ChangeLog revision c6e579a2
1commit a465588218c1643eedc35b3c24409cb775454eee 2Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3Date: Fri Apr 5 15:50:06 2024 -0700 4 5 libX11 1.8.9 6 7 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8 9commit 52a191ee096e93cf711f77b6a56c508b68afc0b6 10Author: Peter Hutterer <peter.hutterer@who-t.net> 11Date: Fri Apr 5 13:15:32 2024 +1000 12 13 Revert "ximcp: Unmark to fabricate key events with XKeyEvent serial" 14 15 This commit causes a regression, see #205, #206, #207, #208. 16 17 This reverts commit 024d229fdf88a7755577b01b46af6ef908d599e0. 18 19commit 3ea9f4f769fecbbc8c5210c5833ad83ab8d9e97e 20Author: Peter Hutterer <peter.hutterer@who-t.net> 21Date: Fri Apr 5 13:14:33 2024 +1000 22 23 Revert "imDefLkup: Commit first info in XimCommitInfo" 24 25 This commit causes a regression, see #205, #206, #207, #208. 26 27 This reverts commit 041b5291f0956c5cda5054be2981c0d02b009a4c. 28 29commit 9afd55ada5d7a3c1d3ed23ff8451629a1d989a7f 30Author: Alan Coopersmith <alan.coopersmith@oracle.com> 31Date: Mon Mar 25 11:51:03 2024 -0700 32 33 xlibi18n: restore parse_line1 for WIN32 builds 34 35 Accidentally removed by __UNIXOS2__ cleanup 36 Closes: #204 37 Fixes: 225a4bbb ("unifdef __UNIXOS2__") 38 39 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 40 41commit e4927d0c4fbf950bc45fc27309290e1911e75f42 42Author: Alan Coopersmith <alan.coopersmith@oracle.com> 43Date: Sun Mar 24 15:02:23 2024 -0700 44 45 libX11 1.8.8 46 47 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 48 49commit 4ab58f26d388c82ffe46ef1c7dd1cee6259eec8a 50Author: Alan Coopersmith <alan.coopersmith@oracle.com> 51Date: Wed Feb 21 18:23:36 2024 -0800 52 53 unifdef NULL_NOT_ZERO 54 55 I can't find any evidence this was ever defined, should only have 56 been needed for odd-ball pre-C89 compilers. 57 58 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 59 60commit 7bb2a5050a2f6049c92b301fae1cf89532b70060 61Author: Alan Coopersmith <alan.coopersmith@oracle.com> 62Date: Wed Feb 21 18:18:46 2024 -0800 63 64 unifdef USL_SHAREDLIB 65 66 I can't find any history of this being set in the imake or autoconf builds 67 68 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 69 70commit 4400a68b3a5ecacd1f997d41a0572544fa8fe544 71Author: Alan Coopersmith <alan.coopersmith@oracle.com> 72Date: Sat Feb 17 15:37:41 2024 -0800 73 74 unifdef Lynx 75 76 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 77 78commit 1e56b27429667e772b15b0f39863a8467297062b 79Author: Alan Coopersmith <alan.coopersmith@oracle.com> 80Date: Sun Feb 11 15:15:00 2024 -0800 81 82 unifdef __QNX__ 83 84 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 85 86commit ab0a3014822687da48cd3784253c34dea0a6ad2c 87Author: Alan Coopersmith <alan.coopersmith@oracle.com> 88Date: Sun Feb 11 15:10:59 2024 -0800 89 90 unifdef __uxp__ 91 92 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 93 94commit 4ce3962b701c502acc96b6eaf104a5ffc317c5d7 95Author: Alan Coopersmith <alan.coopersmith@oracle.com> 96Date: Sun Feb 11 15:07:15 2024 -0800 97 98 unifdef __vax__ 99 100 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 101 102commit 65a6f162fd9fee4e647ac49c546c5d6a1655c58c 103Author: Alan Coopersmith <alan.coopersmith@oracle.com> 104Date: Sun Feb 11 15:03:19 2024 -0800 105 106 unifdef __sgi_not_xconsortium 107 108 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 109 110commit 3296d7b8d14a6c114489d40b76e8da339e11b25c 111Author: Alan Coopersmith <alan.coopersmith@oracle.com> 112Date: Sun Feb 11 15:01:25 2024 -0800 113 114 unifdef __sgi 115 116 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 117 118commit 4322fff7e125c7975dc1b83c433b904c7d3384b5 119Author: Alan Coopersmith <alan.coopersmith@oracle.com> 120Date: Sun Feb 11 15:00:34 2024 -0800 121 122 unifdef sgi 123 124 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 125 126commit 613d3624c2cc0ab2f18206e165a3e4a02e261148 127Author: Alan Coopersmith <alan.coopersmith@oracle.com> 128Date: Sun Feb 11 14:59:28 2024 -0800 129 130 unifdef hpux 131 132 Also removes shl_load() support, which was only buildable for HP-UX 133 134 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 135 136commit 225a4bbbbd174e26cc408640e10f3c37f81f3a3f 137Author: Alan Coopersmith <alan.coopersmith@oracle.com> 138Date: Sun Feb 11 14:56:22 2024 -0800 139 140 unifdef __UNIXOS2__ 141 142 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 143 144commit 0df284b450933e0239d1a85a7ff2d0c9d44d22af 145Author: Alan Coopersmith <alan.coopersmith@oracle.com> 146Date: Sun Feb 11 14:50:29 2024 -0800 147 148 unifdef ultrix 149 150 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 151 152commit c3f3eb1284c59305568288923810dc155df6025c 153Author: Alan Coopersmith <alan.coopersmith@oracle.com> 154Date: Sun Feb 11 14:49:13 2024 -0800 155 156 unifdef AIXV3 157 158 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 159 160commit b35344c9a74b478509e33384cc99bcc4fd092a2d 161Author: Alan Coopersmith <alan.coopersmith@oracle.com> 162Date: Sun Feb 11 14:47:31 2024 -0800 163 164 unifdef __osf__ 165 166 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 167 168commit 041b5291f0956c5cda5054be2981c0d02b009a4c 169Author: Takao Fujiwara <tfujiwar@redhat.com> 170Date: Wed Jan 31 20:27:57 2024 +0900 171 172 imDefLkup: Commit first info in XimCommitInfo 173 174 Xic.private.proto.commit_info can receive multiple XimCommitInfo 175 when typing keys very quickly like an bar code scanner (or evemu-play) 176 and the first info in XimCommitInfo should be committed to keep 177 the typing key order. 178 179 Fixes: #198 180 181commit 024d229fdf88a7755577b01b46af6ef908d599e0 182Author: Takao Fujiwara <tfujiwar@redhat.com> 183Date: Wed Jan 31 20:26:40 2024 +0900 184 185 ximcp: Unmark to fabricate key events with XKeyEvent serial 186 187 _XimProtoKeypressFilter() and _XimProtoKeyreleaseFilter() can 188 receive XKeyEvent from both the typing on the keyboard and the 189 callback of XIM_FORWARD_EVENT. 190 191 If the filter functions unmark to fabricate XKeyEvent from the typing 192 on the keyboard during receiving XKeyEvent from the callback of 193 XIM_FORWARD_EVENT with typing keys very quickly likes an bar code 194 scanner (or evemu-play), XIM server cannot receive some key events and 195 it causes the key typing order to get scrambled. 196 197 Now XIM client saves the serial in XKeyEvent and the filter functions 198 unmark to fabricate XKeyEvent from the callback of XIM_FORWARD_EVENT 199 only. 200 201 Fixes: #198 202 203commit ae3eca18cec44a953789c7f77ffab888713ed132 204Author: Peter Hutterer <peter.hutterer@who-t.net> 205Date: Thu Nov 30 15:05:51 2023 +1000 206 207 Fix _XkbReadGetDeviceInfoReply for nButtons == dev->buttons 208 209 XkbGetDeviceInfo(dpy, XkbXI_ButtonActionsMask, 2, 0, 0) always returns 210 NULL because the number of buttons on the device equals (unsurpisingly) 211 the number of buttons requested (i.e. first + nBtns == dev->nbuttons). 212 213 This currently causes it to bail out and return NULL. 214 215 Fixes f293659d5a4024bda386305bb7ebeb4647c40934 216 217commit 0a951047f64e87c5ab328084310bf27bcd4964f3 218Author: Walter Harms <wharms@bfs.de> 219Date: Mon Jan 8 17:18:19 2024 +0100 220 221 _XimProtoIMFree:no need to check arg for Xfree() 222 223 Xfree() will happily ignore NULL, no need to check 224 225commit dce614623e1f069d79d33a233305e0754fa2fd48 226Author: Walter Harms <wharms@bfs.de> 227Date: Mon Jan 8 17:01:44 2024 +0100 228 229 _XimEncodeString:no need to check arg for Xfree() 230 231 Xfree() will happily ignore NULL, no need to check 232 233commit 4f78b615806fa1cfe2bdcf3f7c868b8a00dede77 234Author: Walter Harms <wharms@bfs.de> 235Date: Mon Jan 8 16:50:52 2024 +0100 236 237 Fix XCreateIC() memory leak (Part 2) 238 239 Direct leak of 12 byte(s) in 2 object(s) allocated from: 240 #0 0x7f4f25c3f7a7 in strdup (/usr/lib64/libasan.so.6+0x5c7a7) 241 #1 0x7f4f252ce6a1 in _XimEncodeString libX11-1.8.3/modules/im/ximcp/imRm.c:818 242 #2 0x7f4f252ce6a1 in _XimEncodeString libX11-1.8.3/modules/im/ximcp/imRm.c:807 243 #3 0x7f4f252d2f0f in _XimSetICValueData libX11-1.8.3/modules/im/ximcp/imRm.c:2912 244 #4 0x7f4f252b536a in _XimLocalCreateIC libX11-1.8.3/modules/im/ximcp/imLcIc.c:176 245 #5 246 247 0x7f4f251f0105 in XCreateIC libX11-1.8.3/src/xlibi18n/ICWrap.c:251 248 249 detected and fix by Patrick Lerda <patrick9876@free.fr> 250 applied with adjustment, do changes when OOM (unlikely but good practise) 251 252commit ed0b97e480d7d07c24719007273989b2eb3e4046 253Author: Walter Harms <wharms@bfs.de> 254Date: Mon Jan 8 16:21:02 2024 +0100 255 256 _XimLocalDestroyIC:fix possible mem leak 257 258 Adapted: 259 Fix XCreateIC() memory leak by Patrick Lerda <patrick9876@free.fr> Part 1 260 261commit 07978634b8f4f7d9537c7f46ddcd48f2ab3efcb8 262Author: Walter Harms <wharms@bfs.de> 263Date: Mon Jan 8 15:16:11 2024 +0100 264 265 _XimLocalCreateIC: get rid of bzero 266 267commit 59c9a89e250d479334bdd4d800ca6cfa919b8fc3 268Author: Walter Harms <wharms@bfs.de> 269Date: Mon Jan 8 15:09:49 2024 +0100 270 271 _XimLocalCreateIC: minor cleanup 272 273 minor cleanup, no code change 274 275commit e5b14e59a16fa49c7e7af0acfc1e4311bebec536 276Author: Walter Harms <wharms@bfs.de> 277Date: Mon Jan 8 13:35:28 2024 +0100 278 279 _XimLocalCreateIC:no need to check arg for Xfree() 280 281 Xfree() will happily ignore NULL, no need to check 282 283commit e6310b5212f1dea184e1626ad5eeec167fcb41ac 284Author: Walter Harms <wharms@bfs.de> 285Date: Mon Jan 8 13:06:03 2024 +0100 286 287 _XimLocalDestroyIC: no need to check arg for Xfree() 288 289 Xfree() will happily ignore NULL, no need to check 290 291commit 1261802fb316f088504b99ffe80095d13825a85b 292Author: Walter Harms <wharms@bfs.de> 293Date: Sun Jan 7 18:30:30 2024 +0100 294 295 fix table width 296 297 the width of first column was to small and 298 caused a hyphenation. there is only one word, 299 fix for me. 300 301commit c745719e23af44a4b40ab4508447637b35d91a1e 302Author: Alan Coopersmith <alan.coopersmith@oracle.com> 303Date: Tue Oct 3 09:00:01 2023 -0700 304 305 libX11 1.8.7 306 307 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 308 309commit 7916869d16bdd115ac5be30a67c3749907aea6a0 310Author: Yair Mizrahi <yairm@jfrog.com> 311Date: Thu Sep 7 16:15:32 2023 -0700 312 313 CVE-2023-43787: Integer overflow in XCreateImage() leading to a heap overflow 314 315 When the format is `Pixmap` it calculates the size of the image data as: 316 ROUNDUP((bits_per_pixel * width), image->bitmap_pad); 317 There is no validation on the `width` of the image, and so this 318 calculation exceeds the capacity of a 4-byte integer, causing an overflow. 319 320 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 321 322commit b4031fc023816aca07fbd592ed97010b9b48784b 323Author: Alan Coopersmith <alan.coopersmith@oracle.com> 324Date: Thu Sep 7 16:12:27 2023 -0700 325 326 XCreatePixmap: trigger BadValue error for out-of-range dimensions 327 328 The CreatePixmap request specifies height & width of the image as CARD16 329 (unsigned 16-bit integer), so if either is larger than that, set it to 0 330 so the X server returns a BadValue error as the protocol requires. 331 332 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 333 334commit 73a37d5f2fcadd6540159b432a70d80f442ddf4a 335Author: Alan Coopersmith <alan.coopersmith@oracle.com> 336Date: Thu Sep 7 15:55:04 2023 -0700 337 338 XPutImage: clip images to maximum height & width allowed by protocol 339 340 The PutImage request specifies height & width of the image as CARD16 341 (unsigned 16-bit integer), same as the maximum dimensions of an X11 342 Drawable, which the image is being copied to. 343 344 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 345 346commit 204c3393c4c90a29ed6bef64e43849536e863a86 347Author: Alan Coopersmith <alan.coopersmith@oracle.com> 348Date: Thu Sep 7 15:54:30 2023 -0700 349 350 CVE-2023-43786: stack exhaustion from infinite recursion in PutSubImage() 351 352 When splitting a single line of pixels into chunks to send to the 353 X server, be sure to take into account the number of bits per pixel, 354 so we don't just loop forever trying to send more pixels than fit in 355 the given request size and not breaking them down into a small enough 356 chunk to fix. 357 358 Fixes: "almost complete rewrite" (Dec. 12, 1987) from X11R2 359 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 360 361commit 6858d468d9ca55fb4c5fd70b223dbc78a3358a7f 362Author: Alan Coopersmith <alan.coopersmith@oracle.com> 363Date: Sun Sep 17 14:19:40 2023 -0700 364 365 CVE-2023-43785: out-of-bounds memory access in _XkbReadKeySyms() 366 367 Make sure we allocate enough memory in the first place, and 368 also handle error returns from _XkbReadBufferCopyKeySyms() when 369 it detects out-of-bounds issues. 370 371 Reported-by: Gregory James DUCK <gjduck@gmail.com> 372 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 373 374commit 78b37accff1abbe713349d59fdefd963ffa04bbc 375Author: Alan Coopersmith <alan.coopersmith@oracle.com> 376Date: Sun Sep 17 10:11:02 2023 -0700 377 378 XOpenDisplay: ensure each screen has a valid root_visual pointer 379 380 Other code assumes this pointer cannot be NULL, so fail the connection 381 if a bug has caused the X server to give a non-existent visual ID for 382 the default visual of any screen. 383 384 Reported-by: Gregory James DUCK <gjduck@gmail.com> 385 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 386 387commit 3962080b7846d79fd3cf10b16eddacf68eb321b7 388Author: Antti Savolainen <antti.savo@gmail.com> 389Date: Tue Aug 8 11:37:03 2023 +0000 390 391 Add two compose sequences for dagger/obelisk symbols 392 393 The dagger symbol has several modern uses such as marking someone as 394 dead or something as extinct. Historically it has been used to indicate 395 a footnote. 396 397commit 15bbc81cf6b7c76b8a59cab6d77c9aba88b8562d 398Author: Christopher Chavez <chrischavez@gmx.us> 399Date: Sat Jul 29 16:58:05 2023 +0000 400 401 Xutil.h: use Bool for XEmptyRegion(), XEqualRegion() 402 403commit 5433cb93513c31e9cba2a447cf48d4b9aeb53285 404Author: Benno Schulenberg <bensberg@telfort.nl> 405Date: Mon Jul 17 08:02:09 2023 +0200 406 407 nls: add two compose sequences for currency symbol `฿` (the Thai baht) 408 409 The compose sequences are obvious: the capital B plus the vertical bar. 410 411 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 412 413commit 2ff401a472c5f210844cc358d5a27bd340546796 414Author: Benno Schulenberg <bensberg@telfort.nl> 415Date: Mon Jul 17 07:56:14 2023 +0200 416 417 nls: reshuffle the compose sequences for currencies into ascending order 418 419 That is: sort them according to code point. 420 421commit c0e2d42d411be3245498ffe6484b35421f096c02 422Author: Benno Schulenberg <bensberg@telfort.nl> 423Date: Mon Jul 17 07:27:14 2023 +0200 424 425 nls: add compose sequences for `₲` (guaraní), `₭` (kip), and `₮` (tugrik) 426 427 When Ghana, Nigeria, Costa Rica and El Salvador have compose sequences 428 for their currency symbols (cedi: `₵`, naira: `₦`, colón: `₡`), then 429 Paraguay, Laos, and Mongolia deserve to have such sequences as well. 430 431 The sequences should be obvious: the relevant capital letter of the 432 name of the currency (G, K, T) plus a vertical bar, a minus sign, 433 and an equals sign, respectively. 434 435 Also add two sequences for `$` (the dollar sign), for consistency. 436 437 Drop the tentative comments for drachma, penny, and austral, as those 438 currencies have been obsolete for more than twenty years. 439 440 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 441 442commit d9fe5dc735c75398e36998c226d2418137229673 443Author: Benno Schulenberg <bensberg@telfort.nl> 444Date: Mon Jul 17 07:06:51 2023 +0200 445 446 nls: change result of sequence `<C> <bar>` from `¢` (CENT) to `₵` (CEDI) 447 448 As a cent is a small coin, it makes no sense to use an uppercase letter 449 to compose the `¢` symbol -- having four sequences with a lowercase `c` 450 plus a `bar` or a `slash` available for composing `¢` should be enough. 451 452 Use the sequence `<C> <bar` (with the uppercase `C`) for composing `₵` 453 (the CEDI SIGN) instead. 454 455 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 456 457commit 031e63060ed11eb8a876aac7515bc5a67371ed6b 458Author: Antti Savolainen <antti.savo@gmail.com> 459Date: Thu Aug 17 18:25:20 2023 +0300 460 461 nls: add a compose sequence with double dead_tilde for ≈ (almost equal to) 462 463 Some keyboards lack direct asciitilde and instead have only dead tilde. 464 This compose sequence allows those keyboards to type the almost equal 465 to sign. 466 467commit df2a18c4643bb4a50f2f795a75a905a1032a004a 468Author: Benno Schulenberg <bensberg@telfort.nl> 469Date: Mon Jul 17 17:03:16 2023 +0200 470 471 nls: delete twenty seven untypable Greek compose sequences 472 473 Only a few keyboard layouts contain the `dead_iota` keysym, and none 474 of those layouts contains the `acute` keysym, so compose sequences 475 that combine the two symbols cannot be typed and are thus useless. 476 477 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 478 479commit 44f908d9283710ffc75b22d2ae7a8948119b3e61 480Author: Alan Coopersmith <alan.coopersmith@oracle.com> 481Date: Sun Aug 20 15:11:04 2023 -0700 482 483 docs: fix names for XkbKeyKeyType & XkbKeyKeyTypeIndex 484 485 Implementation had "Key" twice in these macro names, 486 but the docs had only listed it once. 487 488 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 489 490commit 7b16c55063ba4eb258bf91c170e378c1e2d398bd 491Author: Alan Coopersmith <alan.coopersmith@oracle.com> 492Date: Sun Aug 20 15:01:09 2023 -0700 493 494 docs: remove XkbKeySymsOffset 495 496 This function was documented but never implemented. 497 498 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 499 500commit 41a157dfeefe943f5a79948e4f9ef93473144fd8 501Author: Alan Coopersmith <alan.coopersmith@oracle.com> 502Date: Sun Aug 20 14:52:18 2023 -0700 503 504 docs: remove XkbGetNameChanges 505 506 This function was documented but never implemented. 507 508 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 509 510commit 3442a8cee25dc0254ee69bcf83a89d287a86c97a 511Author: Alan Coopersmith <alan.coopersmith@oracle.com> 512Date: Sun Aug 20 14:44:18 2023 -0700 513 514 docs: remove XkbChangeIndicators and Xkb{Get,Note}IndicatorChanges 515 516 These functions were documented but never implemented. 517 518 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 519 520commit b7a15d6c6efd0bfddc81cf8443073c4bc1c9772d 521Author: Alan Coopersmith <alan.coopersmith@oracle.com> 522Date: Sun Aug 20 14:31:20 2023 -0700 523 524 docs: XkbSAActionSetCtrls is really named XkbActionSetCtrls 525 526 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 527 528commit a1d8dd726757e00e3a774b92456f5c742492475b 529Author: Alan Coopersmith <alan.coopersmith@oracle.com> 530Date: Sun Aug 20 14:18:32 2023 -0700 531 532 docs: remove XkbGetStickyKeysOptions & XkbSetStickyKeysOptions 533 534 These functions were documented but never implemented. 535 536 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 537 538commit bf13f197406fe745c5ff46a4748a41b6646aebf6 539Author: Alan Coopersmith <alan.coopersmith@oracle.com> 540Date: Sun Aug 20 14:09:35 2023 -0700 541 542 docs: remove XkbGetSlowKeysDelay & XkbSetSlowKeysDelay 543 544 These functions were documented but never implemented. 545 546 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 547 548commit caab7321d2ec14ee776ec329ed2facf0816a3136 549Author: Alan Coopersmith <alan.coopersmith@oracle.com> 550Date: Sun Aug 20 14:04:15 2023 -0700 551 552 docs: remove XkbGetAccessXTimeout & XkbSetAccessXTimeout 553 554 These functions were documented but never implemented. 555 556 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 557 558commit 843e89bd2467a58cb454705a09d095726b96d93f 559Author: Alan Coopersmith <alan.coopersmith@oracle.com> 560Date: Sun Aug 20 12:58:47 2023 -0700 561 562 docs: remove XkbAllocDeviceLedInfo 563 564 This function was documented but never implemented. 565 566 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 567 568commit 2686dd7a9d6e8e02ce2868e446cdec11617ffab4 569Author: Alan Coopersmith <alan.coopersmith@oracle.com> 570Date: Sun Aug 20 12:41:42 2023 -0700 571 572 docs: finish removing XkbGetBounceKeysDelay & XkbSetBounceKeysDelay 573 574 These functions were documented but never implemented. 575 576 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 577 578commit 188991fe7e41bbf2f384d0875ab2e25c936b0b41 579Author: G. Branden Robinson <g.branden.robinson@gmail.com> 580Date: Sat Jul 29 10:10:00 2023 -0700 581 582 configure script reports stray, confusing "yes" 583 584 Closes issue #194 585 586 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 587 588commit 70f44d76fd8fad96f33ba5c80a394d68b77dda03 589Author: Benno Schulenberg <bensberg@telfort.nl> 590Date: Sun Jul 16 17:08:52 2023 +0200 591 592 nls: add, correct, and normalize some comments of compose sequences 593 594 Add comments for the Khmer digraphs, correct the comments for the 595 Arabic lam-alef decompositions, and normalize the comments for the 596 Breton digraphs and trigraphs. 597 598 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 599 600commit 2a57c32e14779cf1c0c1d2da98a9cac21c00c095 601Author: Benno Schulenberg <bensberg@telfort.nl> 602Date: Sun Jun 25 10:01:05 2023 +0200 603 604 nls: harmonize the comments for compose sequences with combining accents 605 606 Replace the "WITH" with "plus" and lowercase the "AND" in the comments 607 for sequences with combining accents, to make it slightly clearer that 608 the resulting string consists of multiple code points. Also, use the 609 word "COMBINING" in the names of the accents, so that these sequences 610 can be easily grepped, and drop the redundant word "ACCENT". 611 612 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 613 614commit 757ab5ddd86f6cdce24b8ae2c43206b7beb7ab44 615Author: Alan Coopersmith <alan.coopersmith@oracle.com> 616Date: Mon Aug 14 08:33:28 2023 -0700 617 618 Remove XkbSetBounceKeysDelay.man from list of man pages to build 619 620 Fixes: f90e6bfa ("rm XkbSetBounceKeysDelay.man") 621 622 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 623 624commit f90e6bface92fd192b26f03b7ae745661d9c2690 625Author: Walter Harms <wharms@bfs.de> 626Date: Sun Aug 13 22:42:50 2023 +0200 627 628 rm XkbSetBounceKeysDelay.man 629 630 there is no function XkbSetBounceKeysDelay(). 