f755e5545f63d3c8f299297927238da155773 added generic event
13432    cookie handling.  Bump libX11 version number accordingly.
13433    
13434    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13435
13436commit 640fec5f4feacd01a00eea3dcd4edb220907d3dc
13437Author: Julien Cristau <jcristau@debian.org>
13438Date:   Sun Aug 2 17:18:31 2009 +0200
13439
13440    Add _XFUNCPROTOBEGIN/END to Xlib-xcb.h
13441    
13442    X.Org bug#22252 <https://bugs.freedesktop.org/show_bug.cgi?id=22252>
13443    
13444    Reported-by: Riku Salminen <rsalmin2@cc.hut.fi>
13445    Signed-off-by: Julien Cristau <jcristau@debian.org>
13446
13447commit bc06d49e9dac1836d6824769ddb2ac5ba9f14df7
13448Author: Peter Hutterer <peter.hutterer@who-t.net>
13449Date:   Wed Jul 29 08:44:09 2009 +1000
13450
13451    Fix compiler warning 'unused variable qelt'
13452    
13453    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13454
13455commit 03f4907e14f5755e72309f08742977b871e81e33
13456Author: Peter Hutterer <peter.hutterer@who-t.net>
13457Date:   Wed Jul 29 08:34:57 2009 +1000
13458
13459    Add utlist.h to the Makefile.am
13460    
13461    utlist.h contains the linked list macros, it was added with the recent
13462    addition of event cookies but utlist.h wasn't added to the Makefile.am. As a
13463    result, make dist failed.
13464    
13465    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13466
13467commit 554f755e5545f63d3c8f299297927238da155773
13468Author: Peter Hutterer <peter.hutterer@who-t.net>
13469Date:   Fri Jun 26 11:27:43 2009 +1000
13470
13471    Add generic event cookie handling to libX11.
13472    
13473    Generic events require more bytes than Xlib provides in the standard XEvent.
13474    Memory allocated by the extension and stored as pointers inside the event is
13475    prone to leak by simple 'while (1) { XNextEvent(...); }' loops.
13476    
13477    This patch adds cookie handling for generic events. Extensions may register
13478    a cookie handler in addition to the normal event vectors. If an extension
13479    has registered a cookie handler, _all_ generic events for this extensions
13480    must be handled through cookies. Otherwise, the default event handler is
13481    used.
13482    
13483    The cookie handler must return an XGenericEventCookie with a pointer to the
13484    data.The rest of the event (type, serialNumber, etc.) are to be filled as
13485    normal. When a client retrieves such a cookie event, the data is stored in
13486    an internal queue (the 'cookiejar'). This data is freed on the next call to
13487    XNextEvent().
13488    
13489    New extension interfaces:
13490        XESetWireToEventCookie(display, extension_number, cookie_handler)
13491    
13492    Where cookie_handler must set cookie->data. The data pointer is of arbitray
13493    size and type but must be a single memory block. This memory block
13494    represents the actual extension's event.
13495    
13496    New client interfaces:
13497        XGetEventData(display, *cookie);
13498        XFreeEventData(display, *cookie);
13499    
13500    If the client needs the actual event data, it must call XGetEventData() with
13501    the cookie. This returns the data pointer (and removes it from the cookie
13502    jar) and the client is then responsible for freeing the event with
13503    XFreeEventData(). It is safe to call either function with a non-cookie
13504    event. Events unclaimed or not handled by the XGetEventData() are cleaned up
13505    automatically.
13506    
13507    Example client code:
13508        XEvent event;
13509        XGenericEventCookie *cookie = &ev;
13510    
13511        XNextEvent(display, &event);
13512        if (XGetEventData(display, cookie)) {
13513            XIEvent *xievent = cookie->data;
13514            ...
13515        } else if (cookie->type == GenericEvent) {
13516            /* handle generic event */
13517        } else {
13518            /* handle extension/core event */
13519        }
13520        XFreeEventData(display, cookie);
13521    
13522    Cookies are not multi-threading safe. Clients that use XGetEventData() must
13523    lock between XNextEvent and XGetEventData to avoid other threads freeing
13524    cookies.
13525    
13526    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13527
13528commit d7675cb8fa7155e7aff1459636a117a97aa1bf28
13529Author: Peter Hutterer <peter.hutterer@who-t.net>
13530Date:   Mon Jul 6 13:17:35 2009 +1000
13531
13532    Bump to 1.2.99.1
13533    
13534    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13535
13536commit 75fe48e7a42a685d7098e8d7108b9b956c471563
13537Author: Peter Hutterer <peter.hutterer@who-t.net>
13538Date:   Fri Jul 10 14:07:34 2009 +1000
13539
13540    Bump to 1.2.2
13541    
13542    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13543
13544commit 5d0fe0e0e92759af5667c5dca2eacb1b6f2d66ea
13545Author: Peter Hutterer <peter.hutterer@who-t.net>
13546Date:   Thu Jul 2 09:10:25 2009 +1000
13547
13548    XMaskEvent/XCheckMaskedEvents must not check for GenericEvents.
13549    
13550    GenericEvent cannot be selected for in the core event masks and they must
13551    thus be treated like extension events.
13552    
13553    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13554
13555commit c1bf65b89f4e361f6178a73dd5334c8f2bd95732
13556Author: Peter Hutterer <peter.hutterer@who-t.net>
13557Date:   Thu Jul 2 09:06:05 2009 +1000
13558
13559    XWindowEvent/XCheckWindowEvent must not return GenericEvents.
13560    
13561    GenericEvents have no fixed position for the window, so they must be treated
13562    like extension events.
13563    
13564    Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
13565
13566commit 38f9054554d63525d2dd51aafb5eb57821158ab9
13567Author: Alan Coopersmith <alan.coopersmith@sun.com>
13568Date:   Mon Jun 15 19:00:43 2009 -0700
13569
13570    Drop ancient USG SysV #ifdefs
13571    
13572    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13573
13574commit 7bfe1323f16a1a69cc474659f7ac0c2570b1cf42
13575Author: Adam Jackson <ajax@redhat.com>
13576Date:   Fri Jun 12 12:44:01 2009 -0400
13577
13578    Remove X_NOT_STDC_ENV usage. (#6527)
13579
13580commit 6dd74d7fb414ca1e99bae5c13e333961f396eb36
13581Author: Julien Cristau <jcristau@debian.org>
13582Date:   Fri May 29 11:18:11 2009 +0200
13583
13584    nls: remove more duplicated aliases
13585    
13586    Signed-off-by: Julien Cristau <jcristau@debian.org>
13587
13588commit e4b0899f516da224010e68bd2d953d5293d94993
13589Author: parag <parag@rawhideTM.pnq.redhat.com>
13590Date:   Thu May 28 11:29:35 2009 +0530
13591
13592    nls: Add pa_PK locale information and make pa_IN as default for pa.
13593    
13594    X.Org bug#21954 <http://bugs.freedesktop.org/show_bug.cgi?id=21954>
13595    
13596    [jcristau: removed the pa_PK.UTF-8 alias to itself]
13597    
13598    Signed-off-by: parag <pnemade@redhat.com>
13599    Signed-off-by: Julien Cristau <jcristau@debian.org>
13600
13601commit f0ea1f6d51145592f8617854f9320ec5dbff3299
13602Author: Julien Cristau <jcristau@debian.org>
13603Date:   Fri May 29 10:58:20 2009 +0200
13604
13605    nls: remove broken sd_IN.UTF-8 alias
13606    
13607    Signed-off-by: Julien Cristau <jcristau@debian.org>
13608
13609commit e29e010dabdb17d6498f2ef1786f69b8830c18ca
13610Author: Julien Cristau <jcristau@debian.org>
13611Date:   Fri May 29 10:57:43 2009 +0200
13612
13613    nls: remove duplicated en_US* aliases
13614    
13615    Signed-off-by: Julien Cristau <jcristau@debian.org>
13616
13617commit a89a300d87852c84389ad97db66dcb8930cb45dd
13618Author: Caolan McNamara <caolanm@redhat.com>
13619Date:   Thu May 21 18:41:05 2009 +0200
13620
13621    man: missing space in XAllocColor man page
13622    
13623    X.Org bug#21854 <http://bugs.freedesktop.org/show_bug.cgi?id=21854>
13624
13625commit c1c001e36504fd304f76f69bf6af3643225c49ea
13626Author: James Cloos <cloos@jhcloos.com>
13627Date:   Wed May 13 13:03:54 2009 -0400
13628
13629    [nls] Replace remaining UCS Combining Characters in Compose sequences.
13630    
13631    The replaces the instances of keysyms which match <U03[0-6][0-9A-Fa-f]>,
13632    where the keysym is used as a dead_key, with an actual dead_key symbol.
13633    
13634    The only remaining instances of UCS combining characters in the
13635    compose sequences are of U0338 COMBINING LONG SOLIDUS OVERLAY
13636    used as a suffix in Multi_key-initiated sequences to create
13637    mathematics characters such as ∉ U+2209 NOT AN ELEMENT OF
13638    from ∈ U+2208 ELEMENT OF.
13639
13640commit e2b0bad3d3b9e9ca781fc264eb7584afbe2a1a4f
13641Author: James Cloos <cloos@jhcloos.com>
13642Date:   Wed May 13 09:58:59 2009 -0400
13643
13644    [nls] Remove extraneous instances of UCS Combining Characters in Compose sequences.
13645    
13646    This removes those instances of keysyms which match <U03[0-6][0-9A-Fa-f]>,
13647    where the matching keysym is used as a dead_key, and for which alternative
13648    compose sequences exist.
13649
13650commit 79f47e6dff2f0a0b673bbfecc47528edca814baa
13651Author: James Cloos <cloos@jhcloos.com>
13652Date:   Fri May 8 20:11:54 2009 -0400
13653
13654    [nls] Remove combining_ keysyms from the Compose files
13655    
13656    Some of the UTF-8 Compose tables included combining_ keysyms in
13657    the compose sequences as though they were dead symbols.  This
13658    is contrary to how combining characters are used in the UCS.
13659    Therefore, those lines have been removed from the Compose tables.
13660    
13661    There were also some combining_ keysyms as targets.  As those
13662    are not included in x11proto’s keysymdef.h, and as those do
13663    exist there as Uxxxx keysyms, they are replaced with the Uxxxx
13664    keysym names.
13665    
13666    This addresses http://bugzilla.freedesktop.org/show_bug.cgi?id=5107
13667    and is based on attachment 25644 by samuel.thibault@ens-lyon.org.
13668    
13669    Signed-off-by: James Cloos <cloos@jhcloos.com>
13670
13671commit 4a08a3dfbda497b2be46e3e5fe6b777815ea27f9
13672Author: parag <pnemade@redhat.com>
13673Date:   Tue May 5 16:50:47 2009 +0530
13674
13675    libX11: Add new Indic language information to nls directory files. #21560
13676    
13677    Signed-off-by: parag <pnemade@redhat.com>
13678
13679commit aaf81096eb44b4c2812108721ba02738391884da
13680Author: Alan Coopersmith <alan.coopersmith@sun.com>
13681Date:   Sat May 2 01:38:14 2009 -0700
13682
13683    Correct return type in XkbGetKeyboard man page
13684    
13685    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13686
13687commit 2bef065b70f70af520b5de8fb23529254d15f003
13688Author: Christoph Pfister <christophpfister@gmail.com>
13689Date:   Mon Apr 27 22:32:57 2009 -0700
13690
13691    X.Org Bug #21117: crash in get_rotate_fontname (omGeneric.c)
13692    
13693    http://bugs.freedesktop.org/show_bug.cgi?id=21117
13694    
13695    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13696
13697commit 5c1cde105db10df24d3c532f032cbc59050c7313
13698Author: Alan Coopersmith <alan.coopersmith@sun.com>
13699Date:   Fri Apr 17 22:14:47 2009 -0700
13700
13701    Use AC_USE_SYSTEM_EXTENSIONS instead of hand-rolled check for _GNU_SOURCE
13702    
13703    Raises minimum autoconf version required to 2.60
13704    
13705    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13706
13707commit 84b7a91ef84f345384e4b0e13907385ca3ca3255
13708Author: Julien Cristau <jcristau@debian.org>
13709Date:   Tue Apr 14 15:59:57 2009 +0100
13710
13711    Fix fi_FI.UTF-8, again
13712    
13713    Commit 97fc6babd4ccaf300e25708868aa2a738893dc30 "NLS: Add UTF-8 compose
13714    file for Finnish" made fi_FI.UTF-8 use a broken empty XLC_LOCALE file.
13715    This reverts it back to using the en_US.UTF-8 one.
13716
13717commit 128daff4422f973ea40dd1e31b2db230e643549e
13718Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
13719Date:   Thu Apr 9 12:01:07 2009 +0700
13720
13721    Thai XIM not retrieve MB surrounding on UTF-8 LC
13722    
13723    On th_TH.UTF-8 locale, Thai XIM rejects all combining characters for GTK+ apps
13724    that use X Input Method.
13725    
13726    This is because GTK+ imxim immodule passes surrounding text in locale encoding,
13727    which is UTF-8 for UTF-8 locales. But current Thai XIM in Xlib assumes the
13728    multi-byte StringConversionText response for the StringConversionCallback to
13729    always be TIS-620, by retrieving a single byte and using it as-is.
13730    
13731    If the Thai XIM tries to convert the multi-byte text based on locale codeset
13732    before using it, it will work again.
13733    
13734    X.Org But 12759 <http://bugs.freedesktop.org/show_bug.cgi?id=12759>
13735    
13736    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
13737    Signed-off-by: Julien Cristau <jcristau@debian.org>
13738
13739commit e09f0d227fbf95b6252759af9d426efd57686f9f
13740Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
13741Date:   Thu Apr 9 11:47:55 2009 +0700
13742
13743    Thai XIM not filters when NumLock or CapsLock is on
13744    
13745    The Thai XIM component in libx11 activated on 'th*' locales normally filters
13746    input sequence according to orthographic rules. However, when NumLock/CapsLock
13747    is on, this stops working. All sequences are passed through.
13748    
13749    This is caused by missing masks in _XimThaiFilter(), which normally screens out
13750    certain special keys from entering orthographic rules. Unfortunately, this
13751    included events with NumLock/CapsLock on. Negating the masks from the check
13752    allows the events to be tested by the rules.
13753    
13754    X.Org Bug 12517 <http://bugs.freedesktop.org/show_bug.cgi?id=12517>
13755    
13756    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
13757    Signed-off-by: Julien Cristau <jcristau@debian.org>
13758
13759commit d108d3c706af3502820b5202564488ea19908b77
13760Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
13761Date:   Thu Apr 9 11:25:25 2009 +0700
13762
13763    CharSet-to-CompoundText Conversion Failed for Thai Locales
13764    
13765    SCIM fails to commit Thai input characters on Thai locales, because it commits
13766    string in compound text form, which was converted via
13767    XwcTextListToTextProperty(). But the XLC_LOCALE for th_TH and th_TH.UTF-8
13768    declares cs1's ct_encoding as TIS620-0:GR, which was commented out in
13769    src/xlibi18n/lcCT.c default_ct_data, in favor of ISO8859-11 ESC sequence.
13770    So, declaring cs1 as ISO8859-11:GR instead makes it work.
13771    
13772    Besides, for th_TH.UTF-8, adding cs2 class with ISO10646-1 encoding also adds
13773    support for UTF-8 input.
13774    
13775    And, along discussion in the bug, a similar problem was found for fontset, too,
13776    by causing delays on X apps startups on systems without tis620-0 fonts. This
13777    is normally the case, as mkfontdir and mkfontscale generate iso8859-11 entries
13778    by default for Thai X fonts. So, Thai fontset charset is also patched.
13779    
13780    X.Org Bug 16475 <http://bugs.freedesktop.org/show_bug.cgi?id=16475>
13781    
13782    Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net>
13783    Signed-off-by: Julien Cristau <jcristau@debian.org>
13784
13785commit fd62d3318c846cd43d66a505946e94704d7d83dc
13786Author: Alan Coopersmith <alan.coopersmith@sun.com>
13787Date:   Wed Apr 8 19:42:25 2009 -0700
13788
13789    Revert "Change masculine to ordmasculine in Compose file comments"
13790    
13791    This reverts commit 892b401d5acc055803a20e349ede0d64490f2230.
13792    
13793    As Julien Cristau correctly points out, I misread the Compose file grammar
13794    in modules/im/ximcp/imLcPrs.c, and those are keysyms, not comments, and
13795    the keysym is named XK_masculine in keysymdef.h.   This change is thus a
13796    bug in the Solaris compose tables to be fixed, not an improvement to bring
13797    upstream.
13798
13799commit eac57c77afdf44f50692225b8b0345a7c927bc84
13800Author: Alan Coopersmith <alan.coopersmith@sun.com>
13801Date:   Tue Apr 7 15:38:45 2009 -0700
13802
13803    Version bump: 1.2.1
13804    
13805    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13806
13807commit bfbec08baec33c5024510b0bcbbee6e4a8473e79
13808Author: Yaakov Selkowitz (Cygwin Ports maintainer) <yselkowitz@users.sourceforge.net>
13809Date:   Tue Apr 7 13:46:57 2009 -0700
13810
13811    Bug 20773: Xcursor dynamic loading on Cygwin
13812    
13813    X.Org Bug #20773 <http://bugs.freedesktop.org/show_bug.cgi?id=20773>
13814    Patch #24096 <http://bugs.freedesktop.org/attachment.cgi?id=24096>
13815    
13816    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13817
13818commit c8c41614911be4fa222fa22478677d263b41c751
13819Author: Alan Coopersmith <alan.coopersmith@sun.com>
13820Date:   Mon Apr 6 16:52:46 2009 -0700
13821
13822    Fix a several sparse warnings: Using plain integer as NULL pointer
13823    
13824    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13825
13826commit b336c3d0cc2aefc8926500cff5f76b5a3e803886
13827Author: Alan Coopersmith <alan.coopersmith@sun.com>
13828Date:   Mon Apr 6 16:32:05 2009 -0700
13829
13830    Further ansify prototypes & reduce #ifdefs in locking.c
13831    
13832    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13833
13834commit 892b401d5acc055803a20e349ede0d64490f2230
13835Author: Alan Coopersmith <alan.coopersmith@sun.com>
13836Date:   Mon Apr 6 10:50:09 2009 -0700
13837
13838    Change masculine to ordmasculine in Compose file comments
13839    
13840    Matches the ordfeminine name used for the matching character,
13841    and the ordmasculine name used in many font descriptions of the glyph.
13842    
13843    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13844
13845commit e3198b55dc16ec57346cc28aa8d34165ce8cde96
13846Author: Julien Cristau <jcristau@debian.org>
13847Date:   Fri Mar 13 13:52:33 2009 +0100
13848
13849    Add a ru_RU.UTF-8 locale
13850    
13851    Based on patch by Eugene Konev <ejka@imfi.kspu.ru> for X.Org 6.9.0.
13852    
13853    Debian bug#330144 <http://bugs.debian.org/330144>
13854    X.Org bug#15887 <http://bugs.freedesktop.org/show_bug.cgi?id=15887>
13855
13856commit d239de9452691d6f875e6e5ace3d499ec3bf14d9
13857Author: Alan Coopersmith <alan.coopersmith@sun.com>
13858Date:   Wed Mar 25 17:59:09 2009 -0700
13859
13860    Delete some unused "#ifdef notdef" static functions
13861    
13862    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13863
13864commit 4281892e31058ef3aecc96a5767824b34d88d415
13865Author: Alan Coopersmith <alan.coopersmith@sun.com>
13866Date:   Wed Mar 25 17:52:48 2009 -0700
13867
13868    Remove _XP_PRINT_SERVER_ #ifdefs from Xrm.c
13869    
13870    This copy of Xrm.c is never compiled into the Xprint server any more, so
13871    this old code-sharing #ifdef from the monolith tree isn't needed.
13872    
13873    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13874
13875commit 27a2e16e8ea11c1604e28645fae4d6ba4371d513
13876Author: Alan Coopersmith <alan.coopersmith@sun.com>
13877Date:   Tue Mar 17 18:38:58 2009 -0700
13878
13879    makekeys: combine malloc(strlen)+strcpy into strdup
13880    
13881    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13882
13883commit 78a894231ae8ec4959007b6d9b8d2a15d6333a1e
13884Author: Alan Coopersmith <alan.coopersmith@sun.com>
13885Date:   Tue Mar 17 15:42:19 2009 -0700
13886
13887    When makekeys fails to find a good hash, print error instead of divide-by-zero
13888    
13889    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13890
13891commit c9f84af591d15fbc3fa890bcd955d94f1ff82a0b
13892Author: Alan Coopersmith <alan.coopersmith@sun.com>
13893Date:   Tue Mar 17 14:59:16 2009 -0700
13894
13895    Remove ifdef checks for macII left over from ancient A/UX 3.0 support
13896    
13897    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13898
13899commit 4b90da0ffb32b791d915ecae11996cc2f2cac9a7
13900Author: Alan Coopersmith <alan.coopersmith@sun.com>
13901Date:   Mon Mar 16 18:37:49 2009 -0700
13902
13903    XErrorDB additions for DRI2 requests
13904    
13905    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13906
13907commit 273f45af1441be9d2135e4cac8c46ceb33470236
13908Author: Alan Coopersmith <alan.coopersmith@sun.com>
13909Date:   Mon Mar 16 18:35:44 2009 -0700
13910
13911    XErrorDB additions for XInput errors and new XInput 1.5 requests
13912    
13913    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13914
13915commit 90ef36f6336068183bf9d13ca972202db35b1202
13916Author: Alan Coopersmith <alan.coopersmith@sun.com>
13917Date:   Mon Mar 16 18:23:05 2009 -0700
13918
13919    XErrorDB additions for RANDR 1.3
13920    
13921    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13922
13923commit 124a4efaccf329f1a44f6b824e70278bdd1cad83
13924Author: Lubos Lunak <l.lunak@suse.cz>
13925Date:   Mon Mar 16 18:03:36 2009 -0700
13926
13927    XErrorDB updates for XTEST, RANDR, DAMAGE extensions
13928    
13929    From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html
13930    
13931    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13932
13933commit 80811846e37b805fddb37c71589fd5f6f6037b3f
13934Author: Lubos Lunak <l.lunak@suse.cz>
13935Date:   Mon Mar 16 17:57:52 2009 -0700
13936
13937    XGetErrorText() fails for extension error codes equal to the error base
13938    
13939    From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html
13940    
13941    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13942
13943commit da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd
13944Author: Milos Komarcevic <miloskomarcevic@netscape.net>
13945Date:   Mon Mar 16 17:43:26 2009 -0700
13946
13947    Bug 11456: Serbian locale updates (sr_RS and sr_ME)
13948    
13949    X.Org Bug #11456 <http://bugs.freedesktop.org/show_bug.cgi?id=11456>
13950    Patch #23937 <http://bugs.freedesktop.org/attachment.cgi?id=23937>
13951    
13952    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13953
13954commit 934a6c0519a9e32505beee55b004f36c2a710217
13955Author: Alan Coopersmith <alan.coopersmith@sun.com>
13956Date:   Mon Mar 16 14:55:22 2009 -0700
13957
13958    Bug 10082: Compose entries for some standard mathematical operators
13959    
13960    X.Org Bug #10082 <http://bugs.freedesktop.org/show_bug.cgi?id=10082>
13961    
13962    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13963
13964commit abf4da1ed0f735ca7ce471dc13a0ec3677391486
13965Author: Alan Coopersmith <alan.coopersmith@sun.com>
13966Date:   Mon Mar 16 14:27:46 2009 -0700
13967
13968    Bug 14651: We need to add new locale specification for Belarusian Latin locale
13969    
13970    X.Org Bug #14651 <http://bugs.freedesktop.org/show_bug.cgi?id=14651>
13971    
13972    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13973
13974commit 837703c8651e1321a50147a8311c56e4758ce08a
13975Author: Caolan McNamara <caolanm@redhat.com>
13976Date:   Mon Mar 16 14:15:50 2009 -0700
13977
13978    Bug 20575: man page for XCreatePixmapFromBitmapData doesn't match signature
13979    
13980    X.Org Bug #20575 <http://bugs.freedesktop.org/show_bug.cgi?id=20575>
13981    Patch #23717 <http://bugs.freedesktop.org/attachment.cgi?id=23717>
13982    
13983    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13984
13985commit 22199018879055d8653e59d8236bef57164fac66
13986Author: Alan Coopersmith <alan.coopersmith@sun.com>
13987Date:   Mon Mar 16 13:28:18 2009 -0700
13988
13989    Correct locale alias for sh_BA.ISO8859-2@bosnia (should be sr, not nr)
13990    
13991    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
13992
13993commit 4c63c27eab2b88f5556dbf72c36321f50f6de35e
13994Author: Alan Coopersmith <alan.coopersmith@sun.com>
13995Date:   Thu Mar 12 18:57:20 2009 -0700
13996
13997    Bug 9953: Please provide locale alias hu_HU.utf8
13998    
13999    X.Org Bug #9953 <http://bugs.freedesktop.org/show_bug.cgi?id=9953>
14000    Debian Bug #407573 <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=407573>
14001    
14002    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14003
14004commit 501f4e0ada1690783ada05ad412e4b191ad55336
14005Author: Alan Coopersmith <alan.coopersmith@sun.com>
14006Date:   Thu Mar 12 17:38:21 2009 -0700
14007
14008    Bug 6820: Xlib shouldn't handle EAGAIN as a fatal IO error
14009    
14010    X.Org Bug #6820 <http://bugs.freedesktop.org/show_bug.cgi?id=6820>
14011    Patch #17637 <http://bugs.freedesktop.org/attachment.cgi?id=17637>
14012    
14013    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14014
14015commit 7b9c543e7210c7da204871c31a160e79d3a949b6
14016Author: Paul Bender <pebender@gmail.com>
14017Date:   Thu Mar 12 17:11:42 2009 -0700
14018
14019    Bug 15664: xau & xdmcp not needed in x11.pc dependencies when built with xcb
14020    
14021    X.Org bug #15664 <https://bugs.freedesktop.org/show_bug.cgi?id=15664>
14022    Patch #16128 <https://bugs.freedesktop.org/attachment.cgi?id=16128>
14023    
14024    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14025
14026commit fd2cf1ef66c2aff3dc758956c9e9e567b9892c06
14027Author: Xue Wei <Wei.Xue@Sun.COM>
14028Date:   Wed Mar 4 19:32:29 2009 -0800
14029
14030    Add UTF-8 locale entries for es_US, kk_KZ, mt_MT, and sh_BA
14031    
14032    Sun bug 6809309 Add new utf8 locales supported by Xlib
14033    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6809309>
14034    
14035    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14036
14037commit cb70c9bc43267577859a3674ca9de9be396ba69e
14038Author: Alan Coopersmith <alan.coopersmith@sun.com>
14039Date:   Mon Feb 23 19:29:15 2009 -0800
14040
14041    Add --with-locale-lib-dir configure option to set locale lib install dir
14042    
14043    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14044
14045commit ccd3584f0330db8dac90b9313c33ab8b5b2ec6af
14046Author: Alan Coopersmith <alan.coopersmith@sun.com>
14047Date:   Mon Feb 23 18:33:51 2009 -0800
14048
14049    Incorporate more locale names/aliases from Solaris libX11
14050    
14051    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14052
14053commit 83ce4daefdf544f801c7d666c89162690a36ce41
14054Author: Alan Coopersmith <alan.coopersmith@sun.com>
14055Date:   Mon Feb 23 18:32:34 2009 -0800
14056
14057    Incorporate char range comments from Solaris version of ksc5601.h
14058    
14059    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14060
14061commit ee279c84e34f1ebb8a6ff17d54ee146d11e29764
14062Author: Chris Ball <cjb@laptop.org>
14063Date:   Sat Feb 21 14:48:42 2009 -0500
14064
14065    Fix fi_FI locale install directory.
14066    
14067    fi_FI was setting "x11thislocaledir" to en_US, with the result that its
14068    locale data was written in that locale dir.
14069    
14070    Signed-off-by: Chris Ball <cjb@laptop.org>
14071
14072commit da6bbca07c796c69172a649405474f03bee66754
14073Author: Emilio Jesús Gallego Arias  <egallego@babel.ls.fi.upm.es>
14074Date:   Sat Feb 21 20:17:23 2009 +0100
14075
14076    xcb_io: Avoid datatype overflow on AMD64 and friends.
