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 -> struct references 5328 5329 perl -i -0 -p -e 's{\s*->\s*}{->}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 -> with → 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 => with ⇒ 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 – 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 − 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{^<(.*\.h)>\ *}{<filename class="headerfile"><\1></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 − and \^ to either   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><X11/([^&]+)></term>(\s+)<listitem>(\s+)<para>}{<varlistentry id="Standard_Header_Files:$2">$1<term><filename class="headerfile"><X11/$2></filename></term>$3<listitem>$4<indexterm type="file"><primary><filename class="headerfile">X11/$2</filename></primary></indexterm>$4<indexterm><primary>Files</primary><secondary><X11/$2></secondary></indexterm>$4<indexterm><primary>Headers</primary><secondary><X11/$2></secondary></indexterm>$4<para>}g' CH01.xml 11648 11649 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11650 11651commit bb66e8f1ce330043278d3e8cbc7d5d3bc56ee030 11652Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11653Date: Sun Jul 18 12:47:27 2010 -0700 11654 11655 specs/libX11: Add indexterms for headerfiles that .hN used to provide 11656 11657 Automatic substitution performed via: 11658 perl -i -p -e 's{^<<filename class="headerfile">([^<]+)</filename>>(.*)$}{<filename class="headerfile"><$1></filename>$2\n<indexterm type="file"><primary><filename class="headerfile">$1</filename></primary></indexterm>\n<indexterm><primary>Files</primary><secondary><filename class="headerfile"><$1></filename></secondary></indexterm>\n<indexterm><primary>Headers</primary><secondary><filename class="headerfile"><$1></filename></secondary></indexterm>}' *.xml 11659 11660 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11661 11662commit d0b0d215cb85da48b323b1392149ce181e7ef5ec 11663Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11664Date: Fri Jul 16 21:15:13 2010 -0700 11665 11666 specs/libX11: convert multicolumn lists from tables to <simplelist> 11667 11668 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11669 11670commit e679f80c6210c0aefe3669f8f536353619097a4b 11671Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11672Date: Fri Jul 16 20:51:52 2010 -0700 11673 11674 specs/libX11: Tag WM_* as <property> 11675 11676 Mass substitution done by: 11677 perl -i -p -e 'if ($_ !~ 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 {<<filename class="headerfile">$1</filename>>$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