631 It is listed in the specs but never implemented if someone 632 does it, feel free to get the file back. 633 634 closes issue #105 635 636commit b2c3fb7b4dd4c4c7b9d28c7af7776cccc8e0a98b 637Author: Benno Schulenberg <bensberg@telfort.nl> 638Date: Sat Jul 8 08:11:39 2023 +0200 639 640 nls: use the shorter and more consistent name for the dead capital schwa 641 642 The `dead_SCHWA` name is available since xorgproto-2023.1, which was 643 released last month. 644 645commit e45dff5c2a51ad0392061208a8ad33211d952385 646Author: jmcwilliams403 <jmcwilliams403@gmail.com> 647Date: Fri Jun 16 11:37:20 2023 -0400 648 649 NLS: Drop two math sequences, slightly clean up APL sequences. 650 651 * Two compose sequences containing `leftshoe` and `rightshoe` are 652 dropped as no keyboard layout uses these keysyms. 653 * The compose sequences for `therefore` and `because` in the APL 654 block are moved to ascending Unicode order. 655 * The comments for the compose sequences for `U2299` are corrected to 656 more accurately reflect its sequence nodes and its real name in 657 Unicode. 658 659commit af2b6dfab1616dc85be9c9b196e4c56d00447851 660Author: Benno Schulenberg <bensberg@telfort.nl> 661Date: Sun Jun 25 09:09:21 2023 +0200 662 663 nls: delete twenty eight compose sequences that cannot be typed 664 665 The letters `Ǫ` (U+01EA, O with ogonek), `Ȩ` (U+0228, E with cedilla), 666 `Ȯ` (U+022E, O with dot above), `Ạ` (U+1EA0, A with dot below), and 667 their lowercase forms do not occur in any layout of xkeyboard-config, 668 meaning that the compose sequences that contained these letters could 669 not be typed. Delete their dead weight. 670 671 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 672 673commit ab0442d3fa835ce16559b29532ac7f674f8557f4 674Author: Benno Schulenberg <bensberg@telfort.nl> 675Date: Sat Jun 24 09:46:07 2023 +0200 676 677 nls: remove four unobvious and redundant compose sequences for ¥ (yen) 678 679 Composing ¥ with Y and a minus sign seems to have been added in analogy 680 to composing £ with L and a minus sign. But ¥ clearly has a double line 681 through it, so using the equals sign for this is far more logical (and 682 those compose sequences of course exist). Also, L plus an equals sign 683 produces ₤ (lira), not £ (sterling). So... make these sequences more 684 consistent and allow composing ¥ only with Y/y plus an equals sign. 685 686 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 687 688commit 933912480c08495428d45988eebb48fcbe62f14b 689Author: Benno Schulenberg <bensberg@telfort.nl> 690Date: Sat Jun 24 10:02:45 2023 +0200 691 692 nls: order a few compose sequences left first, and in mirroring pairs 693 694commit 178c4fefd3b0b4be5e69274108b86a51f08ec3a6 695Author: Benno Schulenberg <bensberg@telfort.nl> 696Date: Fri Jun 16 09:33:03 2023 +0200 697 698 nls: remove three duplicated combining acute accents 699 700 For the Cyrillic YU with combining acute accent, the string between 701 the quotes contained two U+0301 code points, whereas one is enough, 702 like for all the neighboring strings. 703 704 (This duplication was found by accident with `nano --mini --cons`.) 705 706commit 695e90ad26f632feb0f58ad94882fb3a263bf114 707Author: Alan Coopersmith <alan.coopersmith@oracle.com> 708Date: Thu Jun 15 09:14:48 2023 -0700 709 710 libX11 1.8.6 711 712 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 713 714commit 304a654a0d57bf0f00d8998185f0360332cfa36c 715Author: Alan Coopersmith <alan.coopersmith@oracle.com> 716Date: Sat Jun 10 16:30:07 2023 -0700 717 718 InitExt.c: Add bounds checks for extension request, event, & error codes 719 720 Fixes CVE-2023-3138: X servers could return values from XQueryExtension 721 that would cause Xlib to write entries out-of-bounds of the arrays to 722 store them, though this would only overwrite other parts of the Display 723 struct, not outside the bounds allocated for that structure. 724 725 Reported-by: Gregory James DUCK <gjduck@gmail.com> 726 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 727 728commit 71b08b8af20474bb704a11affaa8ea39b06d5ddf 729Author: Alan Coopersmith <alan.coopersmith@oracle.com> 730Date: Wed May 31 17:45:40 2023 -0700 731 732 libX11 1.8.5 733 734 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 735 736commit 03ba0140940cc76524d83096a47309f5c398541f 737Author: Benno Schulenberg <bensberg@telfort.nl> 738Date: Mon May 22 12:02:06 2023 +0200 739 740 nls: use a slash instead of a combining solidus in compose sequences 741 742 First: combining diacritics like the combining long solidus (`U+0338`) 743 are not meant to be used in compose sequences. Second: there is just 744 one layout in xkeyboard-config that contains the `U0338` character: 745 the deprecated/obsolete German T3 layout. So, practically speaking, 746 these compose sequences with `U0338` were untypable. So, use a slash 747 instead, that almost all layouts have. This does require that the 748 sequence `<Compose> <less> <slash>` changes its meaning from backslash 749 to not-less-than (`≮`). This seems like an acceptable sacrifice, as 750 the sequence `<Compose> <slash> <slash>` is a faster/easier sequence 751 for the backslash and most layouts contain a backslash already anyway. 752 753 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 754 755commit 616701317e25b2c0f509f4d93af5dee6de09012d 756Author: Benno Schulenberg <bensberg@telfort.nl> 757Date: Mon May 22 11:19:32 2023 +0200 758 759 nls: delete six compose sequences that cannot be typed 760 761 None of the layouts in xkeyboard-config contain `Ohook` or `Uhook` 762 or `U+01D3` (U with caron) nor their lowercase forms, so these six 763 sequences cannot be typed on any known keyboard. 764 765 (Also, the target characters of the sequences with hook and horn 766 seem to occur only in Vietnamese, where the horn is part of the 767 vowel and the hook is a tonal mark. All Vietnamese layouts in 768 xkeyboard-config contain `Ohorn` and `Uhorn`, and then the five 769 tonal marks as dead keys, `dead_hook` among them. It does not 770 make sense to want to apply a horn to a vowel that already has 771 a tonal mark, nor is it possible with the Vietnamese layouts.) 772 773 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 774 775commit 35d9621d78cb196b4ccd3ee1baa1ce2a21e4674b 776Author: Benno Schulenberg <bensberg@telfort.nl> 777Date: Sun May 21 16:17:32 2023 +0200 778 779 nls: sort three sequences alphabetically in their group, like all others 780 781 Also, consistently form uppercase+lowercase pairs, like for most others. 782 783commit 208d550954c7266fa8093b02a2a97047e1478c00 784Author: Benno Schulenberg <bensberg@telfort.nl> 785Date: Thu Apr 13 17:25:35 2023 +0200 786 787 nls: make the Amharic compose sequences use the dead-vowel symbols 788 789 Last year, commit 488b156fe2 changed mistaken ASCII characters to the 790 code points that were actually used in the Ethiopian keyboard layout. 791 792 But those code points were mistaken too: the intention was not to use 793 the code point U+FE68 in the layout but the raw symbol code 0xfe68 -- 794 and so on for the other codes. The symbols that these raw codes were 795 intended to refer to were `dead_a` and friends. But these dead vowels 796 were added under different codes: 0xfe80 to 0xfe8b. 797 798 So, correct the Amharic compose sequences to make use of the dead-vowel 799 symbols that were specifically added for them fifteen years ago. 800 801 (As soon as this is merged, xkeyboard-config wil be changed too to use 802 the dead vowels in the Ethiopian layout.) 803 804 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 805 806commit 96cbac89a02220eb21e13ca5fa7c8c5adf77028a 807Author: jmcwilliams403 <jmcwilliams403@gmail.com> 808Date: Fri Mar 31 09:47:03 2023 -0400 809 810 nls: add compose sequence for capital schwa, and delete a deviant one 811 812 This sequence has already been defined in the iso8859-9e compose 813 file meant for the Azerbaijani locale, yet for the longest time, only 814 the lowercase form has existed in the en_US.UTF-8 master compose 815 file. This patch adds one sequence for Latin Capital Letter Schwa for 816 the Azerbaijani language. 817 818 Additionally, the sequence in the pt_BR.UTF-8 locale for capital 819 schwa has been removed as it uses two lowercase e's as part of its 820 sequence, which in practice would make typing both `Multi_key e e` 821 and `Multi_key E E` output capital schwa. 822 823commit a6528d92b3865adb4f9951df91a714f55f56d59e 824Author: Benno Schulenberg <bensberg@telfort.nl> 825Date: Tue Apr 4 11:35:44 2023 +0200 826 827 XSetScreenSaver.man: restore the part that was accidentally snipped 828 829 Four years ago, while removing private macros, commit fe46eb2871 830 unintentionally truncated the XSetScreenSaver manpage. Restore 831 the snipped part (minus trailing spaces and a mistaken singular). 832 833 This fixes issue #187. 834 835 Reported-by: Sergio Gelato 836 837 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 838 839commit eb166af8f9f4184108fe8e0611a09af03d12059e 840Author: jmcwilliams403 <jmcwilliams403@gmail.com> 841Date: Fri Mar 24 08:37:32 2023 -0400 842 843 NLS: Remove strange dead_cedilla cedi sign sequences 844 845 These sequences appear to be a leftover artifact of auto-generating 846 dead_currency sequences for cent and the cedi sign, where chaining 847 with ascii "c" vs "ccedilla" would output cent vs the cedi sign 848 respectively, but somehow this ended up creating strange sequences 849 with dead_cedilla followed by cent or the colon sign to output the 850 cedi sign, the latter sequence of which was already removed recently. 851 Additionally, sequences beginninng with dead_cedilla followed by 852 currency have been removed as they also appear to be related to these 853 artifacts. The two remaining sequences beginning with dead_currency 854 followed by dead_cedilla are maintained and moved to the stretch of 855 dead_currency sequences next to ccedilla, and sorted in order of 856 unicode decompositional complexity. 857 858commit 6e889f34679bb96f4d61ef069259c042cb4a9a59 859Author: Benno Schulenberg <bensberg@telfort.nl> 860Date: Fri Mar 24 09:41:48 2023 +0100 861 862 nls: remove twenty two untypable Greek compose sequences 863 864 None of the keyboard layouts in xkeyboard-config contains accented 865 Greek characters like `Greek_alphaaccent` or `Greek_iotadieresis` 866 (nor in their Unicode hexadecimal forms, like U03AC or 0x10003CA), 867 so none of these sequences could be typed on an existing layout. 868 869 Also, the two sequences that started with `Greek_accentdieresis` 870 were plain wrong: making a spacing character into a dead key. 871 872 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 873 874commit 98be4cb23a5eadc98f81f05c4696f687119152c8 875Author: Benno Schulenberg <bensberg@telfort.nl> 876Date: Fri Mar 24 09:00:48 2023 +0100 877 878 nls: remove four hundred and sixty untypable Greek compose sequences 879 880 None of the keyboard layouts in xkeyboard-config contains code points 881 in the range U+1F00..U+1FFF (the Greek Extended block), so none of the 882 removed sequences could be typed on an existing layout. 883 884 (Of course it is possible that some people have their own private layout 885 that contains some of the Greek Extended characters, but if they do, 886 then they can also define private compose sequences, when needed.) 887 888 Modern Greek does not use these characters with polytonic diacritics 889 anyway -- they are useful only for typing traditional Greek. 890 891 The useless sequences were removed with this command: 892 893 sed -i '/<U1F..>/d' nls/en_US.UTF-8/Compose.pre 894 895 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 896 897commit ff8706a5eae25b8bafce300527079f68a201d27f 898Author: Benno Schulenberg <bensberg@telfort.nl> 899Date: Wed Jun 15 11:54:54 2022 +0200 900 901 nls: delete compose sequences that pointlessly mix upper and lower case 902 903 It's nice to have compose sequences for things like © and ®, but allowing 904 them to be composed also with letters of different case is unneeded. It 905 is good enough to have sequences of only lowercase and only uppercase. 906 907 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 908 909commit ca99e338a9b8aad300933b1336f9e3c091392213 910Author: jmcwilliams403 <jmcwilliams403@gmail.com> 911Date: Sun Mar 12 15:13:27 2023 +0000 912 913 NLS: move dead-caron subscript compositions to the relevant Unicode block 914 915commit dac2985b347cebc0f631e43e41eaeddb68de1676 916Author: jmcwilliams403 <jmcwilliams403@gmail.com> 917Date: Sun Mar 12 14:49:47 2023 +0000 918 919 NLS: implement the expansion of the six Breton N-graph keysyms 920 921 These one-keysym-to-multiple-characters sequences will allow fully defining 922 the Breton keyboard layout, without needing custom compose sequences. 923 924commit e45ca7b41dcd3ace7681d6897505f85d374640f2 925Author: Alan Coopersmith <alan.coopersmith@oracle.com> 926Date: Tue Mar 14 10:22:48 2023 -0700 927 928 gitlab CI: add workflow rules 929 930 Needed due to the gitlab config changes described in: 931 https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/438 932 933 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 934 935commit 72eeeb6dedc39f68608312eb0001f77fffccd499 936Author: jmcwilliams403 <jmcwilliams403@gmail.com> 937Date: Sun Mar 5 13:33:38 2023 +0000 938 939 NLS: move long S compositions to respective blocks 940 941commit 8bbbed75a20556e41658103265974decee974645 942Author: Alan Coopersmith <alan.coopersmith@oracle.com> 943Date: Sat Mar 4 10:44:45 2023 -0800 944 945 configure: Use LT_INIT from libtool 2 instead of deprecated AC_PROG_LIBTOOL 946 947 AC_PROG_LIBTOOL was replaced by LT_INIT in libtool 2 in 2008, 948 so it's time to rely on it. 949 950 Clears autoconf warnings: 951 952 configure.ac:20: warning: The macro `AC_PROG_LIBTOOL' is obsolete. 953 configure.ac:20: You should run autoupdate. 954 m4/libtool.m4:100: AC_PROG_LIBTOOL is expanded from... 955 configure.ac:20: the top level 956 957 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 958 959commit 53bf8584e8d7d5d4a4a8114bff26a6f631c7fac1 960Author: Alan Coopersmith <alan.coopersmith@oracle.com> 961Date: Mon Feb 6 14:09:45 2023 -0800 962 963 configure: replace deprecated AC_HELP_STRING with AS_HELP_STRING 964 965 This silences an autoconf warning. 966 967 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 968 969commit 31a9451111fb1e3ae1199d44f94cd0b134ca455d 970Author: Alan Coopersmith <alan.coopersmith@oracle.com> 971Date: Mon Feb 6 13:13:21 2023 -0800 972 973 configure: raise minimum autoconf requirement to 2.70 974 975 Needed for builds on NetBSD to work correctly, since it depends on 976 AC_USE_SYSTEM_EXTENSIONS defining _OPENBSD_SOURCE to expose the 977 prototype for reallocarray() in the system headers. 978 979 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 980 981commit 1ba02f878f83e3b42389547e01b24371f0a50e60 982Author: Alan Coopersmith <alan.coopersmith@oracle.com> 983Date: Mon Feb 6 13:32:02 2023 -0800 984 985 gitlab CI: Add libtool to required packages 986 987 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 988 989commit fc5ec8edda1efdf5d66db4e76796e3c407527b3b 990Author: Alan Coopersmith <alan.coopersmith@oracle.com> 991Date: Sat Feb 4 12:02:29 2023 -0800 992 993 libX11 1.8.4 994 995 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 996 997commit ba548ed1688d4d5a83913218155d0ddf616ff176 998Author: Alan Coopersmith <alan.coopersmith@oracle.com> 999Date: Sun Jan 15 11:39:20 2023 -0800 1000 1001 configure: options summary should say "libX11", not just "X11" 1002 1003 We're not configuring the entire window system here, just one library 1004 1005 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1006 1007commit eb1c272ab5230d548077b9f59aca4b3457c3a8f8 1008Author: GaryOderNichts <garyodernichts@gmail.com> 1009Date: Sat Dec 17 16:28:40 2022 +0100 1010 1011 Fix a9e845 and 797755 Allow X*IfEvent() to reenter libX11 1012 1013commit 0c7c4ce875ea65160705180ebb68a2c88afb678e 1014Author: Benno Schulenberg <bensberg@telfort.nl> 1015Date: Thu Nov 3 17:25:29 2022 +0100 1016 1017 nls: add two compose sequences for capital B with stroke, for consistency 1018 1019 Two compose sequences (with plain ASCII characters) existed for 1020 the lowercase b with stroke (ƀ) but not for the capital one (Ƀ). 1021 1022 This addresses part of issue #166. 1023 1024 Reported-by: J. McWilliams 1025 1026 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1027 1028commit a80ef097e713aba0330b9cf9f3c1e309084b926f 1029Author: Benno Schulenberg <bensberg@telfort.nl> 1030Date: Thu Nov 3 16:48:03 2022 +0100 1031 1032 nls: delete twelve compose sequences that cannot be typed 1033 1034 There is not any keyboard layout (in xkeyboard-config) that contains 1035 `obelowdot` or `ubelowdot`, so having compose sequences that use these 1036 symbols is pointless. 1037 1038 There are some layouts that contain `otilde` (and one that contains 1039 `utilde`), but those layouts then do not contain `dead_horn`, so 1040 the compose sequences that combine the two symbols are pointless. 1041 1042 There are a few layouts that contain U+0256, U+025C, U+025F or U+0279, 1043 but those layouts do not contain `dead_hook`, so the compose sequences 1044 that combine the latter with one of the former are pointless. 1045 1046 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1047 1048commit 078cc39f05ddfd6166762f3166129258ddd07dd3 1049Author: Benno Schulenberg <bensberg@telfort.nl> 1050Date: Thu Nov 3 16:31:04 2022 +0100 1051 1052 nls: delete 88 compose sequences that have the accents in the wrong order 1053 1054 The various compose sequences with the accents in their proper order 1055 (the highest placed accent first) continue to exist. 1056 1057 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1058 1059commit aa5ef0ba80558f946871f574d38a15673e66221e 1060Author: Benno Schulenberg <bensberg@telfort.nl> 1061Date: Thu Nov 3 16:27:51 2022 +0100 1062 1063 nls: move the compose sequences for Ș and Ț to the fitting Unicode block 1064 1065 This addresses a side note of issue #166. 1066 1067 Reported-by: J. McWilliams 1068 1069 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1070 1071commit 82c25879b72f1ba7e9abe00e7bad55add33135cf 1072Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1073Date: Wed Jan 4 11:25:48 2023 -0800 1074 1075 xlibi18n: Update Imake instructions to autoconf equivalent in comments 1076 1077 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1078 1079commit 88399e01be679bfcc9a5e8922ffe2c47f0e56dee 1080Author: Yuxuan Shui <yshuiv7@gmail.com> 1081Date: Tue Jan 3 15:09:28 2023 +0000 1082 1083 Revert "Update XPutBackEvent() to support clients that put back unpadded events" 1084 1085 This reverts commit d6d6cba90215d323567fef13d6565756c9956f60. 1086 1087 The reverted commit intended to fix the problem where an unpadded X 1088 event struct is passed into XPutBackEvent, by creating a padded struct 1089 with _XEventToWire and _XWireToEvent. However, _XWireToEvent updates the 1090 last sequence number in Display, which may cause xlib to complain about 1091 lost sequence numbers. 1092 1093 IMO, the problem that commit tried to solve is a bug in the client 1094 library, and workaround it inside Xlib is bad practice, especially given 1095 the problem it caused. Plus, the offender cited in the original commit 1096 message, freeglut, has already fixed this problem. 1097 1098 Fixes: #176 #174 1099 1100 Signed-off-by: Yuxuan Shui <yshuiv7@gmail.com> 1101 1102commit 48487c4b9028e9eec5371742f51e1d64c2e17296 1103Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1104Date: Thu Dec 15 14:52:20 2022 -0800 1105 1106 libX11 1.8.3 1107 1108 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1109 1110commit d6d6cba90215d323567fef13d6565756c9956f60 1111Author: Keith Packard <keithp@keithp.com> 1112Date: Sun Dec 11 10:32:26 2022 -0800 1113 1114 Update XPutBackEvent() to support clients that put back unpadded events 1115 1116 It seems to be common practice of some X11 clients to pass specific event 1117 types into APIs that take XEvent*. For example, freeglut does: 1118 1119 XConfigureEvent fakeEvent = {0}; 1120 ... 1121 XPutBackEvent(fgDisplay.Display, (XEvent*)&fakeEvent); 1122 1123 This can result in reads overflowing the input event when libX11 does: 1124 1125 XEvent store = *event; 1126 1127 ================================================================= 1128 ==75304==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x00016ee4a8e8 at pc 0x000101c54d14 bp 0x00016ee4a0d0 sp 0x00016ee49888 1129 READ of size 192 at 0x00016ee4a8e8 thread T0 1130 #0 0x101c54d10 in __asan_memcpy+0x1a4 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x3cd10) 1131 #1 0x102848a18 in _XPutBackEvent PutBEvent.c:41 1132 #2 0x1028490a4 in XPutBackEvent PutBEvent.c:84 1133 #3 0x1013295c8 in fgOpenWindow freeglut_window.c:1178 1134 #4 0x101321984 in fgCreateWindow freeglut_structure.c:108 1135 #5 0x10132b138 in glutCreateWindow freeglut_window.c:1551 1136 #6 0x100fb7d94 in main+0x78 (checkeredTriangles:arm64+0x100003d94) 1137 #7 0x197de3e4c (<unknown module>) 1138 1139 Address 0x00016ee4a8e8 is located in stack of thread T0 at offset 840 in frame 1140 #0 0x1013282f8 in fgOpenWindow freeglut_window.c:1063 1141 1142 This frame has 8 object(s): 1143 [32, 40) 'title.addr' 1144 [64, 176) 'winAttr' (line 1066) 1145 [208, 240) 'textProperty' (line 1067) 1146 [272, 352) 'sizeHints' (line 1068) 1147 [384, 440) 'wmHints' (line 1069) 1148 [480, 672) 'eventReturnBuffer' (line 1070) 1149 [736, 740) 'num_FBConfigs' (line 1072) 1150 [752, 840) 'fakeEvent' (line 1074) <== Memory access at offset 840 overflows this variable 1151 1152 This change allows XPutBackEvent() to support such clients without 1153 risk of memory read overflow. 1154 1155 Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com> 1156 Tested-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com> 1157 1158commit 496d9bfeedda89a2211ec665f9214c2693c022f4 1159Author: Jeremy Huddleston Sequoia <jeremyhu@apple.com> 1160Date: Sat Dec 3 16:26:47 2022 -0800 1161 1162 ximcp: Address warning found by UBSan when growing an empty tree 1163 1164 imLcPrs.c:681:52: runtime error: applying zero offset to null pointer 1165 1166 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com> 1167 1168commit 7f7bcd7b6f569e9f70e3ddd134924f178e2596b1 1169Author: Takao Fujiwara <tfujiwar@redhat.com> 1170Date: Tue Dec 6 16:59:56 2022 +0000 1171 1172 nls: consecutive cs number in en_US.UTF-8/XLC_LOCALE 1173 1174commit ba095967349dff583bd0989d0ba78aa334e9bedf 1175Author: Ulrich Sibiller <uli42@gmx.de> 1176Date: Wed Nov 30 23:47:29 2022 +0100 1177 1178 ChkIfEv.c: fix wrong handling of dpy->in_ifevent 1179 1180 Is no longer a bool but a counter. 