14077
14078commit 4ef6491afa69e8441caee7bbebc583e6e796275e
14079Author: Chris Ball <cjb@laptop.org>
14080Date:   Sat Feb 21 12:51:03 2009 -0500
14081
14082    Build fix for fi_FI.
14083    
14084    Commit 642c4e928e770e0.. instructs make to enter nls/fi_FI, but no
14085    Makefile is written there by configure.
14086    
14087    Signed-off-by: Chris Ball <cjb@laptop.org>
14088
14089commit 642c4e928e770e012379539a6ce09e11c02f09a6
14090Author: Julien Cristau <jcristau@debian.org>
14091Date:   Sat Feb 21 03:12:05 2009 +0100
14092
14093    nls: actually use the fi_FI.UTF-8 files
14094    
14095    The subdir wasn't added to nls/Makefile.am
14096
14097commit 9bad8309ef289bb943651abf6967b24fa2252aac
14098Author: Alan Coopersmith <alan.coopersmith@sun.com>
14099Date:   Fri Feb 20 14:45:54 2009 -0800
14100
14101    flags member of Display structure needs to be marked volatile
14102    
14103    Since the Xlib multithreaded code checks the flags variable in _XFlushInt
14104    to see if the other threads are done yet, it has to be marked volatile so
14105    the compiler doesn't optimize out re-loading it on each trip through the
14106    while loop and end up in an impossible-to-exit infinite loop of CPU chewing.
14107    
14108    Part of fix for Sun bug 6409332: infinite loop in XFlushInt() on x86/32-bit
14109    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6409332>
14110    
14111    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14112
14113commit ef5c446395aa30d7b1096a112e241a81c5b358e7
14114Author: Xue Wei <Wei.Xue@Sun.COM>
14115Date:   Fri Feb 20 15:12:35 2009 -0800
14116
14117    Add nn_NO.UTF-8 to compose.dir.pre for Norwegian Nynorsk
14118    
14119    Sun bug 6691236: Swing applications dump core when locale is nn_NO.UTF-8
14120    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691236>
14121    
14122    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14123
14124commit 3822f2654e9630167f0c6cae317b472c09771672
14125Author: Xue Wei <Wei.Xue@Sun.COM>
14126Date:   Fri Feb 20 15:03:51 2009 -0800
14127
14128    Add locale aliases for no_NO & sh_BA locale variants
14129    
14130    Fixes Sun bug id 6691219: xterm refuses to start in some locales
14131    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691219>
14132    
14133    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14134
14135commit 63b6b5b5f522c0a606b32163c643edb64ca91d54
14136Author: Xue Wei <Wei.Xue@Sun.COM>
14137Date:   Fri Feb 20 14:23:11 2009 -0800
14138
14139    Add kk_KZ.UTF-8 to locale.dir.pre for Kazakhstan
14140    
14141    Fixes Sun bug id 6737254 ("kk_KZ.UTF-8 locale: In Java applications
14142     changing keyboard layout with gimlet does not work")
14143    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6737254>
14144    
14145    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14146
14147commit d497301707962f3b94542e999a36abffcfbd303d
14148Author: Julien Cristau <jcristau@debian.org>
14149Date:   Tue Feb 17 16:09:41 2009 +0100
14150
14151    Bump to 1.2
14152
14153commit 990e71361d1d7b79bf07b1dc93e4e905d4f1bdaf
14154Author: Julien Cristau <jcristau@debian.org>
14155Date:   Tue Feb 17 15:23:40 2009 +0100
14156
14157    Check Xmalloc return value in _XConnectXCB
14158    
14159    X.Org bug#19137 <http://bugs.freedesktop.org/show_bug.cgi?id=19137>
14160    
14161    Signed-off-by: Julien Cristau <jcristau@debian.org>
14162
14163commit b4b5893f69419ff577bbaa4d18f78e4ffd729a0c
14164Author: James Cloos <cloos@jhcloos.com>
14165Date:   Sat Feb 14 12:35:56 2009 -0500
14166
14167    dolt: allow older versions of bash to compile the library
14168    
14169    Cf xserver commit 7be6520d and bugzilla #19031.
14170
14171commit 20982d6866e24453642b0b592fa0f13a88aa747c
14172Author: Will Thompson <will@willthompson.co.uk>
14173Date:   Thu Feb 5 02:53:06 2009 +1100
14174
14175    NLS: Compose: Non-aliasing CCCP
14176    
14177    Oops, cccp aliased cc for question mark.  Upper-case it to avoid fail.
14178    
14179    Signed-off-by: Will Thompson <will@willthompson.co.uk>
14180    Signed-off-by: Daniel Stone <daniel@fooishbar.org> (sorry)
14181
14182commit f052665394f3f0319e93a98f1d5d4ea287e1dd07
14183Author: Will Thompson <will@willthompson.co.uk>
14184Date:   Wed Feb 4 14:51:11 2009 +0000
14185
14186    Add two essential compose sequences
14187    
14188    Signed-off-by: Will Thompson <will@willthompson.co.uk>
14189    Signed-off-by: Daniel Stone <daniel@fooishbar.org>
14190
14191commit d7bea6fa909bf34c43efe0ca8239ab0f9f3a415f
14192Author: Alan Coopersmith <alan.coopersmith@sun.com>
14193Date:   Mon Feb 2 20:34:31 2009 -0800
14194
14195    Add README with pointers to mailing list, bugzilla & git repos
14196    
14197    Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
14198
14199commit f682c27e93512773122887d2cbabb1657af45d2e
14200Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14201Date:   Mon Feb 2 16:36:39 2009 -0200
14202
14203    Check if a function argument is NULL.
14204    
14205      This was an addition to patch (also by me)
14206    https://bugs.freedesktop.org/attachment.cgi?id=14660
14207    that was not added when rediscovering/correcting the problem.
14208
14209commit 427e9d45d424b84efd9fc499aebf8d72392844c5
14210Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14211Date:   Thu Feb 28 15:58:12 2008 -0300
14212
14213    Allow multiple inclusions of cursorfont.h, cosmetic patch.
14214
14215commit b91524a53e691f6a5d278fd8972b48a14ebeedeb
14216Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14217Date:   Thu Feb 28 15:54:43 2008 -0300
14218
14219    Don't add prototypes for functions that don't exist.
14220    
14221    Note that a full review was not done, only for functions that receive
14222    char/short arguments, or one of it's parameters is a function pointer
14223    that requires char/short arguments.
14224
14225commit 537eb52fe266ac439c4b383bb04a70017b709911
14226Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14227Date:   Thu Feb 28 15:50:27 2008 -0300
14228
14229    WORD64 compile fix. This bug catched on a overview of the code.
14230    
14231    The code is wrong since the first git revision, so it seens that it has
14232    not been compiled with WORD64 for quite some time, there is also another
14233    interesting code in xkb/XKBRdBuf.c:
14234      <hash>ifdef WORD64
14235      _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8
14236      <hash>endif
14237    and possibly there are other similar problems.
14238
14239commit ffd0300fb74c6183208ae599133f2ded09e08d97
14240Author: Brian Rogers <brian@xyzw.org>
14241Date:   Sat Jan 31 10:37:51 2009 -0800
14242
14243    Initialize event_notify after allocating the memory for it.
14244    
14245    An uninitialized or otherwise invalid condition variable can apparently
14246    cause a hang in pthread_cond_broadcast. Ekiga, openoffice, and xine
14247    at least are freezing as a result of event_notify never being initialized.
14248    
14249    Signed-off-by: Brian Rogers <brian@xyzw.org>
14250    Signed-off-by: Bart Massey <bart@cs.pdx.edu>
14251
14252commit 97fc6babd4ccaf300e25708868aa2a738893dc30
14253Author: James Cloos <cloos@jhcloos.com>
14254Date:   Thu Jan 29 20:10:41 2009 -0500
14255
14256    NLS: Add UTF-8 compose file for Finnish
14257    
14258    From bug report:
14259    
14260      https://bugs.freedesktop.org/show_bug.cgi?id=18747
14261
14262commit 1bd2966ed88f83479a066c6ca7da23a515979550
14263Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14264Date:   Thu Jan 29 20:25:15 2009 -0200
14265
14266    patches to avoid gcc warnings for libX11 (#4)
14267    
14268    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14269    Bug #17946, attachment #19443
14270    
14271    This patch avoids the gcc warning
14272            ../../../../libX11-1.1.5/modules/im/ximcp/imDefLkup.c:223: warning: passing arg 1 of `_XimProcSyncReply' from incompatible pointer type
14273    (same as already done at other places)
14274    
14275    BTW: what is the difference between XIM (the type of ic->core.im)
14276    and Xim ?
14277
14278commit f16dd6af3eb17a25b8ee03d6617a7acc6e919fb0
14279Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14280Date:   Thu Jan 29 20:22:21 2009 -0200
14281
14282    patches to avoid gcc warnings for libX11 (#3)
14283    
14284    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14285    Bug #17946, attachment #19441
14286    
14287    This patch avoids the two gcc warnings
14288            ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:413: warning: assignment discards qualifiers from pointer target type
14289            ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:450: warning: assignment discards qualifiers from pointer target type
14290    
14291    Note, that this as a rather crude fix of the problem (and it is really a
14292    shame to cast name_table to non-const).
14293    
14294    The right solution would be to declare XIMValuesList.supported_values
14295    (in include/X11/Xlib.h) as 'const char **' (or '_Xconst char **').
14296    This will, however, require extensive modifications in various places.
14297
14298commit cce75c5dce73fe1f8626ed9e6798138ada09a860
14299Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14300Date:   Thu Jan 29 20:20:18 2009 -0200
14301
14302    patches to avoid gcc warnings for libX11 (#2)
14303    
14304    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14305    Bug #17946, attachment #19440
14306    
14307    Avoid a preprocessor message
14308            <stdin>:194: warning: no newline at end of file
14309    
14310    Two more such warnings (in XkbSAGroup.man and XkbSASetGroup.man)
14311    seem to be caused by a truncated (or otherwise incomplete)
14312    manpage.
14313
14314commit 692baebcc50f1e952800bfe4e2e6bc42f54e62fe
14315Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14316Date:   Thu Jan 29 20:12:24 2009 -0200
14317
14318    patches to avoid gcc warnings for libX11 (#1)
14319    
14320    Author is Peter Breitenlohner <peb@mppmu.mpg.de>
14321    Bug #17946, attachment #19439
14322    
14323    Define as 1 (one) as done by autoconf and the command line
14324    option, e.g. -DX11_t, not as empty.
14325    
14326    This avoids the gcc (3.4.6) warnings:
14327            ../../libX11-1.1.5/src/x11_trans.c:27:1: warning: "X11_t" redefined
14328            <command line>:7:1: warning: this is the location of the previous definition
14329            ../../libX11-1.1.5/src/x11_trans.c:28:1: warning: "TRANS_CLIENT" redefined
14330            <command line>:8:1: warning: this is the location of the previous definition
14331    
14332    Similarly, follow the autoconf convention to define XTHREADS
14333    and XUSE_MTSAFE_API as one.
14334    
14335    This avoids analogous warnings when compiling libXcomposite,
14336    libXcursor, and libXdamage.
14337    
14338    No reason to AC_SUBST XTHREADS and XUSE_MTSAFE_API (unused).
14339
14340commit a1977883c9f5ef0e515569d6e2ebccb07411f98c
14341Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14342Date:   Thu Jan 29 15:01:06 2009 -0200
14343
14344    Janitor: Correct some gcc/sparse warnings.
14345    
14346      Most remaining warnings are about XIM/Xim to/from conversion
14347    and discarding const from pointers.
14348
14349commit 8ba0ca32a63c532f128bdca7f1bf982cab8e12be
14350Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br>
14351Date:   Wed Jan 28 20:31:42 2009 -0200
14352
14353    Janitor: ansification, make distcheck, compiler warnings.
14354    
14355      Only convert to use "ansi prototypes" the functions warned from
14356    compilation with "./autogen.sh --prefix=/usr", on a Linux computer.
14357    
14358      Also, only address "trivial" compiler warning fixes in this commit.
14359    
14360      The new .gitignore is the output of a command like:
14361    % find . -name .gitignore -exec cat {} \; | sort | uniq
14362    and only the toplevel .gitignore file was kept.
14363
14364commit 091c1624fd2f9d933329d6152e4ecd865aa7903a
14365Author: Peter Hutterer <peter.hutterer@who-t.net>
14366Date:   Tue Jan 13 12:05:54 2009 +1000
14367
14368    Fix wrong implies symbol.
14369    
14370    Quote Simos Xenitellis:
14371    > I checked the gtk+ commit logs that go back to 2000, and I see that still it
14372    > was
14373    >
14374    >   { 0x08ce, 0x21d2 }, /*                     implies ⇒ RIGHTWARDS
14375    > DOUBLE ARROW */
14376    >
14377    > In XFree86, it appears there was an error when they converted the
14378    > original table to
14379    > http://cvsweb.xfree86.org/cvsweb/xc/lib/X11/imKStoUCS.c?rev=1.1&content-type=text/vnd.viewcvs-markup
14380    > and the problem still shows up there (November 2000).
14381    
14382    http://lists.freedesktop.org/archives/xorg/2008-October/039743.html
14383    
14384    Reported by Erik Streb del Toro.
14385
14386commit e32521f19e0b07649b7e3a03d56a2bd556b138fb
14387Author: James Cloos <cloos@jhcloos.com>
14388Date:   Sun Dec 7 04:13:34 2008 -0500
14389
14390    [i18n] s/U00DC/Udiaeresis/g
14391    
14392    The xkeyboard-config keyboards generate the symbol Udiaeresis, not
14393    U00DC.  Make sure the relevant Compose sequences expect the symbol
14394    which the keyboards actually send.
14395
14396commit b7502abfe22f8dc009b21cda1172af221d8f9f32
14397Author: James Cloos <cloos@jhcloos.com>
14398Date:   Sun Dec 7 04:09:46 2008 -0500
14399
14400    Revert "For nls/*.pre, allow people to comment lines by starting them with '##'."
14401    
14402    As of commit c9d20e3 the initial double-hashes are replaced with proper C comments.
14403    
14404    This reverts commit a225a0be48770beb689d5ac5da97073634f7deab.
14405
14406commit c9d20e3f697c9cfae5511412023362c1db7449b1
14407Author: James Cloos <cloos@jhcloos.com>
14408Date:   Sun Dec 7 04:08:23 2008 -0500
14409
14410    Use C comments rather than initial doubled hashes to exclude lines from .pre files
14411
14412commit c34ce54d9eac2d8052dc5f205a2ab09866ef5d25
14413Author: vehemens <vehemens@verizon.net>
14414Date:   Sun Dec 7 01:18:26 2008 -0500
14415
14416    [i18n] Distribute new headers which were added for gb18030 support.
14417    
14418    big5hkscs.h and gbk.h, added in 67e34d7a, need to be in SOURCES to
14419    make it into the tar.
14420    
14421    Completes 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51.
14422    
14423    Signed-off-by: James Cloos <cloos@jhcloos.com>
14424
14425commit 418819558d2c60e58b4e3022ce0fadf2143488ac
14426Author: Stefan Dirsch <sndirsch@suse.de>
14427Date:   Sat Nov 22 22:01:07 2008 +0100
14428
14429    Fixed strange font mixups, when fontsets are still used (#2676, Novell #74299).
14430
14431commit 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51a
14432Author: Stefan Dirsch <sndirsch@suse.de>
14433Date:   Sat Nov 22 19:40:54 2008 +0100
14434
14435    Added remaining xlib patch required for gb18030 support (#1573).
14436
14437commit 55782a0a1fe1560f1a9c0ed78bc7f2575c15abcf
14438Author: Stefan Dirsch <sndirsch@suse.de>
14439Date:   Sat Nov 22 17:53:06 2008 +0100
14440
14441    Added remaining hunk of Egbert's patch to prevent XIM deadlocks (#1182).
14442
14443commit c859446c500c883a67f7a86cab1a44844e24dade
14444Author: Ken Thomases <ken@codeweavers.com>
14445Date:   Fri Nov 21 13:58:10 2008 -0500
14446
14447    [i18n] Provide translation from XK_partialderivative (8ef) to Unicode (U2202)
14448    
14449    Signed-off-by: James Cloos <cloos@jhcloos.com>
14450
14451commit 5e68e94d852c730ef9264fc0d8ca61a2ffe98b53
14452Author: Jeremy Huddleston <jeremyhu@freedesktop.org>
14453Date:   Mon Nov 17 20:47:26 2008 -0800
14454
14455    Force local transport when using the launchd socket.
14456    
14457    Fixes a regression due to
14458    bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
14459
14460commit 7aca689ce14d314b5c8c72c8df76f53f76ab467c
14461Author: Peter Hutterer <peter.hutterer@redhat.com>
14462Date:   Fri Nov 14 10:32:50 2008 +1000
14463
14464    Add XF86Suspend, XF86Hibernate to KeysymDB.
14465
14466commit d16b11f25f8265e651def8d80bcd430c0448e664
14467Author: James Cloos <cloos@jhcloos.com>
14468Date:   Tue Nov 11 19:43:39 2008 -0500
14469
14470    [nls] Annotate the Bépo compose sequences
14471    
14472    Add comments with the UCS names.
14473    Add utf-8 strings for each result.
14474    Format for easy reading.
14475
14476commit 730298464240be6f65b32416b3f9b20062c61825
14477Author: James Cloos <cloos@jhcloos.com>
14478Date:   Tue Nov 11 16:41:34 2008 -0500
14479
14480    [nls]  Add some UTF-8 Compose sequences
14481    
14482    As requested for the Bépo keyboard layout (http://clavier-dvorak.org/wiki/).
14483    
14484    Cf. bugs:
14485    
14486      https://bugs.freedesktop.org/show_bug.cgi?id=17821
14487      https://bugs.freedesktop.org/show_bug.cgi?id=17822
14488
14489commit ad6008a0c92733826983eb93f063d3d6276007d8
14490Author: Colin Harrison <colin.harrison@virgin.net>
14491Date:   Thu Nov 6 17:48:21 2008 -0500
14492
14493    Fix copy/paste typo in imLcPrs
14494    
14495    Signed-off-by: James Cloos <cloos@jhcloos.com>
14496
14497commit 0f0168ad18f8a280fc5a689eb02cfaa62d022ea6
14498Author: Adam Jackson <ajax@redhat.com>
14499Date:   Thu Nov 6 14:54:13 2008 -0500
14500
14501    Fix leak in _XimXGetReadData
14502    
14503    Spotted by Denis Dzyubenko
14504
14505commit ae23c25b9349ab1d7ff81f3075c000cf35fc442b
14506Author: Julien Cristau <jcristau@debian.org>
14507Date:   Wed Nov 5 22:09:37 2008 +0100
14508
14509    Bump to 1.1.99.2
14510
14511commit d31e644c65c52828ea3e7abd94a8cf9aee12265c
14512Author: Julien Cristau <jcristau@debian.org>
14513Date:   Wed Nov 5 21:33:13 2008 +0100
14514
14515    Fix distcheck
14516
14517commit e6a7b70cdb2ae8b713012839a0a0bbb93817b8ef
14518Author: Jamey Sharp <jamey@minilop.net>
14519Date:   Wed Oct 29 14:00:33 2008 -0700
14520
14521    Support multiple independent internal sync handlers
14522    
14523    Xlib has several independent tasks that need to be performed with the
14524    display unlocked. It does this by replacing the existing sync handler with
14525    one of a variety of internal sync handlers. However, if multiple internal
14526    sync handlers need to run, then the last one registering wins and
14527    previously registered internal sync handlers are never invoked. This
14528    manifested as a bug with DRI applications on Xlib/XCB as that requires
14529    both an XID handler after every XID allocation, and the periodic sequence
14530    number handler. The XID handler would win, and the sequence number handler
14531    would never be invoked.
14532    
14533    Fix this by unifying the internal sync handler mechanism into a single
14534    function that calls all of the known internal sync handlers. They all need
14535    to deal with being called when not strictly necessary now.
14536    
14537    Signed-off-by: Keith Packard <keithp@keithp.com>
14538    Signed-off-by: Jamey Sharp <jamey@minilop.net>
14539    Signed-off-by: Josh Triplett <josh@freedesktop.org>
14540
14541commit 2dbaaab9c4e3894b33dcae850551dee5473431d5
14542Author: Keith Packard <keithp@keithp.com>
14543Date:   Sat Oct 11 21:44:21 2008 -0700
14544
14545    Ensure that _XReadEvents always leaves an event in the queue on return
14546    
14547    XNextEvent assumes that the event queue will be non-empty on return from
14548    _XReadEvents, but with multiple event readers running, the previous change
14549    could leave the queue empty on return from process_responses. Re-invoke
14550    process_responses until the queue is non-empty.
14551    
14552    Signed-off-by: Keith Packard <keithp@keithp.com>
14553
14554commit bedfe68259037c5564fe52758c92b9c97729640a
14555Author: Keith Packard <keithp@keithp.com>
14556Date:   Sat Oct 11 21:10:23 2008 -0700
14557
14558    Permit only one Xlib thread to block waiting for events
14559    
14560    As Xlib queues events internally, we must prevent multiple Xlib threads from
14561    entering XCB to wait for an event in case the queued event is to be
14562    delivered to the thread which didn't manage to read it. In other words, let
14563    only one Xlib thread into xcb_wait_for_event at a time.
14564    
14565    Jamey Sharp looked over my shoulder while making this fix and, while hating
14566    my whitespace conventions, appears happy enough with the actual code.
14567    
14568    Signed-off-by: Keith Packard <keithp@keithp.com>
14569
14570commit cc19618d2eb3ed92a0b574aee26a7da8b4aed5d2
14571Author: Jamey Sharp <jamey@minilop.net>
14572Date:   Sun Mar 23 16:33:50 2008 -0700
14573
14574    Fix XAllocID race: hold the user display lock until we have a new XID.
14575    
14576    Xlib built --without-xcb is also vulnerable to this race, and a similar
14577    fix might work there too.
14578    
14579    Also, use an XID that's truly invalid while waiting for the next XID to be
14580    requested.
14581
14582commit 54e5c0941b0ded1628d559a9f0a3451ea96c299b
14583Author: Josh Triplett <josh@freedesktop.org>
14584Date:   Sat Mar 15 17:22:23 2008 -0700
14585
14586    Use XCB's new socket handoff mechanism rather than the old XCB Xlib lock.
14587    
14588    Previously, Xlib/XCB used XCB's Xlib lock to prevent XCB from sending
14589    requests between calls to Xlib's LockDisplay and UnlockDisplay macros.
14590    Xlib/XCB then sent all of its requests using XCB's xcb_send_request, and
14591    had to flush its requests when unlocking the display.
14592    
14593    XCB 1.2 adds a new socket handoff mechanism, xcb_take_socket.  Replace
14594    much of the existing Xlib/XCB implementation with the use of
14595    xcb_take_socket to take ownership of the write side of the X connection
14596    socket, and a return_socket callback which writes any outstanding requests
14597    with xcb_writev.  This approach allows Xlib/XCB to use the same buffering
14598    as traditional Xlib did.  In particular, programs which use Xlib/XCB and
14599    never make XCB calls will never need to hand the socket back to XCB, and
14600    vice versa.
14601    
14602    This allows us to discard large quantities of synchronization code from
14603    Xlib/XCB, together with the synchronization bugs present in that code.
14604    Several test cases which previously failed now work perfectly, including
14605    multi-threaded ico.  In addition, the infamous locking correctness
14606    assertions, triggered when double-locking or when unlocking without a
14607    previous lock, no longer exist, because Xlib/XCB no longer has any reason
14608    to care more about application locking than traditional Xlib does.
14609    
14610    Furthermore, the handoff approach provides great improvements to
14611    performance.  Results from x11perf's XNoOp test, which represented the
14612    worst case for the lock-based Xlib/XCB:
14613    
14614    Traditional Xlib:       average 19100000/sec
14615    Lock-based Xlib/XCB:    average  3350000/sec
14616    Handoff-based Xlib/XCB: average 17400000/sec
14617    
14618    Thus, for no-ops, the handoff mechanism provides more than a 4x speedup to
14619    Xlib/XCB, bringing Xlib/XCB within 9% of traditional Xlib no-op
14620    performance.  Of course, real-world workloads do not use no-op, so your
14621    mileage may vary.  In particular, since no-ops represent the worst case,
14622    we expect real workloads to more closely match the performance of
14623    traditional Xlib.
14624    
14625    While removing synchronization code, we changed _XReply to not drop any
14626    locks when calling xcb_wait_for_reply; previously, we had to carefully
14627    avoid a deadlock between the Display lock and the XCB Xlib lock. Holding
14628    the locks reduces implementation complexity and should not impact
14629    applications.
14630    
14631    Commit by Jamey Sharp and Josh Triplett.
14632    XCB's handoff mechanism inspired by Keith Packard.
14633
14634commit 5a19ac473f7a8046b0421fbd5d53da160c22ed75
14635Author: Chris Ball <cjb@laptop.org>
14636Date:   Mon Nov 3 22:57:29 2008 -0500
14637
14638    Remove configure check for xcb-xlib.
14639    
14640    xcb-xlib has been intentionally removed from libxcb; stop checking for
14641    it at configure-time.
14642
14643commit 34b35dda0bb7f3cf0ad9ab95ad7953d35d24f71b
14644Author: Josh Triplett <josh@freedesktop.org>
14645Date:   Wed Oct 29 14:37:44 2008 -0700
14646
14647    .gitignore: Add dolt files
14648
14649commit 1290cccf2d90083eba852f5f413f7e3dff48ccd2
14650Author: Peter Hutterer <peter.hutterer@redhat.com>
14651Date:   Tue Oct 28 11:56:55 2008 +1030
14652
14653    man: fix formatting error in XkbGetIndicatorState man page.
14654
14655commit b1022fa6d7e97640049e93ffa108083fc8d71b05
14656Author: James Cloos <cloos@jhcloos.com>
14657Date:   Sat Oct 25 09:13:08 2008 -0400
14658
14659    Increase size of working arrays in the makekeys utility program.
14660    
14661    Makekeys is used to create an optimal hash of the keysyms defined
14662    in x11proto’s keysymdef.h.
14663    
14664    The recent addition of new keysyms there has triggered a bug in
14665    makekeys where it tries to use a zero on the rhs of the % (mod)
14666    operator (resulting in a divide by zero error) whenever it fails
14667    to find a solution within its constraints.
14668    
14669    Increasing the size of the arrays allows it to find a solution for
14670    the current set of keysyms.
14671    
14672    Makekeys is only run durring the build process, so this has no impact
14673    on users of libX11, only on the amount of VM needed to build it.
14674    
14675    It still needs a more complete fix, but this allows compiles to
14676    progress until that is completed.
14677
14678commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60
14679Author: James Cloos <cloos@jhcloos.com>
14680Date:   Sat Oct 11 01:03:14 2008 -0400
14681
14682    Dolt-ify
14683    
14684    Add dolt to acinclude.m4 and call it it configure.ac to speed compiles.
14685
14686commit 39c0b266cac8cbc15bf501d7869186862f01d823
14687Author: Peter Hutterer <peter.hutterer@redhat.com>
14688Date:   Wed Oct 15 14:30:20 2008 +1030
14689
14690    Add more keysyms for PS3 BD remotes, Ericsson Phones #16519
14691    
14692    X.Org Bug 16519 <https://bugs.freedesktop.org/show_bug.cgi?id=16519>
14693
14694commit d23aad31338e7d869d878d5aa1b6b91d20287005
14695Author: Peter Hutterer <peter.hutterer@redhat.com>
14696Date:   Mon Oct 13 09:41:59 2008 +1030
14697
14698    Add XF86Battery, XF86Bluetooth, XF86WLAN, XF86UWB to keysymdb.
14699
14700commit 214ea6f5fd6aeaa7303ea4a69f9aedabf219ec4c
14701Author: Peter Hutterer <peter.hutterer@who-t.net>
14702Date:   Thu Jul 24 15:44:26 2008 +0930
14703
14704    xkb: fix out-by-1 error in _XkbWriteKeyExplicit.
14705    
14706    Thanks to Michael Meeks, Novell Bug 369263.
14707    https://bugzilla.novell.com/show_bug.cgi?id=369263
14708
14709commit e7ece39afc8e0adc3b6b1e70b337b98376754462
14710Author: Alan Coopersmith <alan.coopersmith@sun.com>
14711Date:   Tue Oct 7 15:41:38 2008 -0700
14712
14713    Sun bug #6739431: double free in _X11TransConnectDisplay()
14714    
14715    Double free() introduced in bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
14716    After copying original_hostname to phostname, set original_hostname
14717    to NULL, so we don't free the same pointer twice when we free both
14718    original_hostname and phostname.
14719    
14720    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6739431>
14721
14722commit 0877bc916afbd1ff8f1833edc930b765ea783576
14723Author: Daniel Stone <daniel@fooishbar.org>
14724Date:   Tue Sep 23 19:02:02 2008 +0300
14725
14726    configure.ac: Fix CC_FOR_BUILD logic error
14727    
14728    Turns out we were accidentally smashing it so that you couldn't set it
14729    externally at all.  Oops.