1181 1182 Signed-off-by: Ulrich Sibiller <uli42@gmx.de> 1183 1184commit 70eaf1174e2809d1a6c43868d53f8cd844ef636a 1185Author: Ulrich Sibiller <uli42@gmx.de> 1186Date: Wed Nov 30 23:42:02 2022 +0100 1187 1188 Indentation fixes around recent dpy->in_ifevent changes 1189 1190 Use the same indentation as the surrounding code. 1191 1192 Signed-off-by: Ulrich Sibiller <uli42@gmx.de> 1193 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1194 1195commit a9e845809bcaae22496bc8aa3ca252b410d5f39b 1196Author: Matthieu Herrb <matthieu@herrb.eu> 1197Date: Fri Nov 11 18:55:23 2022 +0100 1198 1199 Fix 797755 Allow X*IfEvent() to reenter libX11 1200 1201 - the activation logic is reversed 1202 - there is also _XInternalLockDisplay() that needs protection 1203 - I've found cases (in fvwm2) where the callback calls XCheckIfEvent() 1204 recursively. So the flag needs to be a counter. 1205 1206 Reviewed-by: Adam Jackson <ajax@redhat.com> 1207 1208commit bccd787a565d3a88673bfc06574c1939f98d8d72 1209Author: Nia Alarie <nia@NetBSD.org> 1210Date: Thu Nov 10 22:31:47 2022 +0100 1211 1212 Don't use pragma inside a function, it breaks compiling with older GCCs. 1213 1214 XKBBind.c:230: error: #pragma GCC diagnostic not allowed inside functions 1215 1216 Signed-off-by: Thomas Klausner <wiz@gatalith.at> 1217 1218commit 696d19d5db7bcb1c1f582c2b1846520e7e0870cb 1219Author: Oliver <halting@riseup.net> 1220Date: Fri Nov 11 17:04:00 2022 +0000 1221 1222 Add XFreeThreads function. 1223 1224commit b4f24b272c6ef888b6fcfcf80670c196b2e8f755 1225Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1226Date: Thu Nov 10 12:18:17 2022 -0800 1227 1228 libX11 1.8.2 1229 1230 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1231 1232commit 77629ea5e79fa9b1596438bc36cfc7e0c631af43 1233Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1234Date: Thu Nov 3 13:00:50 2022 -0700 1235 1236 README.md: Add 1.8.2 changes 1237 1238 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1239 1240commit d1baf2ec4869dea3732d6fe13ba4935cb6caad01 1241Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1242Date: Thu Nov 3 11:22:37 2022 -0700 1243 1244 Remove NEWS file which only covered 2006 & 2007 releases 1245 1246 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1247 1248commit 1294bfa487bdecfd32621822922fec03203ee188 1249Author: Po Lu <luangruo@yahoo.com> 1250Date: Mon Oct 3 08:44:59 2022 +0800 1251 1252 specs: document change in XIMPreeditCallbacks 1253 1254 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1255 1256commit 1a7e54609ce4dc0f2d3b933502437a413cf3dea8 1257Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1258Date: Thu Nov 3 14:44:22 2022 -0700 1259 1260 man pages: document XCloseIM frees its argument 1261 1262 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1263 1264commit a04b84f0f7bb3a78088756759f4b9b8bb363251e 1265Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1266Date: Thu Nov 3 20:31:31 2022 +0000 1267 1268 Copyright & license cleanup 1269 1270commit 5e41119d9611504ff7e61c6aa4da1df708096801 1271Author: Benno Schulenberg <bensberg@telfort.nl> 1272Date: Mon Oct 31 12:42:51 2022 +0100 1273 1274 nls: reshuffle a few compose sequences, to have similar ones together 1275 1276 It makes more sense to have similar sequences grouped together 1277 than to rigidly follow the order of ascending Unicode codes. 1278 1279commit 65d89342f59cfc2eac9bf8010076b937f22554e8 1280Author: Benno Schulenberg <bensberg@telfort.nl> 1281Date: Mon Oct 31 12:50:11 2022 +0100 1282 1283 nls: remove two compose sequences that use deprecated symbols 1284 1285 The last few occurrences of `leftcaret` and `rightcaret` were replaced 1286 with `less` and `greater` in xkeyboard-config half a year ago. 1287 1288commit 6baccbae5324e7342cb5176cae82f463a6b34eae 1289Author: Benno Schulenberg <bensberg@telfort.nl> 1290Date: Mon Oct 31 12:40:12 2022 +0100 1291 1292 nls: let `<Multi_key> <minus> <underscore>` compose U+2212 (MINUS SIGN) 1293 1294 There was not yet any way to compose the Unicode minus sign, U+2212. 1295 1296commit 79775575418fd6f8ee1c5e5bbe403df4606fb5b6 1297Author: Adam Jackson <ajax@redhat.com> 1298Date: Fri Aug 5 15:19:08 2022 -0400 1299 1300 Allow X*IfEvent() to reenter libX11 1301 1302 The documentation for this family of functions very clearly says not to 1303 call into xlib in your predicate function, but historically single 1304 threaded apps could get away with it just fine, and now that we've 1305 forced thread-safety on the world such apps will now deadlock instead. 1306 That's not an acceptable regression even if the app is technically 1307 broken. This has been reported with XFCE and FVWM, and Motif's 1308 cut-and-paste code has the same bug by inspection, so this does need to 1309 be addressed. 1310 1311 This change nerfs LockDisplay/UnlockDisplay while inside the critical 1312 bit of an IfEvent function. This is still safe in the sense that the 1313 display remains locked and no other thread should be able to change it 1314 from under us, but the loop that scans the event queue might not be 1315 robust against it being modified as a side effect of protocol emitted by 1316 the predicate callback. But that's not new, non-XInitThreads'd xlib 1317 would have the same caveat. 1318 1319 Closes: xorg/lib/libx11#157 1320 1321commit 0d1d65bdd98966f52bcac4077f94827b20b229dd 1322Author: Benno Schulenberg <bensberg@telfort.nl> 1323Date: Fri Oct 21 12:18:25 2022 +0200 1324 1325 nls: change <Compose> <^> <-> to mean superscript minus instead of macron 1326 1327 Several other `<Multi_key> <asciicircum> <symbol>` sequences 1328 produce the superscript equivalent of the given symbol. So, 1329 let `<Multi_key> <asciicircum> <minus>` do the same. 1330 1331 Also, add two other sequences for producing a plain macron, 1332 to compensate a bit the loss of the above sequence. 1333 1334 Additionally, make `<Multi_key> <underscore> <minus>` produce 1335 a subscript minus, for consistency. 1336 1337 This fixes issue #165. 1338 1339 Requested-by: J. McWilliams 1340 1341 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1342 1343commit 310898270787c1a3a6ab50221324f8654dddce8e 1344Author: Jan Tojnar <jtojnar@gmail.com> 1345Date: Thu Oct 20 07:10:58 2022 +0000 1346 1347 nls: Map sr locales to sr_RS compose files 1348 1349 Serbian used sr_YU (Yugoslavia) code in the past. 1350 Then it was succeeded by sr_CS (Serbia and Montenegro). 1351 Finally, it was split into sr_RS (Serbia) and sr_ME (Montenegro). 1352 1353 https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd 1354 introduced the modern sr_RS and sr_ME codes. 1355 1356 Next, https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/40761898692e5063957bfa2518cca3d35b2e354a 1357 added the Serbian compose table but only for the legacy sr_CS entry. 1358 1359 https://gitlab.freedesktop.org/xorg/lib/libx11/-/commit/5cd60398b7787297008b13a848ed3cfbd7ef178d 1360 removed the legacy sr_CS entry, the only one pointing to the correct compose file. 1361 It also renamed the file to sr_RS, but did not update the compose mapping. 1362 1363 Let’s point all Serbian locales to the Compose file again. 1364 1365commit a35d706cd8dc7b498bca2a3fc4a3732047439ed7 1366Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1367Date: Fri Oct 14 13:40:37 2022 -0700 1368 1369 COPYING: remove notice for the removed UIThrStubs.c 1370 1371 Fixes: 701e9e9a ("Use same pthread-stubs as libxcb") 1372 1373 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1374 1375commit 3b1750e982fedce0d61bface68aadcd5b1ea1fae 1376Author: Benno Schulenberg <bensberg@telfort.nl> 1377Date: Fri Oct 14 09:28:42 2022 +0200 1378 1379 nls: delete two compose sequences with an anomalous post-fixed breve 1380 1381 Two years ago, commit b126bfd7fe allowed using also a lowercase `u` 1382 wherever an uppercase `U` was used to represent a breve. But the 1383 commit should have limited itself to only the prefixed uses of `U`, 1384 as that is how most letters with a breve are composed. 1385 1386 Also, group the two compose sequences with an uppercase post-fixed `U` 1387 together with the corresponding other post-fixed sequences. 1388 1389 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1390 1391commit 701e9e9afb88bdc68c84a4611adc533400a3df36 1392Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1393Date: Tue Oct 4 11:24:57 2022 -0700 1394 1395 Use same pthread-stubs as libxcb 1396 1397 Avoid conflicts when libX11 calls libxcb and gets its pthread functions 1398 overriding our ancient stubs. 1399 1400 v2: Keep linking with real threads libraries when thread safety 1401 constructor is enabled. 1402 1403 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1404 1405commit 1d11822601fd24a396b354fa616b04ed3df8b4ef 1406Author: Thomas E. Dickey <dickey@invisible-island.net> 1407Date: Tue Oct 4 18:26:17 2022 -0400 1408 1409 fix a memory leak in XRegisterIMInstantiateCallback 1410 1411 Analysis: 1412 1413 _XimRegisterIMInstantiateCallback() opens an XIM and closes it using 1414 the internal function pointers, but the internal close function does 1415 not free the pointer to the XIM (this would be done in XCloseIM()). 1416 1417 Report/patch: 1418 1419 Date: Mon, 03 Oct 2022 18:47:32 +0800 1420 From: Po Lu <luangruo@yahoo.com> 1421 To: xorg-devel@lists.x.org 1422 Subject: Re: Yet another leak in Xlib 1423 1424 For reference, here's how I'm calling XRegisterIMInstantiateCallback: 1425 1426 XSetLocaleModifiers (""); 1427 XRegisterIMInstantiateCallback (compositor.display, 1428 XrmGetDatabase (compositor.display), 1429 (char *) compositor.resource_name, 1430 (char *) compositor.app_name, 1431 IMInstantiateCallback, NULL); 1432 1433 and XMODIFIERS is: 1434 1435 @im=ibus 1436 1437 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 1438 1439commit 1f8fd7ff1cf688ec1d3b34397c6d58110bb0cae5 1440Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1441Date: Mon Sep 26 16:01:40 2022 -0700 1442 1443 If thread-safety-constructor is enabled, link against pthreads not stubs 1444 1445 Only really makes a difference if pthreads is not in libc. 1446 1447 Fixes: #162 1448 1449 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1450 Reviewed-by: Adam Jackson <ajax@redhat.com> 1451 1452commit bfe5d2dd98f705438b0cdddbbf9c12882da89bae 1453Author: Thomas E. Dickey <dickey@invisible-island.net> 1454Date: Sun Sep 11 11:44:41 2022 -0400 1455 1456 reduce compiler warnings with casts (no object change) 1457 1458 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 1459 1460commit 926f2454bca9a7a7d1f4cf0648b0c193f5cef644 1461Author: Thomas E. Dickey <dickey@invisible-island.net> 1462Date: Sun Sep 11 10:35:31 2022 -0400 1463 1464 use casts to reduce compiler warnings (no object change) 1465 1466 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 1467 1468commit dc553ff908f10832c71dc2b5ba5af7e68589218e 1469Author: Mike FABIAN <mfabian@redhat.com> 1470Date: Fri Jul 22 11:39:01 2022 +0200 1471 1472 Remove KOI8-R character set from en_US.UTF-8/XLC_LOCALE 1473 1474 https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/master/src/xlibi18n/lcCT.c#L58 1475 has no escape sequence for KOI8-R. 1476 That makes Xutf8TextListToTextProperty() sometimes fail when trying to 1477 convert to COMPOUND_TEXT and when the KOI8-R charset is tried. 1478 1479 This is the cause for this bug: 1480 1481 https://github.com/ibus/ibus/issues/2422 1482 1483 Removing the KOI8-R charset entry fixes the problem. 1484 1485 This commit also fixes a few wrong texts in comments. 1486 1487 Signed-off-by: Mike FABIAN <mfabian@redhat.com> 1488 1489commit 1c04ac514196db5e2255d99635fb6e3c36be330d 1490Author: Adam Sampson <ats@offog.org> 1491Date: Sun Jul 4 23:02:04 2021 +0100 1492 1493 xkb: Fix off-by-one error in XKeycodeToKeysym 1494 1495 The code here that made indexes greater than 3 refer to XKB symbol 1496 groups had an off-by-one error, so it would always leave out the symbol 1497 that should have been at index 4. Rewrite the code to fix this and 1498 simplify the logic a bit. 1499 1500 Signed-off-by: Adam Sampson <ats@offog.org> 1501 1502commit ddaacd219609104f138006db6a8f708226db3874 1503Author: Thomas E. Dickey <dickey@invisible-island.net> 1504Date: Fri Sep 9 20:17:35 2022 -0400 1505 1506 reduce compiler warnings for macros BufAlloc, Data and Data32 using casts 1507 1508 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 1509 1510commit 62c3337d89d31e0d3ed807004e73ad711fad3342 1511Author: Kirill Chibisov <contact@kchibisov.com> 1512Date: Thu Sep 8 22:50:30 2022 +0000 1513 1514 ximcp/imRm.c: allow XNSpotLocation with OnTheSpot 1515 1516commit 1272879074ad9612314858a75cf844783c1963b2 1517Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1518Date: Thu Jul 28 17:30:21 2022 -0700 1519 1520 gitlab CI: stop requiring Signed-off-by in commits 1521 1522 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1523 1524commit 38033b073f393564acf6a04501927002b2b4aa86 1525Author: David H. Gutteridge <david@gutteridge.ca> 1526Date: Mon Jul 25 19:30:19 2022 -0400 1527 1528 configure.ac: report thread safety constructor build status 1529 1530 Signed-off-by: David H. Gutteridge <david@gutteridge.ca> 1531 1532commit 9a90b3a63d02da92dc8692ed3f6269d3e0c7c8cc 1533Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1534Date: Sun Jul 10 11:55:11 2022 -0700 1535 1536 doc: Document possible return values for XkbGetKeyboard() 1537 1538 Fixes: #160 1539 1540 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1541 1542commit 0beb0cf46ae48b0d4322ff8a0221f47bdb7d2296 1543Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1544Date: Sun Jul 10 11:42:19 2022 -0700 1545 1546 xkb: set num_keys when filling in keys in _XkbReadGetNamesReply() 1547 1548 Fixes: #160 1549 1550 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1551 1552commit 487e9489d1e086cc62c53045206ab6d6d9d02031 1553Author: Jon Turney <jon.turney@dronecode.org.uk> 1554Date: Wed Jun 22 16:15:37 2022 +0100 1555 1556 Fix Win32 build with -fno-common 1557 1558 /work/xorg/lib/libX11/src/XlibInt.c:1968: multiple definition of `_Xdebug_p'; .libs/globals.o:globals.c:(.bss+0xc): first defined here 1559 1560 Avoid redundant definition of _Xdebug_p in globals.c (which is under the 1561 influence of _Xdebug being #defined to _Xdebug_p. 1562 1563 This appears to be an ancient hack to work around data exports resolving 1564 to the address of the import stub, not the import. (See [1]). 1565 1566 (This is probably no longer needed or can be done in a better way, as 1567 per the discussion under --enable-auto-import in the ld manpage.) 1568 1569 [1] https://cygwin.com/pipermail/cygwin-xfree/2001-May/004606.html 1570 1571 Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk> 1572 1573commit 35ad99fad2f65ec2c69bd84bdeb8138329092b57 1574Author: Benno Schulenberg <bensberg@telfort.nl> 1575Date: Fri Jul 15 11:29:46 2022 +0200 1576 1577 nls: add four sequences for the Samogitian E with dot above and macron 1578 1579 These sequences each produce two code points: the E-with-dot-above and 1580 a combining macron. The XIM input method is required for this to work. 1581 1582 (Also add a missing comment for a Unicode block.) 1583 1584 This fixes issue #54. 1585 1586 Requested-by: Arns Udovīčė 1587 1588 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1589 1590commit 2732dd20cc2694dd5c32fef292e1f48626011032 1591Author: Benno Schulenberg <bensberg@telfort.nl> 1592Date: Wed Jul 6 16:44:50 2022 +0200 1593 1594 nls: delete eight sequences that pointlessly mix upper and lower case 1595 1596 Typing a compose sequence requires some care -- surely the user is able 1597 to either keep holding the Shift key or not touch it at all while typing 1598 the sequence. Also, compose sequences are not an infinite resource AND 1599 take up space and time -- defining redundant ones is a waste. 1600 1601 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1602 1603commit ea3cc474e1acfed47b9fb856ea151008fa6cfd30 1604Author: Benno Schulenberg <bensberg@telfort.nl> 1605Date: Wed Jul 6 16:36:48 2022 +0200 1606 1607 nls: add compose sequences for the double-struck capitals ℕ ℤ ℚ ℝ ℂ 1608 1609 This allows the user to type the symbols for the five number systems. 1610 1611 This fixes the reasonable part of issue #159. 1612 1613 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1614 1615commit 95a03cde434c447fdb41d8cccbbdd41f20750f32 1616Author: Benno Schulenberg <bensberg@telfort.nl> 1617Date: Thu Jul 14 12:00:45 2022 +0200 1618 1619 nls: delete some pointless and wrong compose sequences meant for Bépo 1620 1621 Fourteen years ago, commit 7302984642 added some four hundred compose 1622 sequences for the benefit of the French Bépo layout. But among these 1623 four hundred there are several that use symbols that are not available 1624 in the Bépo layout and are thus impossible to type. Some of the used 1625 symbols, like Ahook, Ehook and Ohook, are not even present in *any* 1626 layout, making these sequences a dead weight in the Compose file. 1627 1628 The Amacron and Omacron are available only in the Latvian, Hawaiian, 1629 and Maori layouts, and the Omacron also in the Silesian layout. But 1630 the Latvian layouts and the Hawaiian do not contain any dead keys. 1631 Only in the Maori and Silesian layouts these sequences with Amacron 1632 and Omacron could be typed, but that was not why they were added. 1633 1634 More importantly, as James Cloos noted in issue #54, sequences like 1635 `<dead_abovedot> <amacron>` for generating `ǡ` (that is: the macron 1636 above the dot) are questionable, as in compose sequences generally 1637 the first accent typed is the uppermost in the composed character. 1638 1639 Reference: 1640 https://gitlab.freedesktop.org/xorg/lib/libx11/-/issues/54#note_17321 1641 1642 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1643 1644commit 9702b836862da6838d725f555e7a1d83faf51740 1645Author: Matthieu Herrb <matthieu@herrb.eu> 1646Date: Fri Jul 1 11:13:13 2022 +0200 1647 1648 Mention that the predicate function is called with the display lock 1649 1650 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 1651 1652commit a5d00f889fadad3f5b11e53fd5c04cf5343fa8a2 1653Author: Benno Schulenberg <bensberg@telfort.nl> 1654Date: Fri Jun 24 16:20:52 2022 +0200 1655 1656 docs: replace three placeholders with something that makes sense 1657 1658 Also, uncapitalize two arguments, to match the style of all others. 1659 1660 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1661 1662commit 9ff8502040e94cb172426d59df722cefa752a401 1663Author: Benno Schulenberg <bensberg@telfort.nl> 1664Date: Fri Jun 24 16:07:22 2022 +0200 1665 1666 docs: remove the unsightly dashes from the overviews of arguments 1667 1668 Seventeen months ago, commits 78027fdb7a and 4f15cfc645 removed 1669 these dashes from two of the man pages. Now, remove them all. 1670 They are unhelpful and just make one wonder why they are there 1671 (probably to function as improvised bullet points). 1672 1673 Also remove four leading spaces and a trailing comma. 1674 1675 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1676 1677commit a7968c47e35703446d67d070234fd9adf07f34b0 1678Author: Benno Schulenberg <bensberg@telfort.nl> 1679Date: Thu Jun 23 15:48:05 2022 +0200 1680 1681 docs: hard-wrap some items so that Table 2 fits within 80 characters 1682 1683 (Table 1 hard-wraps the first-column items in the same way.) 1684 1685 Also, correct the formatting of the subsequent paragraph. 1686 1687 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1688 1689commit c601c779ca8f46dc4335180ec2829851fb8bbc13 1690Author: Benno Schulenberg <bensberg@telfort.nl> 1691Date: Wed Jun 15 10:55:38 2022 +0200 1692 1693 nls: delete compose sequences that mix top-row digits with numpad digits 1694 1695 Compose sequences for circled numbers, like ⑫ or ㉑, are nice to have, 1696 but allowing them to be composed by typing one digit on the top row and 1697 the other on the numerical keypad (or the other way around) is over the 1698 top. Remove these absurd sequences. Keep only the sequences where both 1699 digits are either on the top row or on the numerical keypad. 1700 1701 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1702 1703commit 3a30ada60c5217ada37b143b541c8e6f6284c7fa 1704Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1705Date: Wed Jun 8 17:30:14 2022 -0700 1706 1707 libX11 1.8.1 1708 1709 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1710 1711commit 70f7403fd3bf362fc50b8de4a2e26300f757c68e 1712Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1713Date: Tue May 3 14:45:05 2022 -0700 1714 1715 configure.ac: Fix --enable-thread-safety-constructor 1716 1717 Prior to this, --enable-thread-safety-constructor would disable it, 1718 while --disable worked as expected, and no option left it enabled as 1719 expected by default. This also fixes the --help text to be displayed. 1720 1721 Fixes: afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9 1722 Reported-by: @igor.v.kovalenko 1723 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1724 Reviewed-by: Adam Jackson <ajax@redhat.com> 1725 1726commit d0da5a1e0fb214ff911ecff86d35f6673ff70023 1727Author: Adam Jackson <ajax@redhat.com> 1728Date: Fri Apr 29 11:27:44 2022 -0400 1729 1730 libX11 1.8 1731 1732 Minor number bump to reflect the thread safety default change. 