14730
14731commit 58bf3aa746908f01c37be7045699e43a4e270944
14732Author: John Tapsell <johnflux@gmail.com>
14733Date:   Tue Sep 23 17:30:13 2008 +0300
14734
14735    Build: Use native compiler for makekeys
14736    
14737    makekeys needs to be run during the build process, as opposed to on the
14738    target, so build it with either of gcc or cc to fix cross-compiling.
14739    This can be overridden by setting $CC_FOR_BUILD.
14740
14741commit 340422a5c7a413faef18666cada27cee14615250
14742Author: Adam Jackson <ajax@redhat.com>
14743Date:   Wed Sep 17 12:54:34 2008 -0400
14744
14745    Fix the previous patch for the BadFont case.
14746
14747commit 2335eafe4b53c27f6f9ee1bab3e1f5842f896428
14748Author: Matthias Clasen <mclasen@redhat.com>
14749Date:   Wed Sep 17 10:43:52 2008 -0400
14750
14751    Bug #17616: Fix an XCB leak when the client has a non-fatal error handler.
14752
14753commit db0b85db29699be6bf7e78dede655d59ba926dfc
14754Author: Rafael Ávila de Espíndola < <rafael.espindola@gmail.com>
14755Date:   Sun Sep 14 19:15:26 2008 -0400
14756
14757    Fix problem with <dead_acute> <c> in pt_BR.UTF-8
14758    
14759    The <dead_acute> <C> and <dead_acute> <c> lines in the pt_BR UTF-8
14760    Compose file show "Ç" and "ç" (c with cedilla accent) (akin to the
14761    ISO 8859 pt_BR Compose file) as the string but specify the keysym
14762    and comment for Ć and ć (c with acute accent).
14763    
14764    This commit normalizes those two lines to match the specified string.
14765    
14766    Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=4671
14767    
14768    Signed-off-by: James Cloos <cloos@jhcloos.com>
14769
14770commit b065c011baa69b69e3ea77c30d5e153c0d103e2d
14771Author: Michael Verret <michael.verret@gmail.com>
14772Date:   Mon Sep 8 16:33:35 2008 -0400
14773
14774    Fix documentation typo
14775    
14776    Signed-off-by: James Cloos <cloos@jhcloos.com>
14777
14778commit 4213ea95185377bdd1b51e82933f331fc0f52e5b
14779Author: James Cloos <cloos@jhcloos.com>
14780Date:   Sat Sep 6 04:19:19 2008 -0400
14781
14782    Remove extraneous <angle brackets> from the Ethiopic Compose file.
14783    
14784    The am_ET.UTF-8 Compose file submitted in:
14785    
14786    https://bugs.freedesktop.org/show_bug.cgi?id=11307
14787    
14788    for the OLCP project used incorrect syntax.  (It has angle brackets around the
14789    Uxxxx symbols on the right hand side rather than only on the left hand side).
14790    
14791    This bug is noted in OLPC’s ticket:
14792    
14793    http://dev.laptop.org/ticket/7474
14794    http://dev.laptop.org/attachment/ticket/7474/olpc_7474_dead_vowels_libX11.patch
14795
14796commit 9df84b513dd2b6e65e6d528cfac6d4cc3ea46918
14797Author: James Cloos <cloos@jhcloos.com>
14798Date:   Mon Sep 1 17:49:33 2008 -0400
14799
14800    Complete the set of vulgar fractions
14801    
14802    Unicode 1.1 added thirds, fifths, sixths and eights;
14803    we might as well catch up.
14804    
14805    (Unicode and ISO 10646 have 1/7 (U2150), 1/9 (U2151), 1/10 (U2152)
14806    and 0/3 (U2189) in their pipelines, but those four can be added
14807    here after they are published.)
14808
14809commit a788792e9de95f8db0639557859722a35087481d
14810Author: James Cloos <cloos@jhcloos.com>
14811Date:   Wed Aug 20 15:28:07 2008 -0400
14812
14813    nls (en_US) Re-remove long compositions that override shorter
14814    
14815    As reported in <https://bugs.freedesktop.org/show_bug.cgi?id=17228>:
14816    
14817      Commit a6f4bbf7
14818        nls (en_US): remove long compositions that override shorter [...]
14819        removed some longer compose sequences because there are shorter
14820        ones which take preference over the longer. For example the
14821        sequences:
14822    
14823          <Multi_key> <apostrophe> <comma> <c>   : U1E09 # ḉ
14824          <Multi_key> <apostrophe> <comma> <C>   : U1E08 # Ḉ
14825    
14826        were removed becase there already was:
14827    
14828          <Multi_key> <apostrophe> <comma>       : U201A # ‚
14829    
14830      Then commit 4ba09125
14831        Work on making the en_US and pt_BR UTF-8 Compose as similar as
14832        possible added exactly the same key sequences again. Obviusly
14833        they won't work.
14834
14835commit 55248e5c84c3fd8c349a3bb4cb15a1ec86989d74
14836Author: James Cloos <cloos@jhcloos.com>
14837Date:   Thu Jul 17 21:01:42 2008 -0400
14838
14839    Add more <Multi_key> <cedilla> Compose tuples
14840    
14841    The last commit missed the el_GR UTF-8 Compose.pre as well as
14842    the various ISO 8859 locales which have compose sequences
14843    generating ‘WITH CEDILLA’ characters.
14844    
14845    (Interestingly, some of the 8859 locales already supported
14846    <Multi_key> <cedilla> for some CEDILLA characters, but not
14847    for Ç or ç.)
14848    
14849    This is further work on bug 10397.
14850
14851commit 4ba091255bb953d53078ba5619d6751052c739f7
14852Author: James Cloos <cloos@jhcloos.com>
14853Date:   Thu Jul 17 17:16:50 2008 -0400
14854
14855    Work on making the en_US and pt_BR UTF-8 Compose as similar as possible.
14856    
14857    The eventual goal here is to have a single primary UTF-8 Compose
14858    file which the locale-specific UTF-8 Compose.pre files can #include.
14859
14860commit 254522d3c24e0590732fc03cdd61ff4564819d94
14861Author: James Cloos <cloos@jhcloos.com>
14862Date:   Thu Jul 17 17:13:36 2008 -0400
14863
14864    Add <Multi_key> <cedilla> Compose tuples
14865    
14866    The en_US and pt_BR UTF-8 Compose tables had support for using <comma>
14867    with <Multi_key> to enter CEDILLA characters.  Bug 10397 requests
14868    support for using <cedilla> instead of <comma> in said sequences.
14869    
14870    This commit makes both styles work.
14871
14872commit 7dc907f6032e1d5cbe4da0e414bdf2c569c04b44
14873Author: James Cloos <cloos@jhcloos.com>
14874Date:   Sat Jun 28 15:25:23 2008 -0400
14875
14876    Fix commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01
14877    
14878    The new block was added twice to the en_US.UTF-8 Compose.pre;
14879    delete the duplicate.
14880
14881commit 596e081b7457dcd1c4ad555ac140e6999239bc0d
14882Author: Peter Hutterer <peter@cs.unisa.edu.au>
14883Date:   Sat Jun 28 20:14:05 2008 +0930
14884
14885    Fix unbalanced parenthesis in XKBlib.h # 16551
14886    
14887    X.Org Bug 16551 <http://bugs.freedesktop.org/show_bug.cgi?id=16551>
14888
14889commit f6af6dd2f76c12b56ec166bb771457b9f08fe246
14890Author: Adam Jackson <ajax@redhat.com>
14891Date:   Tue Jun 24 13:16:53 2008 -0400
14892
14893    Bug #14898: Don't abuse the sprintf() implementation.
14894    
14895    The thing you're printing into should not itself appear in the list of
14896    things to print from, that's bad juju.  Just use strcat().
14897
14898commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01
14899Author: Khaled Hosny <khaledhosny@eglug.org>
14900Date:   Thu Jun 19 18:26:11 2008 -0400
14901
14902    NLS: Add Arabic Lam-Alef ligature compose sequences (bug #16426)
14903    
14904    Add some Arabic digraphs to utf-8 locales with a Compose.pre
14905    
14906    Signed-off-by: James Cloos <cloos@jhcloos.com>
14907
14908commit bf53987eaf0fbd7f7627783fc39e7ee99e8361ba
14909Author: Alan Coopersmith <alan.coopersmith@sun.com>
14910Date:   Wed Jun 18 20:00:25 2008 -0700
14911
14912    Rework code to choose local connection types and fallback to others
14913    
14914    Adds --with-local-transport-order configure flag if you don't like the
14915    default ordering (which is platform dependent)
14916    
14917    Includes fixes for these Sun/Solaris bug ids:
14918    6678250 X Commands returning incorrect display value unix:0.0 not <system>:0.0
14919    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6678250>
14920    6716481 libX11 should prefer Unix domain sockets over named pipes on Solaris
14921    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6716481>
14922
14923commit cf49e537014c3cb5aaee07e57400933e0bb72b6b
14924Author: Alan Coopersmith <alan.coopersmith@sun.com>
14925Date:   Tue Jun 17 14:41:17 2008 -0700
14926
14927    Strip whitespace from end of lines in source files
14928
14929commit f76fd81dfbbd5cfae75c87ce0511e88e08529cf3
14930Author: Jeff Smith <whydoubt@yahoo.com>
14931Date:   Sun Jun 15 23:52:20 2008 -0500
14932
14933    Fix memory leak in XOpenDisplay
14934    
14935    Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
14936
14937commit fca0b0ba3f72b7284601d4690bba99fc80a92614
14938Author: Jens Herden <jens@khmeros.info>
14939Date:   Tue Jun 10 20:07:30 2008 +0300
14940
14941    NLS: Add Khmer compose sequences (bug #5706)
14942    
14943    Add some Khmer digraphs to all locales with a Compose.pre.
14944
14945commit e54cffb649b1622c17457e470cfab8cc56d38c97
14946Merge: 19802ccd 721b574d
14947Author: Daniel Stone <daniel@fooishbar.org>
14948Date:   Tue Jun 10 20:04:30 2008 +0300
14949
14950    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
14951
14952commit 721b574d36f1884c3f1bf7bd933646e2ed6680b5
14953Author: Peter Hutterer <peter@cs.unisa.edu.au>
14954Date:   Thu May 29 10:57:21 2008 +0930
14955
14956    Bump to 1.1.99.1
14957
14958commit 631d32d13247d1cf52c0833d438c5b38b01b17a4
14959Author: Peter Hutterer <peter@cs.unisa.edu.au>
14960Date:   Wed May 28 17:31:59 2008 +0930
14961
14962    Require xproto 7.0.13 and libxcb 1.1.90 (for GenericEvents)
14963
14964commit e9195db7257e418f83707233baeeb84b225caf4f
14965Merge: c34f76f4 a7f85567
14966Author: Peter Hutterer <peter@cs.unisa.edu.au>
14967Date:   Thu May 22 12:14:28 2008 +0930
14968
14969    Merge branch 'master' into xge
14970
14971commit 19802ccd3909145e3ba2f6c073271cb5f3701685
14972Author: Daniel Stone <daniel@fooishbar.org>
14973Date:   Mon May 19 19:22:31 2008 +0300
14974
14975    gitignore: Update with loads more bits from server
14976
14977commit a7f85567a3e850fba0c44571453d2852ab1a09be
14978Author: Adam Jackson <ajax@redhat.com>
14979Date:   Tue May 13 10:28:39 2008 -0400
14980
14981    Bug #15884: Remove useless sleep()'s from the connection code.
14982    
14983    For network transports, there's enough delay in the network layer
14984    already without adding more.  For local transports, just hurry up
14985    and fail if the server isn't there.
14986
14987commit c34f76f475bc632490122e67b5a82575d69d5569
14988Author: Peter Hutterer <peter@cs.unisa.edu.au>
14989Date:   Mon May 12 21:46:24 2008 +0930
14990
14991    Pull down extra bytes when reading a GenericEvent (non-xcb).
14992    
14993    I refuse to take any responsibily for this code. It works, I guess.
14994    But - all the flushing is done somewhere before that, so we might need to
14995    flush here. Under some circumstances anyway. Don't ask me, I'm an optical
14996    illusion.
14997    
14998    Build with xcb as transport layer highly recommended.
14999
15000commit c9b2ff1e6a607463993afa4a8d085857d97cc2f3
15001Merge: 17d7dcbf 9129057b
15002Author: Peter Hutterer <peter@cs.unisa.edu.au>
15003Date:   Mon May 12 17:58:37 2008 +0930
15004
15005    Merge branch 'master' into xge
15006
15007commit 9129057bdbff0ec9cd8bb780cf7f85f134a291eb
15008Author: Teemu Likonen <tlikonen@iki.fi>
15009Date:   Wed May 7 21:44:22 2008 +0300
15010
15011    Change <dead_belowdot> to <dead_belowring> for U+1E00 and U+U1E01
15012    
15013    Commit 6b6caeea830a977bdb54688cfb648d879821e752 added <dead_belowdot>
15014    <A> and <dead_belowdot> <a> compose sequences for letters U+1E00 and
15015    U+U1E01 (LATIN CAPITAL/SMALL LETTER A WITH RING BELOW). This caused
15016    duplicate compose sequences since these have already been defined. Also,
15017    using <dead_belowring> is more logical since the diacritic is indeed
15018    a "RING BELOW".
15019
15020commit 01a9cb58888d290cc3d319feec4ee4a0297a844c
15021Author: Daniel Stone <daniel@fooishbar.org>
15022Date:   Wed May 7 20:04:44 2008 +0300
15023
15024    NLS: Make UTF-8 the default for Russian
15025    
15026    No-one uses 8859-5 anymore, so make the default for Russian UTF-8; the
15027    only other possible answer would be KOI8-R.
15028    
15029    Signed-off-by: Sergey V. Udaltsov <sergey.udaltsov@gmail.com>
15030
15031commit 407b81bfbbabf6feb565d6da22f9ef9a69016ab8
15032Author: Ross Burton <ross@burtonini.com>
15033Date:   Tue Apr 29 13:38:10 2008 +0300
15034
15035    NLS: Add interrobang to UTF-8 compose tables (bug #15653)
15036    
15037    It is what it says on the box.
15038
15039commit 0b6682303e9c61fefc3818acfda616b1e3691abf
15040Author: Theppitak Karoonboonyanan <thep@linux.thai.net>
15041Date:   Mon Apr 28 11:51:25 2008 +0300
15042
15043    IM: Respect XMODIFIERS for Thai locale (bug #15719)
15044    
15045    When looking at Thai input methods, make sure XMODIFIERS is checked
15046    before jumping straight into built-in Thai processing, so external XIM
15047    servers such as SCIM can be used with Thai.
15048
15049commit c13aded1b2f830ba5004abb0ec5518f9ea16087e
15050Author: Colin Harrison <colin.harrison-at-virgin.net>
15051Date:   Sat Apr 26 18:56:05 2008 +0100
15052
15053    Fix missing error condition
15054
15055commit f5c5ffc175cb383c92ea0fa8c08cfb087c5f3083
15056Author: Colin Harrison <colin.harrison-at-virgin.net>
15057Date:   Mon Apr 21 17:24:33 2008 +0100
15058
15059    Xlib warning fixes
15060
15061commit 6b6caeea830a977bdb54688cfb648d879821e752
15062Author: James Cloos <cloos@jhcloos.com>
15063Date:   Fri Apr 18 02:50:55 2008 -0400
15064
15065    Add some dead_key sequences to en_US.UTF-8 Compose table
15066    
15067    Make use of the new dead key symbols added to x11proto’s
15068    commit 44e24a27bca023cf7b799f191fe6d52e12efbe5f (which
15069    was in responce to bug #15446).
15070
15071commit 8f9b039580deaf658e464b7d6254064fcf183df6
15072Author: Alan Coopersmith <alan.coopersmith@sun.com>
15073Date:   Mon Apr 14 19:09:42 2008 -0700
15074
15075    Update ac_define_dir macro in acinclude.m4 to 2008-04-12 version
15076
15077commit 9f5e96eb91ab55dd441c3e94b75caf48c588778f
15078Author: Alan Coopersmith <alan.coopersmith@sun.com>
15079Date:   Mon Apr 14 18:21:14 2008 -0700
15080
15081    Fix mismatched brace indenting
15082
15083commit a19f9c65ee9e5e5d783feaa84998c36439b0288b
15084Author: Bart Massey <bart@cs.pdx.edu>
15085Date:   Fri Apr 4 18:58:45 2008 -0700
15086
15087    added error check in Xcms color file parser; closes bug #15305
15088
15089commit 12e8d0d01dd72ce98e7683ddb1bde181b7ed246f
15090Author: Christian Weisgerber <naddy@mips.inka.de>
15091Date:   Tue Mar 18 07:30:05 2008 +0100
15092
15093    ConnDis: properly cast 'addr' before accessing it as a byte array.
15094    
15095    If you use XDM-AUTHORIZATION-1 authorization keys for remote X11
15096    clients over IPv6, the clients are liable to segfaults.
15097
15098commit 64325f38bab082a8e0e9ce779a8e582de5c8588e
15099Author: Josh Triplett <josh@freedesktop.org>
15100Date:   Sat Mar 15 12:29:33 2008 -0700
15101
15102    Fix fd.o bug 15023: make Xlib sync correctly given many void requests
15103    
15104    If given many requests without replies, Xlib may not sync until it flushes
15105    the output buffer.  Thus, if Xlib can fit enough requests in the buffer to
15106    pass by the number of requests it would normally sync after (65536 -
15107    BUFSIZE/sizeof(xReq)), it will sync too late.  The test case in bug 15023
15108    demonstrated this by issuing a request with a reply (ListExtensions) at
15109    just the right time to get confused with the GetInputFocus reply issued in
15110    response to the sync 65,536 requests later; the test case used an async
15111    handler to watch the replies, since otherwise it could not issue a request
15112    without waiting for the response.  When the test case failed, Xlib's sync
15113    handler would eat the ListExtensions reply, and the test case's async
15114    handler would see the GetInputFocus reply.
15115    
15116    Fix this by replacing SEQLIMIT with a function sync_hazard() that uses the
15117    buffer size to figure out when the sequence numbers could potentially wrap
15118    before the next flush.
15119    
15120    With this commit, the test case consistently passed, and the async reply
15121    handler always saw the ListExtensions reply.
15122    
15123    Commit by Jamey Sharp and Josh Triplett.
15124
15125commit a5395563bbee15fabe1e8fd7aa86f9f314d8d30e
15126Author: Colin Harrison <colin.harrison@virgin.net>
15127Date:   Sat Mar 15 13:39:13 2008 -0400
15128
15129    Fix typo
15130    
15131    Signed-off-by: James Cloos <cloos@jhcloos.com>
15132
15133commit f07585ca27a8487bc66dfe41486c823f0fdcea7d
15134Author: Daniel Stone <daniel@fooishbar.org>
15135Date:   Sat Mar 15 17:32:57 2008 +0200
15136
15137    configure.ac: Don't search for legacy X11 headers
15138    
15139    This can actually break cross-compiles, so don't do it anymore.
15140
15141commit bf69541238c7df6606340c0f389e5c47149b29c7
15142Author: Matthieu Herrb <matthieu.herrb@laas.fr>
15143Date:   Sun Mar 9 09:08:07 2008 +0100
15144
15145    nuke RCS Ids
15146
15147commit 5e98aed13e529638df744e45893c471d5f2014fb
15148Author: Adam Jackson <ajax@redhat.com>
15149Date:   Thu Mar 6 16:10:33 2008 -0500
15150
15151    libX11 1.1.4
15152
15153commit 8e085971dc661da9f80ff6b67747459c0fb15c08
15154Author: Alan Coopersmith <alan.coopersmith@sun.com>
15155Date:   Thu Feb 28 20:17:41 2008 -0800
15156
15157    Man page typo fixes
15158
15159commit 1a1a42a3ca1dfaf42f1094936b71c140fc030fcb
15160Author: Søren Sandmann Pedersen  <sandmann@redhat.com>
15161Date:   Sun Feb 24 20:03:35 2008 -0500
15162
15163    XIM: Fix a hand when switching input context.
15164    
15165    Red Hat bug #201284.
15166
15167commit e02e4ccafcaf3eb8993152dfcbfbee0240ea2db2
15168Author: Adam Jackson <ajax@redhat.com>
15169Date:   Sun Feb 24 20:00:43 2008 -0500
15170
15171    Bug #14029: Don't LockDisplay() recursively.
15172    
15173    See also Red Hat bugzilla #326461.
15174
15175commit e5892467ae3308c8651be76e06db322dcbc08522
15176Author: Alan Coopersmith <alan.coopersmith@sun.com>
15177Date:   Fri Feb 15 17:27:53 2008 -0800
15178
15179    Add support for building lint library with --enable-lint-library
15180
15181commit e3eb83ec6a9bffa63cdffd94f077c12f85ad7240
15182Author: Alan Coopersmith <alan.coopersmith@sun.com>
15183Date:   Mon Feb 11 20:11:43 2008 -0800
15184
15185    Spell out number in XkbGetKeyVirtualModMap man page to avoid cpp errors from #
15186
15187commit d5ceed7a73a6b61758ddb6ff4e194955fbd5c185
15188Author: Alan Coopersmith <alan.coopersmith@sun.com>
15189Date:   Fri Feb 8 16:46:46 2008 -0800
15190
15191    Add WM_LOCALE_NAME to list of properties set in XSetWMProperties comment
15192
15193commit 416a812200f24d19149dcc497e5c51a0608120f6
15194Author: Alan Coopersmith <alan.coopersmith@sun.com>
15195Date:   Fri Feb 8 15:31:31 2008 -0800
15196
15197    XErrorDB updates for Render 0.9 & XFixes 4.0
15198
15199commit 16a76091cd632e5a3708e235ff864b58f3e4613e
15200Author: Kim Woelders <kim@woelders.dk>
15201Date:   Sat Dec 22 21:45:23 2007 +0100
15202
15203    Fix bs_BA entries in locale.dir.pre
15204    
15205    X.Org bug#13786 <http://bugs.freedesktop.org/show_bug.cgi?id=13786>
15206
15207commit 32115c563b87d2f37e3f9de70fbd0f4d9e424aea
15208Merge: 8f0bd3f4 e8d4cefa
15209Author: James Cloos <cloos@jhcloos.com>
15210Date:   Fri Dec 14 22:43:47 2007 -0500
15211
15212    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
15213
15214commit 8f0bd3f445cbdcc67650f6e8e3baf2ed89cb3695
15215Author: James Cloos <cloos@jhcloos.com>
15216Date:   Fri Dec 14 22:42:59 2007 -0500
15217
15218    add a comment to en_US.UTF-8/Compose
15219
15220commit 4d6c45e60ed13d3b0fea10413873d6a74f9d6a3b
15221Author: James Cloos <cloos@jhcloos.com>
15222Date:   Fri Dec 14 22:40:13 2007 -0500
15223
15224    Add <dead_stroke> compose sequences
15225    
15226    The added sequences match the existing <Multi_key> <slash> sequences.
15227    
15228    This is related to bug #12765¹.
15229    
15230    1] https://bugs.freedesktop.org/show_bug.cgi?id=12765
15231
15232commit e8d4cefa0837afa149a10e981528b368485a9e38
15233Author: Jeremy Huddleston <jeremy@yuffie.local>
15234Date:   Mon Dec 10 23:00:44 2007 -0800
15235
15236    Added launchd support.
15237
15238commit 17d7dcbfced4a9417b33507bd3fd9b7dd8268242
15239Merge: 5dfefd38 13ac8046
15240Author: Peter Hutterer <peter@cs.unisa.edu.au>
15241Date:   Thu Dec 6 14:08:29 2007 +1030
15242
15243    Merge branch 'master' into xge
15244
15245commit 5dfefd3829d3ba7e41d5db0ad28e9dfee92fadd5
15246Merge: a68a1cd7 eff33ae5
15247Author: Peter Hutterer <peter@cs.unisa.edu.au>
15248Date:   Thu Dec 6 13:57:09 2007 +1030
15249
15250    Merge branch 'master' into xge
15251    
15252    Conflicts:
15253    
15254            src/xcb_io.c
15255
15256commit 13ac80469f6958cabac596834e203bd9cb6d4c94
15257Author: James Cloos <cloos@jhcloos.com>
15258Date:   Wed Dec 5 20:14:03 2007 -0500
15259
15260    Update the currency symbols block of en_US.UTF-8/Compose.pre
15261    Add XCOMM lines a la the existing NEW SHEQEL SIGN entry for
15262    the KIP, TUGRIK, DRACHMA. GERMAN PENNY, PESO, GUARANI,
15263    AUSTRAL, HRYVNIA and CEDI SIGNs.
15264
15265commit b0a8f2ec4ba698841683f8ce389f9d72e6bce53e
15266Author: Anton Zinoviev <anton@lml.bas.bg>
15267Date:   Wed Dec 5 19:56:03 2007 -0500
15268
15269    Additions to the Compose file for UTF-8
15270    From bug #5371¹
15271    
15272    Commit 5cf5bc76642bfece7cb5b76faf414bf445f14489 left out this change
15273    from those in attachment #4122².  The post³ on xorg resulted in only
15274    a positive reply⁴ from Daniel, so this block is now also commited.
15275    
15276    1] https://bugs.freedesktop.org/show_bug.cgi?id=5371
15277    2] https://bugs.freedesktop.org/attachment.cgi?id=4122
15278    3] http://article.gmane.org/gmane.comp.freedesktop.xorg/20628
15279    4] http://article.gmane.org/gmane.comp.freedesktop.xorg/23966
15280    
15281    Signed-off-by: James Cloos <cloos@jhcloos.com>
15282    Acked-by:  Daniel Stone <daniel@fooishbar.org>
15283
15284commit 438d02ebc08ee171cf1d3936f4c81050d428ab92
15285Author: James Cloos <cloos@jhcloos.com>
15286Date:   Tue Dec 4 17:25:39 2007 -0500
15287
15288    Fix the <U\x+> keysyms in the en_US.UTF-8 Compose file
15289    
15290    Based on src/KeysymStr.c and src/StrKeysym.c and comments in
15291    bugs #11930¹ and #5129² it is clear that <U100XXXXX> is invalid;
15292    those should be in the form U plus the hex of the UCS Code Point.
15293    
15294    The 0x01000000 is ORed in by the code.
15295    
15296    This update fixes all of those.
15297    
15298    1] https://bugs.freedesktop.org/show_bug.cgi?id=11930
15299    2] https://bugs.freedesktop.org/show_bug.cgi?id=5129
15300
15301commit 02e04059c89e175f51647e3b031344f743286b34
15302Merge: 1254c57d b57129ef
15303Author: James Cloos <cloos@jhcloos.com>
15304Date:   Tue Dec 4 06:55:04 2007 -0500
15305
15306    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
15307
15308commit 1254c57dd3a8d6ea87041b2f63024f99094f290f
15309Author: James Cloos <cloos@jhcloos.com>
15310Date:   Tue Dec 4 06:53:55 2007 -0500
15311
15312    Use the new dead_psili and dead_dasia keysyms added to proto/x11proto 7.0.11
15313    Inspired by bug 11930¹:
15314    
15315    Commit 40ed4eef92e31fcf7ea0a436e1a00cdf49484c1b to x11proto added dead_psili
15316    and dead_dasia keysyms.  Make use of them in the en_US.UTF-8 and el_GR.UTF-8
15317    Compose files.
15318    
15319    This was done with a pair of perl scripts based on the one quoted in the
15320    log for commit c76d30253f1483ac8200ad5c032a818907e65030.
15321    
15322    1] https://bugs.freedesktop.org/show_bug.cgi?id=11930
15323
15324commit b57129ef324c73ee91c2a796b800c4b45f4d4855
15325Author: Jeremy Huddleston <jeremy@yuffie.local>
15326Date:   Mon Dec 3 20:04:19 2007 -0800
15327
15328    Use __APPLE__ instead of __DARWIN__
15329
15330commit b9c032e1d5ed08510863dfb91b8bda588e6c8c9e
15331Author: Jeremy Huddleston <jeremy@yuffie.local>
15332Date:   Wed Nov 28 16:43:49 2007 -0800
15333
15334    Define __DARWIN__ when host_os == darwin* as needed in SetLocale.c
15335
15336commit 4b91ed099554626f1ec17d5bdf7bd77ce1a70037
15337Author: Alan Coopersmith <alan.coopersmith@sun.com>
15338Date:   Wed Nov 21 16:50:04 2007 -0800
15339
15340    X.Org Bug #4312: incorrect comment asterisk in XAnyEvent.3x man page
15341    
15342    X.Org Bugzilla #4312 <http://bugs.freedesktop.org/show_bug.cgi?id=4312>
15343    
15344    Protect /* sequences from cpp pre-processor removal without transforming
15345    to Unicode mathematical asterisk character
15346
15347commit 2af660c2fcd15c86c66459bfc074c190ea1462e6
15348Author: Jamey Sharp <jamey@minilop.net>
15349Date:   Mon Oct 29 10:46:20 2007 -0700
15350
15351    Two threads can request sequence sync and XID fetch simultaneously.
15352    
15353    So don't assert that they can't.
15354    
15355    This makes the Xlib/XCB implementation of _XAllocID more closely
15356    resemble the traditional Xlib version.
15357
15358commit 6e5485e0a5e3ab738becad12193e760c5fee83a4
15359Author: Samuel Thibault <samuel.thibault@ens-lyon.org>
15360Date:   Sun Oct 28 04:44:00 2007 -0800
15361
15362    X.Org Bug #12983: Typos in ./man/XChangeKeyboardControl.man
15363    
15364    <http://bugs.freedesktop.org/show_bug.cgi?id=12983>
15365
15366commit 24527c92fd1f433ea135e85ec876a94a529fe500
15367Author: Yann Droneaud <ydroneaud@mandriva.com>
15368Date:   Wed Oct 24 19:26:07 2007 +0300
15369
15370    XIM: Properly initialise client event
15371    
15372    Make sure all ClientMessage fields are initialised to 0 before we send it.
15373
15374commit e41477f37b0d9b74a056d22dbf0073a94eecb9d6
15375Author: Daniel Stone <daniel@fooishbar.org>
15376Date:   Wed Oct 24 19:22:22 2007 +0300
15377
15378    Colours: Fix --disable-xcms
15379    
15380    Disable large tracts of colour management code when passing
15381    --disable-xcms.
15382
15383commit 11ea09745efa8de7dc82fe30ebd2393f08390957
15384Author: Eric Anholt <eric@anholt.net>
15385Date:   Fri Aug 31 17:30:33 2007 -0700
15386
15387    Bug #2081: Note the range limitation of XSetScreenSaver arguments.
15388
15389commit 31540f1438ec63faf37044f2fd654b335ddf80f0
15390Author: Kristian Høgsberg  <krh@redhat.com>
15391Date:   Wed Aug 29 19:50:57 2007 -0400
15392
15393    Add GLX 1.4 requests and errors.
15394
15395commit 7c996f78914c77fe17e9f4feede980d895d9df51
15396Author: Eric Anholt <eric@anholt.net>
15397Date:   Tue Aug 28 15:15:11 2007 -0700
15398
15399    Add XF86 keyboard/monitor brightness keysyms to the keysymbdb.
15400
15401commit fa4effe82759f864a22a2dc6c920fa72ddb175a8
15402Author: Eric S. Raymond <esr@thyrsus.com>
15403Date:   Sun Jan 14 10:48:00 2007 -0800
15404
15405    Bug #9658: Bad markup on XIfEvent.3x
15406    
15407    X.Org Bugzilla #9658 <https://bugs.freedesktop.org/show_bug.cgi?id=9658>
15408
15409commit 74cba78daa738ef4d92096107d95c1a585933666
15410Author: Eric S. Raymond <esr@thyrsus.com>
15411Date:   Sun Jan 14 10:43:00 2007 -0800
15412
15413    Bug #9655: Bad markup in XrmUniqueQuark.3x
15414    
15415    X.Org Bugzilla #9655 <https://bugs.freedesktop.org/show_bug.cgi?id=9655>
15416
15417commit 4341d1a34b2a2e460b58131b6fd81935f3355bbc
15418Author: Eric S. Raymond <esr@thyrsus.com>
15419Date:   Sun Jan 14 10:43:00 2007 -0800
15420
15421    Bug #9654: Bad markup in XrmGetFileDatabase.3x
15422    
15423    X.Org Bugzilla #9654 <https://bugs.freedesktop.org/show_bug.cgi?id=9654>
15424
15425commit 2db713252090cae08f0200fecad4fc25fb64c8b1
15426Author: Eric S. Raymond <esr@thyrsus.com>
15427Date:   Sun Jan 14 10:40:00 2007 -0800
15428
15429    Bug #9653: Bad markup in XQueryColor.3x
15430    
15431    X.Org Bugzilla #9653 <https://bugs.freedesktop.org/show_bug.cgi?id=9653>
15432
15433commit 9d3ceea4b902e0471824c1e07ad64342b9a1114e
15434Author: Eric S. Raymond <esr@thyrsus.com>
15435Date:   Sun Jan 14 10:40:00 2007 -0800
15436
15437    Bug #9652: Bad markup in XDrawArc.3x
15438    
15439    X.Org Bugzilla #9652 <https://bugs.freedesktop.org/show_bug.cgi?id=9652>
15440
15441commit 2e7e0748d353d7f53bbd65ec6bf0df8758528ddc
15442Author: Eric S. Raymond <esr@thyrsus.com>
15443Date:   Sun Jan 14 10:39:00 2007 -0800
15444
15445    Bug #9651: Bad markup in XcmsColor.3x
15446    
15447    X.Org Bugzilla #9651 <https://bugs.freedesktop.org/show_bug.cgi?id=9651>
15448
15449commit c316aaf0aab06951db9dc5c9c1148bfea835d885
15450Author: Eric S. Raymond <esr@thyrsus.com>
15451Date:   Sun Jan 14 10:38:00 2007 -0800
15452
15453    Bug #9650: Bad markup in XLoadFont.3x manual page
15454    
15455    X.Org Bugzilla #9650 <https://bugs.freedesktop.org/show_bug.cgi?id=9650>
15456
15457commit ca5d9a625ea0965853fa9e74a448b8c29c78ec95
15458Author: Eric S. Raymond <esr@thyrsus.com>
15459Date:   Tue Jan 2 10:18:00 2007 -0800
15460
15461    Bug 9523: Markup problems in XQueryExtension.3x
15462    
15463    X.Org Bugzilla #9523 <https://bugs.freedesktop.org/show_bug.cgi?id=9523>
15464
15465commit fe713c616e29ba19c179b43c18eca1035079ce18
15466Author: Ian Romanick <idr@us.ibm.com>
15467Date:   Tue Aug 21 14:56:33 2007 -0700
15468
15469    Make sure nls/am_ET.UTF-8/Makefile is created by configure.
15470
15471commit e3430616f26b68e1439143cbe10732f3fc329d20
15472Author: Eric S. Raymond <esr@thyrsus.com>
15473Date:   Tue Jan 2 08:40:00 2007 -0800
15474
15475    Bug #9516: Markup error in XAllocWMHints.3x
15476    
15477    X.Org Bugzilla #9516 <https://bugs.freedesktop.org/show_bug.cgi?id=9516>
15478
15479commit d8fe979fc929833e8c754aed32641786d5a0622b
15480Merge: 21ca9533 4ec1723f
15481Author: James Cloos <cloos@jhcloos.com>
15482Date:   Mon Aug 20 15:34:50 2007 -0400
15483
15484    Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11
15485
15486commit 21ca953337fb221b85345bf35ce1a98a0dcb2bf2
15487Author: James Cloos <cloos@jhcloos.com>
15488Date:   Mon Aug 20 15:34:03 2007 -0400
15489
15490    Fix typo
15491    The code <U1000000D> was used where <U10000DC> was obviously intended.
15492    
15493    It is possible that <Udiaeresis> should be used instead, if that will
15494    not break anyone’s setup.
15495
15496commit 4ec1723fff729440cd3349c1f95d87d2a6ba89cf
15497Author: James Cloos <cloos@jhcloos.com>
15498Date:   Mon Aug 20 15:25:48 2007 -0400
15499
15500    Add compose file for Ethiopic to match new keyboard in xkeyboard-config
15501    From bug report:
15502    
15503      https://bugs.freedesktop.org/show_bug.cgi?id=11307
15504
15505commit eff33ae525337ce2026be135a26464c7b1237113
15506Author: James Cloos <cloos@jhcloos.com>
15507Date:   Sat Aug 18 17:58:23 2007 -0400
15508
15509    Patch for Catalan locales
15510    From bugzilla bug 10943¹:
15511    
15512      There are several Catalan locale codes which presently can
15513      be used in X11 systems; especially after they were accepted
15514      in belocs-locale-data².
15515    
15516      In the following patches, I³ add ca_AD, ca_FR and ca_IT Catalan
15517      locale codes.  For instance, without this, using ca_AD (actually
15518      a quite used locale⁴) some applications (eg. Emacs or Skype)
15519      cannot display Catalan diacritic marks as you type them.
15520    
15521    1] https://bugs.freedesktop.org/show_bug.cgi?id=10943
15522    2] http://lists.debian.org/debian-devel-changes/2005/07/msg01429.html
15523    3] Toni Hermoso Pulido <toniher@softcatala.org>
15524    4] https://launchpad.net/~ubuntu.cat/+members
15525
15526commit 1f980cb7d022f53d0aee9e793b08203fb888e86e
15527Author: James Cloos <cloos@jhcloos.com>
15528Date:   Sat Aug 18 17:47:04 2007 -0400
15529
15530    Add additional Euro signs to compose
15531    Inspired by bug 7419¹ make all of:
15532    
15533        C=, =C, c=, =c, E=, =E, e=, E=
15534    
15535    after <Multi_key> generate € U+20AC EURO SIGN.
15536    
15537    1] https://bugs.freedesktop.org/show_bug.cgi?id=7419
15538
15539commit 4b0a14521449dfce8b4347bd17243efd1d3eae2d
15540Author: James Cloos <cloos@jhcloos.com>
15541Date:   Sat Aug 18 17:29:08 2007 -0400
15542
15543    Compose fix for Latin-1 (from Debian)
15544    The description from bugzilla bug 7417¹ is:
15545    
15546      We've been shipping this patch for some time in Debian now. The
15547      problem description from the patch header is reproduced below. You
15548      may want to note the licensing issue mentioned below, but we've been
15549      shipping it because the method by which this particular patch was
15550      generated and updated was also given below.
15551    
15552      This patch by Denis Barbier.
15553    
15554      The X11 protocol states that Unicode keysyms are in the range
15555      0x01000100 - 0x0110FFFF.  If the result of composing characters is a
15556      Unicode codepoint, X returns the corresponding Unicode keysym, which
15557      is its Unicode codepoint augmented by 0x01000000.  Latin-1
15558      characters must not appear with their Unicode codepoints in compose
15559      files, otherwise the returned composed character lies in the range
15560      0x01000000 - 0x010000FF which is not valid.
15561    
15562      There are two solutions: either fix composing routines to return
15563      0xZZ instead of 0x010000ZZ (where Z is an hexadecimal digit), or
15564      replace U00ZZ by their corresponding keysyms in compose files.  The
15565      latter is more logical and less error prone, so compose files will
15566      be patched.  Many applications accept these invalid Unicode keysyms,
15567      but few of them don't, most notably xemacs.  Only UTF-8 locales are
15568      affected.
15569    
15570      This has been fixed very recently in XFree86 CVS (but not xorg), but
15571      for licensing reasons, this patch is not grabbed.  Instead automatic
15572      conversion is performed by:
15573    
15574        sed -e '/XK_LATIN1/,/XK_LATIN1/!d' /usr/include/X11/keysymdef.h \
15575        | grep -v deprecated | grep 0x0 \
15576        | sed -e 's/0x0/U0/' -e 's/XK_//' \
15577        | awk '{ printf "s/\\b%s\\b/%s/ig\n", $3, $2; }' > sedfile
15578        for f in nls/*.UTF-8/Compose.pre
15579        do
15580          sed -f sedfile $f > $f.tmp && mv $f.tmp $f
15581        done
15582    
15583    [I edited the quoted script to update it for the current location of
15584    the installed keysymdef.h and the current layout of the libX11
15585    repo. -JimC]
15586    
15587    I applied the script, not the patch attached to the bugreport.
15588    
15589    1] https://bugs.freedesktop.org/show_bug.cgi?id=7417
15590
15591commit 5cf5bc76642bfece7cb5b76faf414bf445f14489
15592Author: James Cloos <cloos@jhcloos.com>
15593Date:   Sat Aug 18 17:13:41 2007 -0400
15594
15595    Add some compose sequences
15596    
15597    Add some compose sequences from the patch in bug 5371 (attachment 4122).
15598    
15599    Cf:
15600    
15601    https://bugs.freedesktop.org/show_bug.cgi?id=5371
15602    https://bugs.freedesktop.org/attachment.cgi?id=4122
15603
15604commit d4002e389dd69780dfc7c2f7bd3cb0c57f05d4f8
15605Author: James Cloos <cloos@jhcloos.com>
15606Date:   Sat Aug 18 13:57:31 2007 -0400
15607
15608    Fix SMP Compose targets
15609    The compose targets from the SMP (plane 1) were incorrect.
15610    
15611    At some point the 0x10000 bit had been lost.
15612
15613commit f1ed3da9a30a1f0264fdc7d1c6466f27fe2a3d7d
15614Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
15615Date:   Thu Aug 16 17:37:22 2007 -0500
15616
15617    Fix a mutex reference-counting bug.
15618    
15619    I was told that some systems have a much more permissive libpthread.
15620    I was asked to commit this. This is from NetBSD's X source.
15621    (I didn't receive any feedback on xorg list for over two weeks
15622    about this.)
15623
15624commit ac00a44b4875de70382da5a40dd87f976e5b9327
15625Author: Tilman Sauerbeck <tilman@code-monkey.de>
15626Date:   Wed Aug 1 20:36:03 2007 +0200
15627
15628    Bumped version to 1.1.3.
15629
15630commit 76fae9cba1e9bdf7f0eb2ff2b90153d622136cf0
15631Author: Joerg Sonnenberger <joerg@netbsd.org>
15632Date:   Fri Jul 27 11:15:47 2007 -0700
15633
15634    Fail properly on errors in recursive make.
15635
15636commit 1a18319b3bde08dd9ef69c7cd735a76000cf3177
15637Author: Brice Goglin <Brice.Goglin@ens-lyon.org>
15638Date:   Thu Jul 26 23:31:15 2007 +0200
15639
15640    Add missing override parameter in XrmCombineDatabase prototype in the manpage
15641    
15642    Reported by Arnaud Giersch <arnaud.giersch@iut-bm.univ-fcomte.fr>
15643    in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=393434
15644    and https://bugs.freedesktop.org/show_bug.cgi?id=9948
15645    
15646    Also add the type of the second argument in XrmMergeDatabases.
15647
15648commit 6f0764d4b56f64786b4980839ca262f10a51af6f
15649Author: Brice Goglin <Brice.Goglin@ens-lyon.org>
15650Date:   Thu Jul 26 22:53:52 2007 +0200
15651
15652    Clarify return value of XGetCommand in case of error in the manpage
15653    
15654    Reported by Sean Perry <shalehperry@attbi.com>
15655    in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=133348
15656    and https://bugs.freedesktop.org/show_bug.cgi?id=9828
15657
15658commit b8bef57342632cc2d25580bb7daa3839bae04d89
15659Author: Alan Coopersmith <alan.coopersmith@sun.com>
15660Date:   Wed Jul 25 17:44:06 2007 -0700
15661
15662    Include comment/copyright/license for AC_DEFINE_DIR in acinclude.m4
15663
15664commit 590cde811a79375231c59ed8583e02b111ed567c
15665Author: Kean Johnston <kean@armory.com>
15666Date:   Thu Jul 19 16:44:20 2007 -0700
15667
15668    Reset hostname when falling back from :0 to hostname/localhost:0 as well
15669
15670commit d334665e619e9db657a2ea2764a8b852401d4a3f
15671Author: Kean Johnston <kean@armory.com>
15672Date:   Tue Jul 17 16:34:30 2007 -0700
15673
15674    LOCALCONN fallback changes DisplayString() output, breaks KDE
15675    
15676    See <http://lists.freedesktop.org/archives/xorg/2007-July/026443.html>
15677
15678commit 8e76bcf3cafda85058ce5f35f1f81929f4772f72
15679Author: Olivier Blin <blino@mandriva.org>
15680Date:   Mon Jun 11 18:32:09 2007 +1000
15681
15682    fix XGetMotionEvents arguments order - Fixes bug 11222
15683
15684commit c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8
15685Author: Jamey Sharp <jamey@minilop.net>
15686Date:   Sun Jun 10 16:19:59 2007 -0700
15687
15688    Fix locking in _XimGetWindowEventmask.
15689    
15690    Now that XFilterEvent drops the Display lock before invoking callback
15691    functions, _XimGetWindowEventmask is called without the lock held. So
15692    when it called _XGetWindowAttributes, a variant of XGetWindowAttributes
15693    that does not lock the Display, Xlib/XCB would assert:
15694    
15695    xcb_xlib.c:50: xcb_xlib_unlock: Assertion `c->xlib.lock' failed.
15696    
15697    Should fix Gentoo #156367, Ubuntu #87947, Debian #427296. And others?
15698
15699commit 65f0ab5d46d80d55fc04d4eb14fa05f130eb8b9c
15700Author: Jamey Sharp <jamey@minilop.net>
15701Date:   Sun Jun 3 21:41:47 2007 -0700
15702
15703    Move security fixes to the top of NEWS, and fix spacing.
15704    
15705    Commit by Josh Triplett and Jamey Sharp.
15706
15707commit a549a258b8fcb1ba9d0c1b01b72967e385f67cab
15708Author: Jamey Sharp <jamey@minilop.net>
15709Date:   Sun Jun 3 21:29:40 2007 -0700
15710
15711    Add NEWS item for bugfix in commit e2c1d788d1fe7bd2d34756493951552441e59b8c.
15712    
15713    Commit by Josh Triplett and Jamey Sharp
15714
15715commit 5123b77a3d32d3ad479462f319762c328278aed9
15716Author: Jamey Sharp <jamey@minilop.net>
15717Date:   Sun Jun 3 21:24:54 2007 -0700
15718
15719    Xlib/XCB: Inline and simplify handle_event, since only process_responses calls it now.
15720    
15721    Commit by Josh Triplett and Jamey Sharp.
15722
15723commit 582ca690ea4f3ffd2b94826c4db97229bd3c7238
15724Author: Jamey Sharp <jamey@minilop.net>
15725Date:   Sun Jun 3 20:59:12 2007 -0700
15726
15727    Xlib/XCB: Fix _XReadEvents to always enqueue a new event, even if an error occurs
15728    
15729    Commit c337f9de7cfd89f983f83956b7457a274dd412f3 broke the invariant that
15730    _XReadEvents always enqueues at least one event even if an error occurred,
15731    because the one call to xcb_wait_for_event would then return an error, not an
15732    event, and nothing else ensured that process_responses would obtain an event.
15733    Fix this by reverting most of c337f9de7cfd89f983f83956b7457a274dd412f3 and
15734    f417570735aac865eb6b576d1ea76b5bfcd8573b and implementing the correct fix.  In
15735    process_responses, wait_for_first_event now serves as a flag, cleared when
15736    actually handling an event.
15737    
15738    Commit by Josh Triplett and Jamey Sharp.
15739
15740commit e2c1d788d1fe7bd2d34756493951552441e59b8c
15741Author: Jamey Sharp <jamey@minilop.net>
15742Date:   Sun Jun 3 17:33:23 2007 -0700
15743
15744    Xlib/XCB: Only remove pending_requests when there are provably no more responses.
15745
15746commit 7a6dbd4b07ca0a49c30ca7a1d2437eafb2e15eab
15747Author: Josh Triplett <josh@freedesktop.org>
15748Date:   Sun Jun 3 15:39:39 2007 -0700
15749
15750    Bump version number to 1.1.2, and add NEWS entry for 1.1.2
15751    
15752    Signed-off-by: Josh Triplett <josh@freedesktop.org>
15753
15754commit 416f38f2e67ee1979b3d2feac6f06b3670238804
15755Author: Josh Triplett <josh@freedesktop.org>
15756Date:   Sun Jun 3 12:13:44 2007 -0700
15757
15758    Revert "Revert "include: don't distribute XlibConf.h""
15759    
15760    This reverts commit 79fa3d8070d95b960ba486f2439225872471dadd.
15761    
15762    Re-revert the XlibConf.h change, which prevented distribution, not
15763    installation.
15764
15765commit 79fa3d8070d95b960ba486f2439225872471dadd
15766Author: Josh Triplett <josh@freedesktop.org>
15767Date:   Sat Jun 2 22:05:16 2007 -0700
15768
15769    Revert "include: don't distribute XlibConf.h"
15770    
15771    This reverts commit c9e28e05ae01ce8a29bea09df759b6271865b44c.
15772    
15773    The installed XlibInt.h includes XlibConf.h , so libX11 should ship
15774    XlibConf.h.  (Commit c9e28e05ae01ce8a29bea09df759b6271865b44c didn't actually
15775    prevent automake from shipping XlibConf.h, because it used
15776    nodist_x11include_HEADERS rather than nodist_HEADERS.)
15777
15778commit f417570735aac865eb6b576d1ea76b5bfcd8573b
15779Author: Jamey Sharp <jamey@minilop.net>
15780Date:   Sat Jun 2 17:59:15 2007 -0700
15781
15782    Xlib/XCB: inline wait_or_poll_for_event, which now had only one caller.
15783    
15784    Commit by Josh Triplett and Jamey Sharp.
15785
15786commit c337f9de7cfd89f983f83956b7457a274dd412f3
15787Author: Jamey Sharp <jamey@minilop.net>
15788Date:   Sat Jun 2 17:46:41 2007 -0700
15789
15790    Xlib/XCB: Ensure _XReadEvents reads at least one new event and blocks for exactly one event.
15791    
15792    Commit by Jamey Sharp and Josh Triplett.
15793
15794commit 7f66c897f04806b75e574b55b48921b48045e3f9
15795Author: Jamey Sharp <jamey@minilop.net>
15796Date:   Sat Jun 2 16:43:39 2007 -0700
15797
15798    Update _XReply's copy of _XCBUnlockDisplay's guts.
15799    
15800    We introduced this bug in 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5.
15801    
15802    Also add a comment in _XCBUnlockDisplay to discourage this problem from
15803    respawning.
15804    
15805    Commit by Josh Triplett and Jamey Sharp.
15806
15807commit 740ead23512f8d2eaafaa69e514f1ebafad475b9
15808Author: Jamey Sharp <jamey@minilop.net>
15809Date:   Sat Jun 2 16:01:01 2007 -0700
15810
15811    Xlib/XCB: Avoid re-crashing after _XIOError.
15812    
15813    Commit by Josh Triplett and Jamey Sharp.
15814
15815commit 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5
15816Author: Jamey Sharp <jamey@minilop.net>
15817Date:   Sat Jun 2 12:30:30 2007 -0700
15818
15819    Hold XCB's Xlib lock even when only the user lock (XLockDisplay) is held.
15820    
15821    An Xlib client can query Display state, such as with NextRequest, while
15822    it holds only the Xlib user lock (between XLockDisplay and
15823    XUnlockDisplay), so XCB requests in other threads should be blocked when
15824    the Xlib user lock is held.
15825    
15826    We acquire the lock even when XInitThreads was not called, so that pure
15827    XCB code can use multiple threads even in an otherwise single-threaded
15828    Xlib application.
15829    
15830    Commit by Josh Triplett and Jamey Sharp.
15831
15832commit 95523387d619af5b400748898d722e080b5ce1a6
15833Author: Jamey Sharp <jamey@minilop.net>
15834Date:   Sat Jun 2 11:57:39 2007 -0700
15835
15836    Allow re-entrant Xlib calls from _XIOError.
15837    
15838    Some libraries try to clean up X resources from atexit handlers, _fini,
15839    or C++ destructors. To make these work, the Display lock should be
15840    downgraded to a user lock (as in XLockDisplay) before calling exit(3).
15841    This blocks Xlib calls from threads other than the one calling exit(3)
15842    while still allowing the exit handlers to call Xlib.
15843    
15844    This assumes that the thread calling exit will call any atexit handlers.
15845    If this does not hold, then an alternate solution would involve
15846    registering an atexit handler to take over the lock, which would only
15847    assume that the same thread calls all the atexit handlers.
15848    
15849    Commit by Josh Triplett and Jamey Sharp.
15850
15851commit 91b02b8064f4e0bcc56019f0722914850008a597
15852Author: Tilman Sauerbeck <tilman@code-monkey.de>
15853Date:   Thu May 17 19:07:34 2007 +0200
15854
15855    More constification.
15856
15857commit a4f3841940158351f9424c3f59b305cce877177d
15858Author: Tilman Sauerbeck <tilman@code-monkey.de>
15859Date:   Thu May 17 00:29:43 2007 +0200
15860
15861    Constified composite text charset table.
15862
15863commit 0581c0aa6039e6b2abb9f7b0a4f9904d8e01f00e
15864Author: Tilman Sauerbeck <tilman@code-monkey.de>
15865Date:   Wed May 16 23:19:22 2007 +0200
15866
15867    Have the compiler fill in hexTable so we don't have to do it at runtime.
15868
15869commit 0e8d9ca47dab0d069e305d5784d05f2ade04f0a8
15870Author: Tilman Sauerbeck <tilman@code-monkey.de>
15871Date:   Wed May 16 19:36:23 2007 +0200
15872
15873    More constification.
15874
15875commit 6d2bed8f04942b4de086a519ac693e729c9fdeea
15876Author: Tilman Sauerbeck <tilman@code-monkey.de>
15877Date:   Wed May 16 19:18:20 2007 +0200
15878
15879    Constify and clean up token table.
15880
15881commit 6c508eab5df5d517f7e4cbe6087308cd53a564b2
15882Author: Tilman Sauerbeck <tilman@code-monkey.de>
15883Date:   Wed May 16 18:24:42 2007 +0200
15884
15885    Constified more tables.
15886
15887commit e699c4231c205ef00d687b6412308d031b99806b
15888Author: Tilman Sauerbeck <tilman@code-monkey.de>
15889Date:   Wed May 16 18:01:23 2007 +0200
15890
15891    Constified error list.
15892
15893commit a68a1cd7cb990ba276fbc36a7591044d78b3d3c1
15894Author: Peter Hutterer <peter@cs.unisa.edu.au>
15895Date:   Tue May 15 16:54:01 2007 +0930
15896
15897    Add XGenericEvent definition and handling for long events.
15898
15899commit c76d30253f1483ac8200ad5c032a818907e65030
15900Author: Jan Willem Stumpel <jstumpel@planet.nl>
15901Date:   Fri May 4 12:00:49 2007 -0700
15902
15903    Update el_GR.UTF-8/Compose.pre to match changes in xkeyboard-config cvs
15904    
15905    Cf:
15906    
15907      https://bugs.freedesktop.org/show_bug.cgi?id=10851
15908      https://bugs.freedesktop.org/show_bug.cgi?id=10824
15909      http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=386385
15910    
15911    The greek keyboard definition was changed to replace dead_horn and
15912    dead_ogonek with U0313 COMBINING COMMA ABOVE (aka Psili) and U0314
15913    COMBINING REVERSED COMMA ABOVE (aka Dasia).
15914    
15915    This patch modifies the Greek Compose.pre to match.
15916    
15917    It is generated by the script Jan Willem Stumpel <jstumpel@planet.nl>
15918    posted to 386385@bugs.debian.org:
15919    
15920    #!/usr/bin/perl
15921    while (<>) {
15922      print $_;
15923      if (/dead_horn/) {
15924        s/dead_horn/U0313/;
15925        print $_;
15926      }
15927      elsif (/dead_ogonek/) {
15928        s/dead_ogonek/U0314/;
15929        print $_;
15930      }
15931    }
15932
15933commit a48386ce6b5f8fd2d9dc11a966c9bf5da59f3831
15934Author: Magnus Kessler <Magnus.Kessler@gmx.net>
15935Date:   Tue May 1 15:20:08 2007 +0200
15936
15937    Switched function definitions from K&R to ANSI style.
15938
15939commit 605d357074d556a05a3fba2e85cbea36a3204248
15940Author: Tilman Sauerbeck <tilman@code-monkey.de>
15941Date:   Tue May 1 14:47:03 2007 +0200
15942
15943    Tweaked configure output about the man pages suffix.
15944
15945commit 9824b40d2af4ca2376512c1be7743da0d5065900
15946Author: Alan Coopersmith <alan.coopersmith@sun.com>
15947Date:   Sat Apr 28 00:42:18 2007 -0700
15948
15949    Fix typo in nroff macro in XkbAddGeomOverlayKey.man
15950
15951commit f93849dcc68bd5042ea0884e5190dc7c35b31d68
15952Author: Alan Coopersmith <alan.coopersmith@sun.com>
15953Date:   Sat Apr 28 00:30:55 2007 -0700
15954
15955    Protect C comments and #defines in XKB man pages from being mangled by cpp
15956
15957commit f2f27d4763c7665e422fab10b96b4cf5ad6c0a6f
15958Author: Alan Coopersmith <alan.coopersmith@sun.com>
15959Date:   Sat Apr 28 00:14:50 2007 -0700
15960
15961    Add Makefile to process/install XKB man pages
15962
15963commit d9954c6f6f3a8c406b946acd0d034ff83c656156
15964Author: Dennis Arellano <Dennis.Arellano@Sun.COM>
15965Date:   Thu Aug 19 00:00:00 1999 -0700
15966
15967    Add man pages for XKB API's
15968    
15969    Man pages originally written for X11R6.4 integration to Solaris 7 11/99
15970    Sun bug id 4258344: Add new XKB API manpages for 6.4 upgrade
15971    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4258344>
15972    
15973    (Volunteer needed to convert prototypes in man pages to ANSI C style...)