1733 1734 Signed-off-by: Adam Jackson <ajax@redhat.com> 1735 1736commit afcdb6fb0045c6186aa83d9298f327a7ec1b2cb9 1737Author: Adam Jackson <ajax@redhat.com> 1738Date: Tue Mar 22 18:24:29 2022 -0400 1739 1740 global: call XInitThreads() from the library's constructor 1741 1742 There is really no point in not being thread safe, I measured, all you 1743 can see happen is noop performance gets like twice as slow and you have 1744 thread safety bugs. And we're using xcb as the transport which means we 1745 should expect threads in our clients anyway. Just do it. 1746 1747 This assumes your compiler understands __attribute__((constructor)). If 1748 this is not your compiler, you can disable this with the appropriate 1749 configure flag, but be aware you're asking for bugs. 1750 1751 Signed-off-by: Adam Jackson <ajax@redhat.com> 1752 1753commit 1ab48f3cc966751d86e434808180beb2dc76c4fa 1754Author: Benno Schulenberg <bensberg@telfort.nl> 1755Date: Sun Apr 24 15:06:45 2022 +0200 1756 1757 docs: add release notes for versions 1.7.3, 1.7.4, and 1.7.5 1758 1759 They were forgotten when those releases were made. 1760 1761 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1762 1763commit 35d1513bc7fdd3ac6f5807feb601efc34ac19163 1764Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1765Date: Sat Apr 16 13:04:23 2022 -0700 1766 1767 XkbOpenDisplay: Add _Xconst qualifier to display name argument 1768 1769 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1770 1771commit c88ceaad4aaa0439e5cba13c1f5737ad54566e0b 1772Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1773Date: Sat Apr 16 12:59:39 2022 -0700 1774 1775 XKBgeom.h: Add _Xconst qualifier to char * arguments in functions 1776 1777 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1778 1779commit 0ce4425826074b12a58ced432f9d6fb346474737 1780Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1781Date: Sat Apr 16 12:20:51 2022 -0700 1782 1783 XKBgeom.h: Remove XKB_IN_SERVER section 1784 1785 The Xserver made its own copy of this file in 2008, and the API's are 1786 no longer the same between the server and client forks. 1787 1788 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1789 1790commit be8c01946ab1a1d4b79fdc5358541d630f14dc0a 1791Author: Benno Schulenberg <bensberg@telfort.nl> 1792Date: Wed Apr 20 09:29:09 2022 +0200 1793 1794 lcCT: use the correct index for checking the second byte 1795 1796 (Not that it will make any difference, as the checking of these 1797 high bits looks like an excess of precaution.) 1798 1799 This fixes issue #134. 1800 1801 Reported-by: Rafał Mikrut 1802 1803 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1804 1805commit 488b156fe2cc8aca6946a49236ec7b7698fceda4 1806Author: Benno Schulenberg <bensberg@telfort.nl> 1807Date: Fri Feb 25 16:35:46 2022 +0100 1808 1809 nls: make the Ethiopian compose sequences actually work 1810 1811 In the Ethiopian keyboard layout, the dead vowel keys do not produce <e> 1812 and <u> and so on, but instead produce <U+FE69> and <U+FE75> and so on, 1813 so the compose sequences should use those latter code points. 1814 1815 Also, include the basic compose sequences from en_US.UTF-8/Compose, 1816 so that, when switching to a different layout in the Ethiopian locale, 1817 all the usual compose sequences work too. 1818 1819 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1820 1821commit e6f98a3dbd82928e10adaf9d034dc574af1de6e5 1822Author: Benno Schulenberg <bensberg@telfort.nl> 1823Date: Mon Apr 18 17:24:55 2022 +0200 1824 1825 nls: add Multi-key sequences for abovedot, diaeresis, and ogonek 1826 1827 These accents by themselves could only be produced when one had a 1828 dead key for them, not with the help of the Multi key. 1829 1830 [Note that the sequences <dead_acute> <space> for apostrophe (') 1831 and <dead_diaeresis> <space> for double quote (") are anomalies, 1832 as normally <dead_accent> <space> produces the accent itself.] 1833 1834 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1835 1836commit d323704e7c57f9ac7fdfea3dc9fa73bf943e72d3 1837Author: Benno Schulenberg <bensberg@telfort.nl> 1838Date: Mon Apr 18 16:57:35 2022 +0200 1839 1840 nls: remove eight pointless self-producing compose sequences 1841 1842 Compose sequences are meant to produce certain symbols by combining 1843 certain different symbols, not to produce a symbol with the help of 1844 the symbol itself. 1845 1846 This fixes issue #59. 1847 1848 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1849 1850commit 3b62c0322c112c03597d850197dbce22d1830bb7 1851Author: Benno Schulenberg <bensberg@telfort.nl> 1852Date: Fri Feb 25 14:12:02 2022 +0100 1853 1854 nls: remove misplaced sequences from Ethiopian and Greek compose files 1855 1856 The Khmer digraphs and Arabic ligatures have nothing to do with 1857 Amharic or Greek. 1858 1859 (Also rewrap a comment and correct two others.) 1860 1861 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1862 1863commit 479af78f2e5caaf9b98a35ae5ff4ccdd2f040289 1864Author: Benno Schulenberg <bensberg@telfort.nl> 1865Date: Fri Feb 25 14:05:35 2022 +0100 1866 1867 nls: remove redundant Khmer digraphs, already defined in included file 1868 1869 This fixes issue #160. 1870 1871 Reported-by: Mike Fabian 1872 1873 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1874 1875commit 6d7d08726f4b0c517041842b27cd7e66e8f371eb 1876Author: Alan Coopersmith <alan.coopersmith@oracle.com> 1877Date: Sun Apr 10 14:51:55 2022 -0700 1878 1879 XkbUpdateKeyTypeVirtualMods: always initialize mask 1880 1881 XkbVirtualModsToReal should only fail to set mask if the server does 1882 not support XKB, but it still made Oracle Parfait complain: 1883 1884 Error: Uninitialised memory 1885 Uninitialised memory variable [uninitialised-mem-var] (CWE 457): 1886 Possible access to uninitialised memory referenced by variable 'mask' 1887 at line 863 of lib/libX11/src/xkb/XKBMisc.c in function 'XkbUpdateKeyTypeVirtualMods'. 1888 Path in callee avoiding write at line 862 1889 mask allocated at line 860 1890 1891 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 1892 1893commit 9ac6859c20be2fc5e70c2908de60c6e466ec04e1 1894Author: Matt Turner <mattst88@gmail.com> 1895Date: Sun Apr 3 10:15:30 2022 -0700 1896 1897 libX11 1.7.5 1898 1899 Signed-off-by: Matt Turner <mattst88@gmail.com> 1900 1901commit 76d1cc3c1ce943c6ff81dc8c62a1d1b30fabf02e 1902Author: Simon McVittie <smcv@debian.org> 1903Date: Sun Apr 3 14:23:36 2022 +0100 1904 1905 Don't try to destroy NULL condition variables 1906 1907 This avoids a segfault during error-unwinding if an invalid display name 1908 is passed to XOpenDisplay(). 1909 1910 Fixes: 8a368d80 "Fix two memory leaks in _XFreeX11XCBStructure()" 1911 Resolves: #155 1912 Signed-off-by: Simon McVittie <smcv@debian.org> 1913 1914commit 80b30d12519ac54fc5b2b05500521d9c8269be9c 1915Author: Matt Turner <mattst88@gmail.com> 1916Date: Wed Mar 30 13:54:13 2022 -0700 1917 1918 libX11 1.7.4 1919 1920 Signed-off-by: Matt Turner <mattst88@gmail.com> 1921 1922commit 8a368d808fec166b5fb3dfe6312aab22c7ee20af 1923Author: Hodong <hodong@yozmos.com> 1924Date: Thu Jan 20 00:57:41 2022 +0900 1925 1926 Fix two memory leaks in _XFreeX11XCBStructure() 1927 1928 Even when XCloseDisplay() was called, some memory was leaked. 1929 1930 XCloseDisplay() calls _XFreeDisplayStructure(), which calls 1931 _XFreeX11XCBStructure(). 1932 1933 However, _XFreeX11XCBStructure() did not destroy the condition variables, 1934 resulting in the leaking of some 40 bytes. 1935 1936 Signed-off-by: Hodong <hodong@yozmos.com> 1937 1938commit da97120f2322882a17f14e5d7da00c1e772679e8 1939Author: Adam Jackson <ajax@redhat.com> 1940Date: Tue Mar 22 14:32:49 2022 -0400 1941 1942 xcb: Clarify the XInitThreads error message 1943 1944 XInitThreads has been called if _Xglobal_lock != NULL, we may as well 1945 check that before printing a misleading error message. 1946 1947 Signed-off-by: Adam Jackson <ajax@redhat.com> 1948 1949commit 257927c51b08242aa5bf239346717fc817b2b286 1950Author: Giovanni Mascellani <gmascellani@codeweavers.com> 1951Date: Fri Feb 4 10:49:25 2022 +0100 1952 1953 xcb_io: Allow jumps backwards when widening the request number. 1954 1955 Request numbers are not always seen in the numeric order by widen(), 1956 for example due to Mesa directly calling _XError(). When this happens, 1957 widen() adds 2^32 to the reported widened number, triggering failed 1958 assertions and bad behavior. 1959 1960 With this commit, wrapping of the lower dword is detected in a more 1961 robust way, by requiring that a skip of at least 2^31 is seen. 1962 1963 This fixes issue #152. 1964 1965 Signed-off-by: Giovanni Mascellani <gmascellani@codeweavers.com> 1966 1967commit 918063298cb893bee98040c9dca45ccdb2864773 1968Author: Benno Schulenberg <bensberg@telfort.nl> 1969Date: Thu Mar 3 12:00:45 2022 +0100 1970 1971 nls: add a compose sequence for the peace symbol 1972 1973 As the thing in the circle looks a bit like an upside-down Y, 1974 use <O> <Y> as the sequence, similar to <O> <A> for anarchism. 1975 1976 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1977 1978commit 4f9eb32709dca705f02ca771bbd47d6c08aba5e0 1979Author: Benno Schulenberg <bensberg@telfort.nl> 1980Date: Fri Feb 25 12:45:06 2022 +0100 1981 1982 nls: remove obsolete locale aliases with uppercase in the language code 1983 1984 (These aliases may have been useful on some systems twenty years ago, 1985 but nowadays all should have settled on lowercase language code plus 1986 uppercase country code.) 1987 1988 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1989 1990commit ba0d3b437deb2212bc7fc7c6f1c603fa6ced443c 1991Author: Benno Schulenberg <bensberg@telfort.nl> 1992Date: Fri Feb 25 12:35:28 2022 +0100 1993 1994 nls: map eo.UTF-8 directly to en_US.UTF-8 instead of using a relay alias 1995 1996 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 1997 1998commit cfddb79981f23650feb0f9a0f4442c9cad584b33 1999Author: Benno Schulenberg <bensberg@telfort.nl> 2000Date: Fri Feb 25 12:12:05 2022 +0100 2001 2002 nls: remove the mistaken eo_EO names -- EO is not a valid country code 2003 2004 (Debian has had several of these lines commented out for years.) 2005 2006 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2007 2008commit eb5885245a1ba6955d058a03e31894c23b9f5490 2009Author: iyzana <iyzana@protonmail.com> 2010Date: Mon Feb 21 13:20:23 2022 +0100 2011 2012 nls: add locale alias from eo.UTF-8 to eo_XX.UTF-8 2013 2014 There is no matching locale in the locale.dir and compose.dir files for 2015 the locale eo.UTF-8. Setting it as the locale resulted in compose files 2016 not being loaded. 2017 2018 Signed-off-by: iyzana <iyzana@protonmail.com> 2019 2020commit 24df23dff9d8c1cf5e918a5f040b852bdf2cfb84 2021Author: Benno Schulenberg <bensberg@telfort.nl> 2022Date: Thu Feb 3 17:19:05 2022 +0100 2023 2024 nls: group the compose sequences for tilde, degree, and cedilla together 2025 2026 Also, correct the comment for the double quote character ("), and move 2027 the soft hyphen and the interrobangs to a special punctuation group. 2028 2029 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2030 2031commit c0191f020f94907770937bc8ae67ea8acd7d63be 2032Author: Benno Schulenberg <bensberg@telfort.nl> 2033Date: Thu Feb 3 16:33:05 2022 +0100 2034 2035 nls: use shape-based compose sequences for hammer-and-sickle instead 2036 2037 Using a misspelled abbreviation (CCCP) of an obsolete country's name 2038 to compose a fairly general communist symbol does not seem right. 2039 2040 https://en.wikipedia.org/wiki/Hammer_and_sickle 2041 2042 Many compose sequences instead combine characters that together look 2043 similar to the target character. Do the same here: question mark (?) 2044 plus backslash (\) look a bit like ☭ . 2045 2046 In the bargain, this fixes issue #63. 2047 2048 Reported-by: Marc Mezzarobba 2049 2050 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2051 2052commit 582dc6f89e1f9288710a55cb2b8fbf2af99d7616 2053Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2054Date: Sun Feb 20 11:49:42 2022 -0800 2055 2056 gitlab CI: enable Static Application Security Testing (SAST) 2057 2058 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2059 2060commit 336e8f72559b489396c5cd503c4586ef800c2a8f 2061Author: Benno Schulenberg <bensberg@telfort.nl> 2062Date: Sun Feb 20 15:06:02 2022 +0100 2063 2064 nls: add compose sequences for leftward, upward and downward double arrow 2065 2066 These sequences follow the logic of: the arrow shaft (=) first, the 2067 arrow head (< or > or ^ or v) second. This is so because the sequence 2068 <= is already taken for ≤, and there is no obvious ASCII character for 2069 a double vertical line, so we have to make do with a horizontal one. 2070 2071 Since commit 6101b967b6 from eight years ago, there is a compose sequence 2072 for the rightward double arrow, but not for the other three directions. 2073 2074 This fixes issue #138. 2075 2076 Requested-by: Mélanie Chauvel 2077 2078 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2079 2080commit 623b77d4f30b47258a40f89262e5aa5d25e95fa7 2081Author: Benno Schulenberg <bensberg@telfort.nl> 2082Date: Mon Feb 14 11:33:25 2022 +0100 2083 2084 imDefLkup: verify that a pointer isn't NULL before using it 2085 2086 It is possible for _XimICOfXICID() to return NULL, so it is necessary 2087 to check this isn't actually the case before dereferencing the pointer. 2088 All other callers of _XimICOfXICID() do this check too. 2089 2090 (The check itself is ugly, but it follows the style of the code in the 2091 rest of the module.) 2092 2093 Fixes issue #45. 2094 2095 Reported-by: Bhavi Dhingra 2096 2097 Original-patch-by: Bhavi Dhingra 2098 2099 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2100 2101commit b83938f7f97aa6210b00835002e6e7d614b2374a 2102Author: Christopher Chavez <chrischavez@gmx.us> 2103Date: Mon Feb 14 01:00:45 2022 +0000 2104 2105 Xlib.h: fix spelling in comment 2106 2107 Signed-off-by: Christopher Chavez <chrischavez@gmx.us> 2108 2109commit 4791055682552b4cc8e775479700fb06980d0b59 2110Author: Christopher Chavez <chrischavez@gmx.us> 2111Date: Sun Feb 13 05:26:14 2022 +0000 2112 2113 Xlib.h: spelling fix in comment 2114 2115commit 9304e645bdcdd3a71ff449a1157c4815e6304d7a 2116Author: Denis Drakhnia <numas13@gmail.com> 2117Date: Wed Feb 2 11:54:43 2022 +0200 2118 2119 nls: Add ru_UA.utf8 locale alias. 2120 2121 Signed-off-by: Denis Drakhnia <numas13@gmail.com> 2122 2123commit 9ef2551f0c201f9163b66d17189a98076b5be07a 2124Author: Benno Schulenberg <bensberg@telfort.nl> 2125Date: Wed Feb 2 16:50:35 2022 +0100 2126 2127 nls: do some manual adjustment of whitespace in the compose sequences 2128 2129 This achieves a more consistent vertical alignment (per group) of the 2130 target character and the trailing comments. 2131 2132 (Most blocks were aligned in some fashion, except the "Greek Extended" 2133 block, which has lines of greatly varying lengths.) 2134 2135 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2136 2137commit 0dad289ead6eb3412027c9365192441cc539d747 2138Author: Benno Schulenberg <bensberg@telfort.nl> 2139Date: Wed Feb 2 14:08:22 2022 +0100 2140 2141 nls: among the compose sequences, replace all runs of spaces with tabs 2142 2143 It was done with this script: 2144 2145 cat nls/en_US.UTF-8/Compose.pre | 2146 sed 's!\(^<.\{6,13\}>\)\s*:!\1\t\t\t\t:!' | 2147 sed 's!\(^<.\{14,21\}>\)\s*:!\1\t\t\t:!' | 2148 sed 's!\(^<.\{22,29\}>\)\s*:!\1\t\t:!' | 2149 sed 's!\(^<.\{30,37\}>\)\s*:!\1\t:!' | 2150 sed 's!\(: ".*"\)\s*!\1\t!' | 2151 sed 's!\(\s*# \)! # !' > trimmed && 2152 mv trimmed nls/en_US.UTF-8/Compose.pre 2153 2154 This saves 37 kilobytes of whitespace. 2155 2156 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2157 2158commit fc79e39c236501357e1fdbe8964882864eeef5d3 2159Author: Benno Schulenberg <bensberg@telfort.nl> 2160Date: Wed Feb 2 14:06:29 2022 +0100 2161 2162 docs: add missing word, wrap line, fix typo, and trim an excess space 2163 2164 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2165 2166commit 93446455c46dfc019984ab822ffa662cf6e2949c 2167Author: Benno Schulenberg <bensberg@telfort.nl> 2168Date: Thu Jan 27 15:01:18 2022 +0100 2169 2170 KeyBind: reshuffle two ifs into the general order of ascending codes 2171 2172 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2173 2174commit 93f15f5f29378ec4cc1dd011b155104e9a52b058 2175Author: Benno Schulenberg <bensberg@telfort.nl> 2176Date: Thu Jan 27 14:27:41 2022 +0100 2177 2178 KeyBind: slightly speed up some case conversions by adding missing returns 2179 2180 This avoids pointlessly tumbling through several more ifs when the 2181 conversion has aleady been done. 2182 2183 Also remove two redundant conditions (as lower codes have already 2184 been handled) and fold two other conditions together. 2185 2186 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2187 2188commit 317c7abbcc848011393d1336b9d020af8cabb000 2189Author: Benno Schulenberg <bensberg@telfort.nl> 2190Date: Thu Jan 27 13:54:29 2022 +0100 2191 2192 KeyBind: update the Greek case-conversion tables to Unicode Data 14.0 2193 2194 A few symbols, like Heta and Sampi and dotted lunate Sigma, have been 2195 assigned code points since Unicode Data 4.0, and need their entries. 2196 2197 This fixes issue #132. 2198 2199 Reported-by: Ray Vine 2200 2201 Original-patch-by: Ray Vine 2202 2203 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2204 2205commit 402b843fa72166d9ceefd887f5cea4728cbc6995 2206Author: Benno Schulenberg <bensberg@telfort.nl> 2207Date: Wed Jan 26 17:20:28 2022 +0100 2208 2209 remove a commented-out code fragment, and remove a stray blank line 2210 2211 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2212 2213commit fbbc3cf8b8b03f74f778772d4703fc4fec242282 2214Author: Benno Schulenberg <bensberg@telfort.nl> 2215Date: Wed Jan 26 16:55:50 2022 +0100 2216 2217 XKBCvt: remove an if that never gets triggered 2218 2219 The 'if' at the beginning of _XkbHandleSpecialSym() allows only symbols 2220 from the numeric keypad and some control symbols to pass -- XK_hyphen 2221 is not among them, so the check for XK_hyphen in the later 'if' will 2222 always be false. 2223 2224 (The corresponding conversion in _XTranslateKeySym() in KeyBind.c was 2225 removed in March 1994.) 2226 2227 (Also, several keyboard layouts nowadays contain the 'hyphen' symbol, 2228 allowing the user to enter soft hyphens into a document. So we really 2229 don't want to remap this symbol.) 2230 2231 Fixes issue #48. 2232 2233 Reported-by: Bhavi Dhingra 2234 2235 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2236 2237commit 6786744f636b3fc545157a13cb2f6169b5be9fa5 2238Author: Benno Schulenberg <bensberg@telfort.nl> 2239Date: Sun Jan 23 17:12:03 2022 +0100 2240 2241 nls: add the es_CU locale name, for Cuba 2242 2243 It was added to glibc ten years ago. 2244 2245 This fixes issue #57. 2246 2247 Reported-by: Dominique Michel 2248 2249 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2250 2251commit d241d8af7f24e95949c2cf7ce791e35df7665e98 2252Author: Benno Schulenberg <bensberg@telfort.nl> 2253Date: Sun Jan 23 16:44:03 2022 +0100 2254 2255 nls: remove some obsolete Norwegian and Yugoslavian locale names 2256 2257 They have not existed in glibc for more than fifteen years. 2258 2259 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2260 2261commit 32dd7c03e92e01015744867b9b3148c250cc159a 2262Author: Benno Schulenberg <bensberg@telfort.nl> 2263Date: Sat Jan 22 17:56:32 2022 +0100 2264 2265 nls: put some entries in their alphabetical order, and adjust alignment 2266 2267 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2268 2269commit 46a066f25103bb909259baea25de8dd5c4a2b338 2270Author: Benno Schulenberg <bensberg@telfort.nl> 2271Date: Sat Jan 22 18:00:20 2022 +0100 2272 2273 nls: fix a typo (be_BG => bg_BG) and drop an obsolete comment 2274 2275 The typo was only partially fixed in commit 22a5255b80 sixteen years ago, 2276 and the internet shows that there was at least one user frustrated that 2277 bg_BG.utf8 didn't work but bg_BG.UTF-8 did. 2278 2279 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2280 2281commit 84efc0aa598c8f7d6dd9a49e7450e85c8bc4773b 2282Author: Benno Schulenberg <bensberg@telfort.nl> 2283Date: Wed Jan 19 15:32:15 2022 +0100 2284 2285 nls: remove the compose sequences with the deviant KP_Space symbol 2286 2287 Using this KP_Space symbol as a stand-in for the symbol 2 doesn't make 2288 sense. It looks like a mistake, or as if someone had a broken keyboard 2289 and used KP_Space as a substitute for 2. 2290 2291 Also, no keyboard layout in the last fifteen years has contained the 2292 KP_Space symbol, so I don't see how anyone could type it. 2293 2294 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2295 2296commit b76e661ca223489da9027b0ab1ca1586efd0a280 2297Author: Benno Schulenberg <bensberg@telfort.nl> 2298Date: Wed Jan 19 13:49:56 2022 +0100 2299 2300 nls: move the block of Jamo to its logical place (ascending Unicodes) 2301 2302 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2303 2304commit 21f82fb9c69f8fdbf0bdc1aac6f8c11c2c6d55d1 2305Author: Benno Schulenberg <bensberg@telfort.nl> 2306Date: Wed Jan 19 13:40:56 2022 +0100 2307 2308 nls: delete some ineffective compose sequences (that were commented out) 2309 2310 They have never worked; there is no point in keeping them in the file. 2311 2312 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2313 2314commit 6cc31eafa8095d2117ff1906d7c14c1c01173248 2315Author: Benno Schulenberg <bensberg@telfort.nl> 2316Date: Wed Jan 19 12:18:21 2022 +0100 2317 2318 nls: group some compose sequences better, and add some comments 2319 2320 Also, align a few trailing comments more consistently, and change 2321 some comments so they will be excluded from the installed file -- 2322 there is no need for those comments there. 2323 2324 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2325 2326commit d60ede7843612f3c2967f83d629bf8272ba09ff1 2327Author: ArenaL5 <arenal5@protonmail.com> 2328Date: Thu Jan 13 01:13:56 2022 +0100 2329 2330 nls: add a compose sequence for the copyleft symbol 2331 2332 Probably not used anywhere yet, let alone seriously, as stated in 2333 https://en.wikipedia.org/wiki/Copyleft#Symbol 2334 2335 Signed-off-by: ArenaL5 <arenal5@protonmail.com> 2336 2337commit df4dcdf9125c595f6df3a302356647cb8df38553 2338Author: Benno Schulenberg <bensberg@telfort.nl> 2339Date: Mon Jan 17 17:31:08 2022 +0100 2340 2341 nls: remove redundant compose sequences with deprecated Hebrew symbols 2342 2343 The compose sequences with the valid Hebrew key symbols are right there. 2344 2345 Also, no keyboard layout in the past seventeen years has contained these 2346 deprecated symbols on any key. 2347 2348 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2349 2350commit 4268472bd11a07a799716fb5833ea46cad3dd479 2351Author: Benno Schulenberg <bensberg@telfort.nl> 2352Date: Sun Jan 16 17:51:29 2022 +0100 2353 2354 nls: remove redundant compose sequences with the deprecated 'underbar' 2355 2356 The key symbol 'underbar' is equivalent to 'underscore' -- the mirror 2357 compose sequences with the latter symbol continue to exist. 