15974
15975commit f640a49b5e2ebf29f9d655df544c63bf826f619a
15976Author: Tilman Sauerbeck <tilman@code-monkey.de>
15977Date:   Fri Apr 20 18:39:59 2007 +0200
15978
15979    Markup tweak for XOpenIM.
15980
15981commit e972b0bb255af4f3258217852542faf5afa60b28
15982Author: Julien Cristau <jcristau@debian.org>
15983Date:   Fri Apr 20 18:35:09 2007 +0200
15984
15985    Bug #9695: Fixed a few argument types in the XOpenIM manpage.
15986
15987commit b4e2276f329fa42397cb8609cfcd34ebafd3d96b
15988Author: Julien Cristau <jcristau@debian.org>
15989Date:   Fri Apr 20 18:28:52 2007 +0200
15990
15991    Bug #9697: Fixed documentation of XVisualInfo struct.
15992    
15993    The "depth" member was said to be unsigned int, but it's signed.
15994
15995commit 4068f3dae01c630f825002673b1d3a047ad61863
15996Author: Julien Cristau <jcristau@debian.org>
15997Date:   Fri Apr 20 16:41:21 2007 +0200
15998
15999    Bug #9696: refer to XDefineCursor() instead of XDefineCusor().
16000
16001commit 603c2f88d4e57ce1a3c16e8b6246866e6edd8fa8
16002Author: Tilman Sauerbeck <tilman@code-monkey.de>
16003Date:   Sat Apr 7 14:42:55 2007 +0200
16004
16005    Use unistd.h to get getresuid() and friends.
16006    
16007    This works since we now have _GNU_SOURCE defined.
16008
16009commit 0300f295bbd3a0c7c46baac8e0a27aeaf53c9d9b
16010Author: Tilman Sauerbeck <tilman@code-monkey.de>
16011Date:   Sat Apr 7 14:27:12 2007 +0200
16012
16013    Bug #10562: Define _GNU_SOURCE on glibc systems.
16014
16015commit a225a0be48770beb689d5ac5da97073634f7deab
16016Author: Tilman Sauerbeck <tilman@code-monkey.de>
16017Date:   Fri Apr 6 12:46:25 2007 +0200
16018
16019    For nls/*.pre, allow people to comment lines by starting them with '##'.
16020    
16021    This fixes a bunch of cpp warnings from nls/en_US.UTF-8/Compose.pre.
16022    Fixing that file would have been a larger diff, and using ## may be
16023    nicer to use anyway.
16024
16025commit 1c75a9479011e5f1ad01c950628d0ef5a302d8b6
16026Author: Tilman Sauerbeck <tilman@code-monkey.de>
16027Date:   Fri Apr 6 11:59:31 2007 +0200
16028
16029    Bug #10475: Fixed lots of char*/const char* mixups.
16030    
16031    I didn't fix all of them, as that would require touching
16032    public headers.
16033
16034commit 680dd50193b5b3fcabdd3f1fcbd6a889d5a95c54
16035Author: Matthias Hopf <mhopf@suse.de>
16036Date:   Thu Apr 5 11:53:18 2007 +0200
16037
16038    Fix 64bit issues with reallocation.
16039
16040commit 4d38aeaca42d0bdfe34a833a142ee4d895de03bf
16041Author: Tilman Sauerbeck <tilman@code-monkey.de>
16042Date:   Wed Apr 4 18:41:18 2007 +0200
16043
16044    Fixed a few warnings.
16045
16046commit 7dc7ef398b6ad90ccd1680ed9cd1cfdd47312f5a
16047Author: Matthieu Herrb <matthieu@roadrock.(none)>
16048Date:   Tue Apr 3 15:39:52 2007 +0200
16049
16050    Multiple integer overflows in the XGetPixel() and XInitImage functions
16051    
16052    CVE-2007-1667
16053
16054commit 0284b144340a455a4b5b5011d81ac5a610372291
16055Author: David Baron <dbaron@dbaron.org>
16056Date:   Fri Mar 30 17:07:46 2007 +0200
16057
16058    Bug #7703: Fixed XSetSizeHints() et al wrt use of uninitialized data.
16059    
16060    Now only those fields of the respective hint struct are set that
16061    are actually valid in the input data.
16062    The changed functions are:
16063    XSetSizeHints(), XSetWMHints() and XSetWMSizeHints().
16064
16065commit 0994faa0c76c45b106442db461b8a30a3e1c9395
16066Author: Tilman Sauerbeck <tilman@code-monkey.de>
16067Date:   Thu Mar 29 17:31:25 2007 +0200
16068
16069    Fixed the change from the previous SendEvent commit.
16070    
16071    Testing a different patch than the one you commit is bad, right?
16072
16073commit 398d75528a84f4b8414eb0e363cf53b1b16f6fdf
16074Author: Tilman Sauerbeck <tilman@code-monkey.de>
16075Date:   Wed Mar 28 22:23:44 2007 +0200
16076
16077    Bug #10292: Fixed a memory leak related to XOpenDisplay() in the XCB code.
16078
16079commit ab0bcd07957cecc8e7c0e75d5160a625e91264fe
16080Author: David Baron <dbaron@dbaron.org>
16081Date:   Wed Mar 28 22:21:40 2007 +0200
16082
16083    Bug #7713: Initialize all of the event's fields before sending it.
16084
16085commit bc80f9fe3ccce40ee41246b97470c4f0519756ad
16086Author: Julien Cristau <jcristau@debian.org>
16087Date:   Sun Mar 18 13:14:48 2007 +0100
16088
16089    Bug #9279: Fixed a file descriptor leak.
16090
16091commit c9e28e05ae01ce8a29bea09df759b6271865b44c
16092Author: Daniel Stone <daniel@fooishbar.org>
16093Date:   Sat Dec 16 00:45:19 2006 +0200
16094
16095    include: don't distribute XlibConf.h
16096    
16097    Since XlibConf.h is built by configure, don't distribute it.
16098
16099commit dd1705ced2cac6b4b6b21e79272fcf9bed4bf376
16100Merge: 129bbb9f 769b9854
16101Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
16102Date:   Thu Dec 14 14:23:20 2006 -0600
16103
16104    Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/lib/libX11
16105
16106commit 129bbb9f9114a571556fa3a24f15ba58a5cdb2de
16107Author: Jeremy C. Reed <reed@glacier.reedmedia.net>
16108Date:   Thu Dec 14 14:21:19 2006 -0600
16109
16110    For NetBSD, define the XTHREADLIB and XTHREAD_CFLAGS.
16111
16112commit 769b9854f7eb1d6d20dd0b4a1c1215ad8e1b77b6
16113Author: Daniel Stone <daniel@fooishbar.org>
16114Date:   Wed Dec 6 18:53:00 2006 +0200
16115
16116    Makefile.am: make ChangeLog hook safer
16117    
16118    Make ChangeLog hook as safe as possible.
16119    (cherry picked from f5d6a3d24095c7ffed86705995d0874c885e7676 commit)
16120
16121commit 8a8185a649e93b90ffa820387ffdca831227f5a9
16122Author: Josh Triplett <josh@freedesktop.org>
16123Date:   Thu Nov 30 18:33:13 2006 -0800
16124
16125    Add autogen.sh to EXTRA_DIST.
16126
16127commit efe817f95ef8d05e863c83147e903140bc860de4
16128Author: Jamey Sharp <jamey@minilop.net>
16129Date:   Thu Nov 30 17:58:35 2006 -0800
16130
16131    Release libX11 1.1.1.
16132
16133commit f637a5b03164263a3af2e644cf655e52b015f1bb
16134Author: Ross Combs <rocombs@cs.nmsu.edu>
16135Date:   Sat Nov 25 14:45:17 2006 -0800
16136
16137    Debian bug #354315: Clarify return value in XGetWindowAttributes man page
16138    
16139    This man page does not discuss the actual return values of the
16140    function, but says they are of type "Status".  One might assume
16141    that this means you could compare it with the "Success" macro.
16142    One would be wrong.
16143    
16144    The X functions seem to have two three types representing status.
16145    If it is an "int" there are a number of error codes or "Success"
16146    which can be compared against.  If it is a bool, the result can be
16147    compared with "True" or "False".  If the return type is "Status" it
16148    appears that the return type is either 0 or 1.  Unfortunately the
16149    value for Success is zero, so it is important to distinguish
16150    between the first two types of return values and the third;
16151    otherwise the conditional will be inverted.
16152    
16153    XGetWindowAttributes() is one of the functions which returns zero
16154    for failure.  The man page should make this clear.
16155
16156commit c6a0b0f18ed1242eeb908f5cf767ab8381edd456
16157Author: Jamey Sharp <jamey@minilop.net>
16158Date:   Sat Nov 25 14:23:45 2006 -0800
16159
16160    Bug #9154: Always process an event for _XReadEvents, even if an error occurs
16161    
16162    Previously, process_responses (in the wait_for_first_event case called
16163    from _XReadEvents) considered any return from xcb_wait_for_event
16164    sufficient to think it had processed an event. If xcb_wait_for_event
16165    returned an error, and no more events occurred before process_responses
16166    called xcb_poll_for_event, process_responses would try to return with
16167    dpy->head NULL, and would fail an assertion for the _XReadEvents
16168    postcondition. Now, process_responses continues using xcb_wait_for_event
16169    until it gets an event.
16170
16171commit d56e78acce9b2aa1dd1bf172afedaa3bccd5e1c8
16172Author: Tilman Sauerbeck <tilman@code-monkey.de>
16173Date:   Sat Nov 25 05:29:31 2006 -0800
16174
16175    Bug #9153: Fix access to freed memory.
16176    
16177    The fix for bug #8622 introduced a smaller bug where _XReply would
16178    read memory shortly after freeing it. This patch caches the needed
16179    value in a stack-allocated variable before the heap-allocated memory
16180    is freed.
16181
16182commit 934ca763bbc0dd7ae460469bfc000ba101602bcc
16183Author: Josh Triplett <josh@freedesktop.org>
16184Date:   Fri Nov 24 19:57:58 2006 -0800
16185
16186    libx11 doesn't use inputproto in public headers; don't require it in x11.pc
16187    
16188    Based on a Debian patch.
16189
16190commit 4255997ef2d92740d51f6e63e9eabcfa089683f0
16191Author: Josh Triplett <josh@freedesktop.org>
16192Date:   Thu Nov 23 07:19:32 2006 -0800
16193
16194    Release libX11 1.1
16195
16196commit a1168e11ec9377307c51a7271faec3bf88a63a66
16197Author: Jamey Sharp <jamey@minilop.net>
16198Date:   Tue Nov 21 17:52:34 2006 -0800
16199
16200    Add note in man-page that XListFontsWithInfo is not thread-safe.
16201    
16202    _XReply drops the Display lock, so the value of dpy->request may change
16203    before _XReply is called again.
16204    
16205    I discovered this by inspection a year or two ago. I'm pretty confident
16206    in the claim, and nobody has come up with an argument for why it's safe
16207    despite appearances.
16208
16209commit 67abe024268c6b1fdee516e5d3a046ccffd7e80a
16210Author: Jamey Sharp <jamey@minilop.net>
16211Date:   Sat Nov 18 15:39:26 2006 -0800
16212
16213    Bug #8622: Fix response processing order for threaded apps
16214    
16215    Previously, process_responses (the common code for _XReply,
16216    _XReadEvents, and _XEventsQueued) took the current request sequence
16217    number as an argument, and did some highly complicated processing to
16218    attempt to process responses in order across threads.  This didn't
16219    actually work.
16220    
16221    Now, process_responses handles responses in order, by adding condition
16222    variables to the list of outstanding requests in
16223    dpy->xcb->pending_requests, and blocking on them when those requests
16224    should get processed to allow _XReply to process them; if actually
16225    called from _XReply, it returns when _XReply's request should get
16226    processed.  _XReply broadcasts on the condition variable after it has
16227    read its reply and re-acquired the display lock.
16228    
16229    Another great commit brought to you by Jamey Sharp, Josh Triplett, the
16230    Gobby collaborative text editor, conspy, and ridiculous amounts of SSH
16231    forwarding.
16232    
16233    Signed-off-by: Josh Triplett <josh@freedesktop.org>
16234    Signed-off-by: Jamey Sharp <jamey@minilop.net>
16235
16236commit 941f02ede63baa46f93ed8abccebe76fb29c0789
16237Author: Lars Knoll <lars@trolltech.com>
16238Date:   Wed Nov 8 12:17:41 2006 -0800
16239
16240    Don't hold the display lock around callbacks to the application.
16241    
16242    This fixes an XCB locking assertion failure, particularly with emacs.
16243
16244commit e494ecaac1ec8a22bd9a85f800fca74d02e9d358
16245Author: Diego 'Flameeyes' Pettenò  <flameeyes@gentoo.org>
16246Date:   Tue Nov 7 09:32:00 2006 -0800
16247
16248    Add xcb-xlib dependency to x11.pc when built against XCB.
16249
16250commit 2302008a3793eb4df8ede777d54fe06505c47eaf
16251Author: Eric Anholt <eric@anholt.net>
16252Date:   Mon Nov 6 17:11:42 2006 -0800
16253
16254    XCB: Allocate the right amount of memory for dpy->lock_fns.
16255    
16256    Fixes a crash I was experiencing on startup of anything using gdk.
16257
16258commit 5f860655be88108b03ccd97470a0814819254bf0
16259Author: Jamey Sharp <jamey@minilop.net>
16260Date:   Thu Nov 2 17:55:31 2006 -0800
16261
16262    Release libX11 1.1 RC2 (1.0.99.2).
16263
16264commit a6f4bbf7b1d725b0f04bd660f57b861a76b19831
16265Author: Benno Schulenberg <bensberg@justemail.net>
16266Date:   Sun Oct 29 03:10:30 2006 +0300
16267
16268    nls (en_US): remove long compositions that override shorter (bug #2286)
16269    Remove long compositions that override (or get overriden by) later shorter
16270    compositions, e.g. a four-key compose sequence that gets overriden by a
16271    later three-key compose sequence.
16272
16273commit 0280bf11ef88673a9b5bba3a91a599260f1f0949
16274Author: Benno Schulenberg <bensberg@justemail.net>
16275Date:   Sun Oct 29 03:08:36 2006 +0300
16276
16277    nls: remove shadowed compose entries (bug #2286)
16278    Remove compose entries shadowed by others later on.
16279
16280commit d118f2b1ef10997194b281524177dea7396da7dd
16281Author: Benno Schulenberg <bensberg@justemail.net>
16282Date:   Sun Oct 29 03:07:15 2006 +0300
16283
16284    nls: remove duplicate compose entries (bug #2286)
16285    Remove a bunch of duplicate entries from various Compose files.
16286
16287commit 5e1cc2fe20e5904ca1e05a4cb7be13d450a593bb
16288Author: Caolan McNamara <caolanm@redhat.com>
16289Date:   Sun Oct 29 02:46:15 2006 +0300
16290
16291    XKB geometry: fix leaks in GetKeyboardByName (bug #8099)
16292    Don't leak the name and value of every property we parse, as well as the
16293    name of every colour.
16294
16295commit 686bb8b35acf6cecae80fe89b2b5853f5816ce19
16296Author: Matthias Hopf <mhopf@suse.de>
16297Date:   Wed Oct 18 14:25:04 2006 +0200
16298
16299    Fix double open of compose file.
16300    
16301    Issue found by Kees Cook <kees@canonical.com>.
16302
16303commit d3e65cb8cddf08913d83c9df2bb9b1517f2ad3a8
16304Author: Jamey Sharp <jamey@minilop.net>
16305Date:   Sat Oct 14 21:25:10 2006 -0700
16306
16307    XCB: check for and handle I/O errors in _XGetXCBBuffer.
16308
16309commit 256eba6b40c5f811a03b04abf5f85f728ee3ab5d
16310Author: Jamey Sharp <jamey@minilop.net>
16311Date:   Wed Oct 11 00:06:50 2006 -0700
16312
16313    XKB bugfix: SyncHandle must be called after UnlockDisplay, not before.
16314
16315commit 1eedf1bd033e496843cfde42ae4ae5a119298605
16316Author: Jamey Sharp <jamey@minilop.net>
16317Date:   Tue Oct 10 23:03:28 2006 -0700
16318
16319    Add correct Display locking to XKB functions.
16320    
16321    Some XKB functions didn't correctly call LockDisplay or UnlockDisplay.
16322    This patch fixes at least some instances of that problem.
16323    
16324    Thanks to Magnus Kessler <Magnus.Kessler@gmx.net> for finding these bugs
16325    and proposing a fix, which this patch is based on.
16326
16327commit e17c2cbe9fbaa1600d4b9463ec800a874b0d87cd
16328Author: David Nusinow <dnusinow@debian.org>
16329Date:   Tue Oct 10 22:11:05 2006 -0400
16330
16331    Dynamically generate internal manpage section using __libmanpagesuffix__ so that it actually matches the section if you don't use 3X11
16332
16333commit e53557da969b706dbc843f6fde3db31ffe382e0f
16334Author: Jamey Sharp <jamey@minilop.net>
16335Date:   Sat Oct 7 21:00:36 2006 -0700
16336
16337    Release libX11 1.1 RC1 (1.0.99.1).
16338
16339commit bf237409c5fce32c557d298f62f44d456c2b5bc8
16340Merge: ba477191 2d426d1f
16341Author: Jamey Sharp <jamey@minilop.net>
16342Date:   Sat Oct 7 21:07:16 2006 -0700
16343
16344    As XCB support is about to be released in libX11, stable is now subsumed by master.
16345
16346commit ba477191c67ce93e61423cc1abe35275704cce50
16347Author: Jamey Sharp <jamey@minilop.net>
16348Date:   Sat Oct 7 03:48:13 2006 -0700
16349
16350    XCB: Don't rely on having the definition of struct xcb_setup_t available.
16351
16352commit 117b55cbd0b0ce51362df88363ed83d44a493ac7
16353Author: Jamey Sharp <jamey@minilop.net>
16354Date:   Fri Oct 6 16:27:31 2006 -0700
16355
16356    xcb_poll_for_event no longer takes an 'int *error' out-parameter.
16357
16358commit caaa8e8a55e837b3585c1dee7bef194fc4c79d16
16359Author: Josh Triplett <josh@freedesktop.org>
16360Date:   Fri Oct 6 16:26:11 2006 -0700
16361
16362    Actually ship Xxcbint.h
16363
16364commit 7b027e53b5e393082f4f515c8ba18077eb97163f
16365Author: Josh Triplett <josh@freedesktop.org>
16366Date:   Fri Oct 6 16:25:50 2006 -0700
16367
16368    Clean ChangeLog only in "make maintainer-clean", not "make clean"
16369    
16370    ChangeLog requires a git repo to generate; make clean and make distclean
16371    shouldn't get rid of it.
16372
16373commit cab22e02e78b3e5b8a73d1cd55cf6686426b47e0
16374Author: Josh Triplett <josh@freedesktop.org>
16375Date:   Fri Oct 6 16:24:58 2006 -0700
16376
16377    Add ChangeLog and "make dist"-generated files to .gitignore
16378
16379commit e4c7cfdee4a40e466c0c6b370cabd432e9e855a0
16380Author: Josh Triplett <josh@freedesktop.org>
16381Date:   Fri Oct 6 15:53:27 2006 -0700
16382
16383    Add manual pages for XGetXCBConnection and XSetEventQueueOwner
16384
16385commit 688224cea95e453f94c5a602dc6fce84bc93dfc0
16386Author: Josh Triplett <josh@freedesktop.org>
16387Date:   Fri Oct 6 13:41:16 2006 -0700
16388
16389    Remove unnecessary prototype for _XFreeDisplayStructure in xcb_disp.c
16390
16391commit ab728ca372288d0db1b486c265e34c1376f29104
16392Author: Josh Triplett <josh@freedesktop.org>
16393Date:   Fri Oct 6 12:36:39 2006 -0700
16394
16395    XCL is dead; long live Xlib/XCB
16396    
16397    Rename all instances of "XCL" to Xlib/XCB-derived names.
16398    
16399    The only user-visible change: rename the include file <X11/xcl.h> to
16400    <X11/Xlib-xcb.h>; programs will need to change their #include lines to match.
16401    
16402    Remove the XCL cast inlines from Xlib-xcb.h.
16403
16404commit 5b73093203039d307eb7ab3845c3ced207e9e26c
16405Author: Josh Triplett <josh@freedesktop.org>
16406Date:   Fri Oct 6 11:33:49 2006 -0700
16407
16408    Add XCB developers to AUTHORS
16409
16410commit 1cb71ff139276a0a58c60ea5f261f64b94706b9b
16411Author: Josh Triplett <josh@freedesktop.org>
16412Date:   Fri Oct 6 11:21:28 2006 -0700
16413
16414    Fix email addresses in README
16415
16416commit 12f038669278019594ca0ed53dadcf4e84092422
16417Author: Jamey Sharp <jamey@minilop.net>
16418Date:   Fri Oct 6 02:13:05 2006 -0700
16419
16420    XCB: Handle all responses in order of monotonically increasing sequence number.
16421
16422commit f392680273278b43079302206897f794e60f3c70
16423Author: Josh Triplett <josh@freedesktop.org>
16424Date:   Fri Oct 6 01:11:08 2006 -0700
16425
16426    Actually install x11-xcb.pc, and ship x11-xcb.pc.in
16427
16428commit f1fcad2e3fd17aaf1294f1d8e9f406fd5b32a863
16429Author: Jamey Sharp <jamey@minilop.net>
16430Date:   Thu Oct 5 18:32:29 2006 -0700
16431
16432    XCB: correctly handle failure to connect to X server.
16433
16434commit e754b3b078d556c7861da56aad47d244e9199e06
16435Author: Josh Triplett <josh@freedesktop.org>
16436Date:   Thu Oct 5 17:44:22 2006 -0700
16437
16438    Split public Xlib/XCB functions into libX11-xcb
16439    
16440    We can never change the libX11 soname, and we don't want to commit to never
16441    changing the public Xlib/XCB functions, so split them into a separate library
16442    libX11-xcb.  This also means that a program linked solely against libX11
16443    should work with either Xlib or Xlib/XCB, which will make life easier for
16444    package maintainers.
16445    
16446    Signed-off-by: Josh Triplett <josh@freedesktop.org>
16447    Acked-by: Jamey Sharp <jamey@minilop.net>
16448
16449commit ffd367f708b295abaedf3a23a1bfd4710d171d6f
16450Author: Jamey Sharp <jamey@minilop.net>
16451Date:   Wed Oct 4 17:16:46 2006 -0700
16452
16453    No longer #include Xmd from xcl.h: we do not need it.
16454
16455commit 3aff149d42ba8ed620091971b3766bdf96c62aaf
16456Author: Jamey Sharp <jamey@minilop.net>
16457Date:   Wed Oct 4 16:58:32 2006 -0700
16458
16459    XCB: Revert locking to simple wrapper around libX11's normal locks.
16460    
16461    No more recursive mutexes, no more banging XCB's I/O lock in-place, and
16462    reduces the differences between the previous stable release and an
16463    XCB-enabled one. Sadly, Xlib's pluggable thread functions work again
16464    too, now. I apologize to the world.
16465
16466commit 8ff122fb529bdb1c2b9a86b12d06b6da1b35d708
16467Author: Jamey Sharp <jamey@minilop.net>
16468Date:   Mon Sep 25 04:54:52 2006 -0700
16469
16470    Link explicitly against XCB's Xlib compatibility functions.
16471
16472commit bde3cd123d65a2f36ee0c417f5f231b7e01d0671
16473Author: Jamey Sharp <jamey@minilop.net>
16474Date:   Mon Sep 25 04:13:20 2006 -0700
16475
16476    libxcb now installs header files in <xcb>, not <X11/XCB>.
16477
16478commit 87d00207f5a1f25a45a153618739cd6481814f89
16479Author: Ian Osgood <iano@quirkster.com>
16480Date:   Sun Sep 24 23:39:01 2006 -0700
16481
16482    Track XCB's "Great Renaming".
16483
16484commit 85a5e98dff9b7752fae157fad9b8c9825cb0efab
16485Author: Jamey Sharp <jamey@minilop.net>
16486Date:   Tue Sep 12 23:02:42 2006 -0700
16487
16488    Quit using XCBGetQueuedRequestRead.
16489
16490commit a61936fc4e9bd93b108764bbacd5b8f786e51915
16491Author: Tollef Fog Heen <tfheen@err.no>
16492Date:   Wed Aug 30 00:05:54 2006 +0300
16493
16494    nls: use _XlcUtf8Loader for en_US (bug #7982)
16495    Use _XlcUtf8Loader instead of _XlcUnicodeLoade,r bringing it into line with
16496    every other locale.
16497
16498commit abcc7e1865cdfbd591f6520cfe4257f0b0b1c03e
16499Author: Alan Coopersmith <alan.coopersmith@sun.com>
16500Date:   Wed Aug 23 18:49:30 2006 -0700
16501
16502    When opening display, if LOCALCONN fails, fall back to UNIXCONN, then TCPCONN
16503    
16504    Port to X11R7 of Sun bug fix 4061225 by Alex Chen for X11R6 - when failing to
16505    connect on a named pipe, try a Unix socket first, to better support people who
16506    replace their X servers with ones that don't support named pipe transport.
16507    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4061225>
16508
16509commit 1ddc44c1cada7e926bd4787406444ce7c36b61e1
16510Author: Mark Brown <mark.brown@sun.com>
16511Date:   Thu Jul 27 19:17:10 2006 -0700
16512
16513    Sun bug 1149809: Document event delivery when grab is terminated.
16514
16515commit 171107b03ac89d94f9006c7cda242aeefb9ecd16
16516Author: Dennis Arellano <Dennis.Arellano@Sun.COM>
16517Date:   Thu Jul 27 18:47:06 2006 -0700
16518
16519    Sun bug 4091271: XGetWindowProperty is missing a crucial prop_return description
16520    
16521    Document that 32-bit format properties are always returned in arrays of type
16522    long, even on systems where long is 64-bits.