2358 2359 Also, no keyboard layout in the past fifteen years has contained the 2360 symbol 'underbar' on any key -- except a Telugu layout since a few 2361 years, but that will be corrected soon. 2362 2363 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2364 2365commit 86e9749ef37189486c205ee7f77e43fed655ac70 2366Author: ArenaL5 <arenal5@protonmail.com> 2367Date: Sat Jan 15 16:20:56 2022 +0100 2368 2369 nls: add a compose sequence for the middle dot (·) using dead_circumflex 2370 2371 This makes the middle dot compose sequences consistent with others that 2372 begin with <dead_circumflex> and <Multi_key> <asciicircum>, e.g. 2373 superscripted 3 (³) 2374 2375 Signed-off-by: ArenaL5 <arenal5@protonmail.com> 2376 2377commit 1adb531cd5be2b6489c2a8106e850e7418fae056 2378Author: Benno Schulenberg <bensberg@telfort.nl> 2379Date: Fri Jan 14 13:06:10 2022 +0100 2380 2381 docs: unwrap the comments of structure elements in the man pages 2382 2383 It doesn't matter that these lines get wider than 80 columns, as it's 2384 easy enough to make the pager scroll a bit to the right, or to have a 2385 terminal that is a 100 or more columns wide. It looks better to not 2386 hard-wrap these comments. 2387 2388 Also, fix the layout of two table elements. 2389 2390 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2391 2392commit da330a1644f10f9669c8dd9ee06489c40f9f0bd1 2393Author: Benno Schulenberg <bensberg@telfort.nl> 2394Date: Fri Jan 14 12:33:23 2022 +0100 2395 2396 docs: unwrap some mistakenly wrapped lines, and correct a comment 2397 2398 The wrapped lines messed up the table. Also, add some blank lines 2399 to make the table more readable. 2400 2401 Elsewhere, add spaces around an operator (for consistency), 2402 and correct a copy-paste error. 2403 2404 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2405 2406commit b9e0346abebb0c9e5c80c780bba4372419e63bde 2407Author: Benno Schulenberg <bensberg@telfort.nl> 2408Date: Fri Jan 14 15:57:56 2022 +0100 2409 2410 XKBMAlloc: use the correct first-key variable when extending a key range 2411 2412 When extending the range of changed virtual-modifier keys, one should 2413 pass the two variables relevant to that range. 2414 2415 These two mistakes were found by grepping for 'modmap_key' and then 2416 looking for asymmetries between the handling of the plain modmap thing 2417 and the virtual modmap thing. Inspired by issue #76. 2418 2419 (I suspect these mistakes were at the base of some intractable bugs 2420 reported against xkeyboard-config where virtual modifiers refused 2421 to work and had to be worked around by making other changes too.) 2422 2423 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2424 2425commit f15d2a01f75304bc2d9dac109d9a5a47d1ce28e0 2426Author: Benno Schulenberg <bensberg@telfort.nl> 2427Date: Wed Jan 12 11:34:23 2022 +0100 2428 2429 lcUTF8: remove two unneeded ifs and two elses that are never reached 2430 2431 When there is a break at the end of the while loop, it means that the 2432 loop will finish the first time that _XlcGetCharSetWithSide() returns 2433 a non-NULL result, which means that the first '(last_charset == NULL)' 2434 will always be true and the else part is redundant. 2435 2436 Fixes issue #46. 2437 2438 Reported-by: Bhavi Dhingra 2439 2440 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2441 2442commit 033f37d507801645a8251fd33c25ebd145d22cce 2443Author: Antti Savolainen <antti.savo@gmail.com> 2444Date: Wed Jan 12 16:10:21 2022 +0000 2445 2446 nls: add two compose sequences for an upwards arrow using dead_circumflex 2447 2448 Signed-off-by: Antti Savolainen <antti.savo@gmail.com> 2449 2450commit b7d724df615311c35418afc733e1fcc68c740aba 2451Author: Hodong <hodong@nimfsoft.com> 2452Date: Fri Jan 7 10:41:15 2022 +0000 2453 2454 specs/XIM: fix a missing "EXT" in the documentation 2455 2456 There was an "EXT" in the original text, but it seems to be missing. 2457 2458 See: https://gitlab.freedesktop.org/xorg/lib/libx11/-/blob/4e66da0783b2e5e3b288aaecd3c89396ed425c20/specs/XIM/xim.ms#L693 2459 2460 Signed-off-by: Hodong Kim <hodong@nimfsoft.com> 2461 2462commit 4c96f3567a8d045ee57b886fddc9618b71282530 2463Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2464Date: Thu Dec 9 19:51:52 2021 -0800 2465 2466 libX11 1.7.3.1 2467 2468 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2469 2470commit c1710589bb00d6f442a383fc41f0d6496b436c88 2471Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2472Date: Tue Dec 7 08:29:43 2021 -0800 2473 2474 specs/XIM: Fix "encording" typos 2475 2476 Fixes: #145 2477 2478 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2479 2480commit e30771b4bbdd84fc7ff64a1cda483a46edb5707b 2481Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2482Date: Mon Dec 6 17:04:14 2021 -0800 2483 2484 libX11 1.7.3 2485 2486 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2487 2488commit 934417b1e89fe715b9cf6c0d62a3867fa1f5bccb 2489Author: Liang Chang <l-chang@users.sourceforge.net> 2490Date: Tue Sep 7 18:54:10 2021 +0800 2491 2492 Make Xutf8DrawString works correctly with ISO10646 fonts. 2493 2494 Signed-off-by: Liang Chang <l-chang@users.sourceforge.net> 2495 2496commit 3ad36c5d3ddca3e4c97faaf87b72a53954d1f473 2497Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2498Date: Thu Dec 2 17:48:55 2021 -0800 2499 2500 Build xz tarballs instead of bzip2 2501 2502 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2503 2504commit 93a050c3ad2d2264d3880db3791387b1a9bf2e9e 2505Author: Tatu Frisk <tatu.frisk@ge.com> 2506Date: Tue Mar 14 14:41:27 2017 +0200 2507 2508 Fix hanging issue in _XReply 2509 2510 Assume event queue is empty if another thread is blocking waiting for event. 2511 2512 If one thread was blocking waiting for an event and another thread sent a 2513 reply to the X server, both threads got blocked until an event was 2514 received. 2515 2516 Signed-off-by: Tatu Frisk <tatu.frisk@ge.com> 2517 Signed-off-by: Jose Alarcon <jose.alarcon@ge.com> 2518 2519commit 8382253010267b93c9aeffc769e3f5e43d698094 2520Author: Matthieu Herrb <matthieu@herrb.eu> 2521Date: Mon Aug 30 21:47:09 2021 +0200 2522 2523 Avoid NULL pointer deref. Fixes issue #47. 2524 2525 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 2526 2527commit 58d7b540c453601faf942c6f19f8277a5c5f78cc 2528Author: Peter Hutterer <peter.hutterer@who-t.net> 2529Date: Wed Dec 1 14:31:16 2021 +1000 2530 2531 gitlab CI: update to use the ci-fairy image 2532 2533 Let's use the pre-prepared image instead of building our own. 2534 2535 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 2536 2537commit 150154c994829f9f3734a560f99f70b21d86b49c 2538Author: Hodong <hodong@nimfsoft.com> 2539Date: Sun Nov 14 06:36:42 2021 +0000 2540 2541 Fix a typo 2542 2543commit 2356e59ff24f8d1b25cdc4dffc5171c65dc2b86e 2544Author: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> 2545Date: Sun Sep 5 10:25:38 2021 +0300 2546 2547 libX11: do not crash in GetResReq() macro 2548 2549 When _XGetRequest() detects that requested length exceeds remaining display 2550 output buffer capacity it would return NULL. GetResReq() macro obtains "req" 2551 pointer from a call to _XGetRequest() and then proceeds to assign request id 2552 through "req" pointer which leads to NULL pointer dereference in this case. 2553 2554 Fix this by checking if "req" is valid before assigning request id. 2555 2556 Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com> 2557 2558commit e92efc63acd7b377faa9e534f4bf52aaa86be2a9 2559Author: Peter Hutterer <peter.hutterer@who-t.net> 2560Date: Tue Jul 27 11:46:19 2021 +1000 2561 2562 makekeys: handle the new _EVDEVK xorgproto symbols 2563 2564 These keys are all defined through a macro in the form: 2565 #define XF86XK_BrightnessAuto _EVDEVK(0x0F4) 2566 2567 The _EVDEVK macro is simply an offset of 0x10081000. 2568 Let's parse these lines correctly so those keysyms end up in our 2569 hashtables. 2570 2571 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 2572 2573commit d01d23374107f6fc55511f02559cf75be7bdf448 2574Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk> 2575Date: Wed Jun 16 12:17:04 2021 +0100 2576 2577 Avoid undefined behaviour after realloc() 2578 2579 Adding the offset between the realloc result and the old allocation to 2580 update pointers into the new allocation is undefined behaviour: the 2581 old pointers are no longer valid after realloc() according to the C 2582 standard. While this works on almost all architectures and compilers, 2583 it causes problems on architectures that track pointer bounds (e.g. 2584 CHERI or Arm's Morello): the value_list pointers will still have the 2585 bounds of the previous allocation and therefore any dereference will 2586 result in a run-time trap. 2587 2588 I found this due to a crash (dereferencing an invalid capability) while 2589 trying to run `xev` over SSH on a CHERI-RISC-V system. With these two 2590 realloc changes, and https://gitlab.freedesktop.org/xorg/proto/xorgproto/-/merge_requests/41 2591 I am able to succesfully run `xev` compiled for CHERI-RISC-V. 2592 2593 Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk> 2594 2595commit 1c845834a32421abb59408aee259ac0bef054d30 2596Author: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk> 2597Date: Tue Jun 15 08:46:53 2021 +0100 2598 2599 Fix cross-compilation from macOS 2600 2601 We can't use `LC_CTYPE=C sed` there since /usr/bin/sed is not compatible 2602 with the expressions in nls/ (`sed: RE error: illegal byte sequence`). 2603 To fix this use $(SED) instead which autotools will set to a GNU 2604 version of sed (usually /usr/local/bin/gsed) on macOS. 2605 2606 Signed-off-by: Alex Richardson <Alexander.Richardson@cl.cam.ac.uk> 2607 2608commit 401f58f8ba258d4e7ce56a8f756595b72e544c15 2609Author: Lev Korol <epicatsupercell@gmail.com> 2610Date: Sun Apr 11 21:57:45 2021 +0300 2611 2612 nls: Add en_IL locale 2613 2614 Signed-off-by: Lev Korol <epicatsupercell@gmail.com> 2615 2616commit f906fe8e9769e4313294b68e61c402610ade69da 2617Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2618Date: Sat Jun 5 09:12:13 2021 -0700 2619 2620 libX11 1.7.2 2621 2622 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2623 2624commit 51b73ac0acda65005c8a9f17ca4ea7281b00ca84 2625Author: Tobias Stoeckmann <tobias@stoeckmann.org> 2626Date: Sun May 23 15:05:17 2021 +0200 2627 2628 Protect against overly long strings 2629 2630 Checking against upper limit of USHRT_MAX must happen before truncating 2631 size_t to int. On 64 bit systems with strings larger than 2 GB this 2632 could otherwise lead to negative ints or ints smaller than USHRT_MAX. 2633 2634 In XParseColor this could lead to out of boundary access with strings 2635 starting with a # (color sequence). A modulo 12 operation is performed 2636 to validate the string length, but with an overflown length, the for 2637 loop would eventually read behind terminating '\0' character. 2638 2639 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 2640 2641commit ab2f59530b16bdfbf023b8e025c7c8aba3b6fd0c 2642Author: Matthieu Herrb <matthieu@herrb.eu> 2643Date: Sun May 30 16:33:48 2021 +0200 2644 2645 Check for NULL strings before getting their lengths 2646 2647 Problem reported by Karsten Trulsen 2648 2649 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 2650 2651commit a8216e8653a246db664e1443ecf707b27de65067 2652Author: Matthieu Herrb <matthieu@herrb.eu> 2653Date: Sat May 22 17:35:47 2021 +0200 2654 2655 include <limits.h> always, not if HAVE_CONFIG_H is set. 2656 2657 via Matthew Green from the NetBSD project 2658 2659 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 2660 2661commit 6953a586df4819143c4d55e011b3a5e5377981b8 2662Author: Matthieu Herrb <matthieu.herrb@laas.fr> 2663Date: Tue May 18 15:27:58 2021 +0200 2664 2665 Version 1.7.1 2666 2667 Release notes in README.md, version bump in configure.ac 2668 2669commit 8d2e02ae650f00c4a53deb625211a0527126c605 2670Author: Matthieu Herrb <matthieu@herrb.eu> 2671Date: Fri Feb 19 15:30:39 2021 +0100 2672 2673 Reject string longer than USHRT_MAX before sending them on the wire 2674 2675 The X protocol uses CARD16 values to represent the length so 2676 this would overflow. 2677 2678 CVE-2021-31535 2679 2680 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 2681 2682commit 838ea5a5a0267c25b20c095c9a70684edeeefba4 2683Author: Gaurav Ujjwal <gujjwal00@gmail.com> 2684Date: Sun May 9 05:04:36 2021 +0000 2685 2686 Fix out-of-bound access in KeySymToUcs4() 2687 2688 Array `keysym_to_unicode_590_5fe` is only valid for range [0x590, 0x5fe] but current lower-bound is checked against 0x589. 2689 2690 So invalid values from 0x58a to 0x58f are being allowed by current check. 2691 2692 If any of these invalid value is passed as `keysym`, `keysym - 0x590` would underflow. 2693 2694 Signed-off-by: Gaurav Ujjwal <gujjwal00@gmail.com> 2695 2696commit 32491b02c76ac22f99dc433a32466e22bdc181a7 2697Author: Christopher Chavez <chrischavez@gmx.us> 2698Date: Mon May 3 15:46:44 2021 +0000 2699 2700 Xlib.h: spelling fix in comment 2701 2702commit 5faa8dc0b364f55f19034c370436074646d9ad54 2703Author: Benno Schulenberg <bensberg@telfort.nl> 2704Date: Tue Nov 24 18:54:24 2020 +0100 2705 2706 nls: adjust three comments about the APL compose sequences 2707 2708 Commit 0bbc0d5e605e (from eight years ago) removed the lines that two 2709 of these comments referred to. Without those lines, the comments don't 2710 make sense any more. Reword and shorten them. 2711 2712 Also reword a comment about two sequences that don't work. 2713 2714 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2715 2716commit b126bfd7fe990b8c1df088f55c31d9b3fd440cde 2717Author: Benno Schulenberg <bensberg@telfort.nl> 2718Date: Tue Nov 24 12:23:32 2020 +0100 2719 2720 nls: allow composing all breved letters also with a lowercase "u" 2721 2722 The letters ă and ŭ can already be composed with "u a" and "u u", but 2723 ĕ, ğ, ĭ, and ŏ can be composed only with an uppercase U. Emancipate 2724 the latter four and understand also a lowercase "u" to mean 'breve'. 2725 2726 (Yesterday I needed ğ and was annoyed that "u g" did not work.) 2727 2728 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2729 2730commit 78027fdb7aa07621a3b7cdf37050bae39012f2fb 2731Author: Walter Harms <wharms@bfs.de> 2732Date: Sat Nov 28 21:05:33 2020 +0100 2733 2734 fix same roff code 2735 2736 Signed-off-by: Walter Harms <wharms@bfs.de> 2737 2738commit 4f15cfc64548098fef3ddd24f7359b76cf1bbda7 2739Author: Walter Harms <wharms@bfs.de> 2740Date: Sat Nov 28 20:56:35 2020 +0100 2741 2742 Fix some roff code add see also 2743 2744 Signed-off-by: Walter Harms <wharms@bfs.de> 2745 2746commit daa106927f78e6c5c9c36558a4b951d7cb58a8d1 2747Author: Walter Harms <wharms@bfs.de> 2748Date: Sat Nov 28 17:49:25 2020 +0100 2749 2750 fix broken nroff coding for code comments 2751 2752 the comments /* */ are code as /\(** */ that does not work. 2753 the coding in other X11 man pages is /\&* */ so we do the same here. 2754 2755commit 7bdeae239b17db002964ac73d06f43f216d1cf86 2756Author: Walter Harms <wharms@bfs.de> 2757Date: Fri Nov 27 22:43:21 2020 +0100 2758 2759 FIX: warning: macro `hN' not defined 2760 2761 Signed-off-by: Walter Harms <wharms@bfs.de> 2762 2763commit b7ec67d3ea78ab68857328d8f72d923324e68cc2 2764Author: Walter Harms <wharms@bfs.de> 2765Date: Fri Nov 27 22:26:15 2020 +0100 2766 2767 FIX: warning: macro `IN' not defined 2768 2769 just remove an other dead macro use. 2770 2771commit 7ca3ceb9175a829a40c347cb3b2e98341ec5a2df 2772Author: Walter Harms <wharms@bfs.de> 2773Date: Fri Nov 27 21:58:04 2020 +0100 2774 2775 fix warning: macro `s' not defined 2776 2777 this is caused by bad nroff coding, fix some more issues on the fly 2778 2779commit 14fb4e535626ba934cad5bc4308f511524cd1103 2780Author: Walter Harms <wharms@bfs.de> 2781Date: Fri Nov 27 20:04:22 2020 +0100 2782 2783 FIX: warning: macro `hN' not defined 2784 2785 this was found by checking man pages with 2786 groff -t -mandoc -Z -wmac -Tutf8 $FILE >/dev/null 2787 2788 In most cases .hN could be replaced with .BR 2789 2790 Signed-off-by: Walter Harms <wharms@bfs.de> 2791 2792commit cb03da4447aa53e8e5382f124917d89b50148036 2793Author: Walter Harms <wharms@bfs.de> 2794Date: Fri Nov 27 19:00:00 2020 +0100 2795 2796 FIX: warning: macro `Pn' not defined 2797 2798 The missing macro is found via: 2799 roff -t -mandoc -Z -wmac -Tutf8 XAnyEvent.man >/dev/null 2800 2801 To fix the problem the macro is replaced with .RB. 2802 2803 Signed-off-by: Walter Harms <wharms@bfs.de> 2804 2805commit cc9f8878f2cbe17c7b4035b4ff4352b52ece38e0 2806Author: Benno Schulenberg <bensberg@telfort.nl> 2807Date: Wed Nov 25 16:58:27 2020 +0100 2808 2809 nls: add 'C.utf8' as an alias for 'en_US.UTF-8' 2810 2811 The normal form is 'C.UTF-8', but 'C.utf8' has been seen in the wild. 2812 2813 Fixes #102. 2814 2815 Reported-by: Tomas Korbar 2816 2817 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 2818 2819commit ca8115186f810eccb7d86b0979980eff3ba95f0b 2820Author: Keith Packard <keithp@keithp.com> 2821Date: Thu Nov 19 11:30:41 2020 -0800 2822 2823 Version 1.7.0 2824 2825 Release notes in README.md, version bump in configure.ac 2826 2827 Signed-off-by: Keith Packard <keithp@keithp.com> 2828 2829commit 325ac177fb6d38f4c7689d9ce059792eaef388ef 2830Author: Keith Packard <keithp@keithp.com> 2831Date: Mon Nov 16 15:48:40 2020 -0800 2832 2833 Update libX11 .so version to 6.4.0 2834 2835 There have been API additions since the previous release, so the 2836 .so version needs to change. 2837 2838 Signed-off-by: Keith Packard <keithp@keithp.com> 2839 2840commit 103e2e11519984aa5746c06e7d90d1f5bc8174a3 2841Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2842Date: Sat Oct 31 09:24:01 2020 -0700 2843 2844 Don't leave dangling pointers in Free functions 2845 2846 While these are mostly called during teardown of larger structures 2847 that are about to themselves be freed, there's no guarantee that 2848 will always be the case, so try to be safer here. 2849 2850 [ This bug was found by the Parfait 4.0 bug checking tool. 2851 http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ] 2852 2853 v2: Deduplicate & simplify pointer clearing in _XFreeEventCookies 2854 as suggested by @keithp 2855 2856 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2857 2858commit 7d70e302218e2ec8518548ddcdc02d828e7889af 2859Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2860Date: Sat Oct 31 08:56:34 2020 -0700 2861 2862 XimProto.h: wrap XIM_HEADER_SIZE definition in parens 2863 2864 Resolves parfait warning of potential macro misinterpretation if 2865 expanded in the midst of other arithmetic operations with higher 2866 precedence. 2867 2868 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2869 2870commit a3c0b5dbd6b12ae64bc78b11795647a7f6df0c7a 2871Author: Keith Packard <keithp@keithp.com> 2872Date: Tue Nov 17 13:15:18 2020 -0800 2873 2874 Copy locale modifiers when creating XimInstCallback [v2] 2875 2876 Locale modifiers may be freed whenever XSetLocaleModifiers gets 2877 called, even if the locale hasn't changed. This means that we cannot 2878 save a pointer to those modifiers in the XimInstCallback record and 2879 must, instead, make a copy of them instead. 2880 2881 This fixes a problem uncovered when running wish under libasan as 2882 follows (on current Debian unstable): 2883 2884 $ LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libasan.so.6 wish 2885 2886 Reported-by: Vittorio Zecca <zeccav@gmail.com> 2887 Signed-off-by: Keith Packard <keithp@keithp.com> 2888 2889 v2: 2890 Remove incorrect 'else' token found by @alanc 2891 2892commit 960e2e0cfac12c3477c672d0d40818a0dc74aca5 2893Author: Peter Hutterer <peter.hutterer@who-t.net> 2894Date: Mon Nov 16 09:34:57 2020 +1000 2895 2896 gitlab CI: add a basic build test 2897 2898 Using Arch as base distribution here because we can expect our dependencies to 2899 be up-to-date. We rely on the Arch for our dependencies rather than building 2900 those from git (notably: xorg-macros, xtrans and libxcb). 2901 2902 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 2903 2904commit dbb55e1a5e82870466b095097d9e46046680ec25 2905Author: Frediano Ziglio <fziglio@redhat.com> 2906Date: Wed Jan 29 09:06:54 2020 +0000 2907 2908 Fix poll_for_response race condition 2909 2910 In poll_for_response is it possible that event replies are skipped 2911 and a more up to date message reply is returned. 2912 This will cause next poll_for_event call to fail aborting the program. 2913 2914 This was proved using some slow ssh tunnel or using some program 2915 to slow down server replies (I used a combination of xtrace and strace). 2916 2917 How the race happens: 2918 - program enters into poll_for_response; 2919 - poll_for_event is called but the server didn't still send the reply; 2920 - pending_requests is not NULL because we send a request (see call 2921 to append_pending_request in _XSend); 2922 - xcb_poll_for_reply64 is called from poll_for_response; 2923 - xcb_poll_for_reply64 will read from server, at this point 2924 server reply with an event (say sequence N) and the reply to our 2925 last request (say sequence N+1); 2926 - xcb_poll_for_reply64 returns the reply for the request we asked; 2927 - last_request_read is set to N+1 sequence in poll_for_response; 2928 - poll_for_response returns the response to the request; 2929 - poll_for_event is called (for instance from another poll_for_response); 2930 - event with sequence N is retrieved; 2931 - the N sequence is widen, however, as the "new" number computed from 2932 last_request_read is less than N the number is widened to N + 2^32 2933 (assuming last_request_read is still contained in 32 bit); 2934 - poll_for_event enters the nested if statement as req is NULL; 2935 - we compare the widen N (which now does not fit into 32 bit) with 2936 request (which fits into 32 bit) hitting the throw_thread_fail_assert. 2937 2938 To avoid the race condition and to avoid the sequence to go back 2939 I check again for new events after getting the response and 2940 return this last event if present saving the reply to return it 2941 later. 2942 2943 To test the race and the fix it's helpful to add a delay (I used a 2944 "usleep(5000)") before calling xcb_poll_for_reply64. 