16523    <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4091271>
16524
16525commit 8309efe6550877cd0bf22979904b3f6bd3e6cffa
16526Author: Alan Coopersmith <alan.coopersmith@sun.com>
16527Date:   Mon Jul 24 15:52:00 2006 -0700
16528
16529    Add support for "make lint" to check code with lint/sparse/etc.
16530
16531commit 931e02fbd1acd09aae2b0954c34342c86c72dff6
16532Author: Alan Coopersmith <alan.coopersmith@sun.com>
16533Date:   Mon Jul 24 15:50:52 2006 -0700
16534
16535    ANSIfy some static function definitions
16536
16537commit 20b7abcaac324d90454de63f32f4a2b398e69e63
16538Author: Alan Coopersmith <alan.coopersmith@sun.com>
16539Date:   Mon Jul 24 15:01:40 2006 -0700
16540
16541    Fix sparse warning: Using plain integer as NULL pointer
16542
16543commit d158ab29930513c4097f5b67e7bea08ed2bfd62c
16544Author: Alan Coopersmith <alan.coopersmith@sun.com>
16545Date:   Mon Jul 24 14:00:24 2006 -0700
16546
16547    Remove unused variable
16548
16549commit 30377000375bdb958042dcb1f38503c94ef21eaf
16550Author: Eric Anholt <anholt@FreeBSD.org>
16551Date:   Fri Jul 21 18:55:36 2006 -0400
16552
16553    Bug #7188: Fix the documentation of XUrgencyHint (not UrgencyHint).
16554
16555commit 4eba45879aea4e415ab550ee56b900d060099110
16556Author: Matthieu Herrb <matthieu.herrb@laas.fr>
16557Date:   Sun Jul 16 10:55:39 2006 +0200
16558
16559    set GIT_DIR=${srcdir}/.git for git-log
16560
16561commit 2d426d1f2608fedb77bd7d010dabece76b8d4a60
16562Author: Aaron Plattner <aplattner@nvidia.com>
16563Date:   Tue Jul 11 13:27:49 2006 -0700
16564
16565    Add a .PHONY to ensure the ChangeLog isn't stale.
16566    
16567    Setting the ChangeLog rule as phony forces it to be re-run even when the
16568    ChangeLog file already exists. Research indicates .PHONY is portable to BSD and
16569    Solaris make.
16570    (cherry picked from b8a98809ed81e1226775e6447ef219ffc01334b5 commit)
16571
16572commit b8a98809ed81e1226775e6447ef219ffc01334b5
16573Author: Aaron Plattner <aplattner@nvidia.com>
16574Date:   Tue Jul 11 13:27:49 2006 -0700
16575
16576    Add a .PHONY to ensure the ChangeLog isn't stale.
16577    
16578    Setting the ChangeLog rule as phony forces it to be re-run even when the
16579    ChangeLog file already exists. Research indicates .PHONY is portable to BSD and
16580    Solaris make.
16581
16582commit 8f2be66089b88e4ed6acb0211ea107f4bb910bd3
16583Author: Mayank Jain <majain@redhat.com>
16584Date:   Tue Jul 11 19:42:47 2006 +0100
16585
16586    add Indian language locales
16587    Add as, kn, ml, or, ur, and te locales.
16588
16589commit a92eb6785699bbc9c6c3813f6af3bb9431a3f6b0
16590Author: Matthias Hopf <mhopf@suse.de>
16591Date:   Tue Jul 4 12:16:30 2006 +0200
16592
16593    Fix for autoconf 2.60 issue.
16594    
16595    Updated AC_DEFINE_DIR.
16596    Reverted datarootdir change.
16597
16598commit 644f4828b15bce42b597eb123ba0bbc372c46c03
16599Author: Keith Packard <keithp@neko.keithp.com>
16600Date:   Sat Jul 1 21:31:23 2006 -0700
16601
16602    Work around recent autoconf (2.59?) changes in directory expansion.
16603    
16604    Recent autoconf versions have changed how directory names are managed in the
16605    configure.ac script; automatic 'eval' invocations now occur as a part of the
16606    AC_DEFINE_DIR macro which make it imperative that AC_DEFINE_DIR be executed
16607    before the variables are used in further macro definitions. Also, ${datadir}
16608    is apparantly an old name for ${datarootdir} as ${datadir} doesn't get
16609    expanded correctly by AC_DEFINE_DIR. This looks like an autoconf bug, but it
16610    is easy to work around by just using ${datarootdir} instead of ${datadir}.
16611
16612commit be70a31229aa106aff0a09d78c00812682cd3475
16613Author: Keith Packard <keithp@neko.keithp.com>
16614Date:   Sat Jul 1 01:56:05 2006 -0700
16615
16616    Xlib/XCB: handle 32-bit sequence wrap.
16617    
16618    Replace broken sequence compares with XCB_SEQUENCE_COMPARE (copied from
16619    XCB).
16620    Account for XCB sequence 0 handling.
16621
16622commit 99c711707ad08e1396e123b1c7df687c560a489a
16623Author: Donnie Berkholz <spyderous@gentoo.org>
16624Date:   Thu Jun 29 19:43:20 2006 -0700
16625
16626    Bump version to 1.0.3.
16627
16628commit cde3c0dd72af2b490e80cffca962e3487dd31be4
16629Author: Donnie Berkholz <spyderous@gentoo.org>
16630Date:   Thu Jun 29 19:39:36 2006 -0700
16631
16632    Bug #7349: Missed one of the setuid fixes.
16633    (cherry picked from e9614c963b532f46a7932c2305a4b177a996a222 commit)
16634
16635commit df3fef8983d96c59d481c4cdaf1f271d54a116d2
16636Author: Matthias Hopf <mhopf@suse.de>
16637Date:   Thu Jun 29 18:59:57 2006 +0200
16638
16639    Update to final Compose cache directory location.
16640    (cherry picked from abda4d223e9cce9ac6e7b5d82a5680d9a502e52a commit)
16641
16642commit 912ef198292d3053daa810f842510e5d62ded0f0
16643Author: Matthias Hopf <mhopf@suse.de>
16644Date:   Thu Jun 29 17:41:41 2006 +0200
16645
16646    Fix alignment of trees and wide chars in the cache.
16647    (cherry picked from 40a64c61f8bc33d497e1224e02c41dea2d424d97 commit)
16648
16649commit 2ece832118b3ee5d8ed19f1ee9b1c822b70ec6e9
16650Author: Matthias Hopf <mhopf@suse.de>
16651Date:   Wed Jun 28 19:17:03 2006 +0200
16652
16653    First (dummy) entry of compose caches was not initialized and thus contained varying garbage.
16654    (cherry picked from f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3 commit)
16655
16656commit e9614c963b532f46a7932c2305a4b177a996a222
16657Author: Donnie Berkholz <spyderous@gentoo.org>
16658Date:   Thu Jun 29 19:39:36 2006 -0700
16659
16660    Bug #7349: Missed one of the setuid fixes.
16661
16662commit abda4d223e9cce9ac6e7b5d82a5680d9a502e52a
16663Author: Matthias Hopf <mhopf@suse.de>
16664Date:   Thu Jun 29 18:59:57 2006 +0200
16665
16666    Update to final Compose cache directory location.
16667
16668commit 40a64c61f8bc33d497e1224e02c41dea2d424d97
16669Author: Matthias Hopf <mhopf@suse.de>
16670Date:   Thu Jun 29 17:41:41 2006 +0200
16671
16672    Fix alignment of trees and wide chars in the cache.
16673
16674commit f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3
16675Author: Matthias Hopf <mhopf@suse.de>
16676Date:   Wed Jun 28 19:17:03 2006 +0200
16677
16678    First (dummy) entry of compose caches was not initialized and thus contained varying garbage.
16679
16680commit bdbe464d774e01d317f67c63ebbda2fd0edbbdd1
16681Author: Donnie Berkholz <spyderous@gentoo.org>
16682Date:   Thu Jun 22 23:59:03 2006 -0700
16683
16684    Delete ChangeLog, and add a rule to autogenerate it for distribution. Also
16685    add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a
16686    ChangeLog.
16687    (cherry picked from b0edfb8df16ab8c9959b83a4c966d55a59c4e295 commit)
16688
16689commit b0edfb8df16ab8c9959b83a4c966d55a59c4e295
16690Author: Donnie Berkholz <donnie@comet.(none)>
16691Date:   Thu Jun 22 23:47:38 2006 -0700
16692
16693    Delete ChangeLog, and add a rule to autogenerate it for distribution. Also
16694    add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a
16695    ChangeLog.
16696
16697commit dd54981aa76e9dfdc4c3302d6105b4b229447c84
16698Author: Donnie Berkholz <donnie@comet.(none)>
16699Date:   Thu Jun 22 15:47:38 2006 -0700
16700
16701    Bump version to 1.0.2.
16702
16703commit c93539d974a67f596a5eb5b65042d26602546c72
16704Author: Matthieu Herrb <matthieu@deville.herrb.com>
16705Date:   Tue Jun 20 21:04:03 2006 +0200
16706
16707    Check setuid() return value.
16708    (cherry picked from 5169d0e08ff6acb350a6ea768623f5ff0b85b05f commit)
16709
16710commit 605533f814ab7892991578706a6458f61a89ca4d
16711Author: Derek Wang <derek.wang@sun.com>
16712Date:   Mon Jun 19 11:05:37 2006 -0700
16713
16714    Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded
16715    (cherry picked from eff50c94a07194297e705da53d9fbb3a40fb9ad4 commit)
16716
16717commit 5bbd0822c5a926de0ed293437fb9f2b75cf3c4f4
16718Author: Scott Revelt <scott.revelt@sun.com>
16719Date:   Fri Jun 16 19:11:13 2006 -0700
16720
16721    Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators
16722    based on locale that doesn't match those used in the Xcms.txt
16723    (cherry picked from 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa commit)
16724
16725commit 0b05cd4da6134df527fb010384a9fd569bd5d6a3
16726Author: Alan Coopersmith <alanc@alf.(none)>
16727Date:   Fri Jun 16 18:53:33 2006 -0700
16728
16729    Add *~ to ignore emacs droppings
16730    (cherry picked from c33d7b8282ac196b36229be26442296768c16f3e commit)
16731
16732commit 214658b76b56768f69c3959a11525aae7813f448
16733Author: Matthias Hopf <mhopf@suse.de>
16734Date:   Fri Jun 16 15:36:40 2006 +0200
16735
16736    Bug #3104: Compose table cache for faster X11 application starts.
16737    (cherry picked from 1f4c9893ade08bad30c9bd12a36bee57d30b001e commit)
16738
16739commit 13968a23aaea838ba4b69e42e8900f803499e091
16740Author: Matthias Hopf <mhopf@suse.de>
16741Date:   Tue Jun 13 20:23:46 2006 +0200
16742
16743    Bug #3104: Compose table cache for faster X11 application starts.
16744    (cherry picked from 4fe22647e6010a2886c2f3a7093adeaeb6678faa commit)
16745
16746commit e7f8bca08f5476d0ca262097639ac7d424bb4d10
16747Author: Lubos Lunak <llunak@suse.de>
16748Date:   Mon Jun 12 18:48:08 2006 +0200
16749
16750    Bug #3104: Compose table cache for faster X11 application starts.
16751    (cherry picked from 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be commit)
16752
16753commit f506aaf8ac9aae1ee8daaef6cde34ee85aecd641
16754Author: Matthias Hopf <mhopf@suse.de>
16755Date:   Fri Jun 9 18:24:02 2006 +0200
16756
16757    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.
16758    (cherry picked from 9354351fcb8baeaab85250d14409cfb4fa50f3e9 commit)
16759
16760commit 0e6d5e979aacb0c295ce79369ecc5f22ffa7922c
16761Author: Jamey Sharp <jamey@minilop.net>
16762Date:   Wed Jun 7 20:29:05 2006 -0700
16763
16764    Fix bug #7035: unnecessary memmove in XOpenDisplay.
16765    Using memmove on the connection setup data causes a problem for XCB, but making
16766    Xlib stop doing that should be harmless for non-XCB as well.
16767    (cherry picked from b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0 commit)
16768
16769commit cd7328c46ae72903ed02832828891b2dab4d5ee0
16770Author: Daniel Stone <daniels@preemptive.fooishbar.org>
16771Date:   Sat Jun 3 13:51:51 2006 +0300
16772
16773    Fix threading support on GNU/kFreeBSD systems.  (Robert Millan)
16774    (cherry picked from e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb commit)
16775
16776commit c336eb6b80a6f91da1d0b3d28634a2cfde324670
16777Merge: 33556ca8 ad9ebbd2
16778Author: Donnie Berkholz <donnie@comet.(none)>
16779Date:   Thu Jun 22 14:25:35 2006 -0700
16780
16781    Merge branch 'stable' of http://people.freedesktop.org/~jamey/libX11 into stable
16782
16783commit 213dacad21740466e7ab31b01d3fc513fe4b3e74
16784Author: Daniel Stone <daniel@fooishbar.org>
16785Date:   Thu Jun 22 17:20:59 2006 +0300
16786
16787    Bump to 1.0.99.0 to avoid confusion.
16788
16789commit efedfd68e31bcee2d21ac340be8dc9e1825ec890
16790Merge: e3acee88 4b8eb5d4
16791Author: Daniel Stone <daniel@fooishbar.org>
16792Date:   Thu Jun 22 16:53:45 2006 +0300
16793
16794    Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11
16795
16796commit 4b8eb5d4a1da73a94b5a6ab12e34784aae4c79c5
16797Merge: 5169d0e0 eff50c94
16798Author: Matthieu Herrb <matthieu@deville.herrb.com>
16799Date:   Tue Jun 20 21:05:15 2006 +0200
16800
16801    Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11
16802
16803commit 5169d0e08ff6acb350a6ea768623f5ff0b85b05f
16804Author: Matthieu Herrb <matthieu@deville.herrb.com>
16805Date:   Tue Jun 20 21:04:03 2006 +0200
16806
16807    Check setuid() return value.
16808
16809commit eff50c94a07194297e705da53d9fbb3a40fb9ad4
16810Author: Derek Wang <derek.wang@sun.com>
16811Date:   Mon Jun 19 11:05:37 2006 -0700
16812
16813    Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded
16814
16815commit 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa
16816Author: Scott Revelt <scott.revelt@sun.com>
16817Date:   Fri Jun 16 19:11:13 2006 -0700
16818
16819    Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators
16820    based on locale that doesn't match those used in the Xcms.txt
16821
16822commit c33d7b8282ac196b36229be26442296768c16f3e
16823Author: Alan Coopersmith <alanc@alf.(none)>
16824Date:   Fri Jun 16 18:53:33 2006 -0700
16825
16826    Add *~ to ignore emacs droppings
16827
16828commit 1f4c9893ade08bad30c9bd12a36bee57d30b001e
16829Author: Matthias Hopf <mhopf@suse.de>
16830Date:   Fri Jun 16 15:36:40 2006 +0200
16831
16832    Bug #3104: Compose table cache for faster X11 application starts.
16833
16834commit 4fe22647e6010a2886c2f3a7093adeaeb6678faa
16835Author: Matthias Hopf <mhopf@suse.de>
16836Date:   Tue Jun 13 20:23:46 2006 +0200
16837
16838    Bug #3104: Compose table cache for faster X11 application starts.
16839
16840commit 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be
16841Author: Lubos Lunak <llunak@suse.de>
16842Date:   Mon Jun 12 18:48:08 2006 +0200
16843
16844    Bug #3104: Compose table cache for faster X11 application starts.
16845
16846commit 9354351fcb8baeaab85250d14409cfb4fa50f3e9
16847Author: Matthias Hopf <mhopf@suse.de>
16848Date:   Fri Jun 9 18:24:02 2006 +0200
16849
16850    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.
16851
16852commit b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0
16853Author: Jamey Sharp <jamey@minilop.net>
16854Date:   Wed Jun 7 20:29:05 2006 -0700
16855
16856    Fix bug #7035: unnecessary memmove in XOpenDisplay.
16857    Using memmove on the connection setup data causes a problem for XCB, but making
16858    Xlib stop doing that should be harmless for non-XCB as well.
16859
16860commit ad9ebbd2424bc2699944ffdf4e19e13f9dd8ab84
16861Author: Daniel Stone <daniels@preemptive.fooishbar.org>
16862Date:   Sat Jun 3 12:57:55 2006 +0300
16863
16864    Bug #2186: Add cs_CZ.iso8859-2 alias.
16865
16866commit 9e7765e0b1cbaae6643072d91066ba1201b36227
16867Author: Daniel Stone <daniel@fooishbar.org>
16868Date:   Fri Jun 2 02:46:29 2006 +0300
16869
16870    nls: Serbian (sr_CS) update (#5575)
16871    
16872    Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a
16873    corresponding locale change.  Update compose.dir.pre, locale.alias.pre,
16874    and locale.dir.pre.  (Milos Komarcevic)
16875
16876commit a4ac2242b588da23044a20aa999ae84d4de7b2d8
16877Author: Daniel Stone <daniel@fooishbar.org>
16878Date:   Fri Jun 2 02:24:25 2006 +0300
16879
16880    im: add Braille input method (#6296)
16881    
16882    Bug #6296: Add a Braille input method.  (Samuel Thibault)
16883
16884commit 90de1e2e141ec591048a76cb695579ef809a28d3
16885Author: Daniel Stone <daniel@fooishbar.org>
16886Date:   Fri Jun 2 02:22:17 2006 +0300
16887
16888    xkb support: small typo
16889
16890commit 4c3e34bece7402f08139d34d1ef5834e3cf533c7
16891Author: Daniel Stone <daniel@fooishbar.org>
16892Date:   Fri Jun 2 01:50:24 2006 +0300
16893
16894    en_US.UTF-8 Compose.pre: updates from Simos (#5129)
16895    
16896    Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms,
16897    adding Unicode character names, removing duplicate entries, et al.
16898
16899commit 0c6473dd329c7334ae511884bcb6e73e632c784f
16900Author: Daniel Stone <daniel@fooishbar.org>
16901Date:   Fri Jun 2 01:44:53 2006 +0300
16902
16903    nls: fix use of non-keysym dead_space (#5107)
16904    
16905    Bug #5107: Change users of dead_space to space.
16906
16907commit 6f99f6349de5120f1cb7e02fbc97849341bc48e8
16908Author: Daniel Stone <daniel@fooishbar.org>
16909Date:   Fri Jun 2 01:41:18 2006 +0300
16910
16911    optional XKB support fix
16912    
16913    Fix compilation with --disable-xkb.
16914
16915commit 217d43ed44ced901122093af3ef1294e1736bb77
16916Author: Daniel Stone <daniel@fooishbar.org>
16917Date:   Fri Jun 2 01:39:12 2006 +0300
16918
16919    i18n: separate data and lib directories
16920    
16921    Break out locale data into separate data and library directories, under
16922    $(datadir) and $(libdir), respectively, by default.
16923
16924commit 92fa7fcde8df22830fca7c0275ab201033f7909c
16925Author: Adam Jackson <ajax@benzedrine.nwnk.net>
16926Date:   Thu May 11 14:04:48 2006 -0400
16927
16928    libXcursor.so.1, not libXcursor.so
16929
16930commit 135b4df13ed7c35dbae8975f302fc1fb8412d7c0
16931Author: Daniel Stone <daniels@preemptive.fooishbar.org>
16932Date:   Wed May 10 18:06:03 2006 +0300
16933
16934    XKBMisc.c: use Xfree, not xfree
16935    
16936    Use Xfree() instead of xfree() when freeing interps.
16937
16938commit 3518d772b08e3433bc28b4d8d293fa53ca25f0ee
16939Author: Daniel Stone <daniels@preemptive.fooishbar.org>
16940Date:   Wed May 10 14:51:37 2006 +0300
16941
16942    locale.alias.pre: bg_BG typo fix
16943    
16944    Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre.
16945
16946commit cc533db60cb64dc163c66451933a9bf77c519062
16947Author: Daniel Stone <daniels@endtroducing.localdomain>
16948Date:   Sun Apr 9 22:22:03 2006 +0300
16949
16950    Coverity #203, #204: Fix potential NULL dereferences.
16951
16952commit b83adf7dfd6157694fe4f232012fef36cd9666da
16953Author: Daniel Stone <daniels@endtroducing.localdomain>
16954Date:   Sun Apr 9 22:20:25 2006 +0300
16955
16956    Coverity #205: Fix potential NULL dereference.
16957
16958commit 2d0cd10ad907864d0136739eaac459779c9a5332
16959Author: Daniel Stone <daniels@endtroducing.localdomain>
16960Date:   Sun Apr 9 22:18:20 2006 +0300
16961
16962    Coverity #209: Fix potential NULL dereference. (Alan Coopersmith)
16963
16964commit dc2f3966068d66a564aa452cab9f0c26657fa1df
16965Author: Daniel Stone <daniels@preemptive.research.nokia.com>
16966Date:   Fri Apr 7 18:11:52 2006 +0300
16967
16968    Coverity #826: Fix potential memory leak.
16969
16970commit 23df609ec451a01c77e8f31ecc85c5af7c62efed
16971Author: Daniel Stone <daniels@preemptive.research.nokia.com>
16972Date:   Fri Apr 7 17:49:41 2006 +0300
16973
16974    Bug #1625: Include keysym.h from Xutil.h.
16975
16976commit 5262a1945c543a3419ed626e1deb09ef5b4584c1
16977Author: Eric Anholt <anholt@leguin.anholt.net>
16978Date:   Wed Apr 5 17:12:15 2006 -0700
16979
16980    Check if visualList == NULL, not nVisualsMatched == 0. NULL happens in more
16981    
16982    cases (allocation failure) than nVisualsMatched == 0.  Noticed from inspection
16983    of Coverity #599, #600.
16984
16985commit 8b42635f577468bb143ca593cdd9fb3450ad712c
16986Author: Eric Anholt <anholt@leguin.anholt.net>
16987Date:   Wed Apr 5 16:42:26 2006 -0700
16988
16989    Coverity #558: Free newly-allocated Database in error path.
16990
16991commit 6d06e41d1f431b3f1a1fcf69161e0af411325e9f
16992Author: Eric Anholt <anholt@leguin.anholt.net>
16993Date:   Wed Apr 5 16:38:52 2006 -0700
16994
16995    Coverity #582: Free newly-allocated region in error path.
16996
16997commit 5fd8f79ad3e38df74d9a6cb573617542c101df1a
16998Author: Daniel Stone <daniels@preemptive.research.nokia.com>
16999Date:   Fri Mar 17 15:58:39 2006 +0200
17000
17001    Properly clip bounds when only one point is defining an outline.
17002
17003commit 1e1572eb7f8394ce152e99d96f711ccf9083baf9
17004Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17005Date:   Sat Mar 11 15:39:15 2006 +0100
17006
17007    Fix prototype of XConfigureWindow(). Bugzilla #6023.
17008
17009commit 1da8bd904f1fc79d63d368473531b438d08bfe37
17010Author: Jeremy C. Reed <reed@reedmedia.net>
17011Date:   Tue Feb 21 14:10:22 2006 -0800
17012
17013    Set XTHREADLIB correctly for dragonfly platforms.
17014
17015commit c9768133e3f65ac4bb64e3941e2d6ae36897ec9c
17016Author: Jamey Sharp <jamey@minilop.net>
17017Date:   Sun Feb 19 12:29:27 2006 -0800
17018
17019    Update .gitignores for *.o and nls/locale.dir*.
17020
17021commit efcbde6ba0b770bb0d4c7818e307712161011f10
17022Author: Jamey Sharp <jamey@minilop.net>
17023Date:   Sun Feb 19 12:28:41 2006 -0800
17024
17025    Move .cvsignore to .gitignore.
17026
17027commit e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb
17028Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17029Date:   Sat Jun 3 13:51:51 2006 +0300
17030
17031    Fix threading support on GNU/kFreeBSD systems.  (Robert Millan)
17032
17033commit f1bd3152359ddfadd0d043006036c239f3e2907b
17034Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17035Date:   Sat Jun 3 12:57:55 2006 +0300
17036
17037    Bug #2186: Add cs_CZ.iso8859-2 alias.
17038
17039commit 2b1b79d90db1d7f0472718b8c010c36275994195
17040Author: Daniel Stone <daniel@fooishbar.org>
17041Date:   Fri Jun 2 02:46:29 2006 +0300
17042
17043    nls: Serbian (sr_CS) update (#5575)
17044    Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a
17045    corresponding locale change.  Update compose.dir.pre, locale.alias.pre,
17046    and locale.dir.pre.  (Milos Komarcevic)
17047
17048commit d6fba1f44d404362d3be1b421f57d7ccc3c8cdac
17049Author: Daniel Stone <daniel@fooishbar.org>
17050Date:   Fri Jun 2 02:24:25 2006 +0300
17051
17052    im: add Braille input method (#6296)
17053    Bug #6296: Add a Braille input method.  (Samuel Thibault)
17054
17055commit 0fed7d3185addd610e917dcdaa0676f0256c0ec5
17056Author: Daniel Stone <daniel@fooishbar.org>
17057Date:   Fri Jun 2 02:22:17 2006 +0300
17058
17059    xkb support: small typo
17060
17061commit cf7d9f9e46f3ce01ac04a95978918d5c0f3f3cf9
17062Author: Daniel Stone <daniel@fooishbar.org>
17063Date:   Fri Jun 2 01:50:24 2006 +0300
17064
17065    en_US.UTF-8 Compose.pre: updates from Simos (#5129)
17066    Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms,
17067    adding Unicode character names, removing duplicate entries, et al.
17068
17069commit 332d45fce9fdbf59168d90a133af1f580a589e54
17070Author: Daniel Stone <daniel@fooishbar.org>
17071Date:   Fri Jun 2 01:44:53 2006 +0300
17072
17073    nls: fix use of non-keysym dead_space (#5107)
17074    Bug #5107: Change users of dead_space to space.
17075
17076commit 34f59ce3d1e2eb2971b732d11871c6fff8a1c75b
17077Author: Daniel Stone <daniel@fooishbar.org>
17078Date:   Fri Jun 2 01:41:18 2006 +0300
17079
17080    optional XKB support fix
17081    Fix compilation with --disable-xkb.
17082
17083commit c5940a0b85edec4003f91a59fc3c44f538accfe1
17084Author: Daniel Stone <daniel@fooishbar.org>
17085Date:   Fri Jun 2 01:39:12 2006 +0300
17086
17087    i18n: separate data and lib directories
17088    Break out locale data into separate data and library directories, under
17089    $(datadir) and $(libdir), respectively, by default.
17090
17091commit 33556ca81db2419b9d2a37664c4cea2069414b37
17092Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17093Date:   Fri May 12 14:49:17 2006 -0400
17094
17095    Bump to 1.0.1
17096
17097commit 80d88557626fae9debc404de33d7fb5d69a6049d
17098Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17099Date:   Thu May 11 14:06:28 2006 -0400
17100
17101    libXcursor.so.1, not libXcursor.so
17102
17103commit 5384f27dfe3c94c462c137ab3540eaa5646ca4c6
17104Author: Adam Jackson <ajax@benzedrine.nwnk.net>
17105Date:   Thu May 11 14:04:48 2006 -0400
17106
17107    libXcursor.so.1, not libXcursor.so
17108
17109commit 01f4d433eed6b70c6e9636157acac022054fdeb6
17110Author: Jamey Sharp <jamey@minilop.net>
17111Date:   Wed May 10 17:02:52 2006 -0700
17112
17113    Count any partial request towards the current Xlib sequence number.
17114
17115commit 770cfbd1fcc80a83a9be0c4f68727b8af2c8f4a4
17116Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17117Date:   Wed May 10 18:06:03 2006 +0300
17118
17119    XKBMisc.c: use Xfree, not xfree
17120    Use Xfree() instead of xfree() when freeing interps.
17121
17122commit 22a5255b80b80772612279bc840a953edd0e3442
17123Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17124Date:   Wed May 10 14:51:37 2006 +0300
17125
17126    locale.alias.pre: bg_BG typo fix
17127    Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre.
17128
17129commit 9cac8c9824874ca7d835f001a4efa910b7fdd822
17130Merge: 19b8840a e514bc87
17131Author: Daniel Stone <daniels@preemptive.fooishbar.org>
17132Date:   Wed May 10 14:50:37 2006 +0300
17133
17134    Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11
17135
17136commit e514bc875f27f4bf197b06b8315eeca526195915
17137Author: Jamey Sharp <jamey@minilop.net>
17138Date:   Tue May 9 12:41:59 2006 -0700
17139
17140    Assert that dpy->request does not go backwards. Catches #5839 earlier.
17141
17142commit fc1159137365a599bf611ee001f439416952c4e0
17143Author: Jamey Sharp <jamey@minilop.net>
17144Date:   Sun May 7 17:40:01 2006 -0700
17145
17146    In _XPutXCBBuffer, set aside any trailing partial request until the last byte is available.
17147
17148commit c394480a4247213239822808e3f6e7c6cd6decd9
17149Author: Jamey Sharp <jamey@minilop.net>
17150Date:   Sun May 7 16:58:13 2006 -0700
17151
17152    Update for XCB ConnSetupSuccessRep name change.
17153
17154commit 7672bf93bc1200905461aeb0a2dc2c2696410b93
17155Author: Daniel Stone <daniels@endtroducing.localdomain>
17156Date:   Sun Apr 9 22:22:03 2006 +0300
17157
17158    Coverity #203, #204: Fix potential NULL dereferences.
17159
17160commit cfcafbe48d22d9a0cd50eb9454ce0ff88f0129d3
17161Author: Daniel Stone <daniels@endtroducing.localdomain>
17162Date:   Sun Apr 9 22:20:25 2006 +0300
17163
17164    Coverity #205: Fix potential NULL dereference.
17165
17166commit b6771501feea98d037032f82117b6d4f15779f07
17167Author: Daniel Stone <daniels@endtroducing.localdomain>
17168Date:   Sun Apr 9 22:18:20 2006 +0300
17169
17170    Coverity #209: Fix potential NULL dereference.  (Alan Coopersmith)
17171
17172commit 19b8840af241087bb17b1edabcaa9b28fdd0a1dc
17173Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17174Date:   Fri Apr 7 18:11:52 2006 +0300
17175
17176    Coverity #826: Fix potential memory leak.
17177
17178commit 3a16f262abe48b44ed641525e894bc22e13bf72a
17179Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17180Date:   Fri Apr 7 17:49:41 2006 +0300
17181
17182    Bug #1625: Include keysym.h from Xutil.h.
17183
17184commit 2481b767ae96e2f2503c0390545932c8397b090f
17185Author: Eric Anholt <anholt@leguin.anholt.net>
17186Date:   Wed Apr 5 17:12:15 2006 -0700
17187
17188    Check if visualList == NULL, not nVisualsMatched == 0.  NULL happens in more
17189    cases (allocation failure) than nVisualsMatched == 0.  Noticed from inspection
17190    of Coverity #599, #600.