2945 2946 Original patch written by Frediano Ziglio, see 2947 https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/34 2948 2949 Reworked primarily for readability by Peter Hutterer, see 2950 https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/53 2951 2952 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 2953 2954commit 30ccef3a48029bf4fc31d4abda2d2778d0ad6277 2955Author: Keith Packard <keithp@keithp.com> 2956Date: Sat Nov 7 22:22:47 2020 -0800 2957 2958 Avoid recursing through _XError due to sequence adjustment 2959 2960 This patch is based on research done by Dmitry Osipenko to uncover the 2961 cause of a large class of Xlib lockups. 2962 2963 _XError must unlock and re-lock the display around the call to the 2964 user error handler function. When re-locking the display, two 2965 functions are called to ensure that the display is ready to generate a request: 2966 2967 _XIDHandler(dpy); 2968 _XSeqSyncFunction(dpy); 2969 2970 The first ensures that there is at least one XID available to use 2971 (possibly calling _xcb_generate_id to do so). The second makes sure a 2972 reply is received at least every 65535 requests to keep sequence 2973 numbers in sync (possibly generating a GetInputFocus request and 2974 synchronously awaiting the reply). 2975 2976 If the second of these does generate a GetInputFocus request and wait 2977 for the reply, then a pending error will cause recursion into _XError, 2978 which deadlocks the display. 2979 2980 One seemingly easy fix is to have _XError avoid those calls by 2981 invoking InternalLockDisplay instead of LockDisplay. That function 2982 does everything that LockDisplay does *except* call those final two 2983 functions which may end up receiving an error. 2984 2985 However, that doesn't protect the system from applications which call 2986 some legal Xlib function from within their error handler. Any Xlib 2987 function which cannot generate protocol or wait for events is valid, 2988 including many which invoke LockDisplay. 2989 2990 What we need to do is make LockDisplay skip these two function calls 2991 precisely when it is called from within the _XError context for the 2992 same display. 2993 2994 This patch accomplishes this by creating a list of threads in the 2995 display which are in _XError, and then having LockDisplay check the 2996 current thread against those list elements. 2997 2998 Inspired-by: Dmitry Osipenko <digetx@gmail.com> 2999 Signed-off-by: Keith Packard <keithp@keithp.com> 3000 Tested-by: Dmitry Osipenko <digetx@gmail.com> 3001 Reviewed-by: Dmitry Osipenko <digetx@gmail.com> 3002 3003commit c9c4d6efbf92ab51695e2e740319503221d68eed 3004Author: Benno Schulenberg <bensberg@telfort.nl> 3005Date: Fri Nov 13 11:34:04 2020 +0100 3006 3007 nls: improve the comments for the Serbian compose sequences 3008 3009 Also put an extra space before the lone combining characters 3010 so they have some room to breathe. 3011 3012 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3013 3014commit 3497dae3d9c52a252110b2a0983b82000be0fe8e 3015Author: Benno Schulenberg <bensberg@telfort.nl> 3016Date: Fri Nov 13 11:23:09 2020 +0100 3017 3018 nls: remove mistaken Serbian compose sequences with combining diacritics 3019 3020 Combining characters are not dead keys -- they have an immediate effect 3021 and combine with the preceding character. So they cannot be used in 3022 compose sequences. 3023 3024 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3025 3026commit 5cd60398b7787297008b13a848ed3cfbd7ef178d 3027Author: Benno Schulenberg <bensberg@telfort.nl> 3028Date: Mon Nov 2 16:34:39 2020 +0100 3029 3030 nls: rename the obsolete sr_CS locale to sr_RS 3031 3032 Fixes #107, for the most part. 3033 3034 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3035 3036commit 7cf1729590534e257b851899cf03421b65c538e3 3037Author: Keith Packard <keithp@keithp.com> 3038Date: Sat Nov 7 17:02:56 2020 -0800 3039 3040 Don't attempt to unload locale context data 3041 3042 Most locale context users call _XlcCurrentLC, which returns a pointer 3043 which never needs to be passed to _XCloseLC, meaning it has unbounded 3044 lifetime, so that locale data can never be freed. 3045 3046 Remove all reference counting and just leave all locales that were 3047 ever used in memory. 3048 3049 Signed-off-by: Keith Packard <keithp@keithp.com> 3050 Acked-by: Martin Peres <martin.peres@mupuf.org> 3051 3052commit 87a8fd8051b823636c3c4f3c37c4cfd11fdb880d 3053Author: Jacek Caban <jacek@codeweavers.com> 3054Date: Mon Aug 14 19:20:19 2017 +0200 3055 3056 Don't cache last lcd in _XlcCurrentLC. 3057 3058 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678 3059 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538 3060 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088 3061 3062 The way it's currently cached is not thread safe. As long as locale doesn't change, the same object is reused anyway. 3063 3064 Signed-off-by: Jacek Caban <jacek@codeweavers.com> 3065 Signed-off-by: Keith Packard <keithp@keithp.com> 3066 Acked-by: Martin Peres <martin.peres@mupuf.org> 3067 3068commit 7a38c09b82ee2522d987c3da4af8af186e562c61 3069Author: Keith Packard <keithp@keithp.com> 3070Date: Sat Nov 7 16:59:37 2020 -0800 3071 3072 Don't cache converters in _Xlcmbtowc and Xlcwctomb 3073 3074 These functions were caching encoding conversion functions in static 3075 variables which is not thread safe. Let the conversion loader do its 3076 job and cache locale to converters there. It's less efficient, but 3077 it's also (now) thread safe. 3078 3079 Signed-off-by: Keith Packard <keithp@keithp.com> 3080 Acked-by: Martin Peres <martin.peres@mupuf.org> 3081 3082commit b52ba5c209165fc2c533b77b86147b31763299f3 3083Author: Jacek Caban <jacek@codeweavers.com> 3084Date: Mon Aug 14 19:20:20 2017 +0200 3085 3086 Make conv_list thread safe. 3087 3088 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=55678 3089 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=68538 3090 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69088 3091 Signed-off-by: Jacek Caban <jacek@codeweavers.com> 3092 Signed-off-by: Keith Packard <keithp@keithp.com> 3093 Acked-by: Martin Peres <martin.peres@mupuf.org> 3094 3095commit 8b5ce3a9b1219c81711944880e5a1e1cfad3d956 3096Author: Antti Savolainen <antti.savo@gmail.com> 3097Date: Wed Sep 16 07:47:26 2020 +0000 3098 3099 Added DOWNWARD ARROW support 3100 3101commit 7138826dccba9d538dfb142ce9af1a40d64ea849 3102Author: Antti Savolainen <antti.savo@gmail.com> 3103Date: Tue Sep 8 00:54:23 2020 +0300 3104 3105 Added support for UPWARDS ARROW and ALMOST EQUAL TO 3106 3107 Signed-off-by: Antti Savolainen <antti.savo@gmail.com> 3108 3109commit 56d59299f7433cdeb88bac970b06eb7df44df5bb 3110Author: Jonathan Belsewir <jonbel@jonasund.de> 3111Date: Thu Nov 5 22:57:50 2020 +0000 3112 3113 Add less and greater to characters that are difficult to access on some keyboards 3114 3115commit 5caf45f8d3bebad5b72ac88545adc156ccfd754d 3116Author: Benno Schulenberg <bensberg@telfort.nl> 3117Date: Fri Oct 23 17:12:22 2020 +0200 3118 3119 nls: Allow all letters with a caron to be composed also with v. 3120 3121 It was strange that the accented letters Ž and ž can be composed with 3122 sequences that start with "v" ("v Z" and "v z"), but not Č and č and 3123 Š and š (and other letters with a caron). For these letters, compose 3124 sequences that start with a "c" had to be used, which was frustrating 3125 because it is hard to remember that "c" stands for "caron", AND the 3126 graphically more obvious "v" is right next to it. 3127 3128 (Unfortunately, the sequence "v l" is already taken for vertical line. 3129 Maybe the compose sequences for vertical line could be reduced to just 3130 "V L" and "L V"?) 3131 3132 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3133 3134commit 62de4ac9025e2cb3f62a890241c06ebc75acce62 3135Author: Benno Schulenberg <bensberg@telfort.nl> 3136Date: Sun Nov 1 20:23:55 2020 +0100 3137 3138 nls: vertically align the target column of aliases 3139 3140 Also improve the grammar of the initial comment. 3141 3142 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3143 3144commit f5b26a16ef959fc6860e9cf58b97a042600900b4 3145Author: Benno Schulenberg <bensberg@telfort.nl> 3146Date: Sun Nov 1 20:17:05 2020 +0100 3147 3148 nls: remove some twenty aliases that have been obsolete for fifteen years 3149 3150 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3151 3152commit 3e61f3b49660e081108f06bc514366f1784cb2e0 3153Author: Benno Schulenberg <bensberg@telfort.nl> 3154Date: Sun Nov 1 19:50:56 2020 +0100 3155 3156 nls: remove country-specific aliases for two constructed languages 3157 3158 These artificial languages are meant to be international and are 3159 thus not specific to any country. If one would want to support 3160 aliases like ia_FR or ia_CH, then one would also have to support 3161 ia_AU, ia_DE, ia_ES, et cetera, et cetera. That would be silly. 3162 3163 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3164 3165commit 3f9c43adde22bb2ab4bef5837bc8401af9bb537c 3166Author: Benno Schulenberg <bensberg@telfort.nl> 3167Date: Sun Nov 1 17:28:24 2020 +0100 3168 3169 nls: remove four aliases where the name is identical to the definition 3170 3171 They were found with: 3172 3173 while read one two; do 3174 if [[ $one == $two: ]]; then echo $two; fi; 3175 done <nls/locale.alias.pre 3176 3177 Signed-off-by: Benno Schulenberg <bensberg@telfort.nl> 3178 3179commit 9f9c5365816bdd036fd80d54b22e86764ea4b7a7 3180Author: Carlos Garnacho <carlosg@gnome.org> 3181Date: Fri Jun 14 17:55:14 2019 +0200 3182 3183 Add XSetIOErrorExitHandler() function 3184 3185 This function complements XSetIOErrorHandler(), allowing to override 3186 the default behavior that trusts on I/O errors never coming back 3187 (i.e. exit()ing the process). 3188 3189 This is meant as a mechanism for Wayland compositors (that are too 3190 a X11 client + compositing manager) to unfasten seatbelts and jump 3191 through the car window. It might get lucky and land on a stack of 3192 pillows. 3193 3194 In consequence, some functions labeled as _X_NORETURN can as a 3195 matter of fact return. So those hints were removed. 3196 3197 Signed-off-by: Carlos Garnacho <carlosg@gnome.org> 3198 Reviewed-by: Michel Dänzer <mdaenzer@redhat.com> 3199 3200commit 4cb758019e374fa7c022fe79d28444e13441717b 3201Author: Carlos Garnacho <carlosg@gnome.org> 3202Date: Fri Jun 14 17:54:47 2019 +0200 3203 3204 Prepare for _XIOError() possibly returning 3205 3206 Ensure current state is cut short on _XIOError(), possible reentrancy 3207 should be skipped through the XlibDisplayIOError flag checks. 3208 3209 Signed-off-by: Carlos Garnacho <carlosg@gnome.org> 3210 Reviewed-by: Michel Dänzer <mdaenzer@redhat.com> 3211 3212commit a37986f7a7d4ae9b8292b132dbb0da2885ea919c 3213Author: Ross Burton <ross.burton@intel.com> 3214Date: Wed Jan 22 17:11:23 2020 +0000 3215 3216 cpprules.in: squash whitespace in generated files 3217 3218 CPP is used to generate files, but as cpp reads files from the build host the 3219 output has a number of blank lines at the beginning which varies depending on 3220 what GCC and friends is used. 3221 3222 Pathalogical example: 3223 3224 $ cpp -undef -traditional /dev/null 3225 # 1 "/dev/null" 3226 # 1 "<built-in>" 3227 # 1 "<command-line>" 3228 # 31 "<command-line>" 3229 # 1 "/usr/include/stdc-predef.h" 1 3 4 3230 3231 # 17 "/usr/include/stdc-predef.h" 3 4 3232 3233 [ 40 blank line ] 3234 3235 # 32 "<command-line>" 2 3236 # 1 "/dev/null" 3237 3238 So depending on the content of stdc-predef.h and what other headers CPP will 3239 load, the amount of whitespace in the generates files varies. This can result in 3240 differences in reproducible environments, and file conflicts in multilib 3241 environments. 3242 3243 As whitespace is irrelevant to these machine-readable files, extend the sed to 3244 just delete blank lines. 3245 3246commit 16192ce2a7e462e09f95aa45ce9eaa180e483c43 3247Author: Carmina16 <mistresssilvara@hotmail.com> 3248Date: Sun Jun 16 02:54:25 2019 +0000 3249 3250 Adding ia and ie locales 3251 3252commit caa71668af7fd3ebdd56353c8f0ab90824773969 3253Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3254Date: Mon Sep 28 16:16:07 2020 -0700 3255 3256 Use __builtin_popcountl if available to replace Ones() in GetImage.c 3257 3258 If the compiler knows of a better algorithm for counting the number of 3259 bits set in a word for the target CPU, let it use that, instead of the 3260 classic algorithm optimized for PDP-6. 3261 3262 Based on libXext commit 490a25e6f8a4d2482af4364c700b68ad11a4d10b 3263 3264 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3265 3266commit 1a9269a95c353e3de9f4d33bab9f1363d84e79db 3267Author: Krzesimir Nowak <qdlacz@gmail.com> 3268Date: Mon Jun 1 12:58:30 2020 +0000 3269 3270 Fix leaking modifiers string. 3271 3272 Reported by valgrind: 3273 3274 ``` 3275 ==118175== 17 bytes in 1 blocks are definitely lost in loss record 13 of 1,675 3276 ==118175== at 0x483A809: malloc (vg_replace_malloc.c:307) 3277 ==118175== by 0x5CD1B46: _XlcDefaultMapModifiers (in /usr/lib64/libX11.so.6.3.0) 3278 ==118175== by 0x5CD1F1A: XSetLocaleModifiers (in /usr/lib64/libX11.so.6.3.0) 3279 ==118175== by 0x496841C: X11_InitKeyboard (SDL_x11keyboard.c:324) 3280 ==118175== by 0x496F0CA: X11_VideoInit (SDL_x11video.c:455) 3281 ==118175== by 0x494747B: SDL_VideoInit_REAL (SDL_video.c:532) 3282 ==118175== by 0x489E886: SDL_InitSubSystem_REAL (SDL.c:206) 3283 ==118175== by 0x402634: main (fade.cc:35) 3284 ``` 3285 3286commit d127217f26df1bf7566c1f372d8b5329a06754ea 3287Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3288Date: Sun Aug 16 17:03:24 2020 -0700 3289 3290 Exempt XLookupKeysym from XKeycodeToKeysym deprecation warnings 3291 3292 Gets rid of: 3293 src/xkb/XKBBind.c: In function ‘XLookupKeysym’: 3294 src/xkb/XKBBind.c:234:5: warning: ‘XKeycodeToKeysym’ is deprecated 3295 [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wdeprecated-declarations-Wdeprecated-declarations] 3296 234 | return XKeycodeToKeysym(dpy, event->keycode, col); 3297 | ^~~~~~ 3298 src/xkb/XKBBind.c:96:1: note: declared here 3299 96 | XKeycodeToKeysym(Display *dpy, 3300 | ^~~~~~~~~~~~~~~~ 3301 3302 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3303 3304commit a64eb5f478c6ca84ef3457a01f9a7c3db8ecf1e5 3305Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3306Date: Sun Aug 16 16:50:43 2020 -0700 3307 3308 lcUTF8: Increase string buffer size from 20 to 32 bytes 3309 3310 While we don't expect large enough ints to need it, we don't 3311 enforce a maximum size, so gcc assumes the worst and warns: 3312 3313 ../../../src/xlibi18n/lcUTF8.c: In function ‘create_tofontcs_conv’: 3314 ../../../src/xlibi18n/lcUTF8.c:1736:34: warning: ‘.charset.name’ directive output may be truncated writing 13 bytes into a region of size between 8 and 17 [-Wformat-truncation=] 3315 1736 | snprintf(buf, sizeof(buf), "fs%d.charset.name", i); 3316 | ^~~~~~~~~~~~~ 3317 ../../../src/xlibi18n/lcUTF8.c:1736:2: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20 3318 1736 | snprintf(buf, sizeof(buf), "fs%d.charset.name", i); 3319 | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 3320 ../../../src/xlibi18n/lcUTF8.c:1739:46: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=] 3321 1739 | snprintf(buf, sizeof(buf), "fs%d.charset", i); 3322 | ^ 3323 ../../../src/xlibi18n/lcUTF8.c:1739:6: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20 3324 1739 | snprintf(buf, sizeof(buf), "fs%d.charset", i); 3325 | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 3326 ../../../src/xlibi18n/lcUTF8.c:1754:41: warning: ‘.charset.name’ directive output may be truncated writing 13 bytes into a region of size between 8 and 17 [-Wformat-truncation=] 3327 1754 | snprintf(buf, sizeof(buf), "fs%d.charset.name", i); 3328 | ^~~~~~~~~~~~~ 3329 ../../../src/xlibi18n/lcUTF8.c:1754:9: note: ‘snprintf’ output between 17 and 26 bytes into a destination of size 20 3330 1754 | snprintf(buf, sizeof(buf), "fs%d.charset.name", i); 3331 | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 3332 ../../../src/xlibi18n/lcUTF8.c:1757:53: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=] 3333 1757 | snprintf(buf, sizeof(buf), "fs%d.charset", i); 3334 | ^ 3335 ../../../src/xlibi18n/lcUTF8.c:1757:13: note: ‘snprintf’ output between 12 and 21 bytes into a destination of size 20 3336 1757 | snprintf(buf, sizeof(buf), "fs%d.charset", i); 3337 | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 3338 3339 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3340 3341commit 54925250ad9c825bea671dcade5c913bad88abee 3342Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3343Date: Sun Aug 16 16:33:59 2020 -0700 3344 3345 i18n: use memcpy instead of strncpy on unterminated char arrays 3346 3347 Avoids gcc warnings that we're using strncpy wrong to copy a known-length 3348 set of characters without a terminating '\0' to a buffer whose length we 3349 are checking separately. (Should also be imperceptibly faster since we 3350 no longer check if each byte is '\0' when we already know it won't be.) 3351 3352 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3353 3354commit 8754868dbbbc5fa096282df0504aa7880488eeb9 3355Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3356Date: Sun Aug 16 16:02:01 2020 -0700 3357 3358 Mark FreeBlueTblElements label unused in LRGB.c 3359 3360 Quiets gcc 10.2 warning of: 3361 src/xcms/LRGB.c: In function ‘LINEAR_RGB_InitSCCData’: 3362 src/xcms/LRGB.c:798:1: warning: label ‘FreeBlueTblElements’ defined 3363 but not used 3364 [https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wunused-label-Wunused-label] 3365 798 | FreeBlueTblElements: 3366 | ^~~~~~~~~~~~~~~~~~~ 3367 3368 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3369 3370commit 52b0cd67073f2f4665884b5a7decfedeb3b7ca1e 3371Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3372Date: Sun Aug 16 14:54:47 2020 -0700 3373 3374 Raise minimum required xproto version to 7.0.25 (released 2013-11-23) 3375 3376 Allows us to depend on _X_COLD directly instead of having to check for it. 3377 3378 (Since we also use _X_UNUSED, 7.0.22 or later was implicitly required 3379 already but not checked for.) 3380 3381 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3382 3383commit deafb99fd8faf8b044eaee13d072440ce4db76af 3384Author: Maya Rashish <maya@NetBSD.org> 3385Date: Sun Aug 23 21:19:33 2020 +0300 3386 3387 Avoid the use of "register" keyword in public headers. 3388 3389 This causes issues when compiling code for C++17. 3390 3391commit 505420662577749e36640db48f6b6b9ae0236e09 3392Author: Matthieu Herrb <matthieu@herrb.eu> 3393Date: Mon Aug 24 15:42:25 2020 +0200 3394 3395 libX11 1.6.12 3396 3397 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3398 3399commit acdaaadcb3d85c61fd43669fc5dddf0f8c3f911d 3400Author: Matthieu Herrb <matthieu@herrb.eu> 3401Date: Thu Aug 13 18:02:58 2020 +0200 3402 3403 Fix an integer overflow in init_om() 3404 3405 CVE-2020-14363 3406 3407 This can lead to a double free later, as reported by Jayden Rivers. 3408 3409 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3410 3411commit d15c24c8b44be5e4054c8ecd0ff9dcf2c8e18e5b 3412Author: Niclas Zeising <zeising@daemonic.se> 3413Date: Tue Aug 11 13:44:38 2020 +0200 3414 3415 Fix input clients connecting to server 3416 3417 Fix a bug where some input clients can't connect to the input server. 3418 This fixes #117. 3419 3420 FreeBSD bugzilla reference: 3421 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248549 3422 3423 Signed-off-by: Niclas Zeising <zeising@daemonic.se> 3424 3425commit 6dd618e5ea8e7558dd8e99a2777ab7b69211789c 3426Author: Maya Rashish <maya@NetBSD.org> 3427Date: Sat Aug 15 00:48:56 2020 +0300 3428 3429 Avoid the use of "register" keyword in XkbTranslateKeySym. 3430 3431 This causes issues when compiling code for C++17. 3432 While here, make function prototype match the header with regards 3433 to removal of another register keyword. 3434 3435commit 780d222343098b4bfe3ec529b4a0dc2aadea19bf 3436Author: Christopher Chavez <chrischavez@gmx.us> 3437Date: Mon Aug 10 17:08:39 2020 +0000 3438 3439 Fix typo GCCLipYOrigin -> GCClipYOrigin in XCreateGC() manpage 3440 3441commit 29a8251a456bb7dc10146b4713f499a624d6dacd 3442Author: Felix Yan <felixonmars@archlinux.org> 3443Date: Thu Aug 6 16:03:38 2020 +0000 3444 3445 Correct a typo in GetStCmap.c 3446 3447commit 0d8f038db83f68724a84ef9d4ccd0d5e33ec6a70 3448Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3449Date: Thu Aug 6 08:07:57 2020 -0700 3450 3451 libX11 1.6.11 3452 3453 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3454 3455commit 93fce3f4e79cbc737d6468a4f68ba3de1b83953b 3456Author: Yichao Yu <yyc1992@gmail.com> 3457Date: Sun Aug 2 13:43:58 2020 -0400 3458 3459 Fix size calculation in `_XimAttributeToValue`. 3460 3461 The check here guards the read below. 3462 For `XimType_XIMStyles`, these are `num` of `CARD32` and for `XimType_XIMHotKeyTriggers` 3463 these are `num` of `XIMTRIGGERKEY` ref[1] which is defined as 3 x `CARD32`. 3464 (There are data after the `XIMTRIGGERKEY` according to the spec but they are not read by this 3465 function and doesn't need to be checked.) 3466 3467 The old code here used the native datatype size instead of the wire protocol size causing 3468 the check to always fail. 3469 3470 Also fix the size calculation for the header (size). It is 2 x CARD16 for both types 3471 despite the unused `CARD16` for `XimType_XIMStyles`. 3472 3473 [1] https://www.x.org/releases/X11R7.6/doc/libX11/specs/XIM/xim.html#Input_Method_Styles 3474 3475 This fixes a regression caused by 388b303c62aa35a245f1704211a023440ad2c488 in 1.6.10. 3476 3477 Fix #116 3478 3479commit 9949364ea761ab7efa2a9930ec7718f966a740c0 3480Author: Matthieu Herrb <matthieu@herrb.eu> 3481Date: Thu Jul 30 21:36:22 2020 +0200 3482 3483 libX11 1.6.10 3484 3485 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3486 3487commit 1703b9f3435079d3c6021e1ee2ec34fd4978103d 3488Author: Matthieu Herrb <matthieu@herrb.eu> 3489Date: Fri Jul 24 21:09:10 2020 +0200 3490 3491 Change the data_len parameter of _XimAttributeToValue() to CARD16 3492 3493 It's coming from a length in the protocol (unsigned) and passed 3494 to functions that expect unsigned int parameters (_XCopyToArg() 3495 and memcpy()). 