17191
17192commit 152b17e47d878c2d928eb74581aa69d925a29123
17193Author: Eric Anholt <anholt@leguin.anholt.net>
17194Date:   Wed Apr 5 16:42:26 2006 -0700
17195
17196    Coverity #558: Free newly-allocated Database in error path.
17197
17198commit bc62b99ef36edb34035911c42104be7f6f9d2333
17199Author: Eric Anholt <anholt@leguin.anholt.net>
17200Date:   Wed Apr 5 16:38:52 2006 -0700
17201
17202    Coverity #582: Free newly-allocated region in error path.
17203
17204commit d47f0b3cec1388f7ce60ab2af91df0dea0f221c5
17205Author: Jamey Sharp <jamey@minilop.net>
17206Date:   Fri Mar 31 22:53:07 2006 -0800
17207
17208    Fix buggy interaction with XCB when running out of XIDs.
17209
17210commit e3f452571824d6a875bbf582946de185de9e01e9
17211Author: Jamey Sharp <jamey@minilop.net>
17212Date:   Fri Mar 31 22:52:14 2006 -0800
17213
17214    Add explicit include of Xmd.h to work around bug including both xcb.h and Xmd.h simultaneously.
17215
17216commit 2363b74ca795c1b3a73c9e572532ba5191adec5b
17217Author: Daniel Stone <daniels@preemptive.research.nokia.com>
17218Date:   Fri Mar 17 15:58:39 2006 +0200
17219
17220    Properly clip bounds when only one point is defining an outline.
17221
17222commit e876efb8aa410f2f5c87aaaa7042f847c4ff96f7
17223Merge: be266b20 f71ea0bc
17224Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17225Date:   Sat Mar 11 15:40:35 2006 +0100
17226
17227    Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11
17228
17229commit be266b201dc13530a302a7572283ccd3f32aad87
17230Author: Matthieu Herrb <matthieu@bluenote.herrb.com>
17231Date:   Sat Mar 11 15:39:15 2006 +0100
17232
17233    Fix prototype of XConfigureWindow(). Bugzilla #6023.
17234
17235commit f71ea0bc737c5a42e9e022b86e7ec3b4f846d31c
17236Author: Jamey Sharp <jamey@minilop.net>
17237Date:   Fri Mar 3 11:08:41 2006 -0800
17238
17239    Update for XCBSendRequest sequence number API changes.
17240
17241commit a11d1b0ae674320cf9897f6a83ec08c65eca8d9b
17242Author: Jamey Sharp <jamey@minilop.net>
17243Date:   Fri Mar 3 01:42:49 2006 -0800
17244
17245    Use the full_sequence from XCBGenericError/Event for setting last_request_read, and quit replacing _XSetLastRequestRead with an XCB-specific version.
17246
17247commit d8ba4ae7045b227f8b675628b9094dded02f1c08
17248Author: Jamey Sharp <jamey@minilop.net>
17249Date:   Thu Mar 2 23:43:26 2006 -0800
17250
17251    Bugfix: Rely on XCBSendRequest to leave iov in a well-defined state, and place the spare iovecs at the beginning of the array.
17252
17253commit f9afb5a54435c30961306080e9358d4240ecb844
17254Author: Jamey Sharp <jamey@minilop.net>
17255Date:   Thu Mar 2 23:34:19 2006 -0800
17256
17257    assert() that we will not infinite loop or read uninitialized memory.
17258
17259commit d3512ef3aae5b036a8ce6579318108f1ec20ee22
17260Author: Jamey Sharp <jamey@minilop.net>
17261Date:   Thu Mar 2 15:58:52 2006 -0800
17262
17263    Quit relying on XCBSendRequest to pad to 4-byte boundaries and do it ourselves.
17264
17265commit fb590c15a740264ee867d15a2547072e43b21eed
17266Author: Jamey Sharp <jamey@minilop.net>
17267Date:   Thu Mar 2 12:06:04 2006 -0800
17268
17269    Handle errors correctly when Xlib owns the event queue and XCB has the checked error feature.
17270
17271commit 8356ba37d307a9eda895a6bf41ef727bbfc9a695
17272Author: Jamey Sharp <jamey@minilop.net>
17273Date:   Mon Feb 27 11:51:47 2006 -0800
17274
17275    Use the new XCBSendRequest flag, XCB_REQUEST_RAW, to hand a bag-o-bytes down uninterpreted.
17276
17277commit 07bdf1fbbf2418f866df1a2140d514dd3f035139
17278Author: Jamey Sharp <jamey@minilop.net>
17279Date:   Sun Feb 26 15:46:01 2006 -0800
17280
17281    Update for new XCBSendRequest API.
17282
17283commit 409a08cff8347d39e0e6c53c9f380d21f221f5ac
17284Author: Jamey Sharp <jamey@minilop.net>
17285Date:   Thu Feb 23 18:12:31 2006 -0800
17286
17287    Performance fix: Replace calls to XCBGetRequestRead with the new XCBGetQueuedRequestRead. Cuts a lot of syscalls.
17288
17289commit ec30a27341b97620b07dd886f98d1d7664a67685
17290Author: Jamey Sharp <jamey@minilop.net>
17291Date:   Thu Feb 23 18:01:46 2006 -0800
17292
17293    Minor performance fix: Access dpy->xcl->connection directly instead of calling XCBConnectionOfDisplay. It happens a lot.
17294
17295commit 53c471c6a835d5cedeca99f2c97058d196a3fd7e
17296Author: Jamey Sharp <jamey@minilop.net>
17297Date:   Thu Feb 23 11:46:09 2006 -0800
17298
17299    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.
17300
17301commit 41c0121a8718b530feaf7fe315b673d9b8defce2
17302Author: Jamey Sharp <jamey@minilop.net>
17303Date:   Tue Feb 21 21:25:41 2006 -0800
17304
17305    Refactor the code that passes requests down to XCB into a separate issue_complete_request function.
17306
17307commit 67d06e0fe468dca22847aa14d3f917128f89f9cf
17308Author: Jamey Sharp <jamey@minilop.net>
17309Date:   Tue Feb 21 15:33:05 2006 -0800
17310
17311    If we have not actually put the buffer back, deferred invariants may not hold. This is OK.
17312
17313commit 434bf80b4053ff1ba82adf65de1f76b4d3731bf1
17314Author: Jeremy C. Reed <reed@reedmedia.net>
17315Date:   Tue Feb 21 14:10:22 2006 -0800
17316
17317    Set XTHREADLIB correctly for dragonfly platforms.
17318
17319commit b24834762e975bd319f9ab5c7cf790b2a02a9474
17320Author: Jamey Sharp <jamey@minilop.net>
17321Date:   Tue Feb 21 14:03:26 2006 -0800
17322
17323    Sometimes functions other than _XUnlockDisplay call _XPutXCBBuffer. Some invariants appropriate for Unlock are not appropriate otherwise: move them to _XUnlockDisplay.
17324
17325commit 99b8defd0d5e6993071e21638128c9de2574b37d
17326Author: Jamey Sharp <jamey@minilop.net>
17327Date:   Tue Feb 21 13:04:21 2006 -0800
17328
17329    Execute BeforeFlush hooks on complete buffers, not request-at-a-time.
17330    Traditional Xlib worked this way; I dunno why I changed it.
17331
17332commit 9b01e7849775749182052fe324df9d8e6ceeee99
17333Author: Jamey Sharp <jamey@minilop.net>
17334Date:   Tue Feb 21 12:51:44 2006 -0800
17335
17336    Factor the XCBSendRequest call out of the conditional in _XPutXCBBuffer.
17337
17338commit 7ce7ac882de128955751a5307889db9d712d8a72
17339Author: Jamey Sharp <jamey@minilop.net>
17340Date:   Tue Feb 21 12:44:30 2006 -0800
17341
17342    Quit using a triple-pointer. Almost as if I were a sensible person.
17343
17344commit 35a858be218cdbfa4593d44a67663d5c25297016
17345Author: Jamey Sharp <jamey@minilop.net>
17346Date:   Tue Feb 21 12:21:01 2006 -0800
17347
17348    Remove the XCL_PAD macro.
17349
17350commit e741b70ed2542c5463c57dac44bc37328616733b
17351Author: Jamey Sharp <jamey@minilop.net>
17352Date:   Sun Feb 19 12:29:27 2006 -0800
17353
17354    Update .gitignores for *.o and nls/locale.dir*.
17355
17356commit f25b4b00e1683b0d97dba46dac46d65a9c2270a6
17357Author: Jamey Sharp <jamey@minilop.net>
17358Date:   Sun Feb 19 12:28:41 2006 -0800
17359
17360    Move .cvsignore to .gitignore.
17361
17362commit c7cda56eebaf6ab11403363be14d4948d7d8be38
17363Author: Jamey Sharp <jamey@minilop.net>
17364Date:   Sun Feb 19 11:49:15 2006 -0800
17365
17366    Land XCB support on X.org HEAD.
17367
17368commit 881467b3032261791ef5ec61b3879bb68d0a3d8c
17369Author: Jamey Sharp <jamey@minilop.net>
17370Date:   Sun Feb 19 02:14:11 2006 +0000
17371
17372    Merge the X11 and BIGREQS package sets: they were used in exactly the same places.
17373
17374commit 6b0158dfad714db5b89c04dbea3aedeafa0fb146
17375Author: Jamey Sharp <jamey@minilop.net>
17376Date:   Tue Feb 14 19:37:36 2006 +0000
17377
17378    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.
17379
17380commit 010c3acbb3a6993d39274f42d88c00849acb0fb0
17381Author: Alan Coopersmith <alan.coopersmith@sun.com>
17382Date:   Sun Feb 12 18:19:17 2006 +0000
17383
17384    Bug #5628 <https://bugs.freedesktop.org/show_bug.cgi?id=5628> Shadow pages not created correctly when MANDIR & MANSUFFIX don't match.
17385
17386commit fe8c01c80263457f01b70dc1511b2bd9466b7c96
17387Author: Alan Coopersmith <alan.coopersmith@sun.com>
17388Date:   Fri Feb 3 23:34:43 2006 +0000
17389
17390    Fix typo in .TH line
17391
17392commit b091c217f3c3f60dde78b09a95c150df6c83d7ba
17393Author: Kevin E Martin <kem@kem.org>
17394Date:   Thu Dec 15 00:24:28 2005 +0000
17395
17396    Update package version number for final X11R7 release candidate.
17397
17398commit e7c04e0e65a2a0c70c6ad29ec2d6f4350fd81c2a
17399Author: Kevin E Martin <kem@kem.org>
17400Date:   Tue Dec 6 22:48:41 2005 +0000
17401
17402    Change *man_SOURCES ==> *man_PRE to fix autotools warnings.
17403
17404commit b5c495854d5270e64e6d588388ffa906bfcaac22
17405Author: Kevin E Martin <kem@kem.org>
17406Date:   Sat Dec 3 05:49:42 2005 +0000
17407
17408    Update package version number for X11R7 RC3 release.
17409
17410commit 9e96dbc343c7f27ff47607acd75378ab23903e2a
17411Author: Kevin E Martin <kem@kem.org>
17412Date:   Sat Dec 3 04:41:47 2005 +0000
17413
17414    Add check and cflags for malloc(0) returning NULL.
17415
17416commit 649c37b47909620ccafde3e983de8321cddd74ce
17417Author: Alan Coopersmith <alan.coopersmith@sun.com>
17418Date:   Mon Nov 28 22:03:04 2005 +0000
17419
17420    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)
17421
17422commit 8d1500df66d796ebff2b0d8c02205e5fa6796d4a
17423Author: Alan Coopersmith <alan.coopersmith@sun.com>
17424Date:   Wed Nov 23 22:33:06 2005 +0000
17425
17426    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
17427
17428commit 93cf3747f9ae8d30bd485b41c5ff10397f68f078
17429Author: Eric Anholt <anholt@freebsd.org>
17430Date:   Sun Nov 20 23:17:39 2005 +0000
17431
17432    Add/improve libs .cvsignores.
17433
17434commit fa1f4a08112bfa14d3758f4702733dd3892966c3
17435Author: Kevin E Martin <kem@kem.org>
17436Date:   Sat Nov 19 07:15:39 2005 +0000
17437
17438    Update pkgconfig files to separate library build-time dependencies from application build-time dependencies, and update package deps to work with separate build roots.
17439
17440commit 7012f9b56c594cf40855ba0dbf93af1263417a8c
17441Author: Kevin E Martin <kem@kem.org>
17442Date:   Mon Nov 14 21:51:07 2005 +0000
17443
17444    Fix xf86bigfont pkgconfig dep typo.
17445
17446commit 60a72f47951f46b4e0505d9903a94af3b6bed8ca
17447Author: Kevin E Martin <kem@kem.org>
17448Date:   Wed Nov 9 21:19:12 2005 +0000
17449
17450    Update package version number for X11R7 RC2 release.
17451
17452commit 6bb0c3796b6e1beddc376a896e865704886e1462
17453Author: Kean Johnson <kean@armory.com>
17454Date:   Tue Nov 8 06:33:25 2005 +0000
17455
17456    See ChangeLog entry 2005-11-07 for details.
17457
17458commit 14be0098ad90c3e68bd2d21b00ffabb76f1fd780
17459Author: Kevin E Martin <kem@kem.org>
17460Date:   Tue Nov 1 15:11:50 2005 +0000
17461
17462    Update pkgcheck dependencies to work with separate build roots.
17463
17464commit 1bf71462a972e0fc56de63f5c7fd613b37fc70f1
17465Author: Donnie Berkholz <spyderous@gentoo.org>
17466Date:   Fri Oct 28 10:44:03 2005 +0000
17467
17468    Revert that, it's redundant. But it is worth noting that --disable-xkb is broken, if anyone cares.
17469
17470commit be627a39fe373e2e81fdc263780e70a271d9e0c5
17471Author: Donnie Berkholz <spyderous@gentoo.org>
17472Date:   Fri Oct 28 08:28:08 2005 +0000
17473
17474    Add dependency on inputproto for XI.h.
17475
17476commit 5fae4cb456cb03fb70cd065dbc2ca94c8ed99082
17477Author: Adam Jackson <ajax@nwnk.net>
17478Date:   Fri Oct 21 18:44:24 2005 +0000
17479
17480    Bug #4736: Error decoding for Damage extension. (Jonathan Lennox)
17481
17482commit 1171fa9dc77413f0e90933a565ec07068052afb4
17483Author: Kevin E Martin <kem@kem.org>
17484Date:   Wed Oct 19 02:48:08 2005 +0000
17485
17486    Update package version number for RC1 release.
17487
17488commit 2a2d905706308b9d5a1c16af1067fb390f43850c
17489Author: Alan Coopersmith <alan.coopersmith@sun.com>
17490Date:   Tue Oct 18 00:00:08 2005 +0000
17491
17492    Use @LIB_MAN_SUFFIX@ instead of $(LIB_MAN_SUFFIX) in macro substitutions to work better with BSD make
17493
17494commit 66d35b6971fb26762392a2a8e2c47db46c11116a
17495Author: Alan Coopersmith <alan.coopersmith@sun.com>
17496Date:   Mon Oct 17 21:13:15 2005 +0000
17497
17498    Rename .shadows.DONE to shadows.DONE to avoid some make's thinking it's a suffix rule (reported by Matthieu Herrb)
17499
17500commit a316995a17c084e98ef1b7f25d287c2c08b6d749
17501Author: Donnie Berkholz <spyderous@gentoo.org>
17502Date:   Sun Oct 16 03:03:35 2005 +0000
17503
17504    Change '==' to portable '='.
17505
17506commit b76a072530e884bcbea6ed8fed5aef39361dcfc0
17507Author: Alan Coopersmith <alan.coopersmith@sun.com>
17508Date:   Wed Oct 12 00:04:50 2005 +0000
17509
17510    configure.ac Use XORG_MAN_SECTIONS instead of custom man section configuration. Add shadow man pages for man pages that document multiple functions.
17511
17512commit 8ee5c1429af98206e05a0536f87c0f21a529cdf0
17513Author: Eric Anholt <anholt@freebsd.org>
17514Date:   Tue Oct 11 02:18:36 2005 +0000
17515
17516    Add appropriate pthread libs/flags for FreeBSD, fixing the build of ico and probably others.
17517
17518commit 41ff3b9d1f194a7b56437b650d5f589225c078c6
17519Author: Alan Coopersmith <alan.coopersmith@sun.com>
17520Date:   Sun Oct 9 22:28:39 2005 +0000
17521
17522    Bug #3021 <https://bugs.freedesktop.org/show_bug.cgi?id=3021> Requests and Errors for XFixes are not in XErrorDB (Jonathan Lennox)
17523
17524commit 91ed79852e790049ab54e68f288afb3c953194c9
17525Author: Kevin E Martin <kem@kem.org>
17526Date:   Fri Oct 7 15:00:00 2005 +0000
17527
17528    Clean up generated files
17529    Add missing dist tarball files
17530
17531commit 440399b470c97b159a530602fff11c315aca8d97
17532Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
17533Date:   Wed Oct 5 18:27:10 2005 +0000
17534
17535    Add el_GR.UTF-8 compose file
17536
17537commit 010f0647e25ac617d0f92c8d2b8dda684da545db
17538Author: Daniel Stone <daniel@fooishbar.org>
17539Date:   Fri Sep 30 07:52:46 2005 +0000
17540
17541    Bug #2609: Add Kyrgyz locale (Ilyas Bakirov).
17542
17543commit 3ef2fb67bd8905b208ad7eb790c3843e14cea7ed
17544Author: Daniel Stone <daniel@fooishbar.org>
17545Date:   Fri Sep 30 07:47:55 2005 +0000
17546
17547    Bug #1640: Kinyarwanda locale support. (Steve Murphy)
17548
17549commit 4ae0decabe0960870df0ec165f495166c10a053c
17550Author: Daniel Stone <daniel@fooishbar.org>
17551Date:   Fri Sep 30 07:40:03 2005 +0000
17552
17553    Bug #2268: Add South African locales (Dwayne Bailey). Some whitespace cleanups, as the parser is a little touchy.
17554
17555commit 3f79eb4c99844f618f1889741d1631c2ffe5385f
17556Author: Daniel Stone <daniel@fooishbar.org>
17557Date:   Fri Sep 30 07:11:19 2005 +0000
17558
17559    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.
17560
17561commit d1237d1483ff972c76a0ac344ec97d5280db0007
17562Author: Alan Coopersmith <alan.coopersmith@sun.com>
17563Date:   Thu Sep 29 21:27:12 2005 +0000
17564
17565    Add Xcms.txt to lib/X11
17566
17567commit 4a86f299693f7376cbe98175f0b0c44d691802b3
17568Author: Alan Coopersmith <alan.coopersmith@sun.com>
17569Date:   Sat Sep 24 20:11:06 2005 +0000
17570
17571    Add XQueryExtension.man
17572
17573commit a057a66e2041d45198a13a4ece7c07068f76f21b
17574Author: Alan Coopersmith <alan.coopersmith@sun.com>
17575Date:   Sat Sep 24 00:16:32 2005 +0000
17576
17577    Add XTHREAD_CFLAGS for platforms that need special defines like
17578    -D_REENTRANT or -D_POSIX_whatever to get re-entrant function definitions. Set XDMCP_LIBS correctly for later libXdmcp tests.
17579
17580commit 3e920a65a7c376ad63eae2240fd06904d25d18bf
17581Author: Alan Coopersmith <alan.coopersmith@sun.com>
17582Date:   Fri Sep 2 23:00:30 2005 +0000
17583
17584    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
17585
17586commit 3eb9f2d693af89d04e2fd92492c8205dce332c9c
17587Author: Kristian Høgsberg  <krh@redhat.com>
17588Date:   Thu Sep 1 19:24:13 2005 +0000
17589
17590    Use $(X11_LOCALEDATADIR) instead of @X11_LOCALEDATADIR@ so this install destination can be overridden at make install time.
17591
17592commit 7afa64325183b78d2d6a4862821f8b3e9866105c
17593Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17594Date:   Sun Aug 28 19:45:48 2005 +0000
17595
17596    OpenBSD needs -lpthread for threaded applications too.
17597
17598commit 2b2f3d3877cb7927f196d01a5df6a27bf8d0518a
17599Author: Daniel Stone <daniel@fooishbar.org>
17600Date:   Fri Aug 26 05:16:46 2005 +0000
17601
17602    Define ERRORDB/KEYSYMDB to XERRORDB/XKEYSYMDB if the former is undefined but the latter is. (Gerte Hoogewerf)
17603
17604commit 07066da0902df91c71f2adb81d1a17ec29165553
17605Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17606Date:   Sun Aug 21 15:45:04 2005 +0000
17607
17608    update
17609
17610commit de44d8b111f57bd2f015e085fd8298c5f2a15ef3
17611Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17612Date:   Sun Aug 21 15:38:39 2005 +0000
17613
17614    Threads support for BSD systems:
17615    - need to check for gewtpwuid_r to define mtsafeapi
17616    - build UIThrstubs if needed.
17617
17618commit 60217fdb918bafb2082519efe5cba3b13ad3082a
17619Author: Adam Jackson <ajax@nwnk.net>
17620Date:   Wed Aug 17 19:46:08 2005 +0000
17621
17622    Add xthreadlib variable to x11.pc. Bump to 0.99.1.
17623
17624commit e1f4c6f5e36c1511f66fa1fac76520fd97eecbad
17625Author: Alan Coopersmith <alan.coopersmith@sun.com>
17626Date:   Wed Aug 17 01:27:08 2005 +0000
17627
17628    Fix more broken multi-line .ds macros. Remove extraneous ;'s .
17629
17630commit 1909786f4a7d686369edcfc05a938df115fab37c
17631Author: Alan Coopersmith <alan.coopersmith@sun.com>
17632Date:   Tue Aug 16 19:23:15 2005 +0000
17633
17634    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)
17635
17636commit 83406d69c62070d2eeef23eb47f1ca887f711ee5
17637Author: Alan Coopersmith <alan.coopersmith@sun.com>
17638Date:   Mon Aug 15 19:53:37 2005 +0000
17639
17640    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
17641
17642commit 5bb43de17de8e71d967488a713bf2b3448533444
17643Author: Alan Coopersmith <alan.coopersmith@sun.com>
17644Date:   Sat Aug 6 18:59:49 2005 +0000
17645
17646    Typo fix in output message
17647
17648commit afe34b95862bb3c06cdbe724cb5ec3001a4a5215
17649Author: Alan Coopersmith <alan.coopersmith@sun.com>
17650Date:   Thu Aug 4 02:55:49 2005 +0000
17651
17652    //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)
17653
17654commit 3979a0b88edf6475ce5cfaa386e18ef980bda13c
17655Author: Alan Coopersmith <alan.coopersmith@sun.com>
17656Date:   Thu Aug 4 02:51:30 2005 +0000
17657
17658    //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)
17659
17660commit 34b454df192f4563499c453ccdb8c079f4a20cbe
17661Author: Alan Coopersmith <alan.coopersmith@sun.com>
17662Date:   Sat Jul 30 20:30:46 2005 +0000
17663
17664    Include config.h so Xtrans knows which transport types to build code for
17665
17666commit cd9c9936b49c125eda779b99887d7e6ae4cf56cd
17667Author: Alan Coopersmith <alan.coopersmith@sun.com>
17668Date:   Sat Jul 30 19:15:16 2005 +0000
17669
17670    Add -D flags to clear various warnings (Stefan Dirsch)
17671
17672commit e7fef67b4531faddd805d8f2157903006d3117ed
17673Author: Kevin E Martin <kem@kem.org>
17674Date:   Fri Jul 29 21:22:50 2005 +0000
17675
17676    Various changes preparing packages for RC0:
17677    - Verify and update package version numbers as needed
17678    - Implement versioning scheme
17679    - Change bug address to point to bugzilla bug entry form
17680    - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to reenable it)
17681    - Fix makedepend to use pkgconfig and pass distcheck
17682    - Update build script to build macros first
17683    - Update modular Xorg version
17684
17685commit 2ebb00244928237088e68325b1032b3550455ce9
17686Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17687Date:   Sat Jul 23 20:19:31 2005 +0000
17688
17689    remove orphan TAB at begin of line
17690
17691commit dd7a9cdecda73e024ca84c5b9a22b18688038d94
17692Author: Kevin E Martin <kem@kem.org>
17693Date:   Sat Jul 23 18:09:39 2005 +0000
17694
17695    Modify modular libs to use Xregion.h instead of region.h
17696
17697commit 6d635a88d91647b1b63611c3591f74916f88cd1c
17698Author: Kevin E Martin <kem@kem.org>
17699Date:   Sat Jul 23 18:06:16 2005 +0000
17700
17701    lib/Xrender/Picture.c Change region.h to Xregion.h and modify internal references to include <X11/Xregion.h>.
17702
17703commit 36283f50fd9748733ae84cb7fb52ca8d9e661c15
17704Author: Daniel Stone <daniel@fooishbar.org>
17705Date:   Sat Jul 16 06:25:35 2005 +0000
17706
17707    Set soversion to 6.2.0.
17708
17709commit 75fd5ae6e4683b9b9dcc13bc2f0faf223610a74f
17710Author: Keith Packard <keithp@keithp.com>
17711Date:   Fri Jul 15 04:27:32 2005 +0000
17712
17713    Add missing Makefile.am to lib/X11/modules
17714
17715commit 6d84a8b1329af1fcfe86b198f1a6e7dd6ff616c3
17716Author: Keith Packard <keithp@keithp.com>
17717Date:   Fri Jul 15 04:08:51 2005 +0000
17718
17719    Move i18n modules to top-level so they can be built in the right order (before xlib for non-loadable, after xlib for loadable).
17720    Link i18n modules against xlib to resolve Xlib symbols used by them.
17721
17722commit 45f40126a73295345bb5eb187b1167874842ab6e
17723Author: Alexander Gottwald <ago@freedesktop.org>
17724Date:   Thu Jul 14 19:50:00 2005 +0000
17725
17726    Add $(top_srcdir)/src to include list
17727
17728commit 0aed7d91f5928d09d541617aad03709b5090658d
17729Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17730Date:   Thu Jul 14 17:04:49 2005 +0000
17731
17732    Build fix for non-GNU make.
17733
17734commit b79422ccb02ab44548d1038956ab0cd4e2638645
17735Author: Adam Jackson <ajax@nwnk.net>
17736Date:   Thu Jul 14 15:12:44 2005 +0000
17737
17738    typo fixes (Matthieu Herrb)
17739
17740commit 419304cde2fda19457c667870edefc0b227651b3
17741Author: Alan Coopersmith <alan.coopersmith@sun.com>
17742Date:   Wed Jul 13 02:41:36 2005 +0000
17743
17744    Add missing backslashes to xlocale_la_SOURCES
17745
17746commit d14cc5c8964c4539b57c6cb51ef653292e410b79
17747Author: Keith Packard <keithp@keithp.com>
17748Date:   Mon Jul 11 20:32:55 2005 +0000
17749
17750    Ammend AM_CFLAGS with all needed includes; cannot figure out an easy way to automate this.
17751
17752commit 424c2d8905eb2ad0a2df15b1da2f96140bfbcffb
17753Author: Lars Knoll <lars@trolltech.com>
17754Date:   Mon Jul 11 15:24:32 2005 +0000
17755
17756    compile
17757
17758commit de9784eb1bde46efc316da279e3da27c6cc288a7
17759Author: Keith Packard <keithp@keithp.com>
17760Date:   Mon Jul 11 09:26:40 2005 +0000
17761
17762    Ignore built man page files
17763
17764commit 0c258c36d1523113790c599b16d2947d7aa6469a
17765Author: Keith Packard <keithp@keithp.com>
17766Date:   Mon Jul 11 09:18:31 2005 +0000
17767
17768    Minor changes to help modular Xlib build i18n modules
17769
17770commit 6e752ea1203b786423e40f43340bce15ca3de0f0
17771Author: Keith Packard <keithp@keithp.com>
17772Date:   Mon Jul 11 08:29:18 2005 +0000
17773
17774    Enable loadable i18n modules, making them configurable on the configure command line.
17775    Clean up conditionals for XKB, XCMS, XLOCALEDIR
17776    Create new lib directory for locale modules in ${X11_LIBDIR}/locale/lib. Add this to the default XLOCALEDIR search path.
17777    Create separate X11_LOCALEDATADIR variable pointing at ${datadir}/X11/locale for installing locale data.
17778    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.
17779    Display message at end of configure script with selected options.
17780    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.
17781
17782commit b46cf0d879f1dbf92dcf5a0305d18986c766ed84
17783Author: Daniel Stone <daniel@fooishbar.org>
17784Date:   Sun Jul 10 22:37:33 2005 +0000
17785
17786    Fix segfault when _XimProtoCreateIC() fails to create a context; Debian #239991. (Chung-chieh Shan)
17787
17788commit 6f2132b18e61ca9755e4b45550f3f5097dd1fbb7
17789Author: Alan Coopersmith <alan.coopersmith@sun.com>
17790Date:   Sat Jul 9 20:06:04 2005 +0000
17791
17792    Set __libmansuffix__ & __xorgversion__ correctly when cpp processing man pages
17793
17794commit 3939ac4410446b46071c9d714f4270c12bf904fe
17795Author: Alan Coopersmith <alan.coopersmith@sun.com>
17796Date:   Sat Jul 9 18:44:14 2005 +0000
17797
17798    - 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
17799    - Don't assume $(CPP) can take gcc-only -traditional flag
17800    - CPP process man pages as is done in the monolithic tree
17801
17802commit 19ba9d0df86c688319377467254b9ea9c4b0eccc
17803Author: Keith Packard <keithp@keithp.com>
17804Date:   Sat Jul 9 06:01:49 2005 +0000
17805
17806    Clean up .cvsignore files
17807
17808commit f4bba6fa9b9ce637be7662754750054567e9250a
17809Author: Adam Jackson <ajax@nwnk.net>
17810Date:   Sat Jul 9 02:53:29 2005 +0000
17811
17812    typo fix. i suck.
17813
17814commit 3305da61a61695e24c1ea11d3f59dddb52873d47
17815Author: Adam Jackson <ajax@nwnk.net>
17816Date:   Sat Jul 9 02:41:18 2005 +0000
17817
17818    Check for bigfont proto headers via pkgconfig (Arwed von Merkatz)
17819
17820commit afdae2e8ccb93de2987b5a5e850695af15ece7e0
17821Author: Keith Packard <keithp@keithp.com>
17822Date:   Fri Jul 8 07:11:56 2005 +0000
17823
17824    Create and use XlibConf.h to match modular build which needs XTHREADS defined when building extensions
17825
17826commit 86fa88dc594ca2093030acf5c84973ee2b33eb4c
17827Author: Keith Packard <keithp@keithp.com>
17828Date:   Fri Jul 8 06:57:06 2005 +0000
17829
17830    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.
17831    Add a bunch of .cvsignore files
17832
17833commit 30f6ffedeaf899e699f0c6b14c9471ce9bcc430c
17834Author: Chris Less <clee@c133.org>
17835Date:   Mon Jul 4 23:01:48 2005 +0000
17836
17837    Fixing bug #380 - add a man page for XQueryExtension, XListExtensions, and XFreeExtensionList.
17838
17839commit 55328bb6caabd37885cd19d7a5821ed425daf321
17840Author: Eric Anholt <anholt@freebsd.org>
17841Date:   Sun Jul 3 10:17:38 2005 +0000
17842
17843    Avoid a gmake-specific construct, and instead just write the name of the current directory into each Makefile.am.
17844
17845commit 726538ba21b631bfb0b8094a5546acdaf57379ff
17846Author: Daniel Stone <daniel@fooishbar.org>
17847Date:   Sun Jul 3 07:37:33 2005 +0000
17848
17849    Fix more include paths; add dix-config.h to XKB code.
17850
17851commit 8c7677138e70e14eca0255f2168583f0ecc21994
17852Author: Daniel Stone <daniel@fooishbar.org>
17853Date:   Sun Jul 3 07:00:55 2005 +0000
17854
17855    Add Xtrans definitions (FONT_t, TRANS_CLIENT) to clean up warnings.
17856    Add XSERV_t, TRANS_SERVER, TRANS_REOPEN to quash warnings.
17857    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>.
17858
17859commit 197697c92a63091a4cc3cc04dcb7fa29d2655758
17860Author: Daniel Stone <daniel@fooishbar.org>
17861Date:   Fri Jul 1 22:13:35 2005 +0000
17862
17863    Fix objdir != srcdir, as well as make distcheck.
17864    Don't attempt to create Compose.pre files; formatting fixes.
17865    Added if not already present.
17866
17867commit c162d60ad8f124563f94a2a266de59373936266c
17868Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
17869Date:   Wed Jun 22 22:46:31 2005 +0000
17870
17871    Apply these patches from Theo van Klaveren:
17872    lib-dmx.patch lib-FS.patch lib-X11.patch lib-XRes.patch lib-XScrnSaver.patch lib-xtrans.patch
17873    to make the libraries distcheck.
17874
17875commit 9ee8abdab03ea605a6327118ab7dacab6adf8876
17876Author: Alan Coopersmith <alan.coopersmith@sun.com>
17877Date:   Sat Jun 18 07:48:43 2005 +0000
17878
17879    Move Secure RPC flags from X11/configure.ac to xtrans/xtrans.m4 since multiple modules will need them
17880
17881commit 12afc57b7d455781eee305e9ed6a899ceec8729e
17882Author: Daniel Stone <daniel@fooishbar.org>
17883Date:   Wed Jun 15 16:50:47 2005 +0000
17884
17885    Typo fix to locale/error/keysym location declarations.
17886
17887commit 9a895777e30762f61b98d25be3b5d8b1169baa17
17888Author: Daniel Stone <daniel@fooishbar.org>
17889Date:   Wed Jun 15 13:37:43 2005 +0000
17890
17891    Typo in ImUtil.h commit -- I AM CAPTAIN SKILL.
17892
17893commit add49285663684875ab7a5c58ec7a2cf8b775f67
17894Author: Daniel Stone <daniel@fooishbar.org>
17895Date:   Wed Jun 15 13:32:35 2005 +0000
17896
17897    Define locations for XErrorDB, XKeysymDB, and locale data in configure.ac. Add AC_DEFINE_DIR macro from autoconf-archive.cryp.to towards this end.
17898    Move ImUtil.h from src/ to include/X11/.
17899
17900commit 845dfc6b42b950890866ee4df27761e086f50dca
17901Author: Daniel Stone <daniel@fooishbar.org>
17902Date:   Wed Jun 15 13:27:48 2005 +0000
17903
17904    Move ImUtil.h from src/ to include/X11/. Additionally, copy Cmap.h as a distribution file.
17905
17906commit bba117f0d98f62cfb060d0fab97b407a3a0bfda9
17907Author: Daniel Stone <daniel@fooishbar.org>
17908Date:   Fri Jun 10 14:11:36 2005 +0000
17909
17910    Remove pointless include of Xlib.h.
17911    Fix #include path to bigreqstr.h.
17912
17913commit 1a0de49da1274882bab05b0f7240936b37955e5c
17914Author: Alexander Gottwald <ago@freedesktop.org>
17915Date:   Thu Jun 9 21:30:15 2005 +0000
17916
17917    Use $(srcdir) for Compose.pre and XLC_LOCALE.pre
17918
17919commit af4f0f302644ebfbb0ca9f4016a4aee85c973d37
17920Author: Alexander Gottwald <ago@freedesktop.org>
17921Date:   Thu Jun 9 15:55:33 2005 +0000
17922
17923    Replace <X11/transport.c> with <X11/Xtrans/transport.c>
17924
17925commit fd5f58e0baf692e34b9b622286f18762cc2500d3
17926Author: Alexander Gottwald <ago@freedesktop.org>
17927Date:   Thu Jun 9 15:52:02 2005 +0000
17928
17929    Replace <X11/Xtrans.h> with <X11/Xtrans/Xtrans.h>
17930    Copy Xtrans.h to exports/include/X11/Xtrans only
17931
17932commit 44538f9940f969d46c0e5e4b201c684cde2ba611
17933Author: Alan Coopersmith <alan.coopersmith@sun.com>
17934Date:   Sun Jun 5 03:29:33 2005 +0000
17935
17936    Port Imake flags to autoconf tests & --enable-* flags: HASSETUGID, HASGETRESUID, NO_XLOCALEDIR, HAS_SHM and XF86BIGFONT
17937
17938commit 588e30e9ec65fa6205a34be650b79d5e2243edec
17939Author: Alan Coopersmith <alan.coopersmith@sun.com>
17940Date:   Sat Jun 4 22:53:21 2005 +0000
17941
17942    Add --enable-secure-rpc flag and checks for needed functions for Secure RPC ("SUN-DES-1") authentication method
17943
17944commit a547afee2ef49cc41bbb67f9cff5a52a283c0854
17945Author: Alan Coopersmith <alan.coopersmith@sun.com>
17946Date:   Sat Jun 4 21:20:20 2005 +0000
17947
17948    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)
17949
17950commit ca93c761559ae464189c7ea7cf11c6a6679f2431
17951Author: Matthieu Herrb <matthieu.herrb@laas.fr>
17952Date:   Sat May 28 01:02:32 2005 +0000
17953
17954    Don't use $< in explicit rules. This only works with GNU make. <https://bugs.freedesktop.org/show_bug.cgi?id=3383>
17955
17956commit 09ebb349359e3dd9131fa2fa8b07559faa173654
17957Author: Alan Coopersmith <alan.coopersmith@sun.com>
17958Date:   Sun May 22 19:05:11 2005 +0000
17959
17960    Convert man pages to long file names in lib/X11, lib/Xt, & lib/Xext
17961
17962commit 761219b1ef9befc350c8a35b6f96d047e5f008cc
17963Author: Alan Coopersmith <alan.coopersmith@sun.com>
17964Date:   Sat May 21 23:07:48 2005 +0000
17965
17966    xtrans:
17967    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
17968    ICE:
17969    Add ICE_t #define required by Xtrans headers. Replace static defines of LOCALCONN & UNIXCONN with new XTRANS_CONNECTION_FLAGS macro.
17970    X11:
17971    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.
17972
17973commit 4b1ba6eb975b547b808f4d5c8825a3261de5e02a
17974Author: Alan Coopersmith <alan.coopersmith@sun.com>
17975Date:   Sat May 21 04:26:12 2005 +0000
17976
17977    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)
17978    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).
17979
17980commit 1d425d5e2092dd18a7dd599b37ed9af61cf59819
17981Author: Adam Jackson <ajax@nwnk.net>
17982Date:   Thu May 19 00:22:32 2005 +0000
17983
17984    revert last change, didn't do right thing at all, sorry for the noise
17985
17986commit 1b0c46c1ae61d751dd3ca96de8e2c3fe21c5f4f2
17987Author: Adam Jackson <ajax@nwnk.net>
17988Date:   Thu May 19 00:10:07 2005 +0000
17989
17990    Require automake 1.7 in AM_INIT_AUTOMAKE
17991
17992commit cd4657c175dbab6aaca36f18a0ca92c95b5567dd
17993Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
17994Date:   Tue May 17 22:32:09 2005 +0000
17995
17996    - Check for xproto as its CFLAGS are needed in the .pc file
17997
17998commit 46e8d8a65430dd87c10b066b5cff99a689c22241
17999Author: Egbert Eich <eich-at-freedesktop-dot-org>
18000Date:   Tue May 17 08:10:10 2005 +0000
18001
18002    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).
18003
18004commit 23198d2bfbf0049b2630235cd4d4a4ffba7ec6c1
18005Author:  <ssp@aware-of-vacuity.boston.redhat.com>
18006Date:   Mon May 16 22:35:27 2005 +0000
18007
18008    Make Xdmcp unconditionally required, require various protocol modules.
18009    Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com>
18010    Check for kbproto if using XKB.
18011    Require xextproto rather than xextensions
18012    Remove the entries from the xlibs tree, as they are not relevant here.
18013
18014commit 8bd3aea84ce54b8b76a898f3ae00e2b499c14a5e
18015Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
18016Date:   Mon May 16 21:48:36 2005 +0000
18017
18018    Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com>
18019    Check for kbproto if using XKB.
18020
18021commit 9b1fa9ca3ed852ed40860f137511683f4bdae6fa
18022Author:  <ssp@aware-of-vacuity.boston.redhat.com>
18023Date:   Mon May 16 21:27:35 2005 +0000
18024
18025    Require xextproto rather than xextensions
18026    Remove the entries from the xlibs tree, as they are not relevant here.
18027
18028commit 7eee605e3aeed549d1053325a03027c5e8cbf71b
18029Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
18030Date:   Fri May 13 22:53:36 2005 +0000
18031
18032    - For now put xtrans in X11/Xtrans/X11, since libX11 is looking for it in <X11/...>
18033    - For Xcomposite and Xdamage, don't link the build system out of the xc tree
18034    - Link the public X11 headers into their own directory
18035    - Add links to XKeysymDB and XErrorDB
18036    - Add links to all the Xlib man pages
18037    - Add links to the lcUniConv subdirectory
18038    - Conditionally include config.h in Xlib source
18039
18040commit 6769ccda88caf27d1441d335ef2b318a047a612b
18041Author: Alan Coopersmith <alan.coopersmith@sun.com>
18042Date:   Mon Mar 21 04:58:21 2005 +0000
18043
18044    xc/lib/X11/ErrDes.c
18045    //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>
18046
18047commit 3b9e8ece93b916c55a82df53e85f097418edf471
18048Author: Roland Mainz <roland.mainz@nrubsig.org>
18049Date:   Sat Mar 19 22:04:55 2005 +0000
18050
18051    xc/nls/Compose/iso8859-2
18052    xc/nls/Compose/iso8859-3
18053    xc/nls/Compose/iso8859-9
18054    //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>.
18055
18056commit 0ce5950a08b9ab23ca8a32effdd40c421e92df84
18057Author: Alan Coopersmith <alan.coopersmith@sun.com>
18058Date:   Tue Mar 8 02:53:36 2005 +0000
18059
18060    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>.
18061
18062commit df341cd2f2f263f13323e8c0936ea8aa0d7fbba1
18063Author: Roland Mainz <roland.mainz@nrubsig.org>
18064Date:   Tue Feb 1 03:12:28 2005 +0000
18065
18066    xc/nls/Compose/pt_BR.UTF-8
18067    //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>.
18068
18069commit 483d3973f56dda51f9251d3ea808aaf7f3b48c10
18070Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18071Date:   Fri Jan 28 18:31:31 2005 +0000
18072
18073    cleaned up boundary-case handling for Uxxxx Unicode keysym notation
18074
18075commit f234188a4c1e6b655aef0e3957ccad20d4c5847f
18076Author: Roland Mainz <roland.mainz@nrubsig.org>
18077Date:   Wed Jan 19 01:53:55 2005 +0000
18078
18079    xc/nls/compose.dir
18080    xc/nls/locale.dir
18081    xc/nls/Compose/Imakefile
18082    xc/nls/Compose/pt_BR.UTF-8
18083    xc/nls/XI18N_OBJS/Imakefile
18084    xc/nls/XI18N_OBJS/pt_BR.UTF-8
18085    xc/nls/XLC_LOCALE/Imakefile
18086    xc/nls/XLC_LOCALE/pt_BR.UTF-8
18087    //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.
18088
18089commit 7448ea7ef425d35cfc31eb41d46f4d879774f376
18090Author: Egbert Eich <eich-at-freedesktop-dot-org>
18091Date:   Fri Jan 14 18:03:09 2005 +0000
18092
18093    Made some security enhancements:
18094    - no writing past end of buffer caused by bogus locale.
18095    - explicitely add a \0 character at end of string. (Bugzilla #2262)
18096
18097commit 5557d47fcf22a6f3adf327691158f2270e3d5094
18098Author: Roland Mainz <roland.mainz@nrubsig.org>
18099Date:   Fri Jan 14 04:59:05 2005 +0000
18100
18101    xc/nls/compose.dir
18102    xc/nls/locale.alias
18103    xc/nls/locale.dir
18104    //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>.
18105
18106commit 1fa3737f042f798fa11a9ff5b03f3b3ba3529824
18107Author: Egbert Eich <eich-at-freedesktop-dot-org>
18108Date:   Tue Jan 11 17:37:57 2005 +0000
18109
18110    '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.
18111
18112commit 642cd269f94a234aa470a1d43385bd42625d89a1
18113Author: Egbert Eich <eich-at-freedesktop-dot-org>
18114Date:   Tue Dec 14 08:59:20 2004 +0000
18115
18116    Removed #ifdef'ed out code together with the comment explaining why it was #ifdef'ed out.
18117    Fixed typo.
18118    Added comment to a changed that's been committed with one of the previous commits.
18119
18120commit a07ccae36e629741d2e48de7730114d30a975a41
18121Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18122Date:   Sun Dec 12 08:42:50 2004 +0000
18123
18124    Fix missing XChangeProperty() prototype missing in synopsis section.
18125
18126commit c73adb1c8ad806c2f9f0ba32f4b449e7a871501c
18127Author: Alexander Gottwald <ago at freedesktop dot org>
18128Date:   Wed Dec 8 13:42:01 2004 +0000
18129
18130    Bugzilla #1980 (https://bugs.freedesktop.org/show_bug.cgi?id=1980) Handle XERRORDB only on WIN32 platform
18131
18132commit d5e7ab194103e22fd6c5094aaa107d90210bb600
18133Author: Jim Gettys <jg@freedesktop.org>
18134Date:   Thu Dec 2 16:18:16 2004 +0000
18135
18136    fix comment to indicate additional possible mode. i bug 1756 reported by Owen Taylor.
18137
18138commit 03940d7330cb1bbf93d49c650aefb19de457da7c
18139Author: Alexander Gottwald <ago at freedesktop dot org>
18140Date:   Wed Dec 1 13:06:55 2004 +0000
18141
18142    Bugzilla #1980, https://bugs.freedesktop.org/show_bug.cgi?id=1980 Make location of XErrorDB configurable
18143
18144commit 11a03ab908bfcfab8a3492684e8e9320e492d552
18145Author: Alexander Gottwald <ago at freedesktop dot org>
18146Date:   Wed Dec 1 12:42:17 2004 +0000
18147
18148    Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory
18149
18150commit f3d83ee153f42e8899b844377e6b842d93411e62
18151Author: Alexander Gottwald <ago@freedesktop.org>
18152Date:   Mon Nov 15 15:06:54 2004 +0000
18153
18154    Bufzilla #1802, http://freedesktop.org/bugzilla/show_bug.cgi?id=1802 Added mingw (Win32) port
18155
18156commit b798ea11911ac58a8e6e7d15a2a643b023859749
18157Author: Alexander Gottwald <ago at freedesktop dot org>
18158Date:   Mon Nov 15 13:29:56 2004 +0000
18159
18160    Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory
18161
18162commit 436108cd6c84053698e5ca629096f59b34f50c2a
18163Author: Kristian Høgsberg  <krh@redhat.com>
18164Date:   Thu Nov 11 15:37:01 2004 +0000
18165
18166    Fix #1818
18167
18168commit bf2e6ef66ba55f90efa4a4ba8c8b6d3ec0d1531c
18169Author: Roland Mainz <roland.mainz@nrubsig.org>
18170Date:   Tue Nov 9 00:56:56 2004 +0000
18171
18172    xc/nls/compose.dir
18173    xc/nls/locale.alias
18174    xc/nls/locale.dir
18175    //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>.
18176
18177commit 443890ceefbd6dafe68e30d103ec4f9d316ed655
18178Author: Roland Mainz <roland.mainz@nrubsig.org>
18179Date:   Fri Nov 5 00:58:49 2004 +0000
18180
18181    xc/nls/compose.dir
18182    xc/nls/locale.alias
18183    xc/nls/locale.dir
18184    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>.
18185
18186commit af7467ec734321f127b957921cce7792902b6794
18187Author: Egbert Eich <eich-at-freedesktop-dot-org>
18188Date:   Mon Oct 18 17:29:03 2004 +0000
18189
18190    Correcting font encodings for GB18030, GBK and BIG5-HKSCS. Adding nls support for those encodings (Bugzilla 1573, James Su).
18191
18192commit 1b900b59cb24fe8be3db0d434b04d462c0eeb206
18193Author: Egbert Eich <eich-at-freedesktop-dot-xorg>
18194Date:   Mon Oct 18 14:21:45 2004 +0000
18195
18196    Made handling of DevelDrivers for x86-64 more conformant to other platforms.
18197    Compress all font encodings (Stefan Dirsch).
18198    Fixed warnings.
18199    Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and Root mode (bugzilla #1580, James Su).
18200    Added another compose key combination for the Euro symbol (Stefan Dirsch).
18201    Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian).
18202    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.
18203    Added explanation for DESTDIR to install to a different directory than /.
18204    Added some early bailouts to atiprobe if PCI structure pointer is NULL to prevent sig11.
18205    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.
18206    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().
18207    Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink).
18208    Create missing links for backward compatibility to XFree86 (Stefan Dirsch).
18209    Changed comment to mention xorg.
18210    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).
18211    Added 'pic' to the man page rendering command pipeline (Werner Fink).
18212    Added missing return value (Stefan Dirsch, Roland Mainz)
18213
18214commit 1ba103c3cad2329db3a31f88d7798b913affc570
18215Author: Eric Anholt <anholt@freebsd.org>
18216Date:   Wed Oct 13 07:38:00 2004 +0000
18217
18218    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>)
18219
18220commit d80237db627bf71ad5615ff4ba05e6ab436aa740
18221Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
18222Date:   Fri Oct 8 22:57:56 2004 +0000
18223
18224    Fri Oct 8 18:53:11 2004 Soeren Sandmann <sandmann@redhat.com>
18225    Move iso10646 last so the "fallback" fonts will actually be used if they are better matches.
18226
18227commit d1f76d17ecf418396627b1a58319f2b9b255548a
18228Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18229Date:   Sun Sep 26 22:54:57 2004 +0000
18230
18231    bug fix for previous patch
18232
18233commit 48932d9b71d10e15812f47d5b842ab6aa8dac625
18234Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
18235Date:   Sun Sep 26 20:46:17 2004 +0000
18236
18237    The big keysym cleanup, to bring implementation in line with the recent revision of Appendix A of the protocol spec. (Markus Kuhn)
18238
18239commit 2e02a95dcc43dd3ec7bbaf4675ffb94f5074f543
18240Author: Egbert Eich <eich@freedesktop.org>
18241Date:   Wed Sep 15 09:05:22 2004 +0000
18242
18243    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).
18244    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).
18245
18246commit 2d3afb68a104a80a21ee622b9abb9c95e83505d3
18247Author: Egbert Eich <eich@freedesktop.org>
18248Date:   Tue Aug 31 11:37:03 2004 +0000
18249
18250    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).
18251
18252commit e689746c8d0e21e9011e8b91a3071d235d3a2a74
18253Author: Kevin E Martin <kem@kem.org>
18254Date:   Thu Aug 19 06:48:06 2004 +0000
18255
18256    Fix header file to #ifdef the XKB keysyms when they are used. This fixes the X test suite build failure.
18257
18258commit d558a53a6f57eecfcaadce5141fe3a08860defcb
18259Author: Keith Packard <keithp@keithp.com>
18260Date:   Sat Aug 14 07:12:36 2004 +0000
18261
18262    Use XLIB_SKIP_ARGB_VISUALS environment variable to disable all depth 32 visuals. Necessary to keep Flash from crashing.
18263    Must call ValidateGC/ValidatePicture on "real" GC/Picture to ensure pCompositeClip is set correctly.
18264    Need to take the composite clip from the "real" GC/Picture and turn it into the clientClip for the backing version.
18265    Adjust pixmap screen origin to account for drawable->x/y Change debugging output a bit (disabled by default)
18266
18267commit 85c2d81f299ed3444658011b7d6fb0a7ab8a6f55
18268Author: Alexander Gottwald <ago@freedesktop.org>
18269Date:   Fri Aug 13 16:28:19 2004 +0000
18270
18271    Set most significant bit to be a one. (Bug #1024, Kensuke Matsuzaki)
18272    Fix conversion from sjis and euc. (Bug #1024, Toshio Takabe)
18273
18274commit c4d56e4e288d4e48b84b021a61638f46e9a45e27
18275Author: Adam Jackson <ajax@nwnk.net>
18276Date:   Wed Aug 11 05:25:13 2004 +0000
18277
18278    Bug #372: Prevent a crash in XPolygonRegion when called with a bogus point count. Reported by Andreas Luik.
18279
18280commit 92487437173f600f208d825f65756d3ad14a4f7e
18281Author: Kevin E Martin <kem@kem.org>
18282Date:   Mon Aug 9 22:37:22 2004 +0000
18283
18284    Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, Harold L. Hunt II, Alexander Gottwald).
18285    Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, David Dawes).
18286    Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson).
18287    HP-PA build fix (Bug #828, Guy Martin, Paul Anderson).
18288    Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine).
18289    Fix attempt to read video ROM before enabling it (Bug #843, Ivan Kokshaysky, Mike A. Harris).
18290    Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. Harris).
18291    Clarify xset man page description of how to use the keyboard repeat rate settings (Bug #846, Mike A. Harris).
18292    Fix problem where print-screen key would get remapped to sys-req in certain keymaps, which broke GNOME printscreen functionality (Bug #847, Owen Taylor).
18293    Fix several render problems:
18294    - MMIO mode support
18295    - Hang on IGP chips
18296    - VT switching hang
18297    - 3D render corruption (Bug #922, Hui Yu).
18298
18299commit 55c2ee568e7d3903258286a13bdf96ce5348ffda
18300Author: Matthieu Herrb <matthieu.herrb@laas.fr>
18301Date:   Tue Jul 27 06:06:05 2004 +0000
18302
18303    - remove remaining AMOEBA references.
18304    - remove unused file.
18305
18306commit 6e884b12911eedfb003e90a3829ce66f7fc9cf2d
18307Author: Søren Sandmann Pedersen  <sandmann@daimi.au.dk>
18308Date:   Tue Jul 20 17:48:09 2004 +0000
18309
18310    Tue Jul 20 19:38:06 2004 Soeren Sandmann <sandmann@daimi.au.dk>
18311    Set font_data->xlfd_data to NULL after XFree(). (#837, patch from Bastien Nocera).
18312    Tue Jul 20 18:23:32 2004 Soeren Sandmann <sandmann@daimi.au.dk>
18313    Use /dev/urandom on Linux.(#761).
18314
18315commit 6f0bc97aa4e6de5a4b001f40ac10795cfdf09fc1
18316Author: Eric Anholt <anholt@freebsd.org>
18317Date:   Fri May 28 23:26:44 2004 +0000
18318
18319    Forced commit to note repocopy from xc/lib/XThrStub, will be connected to the build after this.
18320
18321commit aa7010c43ae9f39fb84b5ff155f76117c9e527a0
18322Author: Egbert Eich <eich@pdx.freedesktop.org>
18323Date:   Mon May 24 19:02:11 2004 +0000
18324
18325    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.
18326
18327commit 720702da29769d80ad1254d92edbad5b30f8a8da
18328Author: Alan Coopersmith <alan.coopersmith@sun.com>
18329Date:   Sat May 22 03:47:42 2004 +0000
18330
18331    Bugzilla #658: XStringToKeysym fails for Greek_IOTAdiaeresis (Fixes VSW5 testcase XStringToKeysym-7 failure)
18332
18333commit 3aed873292424b497d9a7dcee2975b95bf5ac966
18334Author: Alan Coopersmith <alan.coopersmith@sun.com>
18335Date:   Sat Apr 24 23:39:25 2004 +0000
18336
18337    XOpenDisplay should try tcp connection if local connections fail (aka Sun bug id #4624183).
18338    Also includes fix from NetBSD Problem Report #25098 (Michael van Elst) (Xlib segfaults with IPv6 if compiled with HASXDMAUTH).
18339    xc/config/cf/sunLib.tmpl
18340    xc/lib/FS/Imakefile Add missing shared library dependencies for Solaris
18341
18342commit c6349f43193b74a3c09945f3093a871b0157ba47
18343Author: Egbert Eich <eich@freedesktop.org>
18344Date:   Fri Apr 23 18:42:09 2004 +0000
18345
18346    Merging XORG-CURRENT into trunk
18347
18348commit c3c4ddc682950a01b80825021f3e2503ab01ea7f
18349Author: Kaleb Keithley <kaleb@freedesktop.org>
18350Date:   Tue Nov 25 19:28:07 2003 +0000
18351
18352    Initial revision
18353
18354commit dc4268a7dadc8da0d561757a68461246728613d3
18355Author: Kaleb Keithley <kaleb@freedesktop.org>
18356Date:   Fri Nov 14 16:48:47 2003 +0000
18357
18358    Initial revision
18359
18360commit deae12c6b683898f5213992d561a59d4ea889cca
18361Author: Kaleb Keithley <kaleb@freedesktop.org>
18362Date:   Fri Nov 14 15:54:30 2003 +0000
18363
18364    R6.6 is the Xorg base-line
18365