3496 3497 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3498 Reviewed-by: Todd Carson <toc@daybefore.net> 3499 3500commit 1a566c9e00e5f35c1f9e7f3d741a02e5170852b2 3501Author: Todd Carson <tc@daybefore.net> 3502Date: Fri Jul 24 20:59:32 2020 +0200 3503 3504 Zero out buffers in functions 3505 3506 It looks like uninitialized stack or heap memory can leak 3507 out via padding bytes. 3508 3509 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3510 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 3511 3512commit 2fcfcc49f3b1be854bb9085993a01d17c62acf60 3513Author: Todd Carson <tc@daybefore.net> 3514Date: Fri Jul 24 20:22:44 2020 +0200 3515 3516 Fix more unchecked lengths 3517 3518 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3519 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 3520 3521commit 388b303c62aa35a245f1704211a023440ad2c488 3522Author: Todd Carson <tc@daybefore.net> 3523Date: Fri Jul 24 19:36:51 2020 +0200 3524 3525 fix integer overflows in _XimAttributeToValue() 3526 3527 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3528 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 3529 3530commit 0e6561efcfaa0ae7b5c74eac7e064b76d687544e 3531Author: Todd Carson <tc@daybefore.net> 3532Date: Fri Jul 24 19:33:30 2020 +0200 3533 3534 Fix signed length values in _XimGetAttributeID() 3535 3536 The lengths are unsigned according to the specification. Passing 3537 negative values can lead to data corruption. 3538 3539 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 3540 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 3541 3542commit 2b7598221d87049d03e9a95fcb541c37c8728184 3543Author: Alan Coopersmith <alan.coopersmith@oracle.com> 3544Date: Wed Jul 22 15:19:58 2020 -0700 3545 3546 Fix spelling/wording issues 3547 3548 Found by using: 3549 codespell --builtin clear,rare,usage,informal,code,names 3550 3551 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3552 3553commit f9e3378ebcfc79ff00591b8617a68e97580a741e 3554Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 3555Date: Sun Jul 19 22:19:24 2020 +0200 3556 3557 xlibi18n: Fix converting horizline keysyms to UCS 3558 3559 U+23BA - U+23BD are meant to represent the scan lines, and U+2500 is 3560 unified with scan line 5. 3561 3562 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 3563 3564commit 4385a84c4a79c1b60bf12839899a1544198de975 3565Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 3566Date: Thu Jun 4 02:19:13 2020 +0200 3567 3568 Braille: Fix default lookup 3569 3570 cbb59d172 ('Braille: Fix typing quickly') broke the default lookup that 3571 translates Braille keysym patterns to Braille Unicode patterns since it 3572 rightfully clears brl_committing, but then we do not have it any more to 3573 fill brl_committed. 3574 3575 This change saves the committed pattern so we can return it in the 3576 default lookup. 3577 3578 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 3579 3580commit fb65b67169105c82c243f265e0f993cf72efdacd 3581Author: Marko Myllynen <myllynen@redhat.com> 3582Date: Fri Feb 28 13:59:36 2020 +0200 3583 3584 Update Finnish compose sequences for SFS 5966:2019 standard 3585 3586 U0219, U0218, U021B, U021A are now part of the standard. 3587 3588 Fixes #70 which was printed incorrectly in the earlier standard version. 3589 3590commit 07724ce046348a21cc82a874853b14b9a285df75 3591Author: Tristan Miller <psychonaut@nothingisreal.com> 3592Date: Wed Sep 25 13:34:01 2019 +0200 3593 3594 Add acute accent compose sequences for Russian vowels. 3595 3596 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/104 3597 3598 Signed-off-by: Tristan Miller <psychonaut@nothingisreal.com> 3599 3600commit 3ec9264d01a3a66c0fa0145a7c1db06a64be1507 3601Author: Thomas E. Dickey <dickey@invisible-island.net> 3602Date: Sat Apr 25 14:40:54 2020 -0400 3603 3604 document an ancient performance problem (see Debian #954845). 3605 3606 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3607 3608commit 791c93816ce0520d54925b0b3ccbc29823370041 3609Author: Thomas E. Dickey <dickey@invisible-island.net> 3610Date: Sat Apr 25 14:22:11 2020 -0400 3611 3612 update/correct function prototypes in manpages to match headers 3613 3614 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3615 3616commit 844271305a28e87031562dfeb43defbea02960a2 3617Author: Thomas E. Dickey <dickey@invisible-island.net> 3618Date: Tue Apr 21 20:53:35 2020 -0400 3619 3620 reduce gcc-normal warnings using casts (no object change) 3621 3622 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3623 3624commit 53b59ddfc114a55d26e9717a1b4822f1b18c40ca 3625Author: Thomas E. Dickey <dickey@invisible-island.net> 3626Date: Tue Apr 21 18:13:22 2020 -0400 3627 3628 reduce gcc-normal warnings using casts (no object change) 3629 3630 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3631 3632commit 68c72a7341b114277ab232f2499ee3bd035af8a0 3633Author: Adam Jackson <ajax@redhat.com> 3634Date: Wed Dec 11 11:53:11 2019 -0500 3635 3636 Fix XTS regression in XCopyColormapAndFree 3637 3638 XCopyColormapAndFree/5 threw an assertion: 3639 3640 520|4 5 00014017 1 2|Assertion XCopyColormapAndFree-5.(A) 3641 520|4 5 00014017 1 3|When a colourmap argument does not name a valid colourmap, 3642 520|4 5 00014017 1 4|then a BadColor error occurs. 3643 520|4 5 00014017 1 5|METH: Create a bad colourmap by creating and freeing a colourmap. 3644 520|4 5 00014017 1 6|METH: Call test function using bad colourmap as the colourmap argument. 3645 520|4 5 00014017 1 7|METH: Verify that a BadColor error occurs. 3646 520|4 5 00014017 1 8|unexpected signal 6 (SIGABRT) received 3647 220|4 5 2 15:05:53|UNRESOLVED 3648 410|4 5 1 15:05:53|IC End 3649 510|4|system 0: Abandoning testset: caught unexpected signal 11 (SIGSEGV) 3650 3651 More specifically: 3652 3653 lt-XCopyColormapAndFree: xcb_io.c:533: _XAllocID: Assertion `ret != inval_id' failed. 3654 3655 This bug was introduced (by following my advice, d'oh) in: 3656 3657 commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d 3658 Author: Tapani Pälli <tapani.palli@intel.com> 3659 Date: Mon May 13 08:29:49 2019 +0300 3660 3661 Protect colormap add/removal with display lock 3662 3663 In that patch we moved the call to _XcmsCopyCmapRecAndFree inside the 3664 display lock. The problem is said routine has side effects, including 3665 trying to implicitly create a colormap in some cases. Since we don't run 3666 the XID handler until SyncHandle() we would see inconsistent internal 3667 xlib state, triggering the above assert. 3668 3669 Fix this by dropping and re-taking the display lock before calling into 3670 XCMS. 3671 3672 Reviewed-by: Tapani Pälli <tapani.palli@intel.com> 3673 3674commit 7f46a38139f66fda734f3a6c445b84ea89c8f310 3675Author: Alex Henrie <alexhenrie24@gmail.com> 3676Date: Wed Jan 1 14:54:06 2020 -0700 3677 3678 Handle small final sigma in XConvertCase 3679 3680 lowercase: GREEK SMALL LETTER FINAL SIGMA (U+03C2) 3681 uppercase: GREEK CAPITAL LETTER SIGMA (U+03A3) 3682 3683 This mapping was correct in UCSConvertCase, but the "legacy" mapping 3684 must also be correct for Caps Lock to work with the final sigma key. 3685 3686 https://gitlab.freedesktop.org/xorg/lib/libx11/issues/5 3687 3688 Signed-off-by: Alex Henrie <alexhenrie24@gmail.com> 3689 3690commit a48787d3d155c8a09cead345f5055da30c1db537 3691Author: Peter Hutterer <peter.hutterer@who-t.net> 3692Date: Wed Dec 11 14:12:27 2019 +1000 3693 3694 Handle ssharp in XConvertCase() 3695 3696 lowercase: LATIN SMALL LETTER SHARP S (U+00DF) 3697 uppercase: LATIN CAPITAL LETTER SHARP S (U+1E9E) 3698 3699 The uppercase sharp s (XK_ssharp) is a relatively recent addition to unicode 3700 but was added to the relevant keyboard layouts in xkeyboard-config-2.25 3701 (d1411e5e95c) 3702 https://gitlab.freedesktop.org/xkeyboard-config/xkeyboard-config/issues/144 3703 3704 Alas, the CapsLock behavior was broken on the finnish layout (maybe others). 3705 This was due to xkbcomp using XConvertCase() to determine whether a key 3706 requires the type FOUR_LEVEL_ALPHABETIC or FOUR_LEVEL_SEMIALPHABETIC. 3707 3708 Let's make this function return the right lower/upper symbols for the sharp s 3709 and hope that the world won't get any worse because of it. 3710 3711 https://gitlab.freedesktop.org/xorg/lib/libx11/issues/110 3712 3713 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 3714 3715commit b8766a438461e20b5c6f7a6111b703d4cb71501f 3716Author: Raul Fernandes <rgfernandes@gmail.com> 3717Date: Wed Oct 9 18:39:26 2019 +0000 3718 3719 Use memcmp and memcpy 3720 3721commit db7cca17ad7807e92a928da9d4c68a00f4836da2 3722Author: Adam Jackson <ajax@redhat.com> 3723Date: Wed Oct 9 13:41:29 2019 -0400 3724 3725 libX11 1.6.9 3726 3727 Signed-off-by: Adam Jackson <ajax@redhat.com> 3728 3729commit f5ba2c632a4c84279bdd48c687271292f5dc4f87 3730Author: Dmitry Osipenko <digetx@gmail.com> 3731Date: Sat May 18 19:10:32 2019 +0300 3732 3733 Fix lockup in _XReply() caused by recursive synchronization 3734 3735 This patch is based on a suggestion made by Uli Schlachter in a comment 3736 to the bug report https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93. 3737 3738 Explanation of the bug (given by Uli Schlachter as well): 3739 3740 An error was received and handled. Since there was an error callback set, 3741 Xlib unlocks the display, runs the error callback, and then locks the display 3742 again. This goes through _XLockDisplay and then calls _XSeqSyncFunction. 3743 On this "lock the thing"-path, Xlib notices that sequence numbers are close to 3744 wrap-around and tries to send a GetInputFocus request. However, the earlier 3745 calls already registered themselves as "we are handling replies/errors, do 3746 not interfere!" and so the code here waits for "that other thread" to be done 3747 before it continues. Only that there is no other thread, but it is this thread 3748 itself and thus a deadlock follows. 3749 3750 The bug is relatively easy to reproduce on any desktop environment by 3751 using actively a touchscreen input that supports multitouch, i.e. practically 3752 all mobile devices are affected. 3753 3754 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/93 3755 Suggested-by: Uli Schlachter <psychon@znc.in> 3756 Tested-by: Dmitry Osipenko <digetx@gmail.com> 3757 Reported-by: Dmitry Osipenko <digetx@gmail.com> 3758 Signed-off-by: Dmitry Osipenko <digetx@gmail.com> 3759 3760commit 1f1ca0863fd81fd0538b104eb9d9841ddbc7c905 3761Author: Adam Jackson <ajax@redhat.com> 3762Date: Fri Sep 13 16:44:32 2019 -0400 3763 3764 xkb: Provide <X11/extensions/XKBgeom.h> ourselves 3765 3766 This header refers to libX11 types, it doesn't belong in xorgproto and 3767 has been moved to legacy in 2019.2. Supply it ourselves. 3768 3769commit edc7680ed5a03cedb5facf14693823455e12c29c 3770Author: Ross Burton <ross.burton@intel.com> 3771Date: Tue Aug 6 14:53:43 2019 +0100 3772 3773 src/util/Makefile: explicitly reset LINK to not use libtool 3774 3775 Simply looking at libtool redefines LINK globally to use libtool, which when 3776 you're trying to cross-compile to Windows can cause complications. 3777 3778 As in src/util/ we're simply building a small binary for the build host, reset 3779 LINK to the automake default so that the traditional compile/link steps occur 3780 without libtool. 3781 3782 Also remove -all-static from LDFLAGS as that is a libtool-specific argument 3783 intended to solve this problem. 3784 3785 Closes: #100 3786 Signed-off-by: Ross Burton <ross.burton@intel.com> 3787 3788commit f111bf3ac091a3a6d1f094172621d9d033c3833c 3789Author: Thomas E. Dickey <dickey@invisible-island.net> 3790Date: Tue Jul 30 16:53:27 2019 -0400 3791 3792 improve some formatting 3793 3794 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3795 3796commit 1d0a6fe9126c7ce61c6648cabf43b2de4d6d3ba9 3797Author: Thomas E. Dickey <dickey@invisible-island.net> 3798Date: Tue Jul 30 15:49:09 2019 -0400 3799 3800 fix a substitution error from recent commit, e.g, "s/^\.EE/XDe/" 3801 3802 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3803 3804commit 8fe8aafe7e9beffaff0460fe08354dfcde1efe98 3805Author: Thomas E. Dickey <dickey@invisible-island.net> 3806Date: Tue Jul 30 05:07:40 2019 -0400 3807 3808 split lines at sentence endings 3809 3810 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3811 3812commit 6f2cbcea5d6dcd1df7ef309481cad1663284c061 3813Author: Thomas E. Dickey <dickey@invisible-island.net> 3814Date: Tue Jul 30 04:52:25 2019 -0400 3815 3816 trim trailing whitespace from manpages 3817 3818 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3819 3820commit 4abf482d8e6b0b1ae84e0f528e71066afd6d18b5 3821Author: Thomas E. Dickey <dickey@invisible-island.net> 3822Date: Tue Jul 30 04:36:24 2019 -0400 3823 3824 another fake-quote fix 3825 3826 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3827 3828commit c0f0651f2b4dfa9fef86d1374689220125631993 3829Author: Thomas E. Dickey <dickey@invisible-island.net> 3830Date: Tue Jul 30 04:34:54 2019 -0400 3831 3832 the last commit overlooked some fake-quote pairs 3833 3834 Signed-off-by: Thomas E. Dickey <dickey@invisible-island.net> 3835 3836commit ef31143c9decda54b3ab2937d3300fcf5b493fd6 3837Author: Walter Harms <wharms@bfs.de> 3838Date: Mon Jul 29 18:50:19 2019 +0200 3839 3840 fix ``fake quotes'' in text 3841 3842 to remove the fake quotes replace them with propper 3843 predefined macros \*(lq and \*(rq. this will allow 3844 nroff to choose the propper characters when using ps etc. 3845 3846 Signed-off-by: Walter Harms <wharms@bfs.de> 3847 3848commit 705027ea9b17d6552388d2067736ab4b0bf9dfb5 3849Author: Walter Harms <wharms@bfs.de> 3850Date: Mon Jul 29 18:27:12 2019 +0200 3851 3852 get rid of ``fake quotes'' 3853 3854 these fake quotes are inside a comment and can be replaced with 3855 some double quotes 3856 3857 Signed-off-by: Walter Harms <wharms@bfs.de> 3858 3859commit fe46eb287106ad05538e896783a5cf877a4138df 3860Author: Walter Harms <wharms@bfs.de> 3861Date: Mon Jul 29 17:57:02 2019 +0200 3862 3863 remove all private macro defines 3864 3865 after converting everything to st. man page macros there is 3866 no need to maintain X11 private nroff macros, so remove them. 3867 3868 Signed-off-by: Walter Harms <wharms@bfs.de> 3869 3870commit b7f09ae19c9a098ecd545b4d4c45a6afe39a8764 3871Author: Walter Harms <wharms@bfs.de> 3872Date: Mon Jul 29 17:47:35 2019 +0200 3873 3874 Replace home-grown .Ds .De macro with man page .EX/.EE macro 3875 3876 Same pages use the man page .EX/.EE macro. Replace all occurences 3877 of .De/.Ds with the std. macros to make the code better to maintain. 3878 3879 Signed-off-by: Walter Harms <wharms@bfs.de> 3880 3881commit 951e772bc8d89bc62c4bbac14adc5d4f5027419f 3882Author: Walter Harms <wharms@bfs.de> 3883Date: Wed Jul 3 18:23:51 2019 +0200 3884 3885 fix TBL format 3886 3887 Signed-off-by: Walter Harms <wharms@bfs.de> 3888 3889commit 4eb167f5aca4cca05a230c728ffdb4fd382a4e1e 3890Author: Walter Harms <wharms@bfs.de> 3891Date: Wed Jul 3 18:18:09 2019 +0200 3892 3893 replace home grown .ZN with std, .B and .BR 3894 3895 Replace the home grown macro .ZN with std. macros 3896 from man macro paket. So we can get rid of the 3897 definition an get a clean header. 3898 3899 Signed-off-by: Walter Harms <wharms@bfs.de> 3900 3901commit 08e6b36292ebf01ba2dcc4a48ec371c6f5d13458 3902Author: Walter Harms <wharms@bfs.de> 3903Date: Sat Jun 22 18:33:48 2019 +0200 3904 3905 remove in-text macros 3906 3907 remove in-text macros an fix a few typos on the way. 3908 hopefuly that will make maintenance more easy. 3909 3910 Signed-off-by: Walter Harms <wharms@bfs.de> 3911 3912commit cb0e894fe9927b07bd9d97c067fd0d007accb105 3913Author: Walter Harms <wharms@bfs.de> 3914Date: Fri Jun 21 18:01:16 2019 +0200 3915 3916 note that we can handle kbd==NULL 3917 3918 Signed-off-by: Walter Harms <wharms@bfs.de> 3919 3920commit 0327c427d62f671eced067c6d9b69f4e216a8cac 3921Author: Adam Jackson <ajax@redhat.com> 3922Date: Mon Jun 17 13:36:08 2019 -0400 3923 3924 makekeys: Detach ourselves from X headers entirely 3925 3926 Subsequent to a121b7b0c210efe10bf93453b29050282324c906 we are no longer 3927 building makekeys with enough -I/foo/bar to find the X11 headers, so if 3928 they're not in a system include path, things fail. Since this utility is 3929 only needed at build time, there's no real reason to demand the X 3930 headers be installed for both the build and target machines if cross- 3931 compiling, we can just assume a vaguely ANSI environment instead. 3932 3933 Tested-by: Niclas Zeising <zeising@daemonic.se> 3934 Reviewed-by: Keith Packard <keithp@keithp.com> 3935 Reviewed-by: Matt Turner <mattst88@gmail.com> 3936 3937commit 57f0d421a1320090d3a4948181fe228d86cc8977 3938Author: Matt Turner <mattst88@gmail.com> 3939Date: Mon Jun 17 10:45:45 2019 -0400 3940 3941 libX11 1.6.8 3942 3943 Signed-off-by: Matt Turner <mattst88@gmail.com> 3944 3945commit 5464b302bc0c4a055d3eac06c6ab15e209ff7668 3946Author: Matt Turner <mattst88@gmail.com> 3947Date: Sun Jun 9 14:00:15 2019 -0700 3948 3949 Use AC_SYS_LARGEFILE 3950 3951 ... and include config.h in makekeys.c to get the definition of 3952 _FILE_OFFSET_BITS. Without it, libX11 can fail to build on a file 3953 system with 64-bit inode numbers. 3954 3955 Bug: https://bugs.gentoo.org/550502 3956 Bug: https://bugs.gentoo.org/616140 3957 Signed-off-by: Matt Turner <mattst88@gmail.com> 3958 3959commit 33b9148a833ab2ea46e44871da3c031bf1b78f77 3960Author: Ran Benita <ran234@gmail.com> 3961Date: Thu Feb 13 20:55:08 2014 +0200 3962 3963 Compose.man: fix escaped hexadecimal char description 3964 3965 The man page says: 3966 Strings may be direct text encoded in the locale for which the 3967 compose file is to be used, or an escaped octal or hexadecimal 3968 character code. Octal codes are specified as "\123" and 3969 hexadecimal codes as "\0x123a". 3970 3971 But the grammar in the parser and the implementation say: 3972 ESCAPED_CHAR ::= ('\\' | '\"' | OCTAL | HEX ) 3973 HEX ::= '\' (x|X) HEX_CHAR [HEX_CHAR]] 3974 HEX_CHAR ::= (0|1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|a|b|c|d|e|f) 3975 3976 So "\0x123a" -> "\x3a". 3977 3978 Signed-off-by: Ran Benita <ran234@gmail.com> 3979 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3980 3981commit d9b2cc35edf01e4f751a9ccb5ac185745184cec0 3982Author: ojab <ojab@ojab.ru> 3983Date: Mon Apr 24 20:12:17 2017 +0000 3984 3985 Compose sequences for rouble sign 3986 3987 Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout. 3988 3989 Signed-off-by: Slava Kardakov <ojab@ojab.ru> 3990 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3991 3992commit 3f211616f6414bd9958566c0ca137e87a2b6df95 3993Author: Pavel Labath <pavelo@centrum.sk> 3994Date: Fri Nov 30 20:45:04 2012 +0100 3995 3996 Fix a leak in XCreateFontSet 3997 3998 a simple snippet like XFreeFontSet(d, XCreateFontSet(d, ...)) will generate lots of memory leaks, 3999 as evidenced by the following valgrind output: 4000 ==983== HEAP SUMMARY: 4001 ==983== in use at exit: 39,409 bytes in 341 blocks 4002 ==983== total heap usage: 4,795 allocs, 4,454 frees, 489,086 bytes allocated 4003 ==983== 4004 ==983== 1,688 (136 direct, 1,552 indirect) bytes in 1 blocks are definitely lost in loss record 4005 40 of 46 4006 ==983== at 0x4C2B042: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) 4007 ==983== by 0x56D5A93: add_codeset.clone.9 (in /usr/lib64/libX11.so.6.3.0) 4008 ==983== by 0x56D5FE0: load_generic (in /usr/lib64/libX11.so.6.3.0) 4009 ==983== by 0x56D7612: initialize (in /usr/lib64/libX11.so.6.3.0) 4010 ==983== by 0x56D7E75: _XlcCreateLC (in /usr/lib64/libX11.so.6.3.0) 4011 ==983== by 0x56F9A5F: _XlcUtf8Loader (in /usr/lib64/libX11.so.6.3.0) 4012 ==983== by 0x56DF815: _XOpenLC (in /usr/lib64/libX11.so.6.3.0) 4013 ==983== by 0x56B255A: XOpenOM (in /usr/lib64/libX11.so.6.3.0) 4014 ==983== by 0x56A665A: XCreateFontSet (in /usr/lib64/libX11.so.6.3.0) 4015 ==983== by 0x4FCA80: conky::x11_output::create_gc() (x11.cc:746) 4016 ==983== by 0x4FC3B4: conky::x11_output::use_own_window() (x11.cc:602) 4017 ==983== by 0x4FAD42: conky::priv::own_window_setting::set(bool const&, bool) (x11.cc:92) 4018 ==983== 4019 ==983== LEAK SUMMARY: 4020 ==983== definitely lost: 136 bytes in 1 blocks 4021 ==983== indirectly lost: 1,552 bytes in 34 blocks 4022 ==983== possibly lost: 0 bytes in 0 blocks 4023 ==983== still reachable: 37,721 bytes in 306 blocks 4024 ==983== suppressed: 0 bytes in 0 blocks 4025 4026 This patch makes the leak dissappear (Well, at least the "definitely lost part". The "still 4027 reachable" thingy remains). After some analysis, I've discovered that the XLCd structure is 4028 destroyed improperly. The "constructor" is in lcGeneric.c, but the structure is destroyed using 4029 code from lcPublic.c. I've found that changing the destructor call to _XlcDestroyLC executes the 4030 correct code path, and I'm pretty sure this is correct (the object was constructed using 4031 _XlcCreateLC, it make sense to destroy it using its conterpart). 4032 4033 So far I haven't observed any strange behaviour on my system caused by this change (although, I'm 4034 not sure, how many programs actually use this function). 4035 4036 Signed-off-by: Pavel Labath <pavelo@centrum.sk> 4037 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4038 4039commit 4645e219133458781e3fb48eaea6a74cccb1b9aa 4040Author: Jon Turney <jon.turney@dronecode.org.uk> 4041Date: Tue Apr 30 14:39:06 2019 +0100 4042 4043 Avoid using libtool wrapper for makekeys 4044 4045 For Windows targets, libtool uses a wrapper executable, not a wrapper 4046 script (see [1]), which it compiles with the host compiler. This 4047 doesn't work when cross-compiling. 4048 4049 Since we don't actually need to link with anything, use the libtool flag 4050 -all-static to tell it to stay completely out of this. 4051 4052 [1] https://www.gnu.org/software/libtool/manual/html_node/Wrapper-executables.html 4053 4054commit 6886d9ba06826785a8ccb312c04ea82b91bb6a25 4055Author: Jon Turney <jon.turney@dronecode.org.uk> 4056Date: Wed Apr 24 12:50:41 2019 +0100 4057 4058 Use EXEEXT_FOR_BUILD for makekeys 4059 4060 Use EXEXT_FOR_BUILD, to fix cross-compiling where EXEEXT differs from 4061 EXEEXT_FOR_BUILD, such as when building for Windows from unix. 4062 4063 (Note: As written, this assumes EXEEXT_FOR_BUILD is always empty when 4064 cross-compiling. There could be some elaborate autodetection for 4065 EXEXT_FOR_BUILD, but for the moment, if you are cross-compiling from 4066 Windows to Unix, you'll need to set EXEEXT_FOR_BUILD explicity...) 4067 4068commit a121b7b0c210efe10bf93453b29050282324c906 4069Author: Jon Turney <jon.turney@dronecode.org.uk> 4070Date: Tue Apr 30 14:30:41 2019 +0100 4071 4072 Remove makekeys dependency on X headers 4073 4074 This is the patch from https://bugs.freedesktop.org/show_bug.cgi?id=6669 4075 by Pierre Ossman, reworked for master. 4076 4077 Avoid using LIBS (which are for host, but we don't need) and rewrite 4078 makekeys slightly to avoid needing to include any X headers, which 4079 avoids potentially having -I with host paths in CFLAGS, which can cause 4080 standard headers e.g. stdio.h for the host to also be used, which can 4081 break things... 4082 4083commit cbb59d1727cb1bfcff91b54fd55f4139e360b7ab 4084Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 4085Date: Fri Jun 7 23:35:33 2019 +0200 4086 4087 Braille: Fix typing quickly 4088 4089 On finishing releasing Braille keys, we should clear the just-commited 4090 pattern, to reset the state to initial state, and avoid having to wait for 4091 0.3s before typing the next pattern. 4092 4093 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 4094 Tested-by: Jean-Philippe Mengual <jpmengual@hypra.fr> 4095 4096commit 99a2cf1aa0b58391078d5d3edf0a7dab18c7745d 4097Author: Tapani Pälli <tapani.palli@intel.com> 4098Date: Mon May 13 08:29:49 2019 +0300 4099 4100 Protect colormap add/removal with display lock 4101 4102 This fixes a bug where concurrent threads call XCreateColormap and 4103 XFreeColormap corrupting a linked list where colormap structures 4104 are stored. 4105 4106 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/94 4107 4108 v2: handle XCopyColormapAndFree also (Adam Jackson) 4109 4110 Signed-off-by: Tapani Pälli <tapani.palli@intel.com> 4111 4112commit fea80d03c381a4de379863c0d41f6d4c00f15e1e 4113Author: Jon Turney <jon.turney@dronecode.org.uk> 4114Date: Wed Mar 13 13:12:20 2019 +0000 4115 4116 Add autoconf checks for headers we include for FIONREAD 4117 4118 Add autoconf checks for the extra headers we include to define FIONREAD. 4119 This needs sys/socket.h on Cygwin, and none of the alternatives on 4120 Windows. 4121 4122commit 87c77a1e6d7034536e9d25ce24a667ebf53486a7 4123Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org> 4124Date: Mon Mar 11 14:19:52 2019 +0000 4125 4126 locale.alias.pre - add the de_IT locale 4127 4128 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4129 4130commit 568f77f414072506bbe48646637c872987c97e63 4131Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org> 4132Date: Mon Mar 11 14:17:10 2019 +0000 4133 4134 locale.dir.pre - add the de_IT locale 4135 4136 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4137 4138commit 38c0c92feaef7dc4a72920b1dd79872e6578086c 4139Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org> 4140Date: Mon Mar 11 14:12:07 2019 +0000 4141 4142 compose.dir.pre - add de_IT locale 4143 4144 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4145 4146commit 49f835611ba6ab82e68b7b851507931c142db7a5 4147Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org> 4148Date: Tue Jan 29 15:34:34 2019 +0000 4149 4150 Update locale.alias.pre - introduce the de_LI aliases (equivalent to de_CH) 4151 4152 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4153 4154commit ed669234d0cd5933618aa5fd2da697b59845b9be 4155Author: Matthias Dieter Wallnöfer <matthias.wallnoefer@lugbz.org> 4156Date: Tue Jan 29 15:28:18 2019 +0000 4157 4158 locale.dir.pre - fix the de_LI latin15 locale 4159 4160 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4161 4162commit 82b5c1454a49333b4d94984d40db79cff386d37f 4163Author: Josh Triplett <josh@joshtriplett.org> 4164Date: Sun Feb 3 08:31:41 2019 +0100 4165 4166 compose.dir.pre: Give the C.UTF-8 locale a full set of UTF-8 compose sequences 4167 4168 Signed-off-by: Josh Triplett <josh@joshtriplett.org> 4169 4170commit 603c3ab25a692194805fd4b0c62afac8093897fe 4171Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4172Date: Sat Mar 9 20:58:07 2019 -0800 4173 4174 Add missing close paren to XrmGetStringDatabase synopsis in man page 4175 4176 Fixes: commit 0642e99ac05afb0c 4177 4178 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4179 4180commit 6590cc41e3c5c5c836a49a2c82774d7cc9ab3c85 4181Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4182Date: Sat Feb 23 11:12:27 2019 -0800 4183 4184 Include reallocarray.h in tarballs so "make distcheck" passes 4185 4186 Fixes: commit bcf7b5aa06c23aee 4187 4188 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4189 4190commit a4a7df06351ce9565d298a84547ff71f2d220fbf 4191Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4192Date: Sat Feb 23 11:04:41 2019 -0800 4193 4194 XlibInt.c: include headers needed for ioctl(...FIONREAD...) on Solaris 4195 4196 Fixes: commit 5538b3e4ae6dee 4197 4198 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4199 4200commit 38ec04de7e537dde6b7b1eaca46b1a24fcdbc044 4201Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4202Date: Sat Feb 23 10:48:26 2019 -0800 4203 4204 specs/XKB: Fix misprint in documentation of XkbFreeGeomOverlayKeys() 4205 4206 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/86 4207 aka: https://bugs.freedesktop.org/show_bug.cgi?id=23550 4208 4209 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4210 4211commit 74f2f26811ae82b15dede31504b0eb541f7a3293 4212Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4213Date: Sat Feb 23 10:46:05 2019 -0800 4214 4215 specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayKeys() 4216 4217 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/85 4218 aka: https://bugs.freedesktop.org/show_bug.cgi?id=23549 4219 4220 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4221 4222commit 66b039a41690f076e962ac6bd46cca017360a7d1 4223Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4224Date: Sat Feb 23 10:44:15 2019 -0800 4225 4226 specs/XKB: Fix misprint in documentation of XkbAllocGeomOverlayRows() 4227 4228 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/84 4229 aka: https://bugs.freedesktop.org/show_bug.cgi?id=23548 4230 4231 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4232 4233commit cdea0e840315171f8390adc63a2397ddf111eea4 4234Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4235Date: Sat Feb 23 10:38:08 2019 -0800 4236 4237 specs/XKB: Fix misprint in documentation of XkbGetNamedGeometry() 4238 4239 Fixes: https://gitlab.freedesktop.org/xorg/lib/libx11/issues/82 4240 aka: https://bugs.freedesktop.org/show_bug.cgi?id=23520 4241 4242 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4243 4244commit c72d17ada98723cd9e03ddf06680a0e54b2c3089 4245Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4246Date: Sat Feb 23 10:30:32 2019 -0800 4247 4248 XkbAllocGeomOverlayKey man page should be named XkbAllocGeomOverlayKeys 4249 4250 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4251 4252commit 05002b8aef9544c8545e0823a02cf14f335d2b68 4253Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4254Date: Sun Feb 17 15:13:25 2019 -0800 4255 4256 Fix man page typo: XkbAllBellNotifyMask -> XkbAllBellEventsMask 4257 4258 Reported-by: Daniel Hahler <git@thequod.de> 4259 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4260 4261commit 5538b3e4ae6dee32c47db9dfc85b07bbe7b90f6c 4262Author: Adam Jackson <ajax@redhat.com> 4263Date: Fri Mar 24 11:07:36 2017 -0400 4264 4265 _XDefaultIOError: Do better at detecting explicit shutdown 4266 4267 Currently, when the X server crashes or a client is disconnected with 4268 XKillClient, you get a somewhat confusing error message from libX11 4269 along the lines of: 4270 4271 XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0" 4272 after 98 requests (40 known processed) with 0 events remaining. 4273 4274 What's happening here is the previous recvmsg has thrown EAGAIN, since 4275 the socket is non-blocking. In this case, check whether the socket has 4276 any more data to read, and if not treat it like EPIPE. 4277 4278 Signed-off-by: Adam Jackson <ajax@redhat.com> 4279 4280commit 6d2cde9633b5ee020cb60caea1cf61e090b86dd2 4281Author: Adam Jackson <ajax@redhat.com> 4282Date: Fri Mar 24 11:07:35 2017 -0400 4283 4284 _XDefaultIOError: Reformat to be less ugly 4285 4286 Signed-off-by: Adam Jackson <ajax@redhat.com> 4287 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4288 4289commit 003e30a66a249f5c70b30d1c187385124cd4cdad 4290Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4291Date: Tue Jan 1 14:34:04 2019 -0800 4292 4293 Avoid use-after-free in _XimProtoSetIMValues() 4294 4295 Fixes gitlab issue #49 4296 4297 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4298 4299commit 30656fd66ad47aeaca0bfc237a818f189554f07e 4300Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4301Date: Tue Jan 1 12:21:04 2019 -0800 4302 4303 Fix implicit conversion warnings in _XlcCreateDefaultCharSet 4304 4305 lcCharSet.c:187:50: warning: implicit conversion changes signedness: 4306 'int' to 'unsigned long' [-Wsign-conversion] 4307 tmp = Xmalloc(name_len + 1 + ct_sequence_len + 1); 4308 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~ 4309 ../../include/X11/Xlibint.h:453:32: note: expanded from macro 'Xmalloc' 4310 ~~~~~~ ^~~~ 4311 lcCharSet.c:192:31: warning: implicit conversion changes signedness: 4312 'int' to 'unsigned long' [-Wsign-conversion] 4313 memcpy(tmp, name, name_len+1); 4314 ~~~~~~ ~~~~~~~~^~ 4315 lcCharSet.c:216:45: warning: implicit conversion changes signedness: 4316 'int' to 'unsigned long' [-Wsign-conversion] 4317 memcpy(tmp, ct_sequence, ct_sequence_len+1); 4318 ~~~~~~ ~~~~~~~~~~~~~~~^~ 4319 lcCharSet.c:183:16: warning: implicit conversion loses integer precision: 4320 'unsigned long' to 'int' [-Wshorten-64-to-32] 4321 name_len = strlen(name); 4322 ~ ^~~~~~~~~~~~ 4323 lcCharSet.c:184:23: warning: implicit conversion loses integer precision: 4324 'unsigned long' to 'int' [-Wshorten-64-to-32] 4325 ct_sequence_len = strlen(ct_sequence); 4326 ~ ^~~~~~~~~~~~~~~~~~~ 4327 lcCharSet.c:198:37: warning: implicit conversion loses integer precision: 4328 'long' to 'unsigned int' [-Wshorten-64-to-32] 4329 unsigned int length = colon - charset->name; 4330 ~~~~~~ ~~~~~~^~~~~~~~~~~~~~~ 4331 4332 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4333 4334commit 2e6300905784fa8adfcea019262ea58b3e1ed4ad 4335Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4336Date: Tue Jan 1 11:41:47 2019 -0800 4337 4338 Remove no-longer-used name variable in _XGetAtomName 4339 4340 Fixes gcc warning: 4341 GetAtomNm.c: In function ‘_XGetAtomName’: 4342 GetAtomNm.c:39:11: warning: unused variable ‘name’ [-Wunused-variable] 4343 char *name; 4344 ^~~~ 4345 4346 Introduced by commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96 4347 4348 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4349 4350commit f077871c2a12cf098b0c50572e112c6dc5e04de0 4351Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4352Date: Sat Dec 29 11:39:29 2018 -0800 4353 4354 Fix out-of-tree builds after commit 8a62e26515e528bb47917ec4a55cc2b492419eeb 4355 4356 In-tree builds found reallocarray.h in $(top_builddir)/src but the 4357 out-of-tree build didn't find it at all. 4358 4359 Reported-by: Emmanuele Bassi <ebassi@gmail.com> from GNOME continuous integration pipeline 4360 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4361 4362commit ba7f7cc783672784bebc6e69349b55c6c75c1973 4363Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4364Date: Sun Sep 27 18:35:27 2015 -0700 4365 4366 Convert src/xlibi18n to use reallocarray() 4367 4368 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4369 4370commit 336c1e7a505698a8f3bcb6eb96dbbb8ed21ade96 4371Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4372Date: Sat Sep 26 11:12:47 2015 -0700 4373 4374 Replace Xmalloc+strcpy pairs with strdup calls 4375 4376 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4377 4378commit 258a8ced681dc1bc50396be7439fce23f9807e2a 4379Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4380Date: Sun Jul 19 11:58:50 2015 -0700 4381 4382 Replace _XkbTypedRealloc & _XkbClearElems with new _XkbResizeArray 4383 4384 Makes resizing & clearing more consistent and gets rid of some weird 4385 quirks like always subtracting 1 from the size passed to _XkbClearElems 4386 so it could always add 1 to the size passed in. 4387 4388 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4389 4390commit f94273e2f4284040306b888e5ca0381ee00591af 4391Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4392Date: Sun Jul 19 09:47:29 2015 -0700 4393 4394 Add some missing allocation failure checks in src/xcms 4395 4396 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4397 4398commit 76850471b4a5a6d30f5660234e87ec2772e7b426 4399Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4400Date: Sun Jul 19 09:33:21 2015 -0700 4401 4402 Convert src/xcms to use reallocarray() 4403 4404 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4405 4406commit 8a62e26515e528bb47917ec4a55cc2b492419eeb 4407Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4408Date: Sun Jul 19 09:24:02 2015 -0700 4409 4410 Convert src/XKB to use reallocarray() 4411 4412 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4413 4414commit 6d1dc1f6169ebf0ba71785d461bd98129c65c862 4415Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4416Date: Thu Jun 4 23:30:43 2015 -0700 4417 4418 Convert main src directory to use reallocarray() 4419 4420commit bcf7b5aa06c23aee00af7999b58cb96a8571ac42 4421Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4422Date: Thu Jun 4 23:18:23 2015 -0700 4423 4424 Import reallocarray() from OpenBSD 4425 4426 Wrapper for realloc() that checks for overflow when multiplying 4427 arguments together, so we don't have to add overflow checks to 4428 every single call. For documentation on usage, see: 4429 http://www.openbsd.org/cgi-bin/man.cgi/OpenBSD-current/man3/calloc.3 4430 4431 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4432 4433commit 9bdfe9c9af11d77d66e5ff651e82b20e695cb460 4434Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4435Date: Mon Nov 19 21:26:23 2018 -0800 4436 4437 Update README for gitlab migration 4438 4439 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4440 4441commit c8bcc19970a2cc4ad337de1338c63a07325eefcd 4442Author: Albert Astals Cid <aacid@kde.org> 4443Date: Fri Nov 9 22:24:52 2018 +0100 4444 4445 locale.dir: Add ast_ES.UTF-8 4446 4447 With this patch xev properly reports 4448 XLookupString gives 2 bytes: (c2 b4) "´" 4449 for the dead_acute key when using the Asturian locale 4450 4451commit 117bb310b44ce03802c91a70949eb58dcc702a3c 4452Author: Albert Astals Cid <aacid@kde.org> 4453Date: Fri Nov 9 22:18:43 2018 +0100 4454 4455 compose.dir: Add ast_ES.UTF-8 4456 4457 With this patch one can properly type dead keys like á when using the 4458 Asturian locale 4459 4460commit d7e22252d68258a63ca537827c99d0ac0bc4b9d8 4461Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4462Date: Sat Nov 10 13:30:47 2018 -0800 4463 4464 Remove documentation of obsolete B16 & B32 tags in specs/libX11 4465 4466 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4467 4468commit f3c978476e0be6813268af494efb7ac507451116 4469Author: Matt Turner <mattst88@gmail.com> 4470Date: Tue Oct 9 10:26:05 2018 -0400 4471 4472 libX11 1.6.7 4473 4474 Signed-off-by: Matt Turner <mattst88@gmail.com> 4475 4476commit 823a0f8a820247b6c1e092f679b49cbdc2ea5c95 4477Author: Michel Dänzer <michel.daenzer@amd.com> 4478Date: Fri Sep 28 17:24:17 2018 +0200 4479 4480 poll_for_event: Allow using xcb_poll_for_queued_event 4481 4482 It avoids reading from the display connection again in cases where that 4483 was already done. 4484 4485 Suggested-by: Uli Schlachter <psychon@znc.in> 4486 Reviewed-by: Uli Schlachter <psychon@znc.in> 4487 4488commit 406afe4b0f1b655c0db19bbc9a0c48da9a46acf5 4489Author: Michel Dänzer <michel.daenzer@amd.com> 4490Date: Tue Sep 25 17:10:58 2018 +0200 4491 4492 poll_for_response: Call poll_for_event again if xcb_poll_for_reply fails 4493 4494 If xcb_poll_for_reply fails to find a reply, poll_for_response would 4495 always return NULL. However, xcb_poll_for_reply may have read events 4496 from the display connection while looking for a reply. In that case, 4497 returning NULL from poll_for_response is wrong and can result in the 4498 client hanging, e.g. because it returns to waiting for the display 4499 connection file descriptor becoming readable after XPending incorrectly 4500 returned 0 pending events. 4501 4502 The solution is to call poll_for_event again after xcb_poll_for_reply 4503 returned 0. This will return the first of any events read by 4504 xcb_poll_for_reply. 4505 4506 Fixes issue #79. 4507 4508 Reported-by: Yuxuan Shui <yshuiv7@gmail.com> 4509 Bugzilla: https://bugs.freedesktop.org/108008 4510 Bugzilla: https://bugs.freedesktop.org/107992 4511 Reviewed-by: Adam Jackson <ajax@redhat.com> 4512 4513commit d0416863d5bf75af54ce81f6c30d4c1476b5e04f 4514Author: Bhavi Dhingra <b.dhingra@samsung.com> 4515Date: Mon Sep 28 08:33:40 2015 +0000 4516 4517 XcmsLookupColor: fully initialize XColor structs passed to _XColor_to_XcmsRGB 4518 4519 Fixes https://gitlab.freedesktop.org/xorg/lib/libx11/issues/44 4520 aka https://bugs.freedesktop.org/show_bug.cgi?id=92154 4521 4522 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4523 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4524 4525commit 733f64bfeb311c1d040b2f751bfdef9c9d0f89ef 4526Author: Matthieu Herrb <matthieu@herrb.eu> 4527Date: Tue Aug 21 16:54:50 2018 +0200 4528 4529 libX11 1.6.6 4530 4531 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 4532 4533commit 173704243fbcbea0b6de2068b4fb403c7408fb54 4534Author: Matthieu Herrb <matthieu@herrb.eu> 4535Date: Tue Aug 21 16:53:40 2018 +0200 4536 4537 Remove statement with no effect. 4538 4539 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> 4540 4541commit e83722768fd5c467ef61fa159e8c6278770b45c2 4542Author: Tobias Stoeckmann <tobias@stoeckmann.org> 4543Date: Fri Jul 27 16:38:00 2018 +0200 4544 4545 Fixed crash on invalid reply (CVE-2018-14598). 4546 4547 If the server sends a reply in which even the first string would 4548 overflow the transmitted bytes, list[0] (or flist[0]) will be set to 4549 NULL and a count of 0 is returned. 4550 4551 If the resulting list is freed with XFreeExtensionList or 4552 XFreeFontPath later on, the first Xfree call: 4553 4554 Xfree (list[0]-1) 4555 turns into 4556 Xfree (NULL-1) 4557 4558 which will most likely trigger a segmentation fault. 4559 4560 I have modified the code to return NULL if the first string would 4561 overflow, thus protecting the freeing functions later on. 4562 4563 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 4564 4565commit dbf72805fd9d7b1846fe9a11b46f3994bfc27fea 4566Author: Tobias Stoeckmann <tobias@stoeckmann.org> 4567Date: Fri Jul 27 16:37:17 2018 +0200 4568 4569 Fixed out of boundary write (CVE-2018-14600). 4570 4571 The length value is interpreted as signed char on many systems 4572 (depending on default signedness of char), which can lead to an out of 4573 boundary write up to 128 bytes in front of the allocated storage, but 4574 limited to NUL byte(s). 4575 4576 Casting the length value to unsigned char fixes the problem and allows 4577 string values with up to 255 characters. 4578 4579 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 4580 4581commit b469da1430cdcee06e31c6251b83aede072a1ff0 4582Author: Tobias Stoeckmann <tobias@stoeckmann.org> 4583Date: Fri Jul 27 16:36:34 2018 +0200 4584 4585 Fixed off-by-one writes (CVE-2018-14599). 4586 4587 The functions XGetFontPath, XListExtensions, and XListFonts are 4588 vulnerable to an off-by-one override on malicious server responses. 4589 4590 The server replies consist of chunks consisting of a length byte 4591 followed by actual string, which is not NUL-terminated. 4592 4593 While parsing the response, the length byte is overridden with '\0', 4594 thus the memory area can be used as storage of C strings later on. To 4595 be able to NUL-terminate the last string, the buffer is reserved with 4596 an additional byte of space. 4597 4598 For a boundary check, the variable chend (end of ch) was introduced, 4599 pointing at the end of the buffer which ch initially points to. 4600 Unfortunately there is a difference in handling "the end of ch". 4601 4602 While chend points at the first byte that must not be written to, 4603 the for-loop uses chend as the last byte that can be written to. 4604 4605 Therefore, an off-by-one can occur. 4606 4607 I have refactored the code so chend actually points to the last byte 4608 that can be written to without an out of boundary access. As it is not 4609 possible to achieve "ch + length < chend" and "ch + length + 1 > chend" 4610 with the corrected chend meaning, I removed the inner if-check. 4611 4612 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 4613 4614commit d81da209fd4d0c2c9ad0596a8078e58864479d0d 4615Author: Tobias Stoeckmann <tobias@stoeckmann.org> 4616Date: Tue Jul 3 22:31:37 2018 +0200 4617 4618 Validation of server response in XListHosts. 4619 4620 If a server sends an incorrect length in its response, a client is prone 4621 to perform an out of boundary read while processing the data. 4622 4623 The length field of xHostEntry is used to specify the amount of bytes 4624 used to represent the address. It is 16 bit, which means that it is not 4625 possible to perform an arbitrary memory access, but it might be enough 4626 to read sensitive information, e.g. malloc-related pointers and offsets. 4627 4628 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 4629 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 4630 4631commit b676e62377483df77bcb6472d26b24f901323fa9 4632Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 4633Date: Wed Jun 13 15:46:58 2018 +0200 4634 4635 XkbOpenDisplay.3: fix typo 4636 4637 XkbOpenDisplay returns a pointer to Display, not a Display. 4638 4639 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 4640 4641commit 9c5845ff0dc080ff10bd68af4fc40fcd805728ca 4642Author: Martin Natano <natano@natano.net> 4643Date: Sat Oct 8 19:57:50 2016 +0200 4644 4645 Don't rebuild ks_tables.h if nothing changed. 4646 4647 ks_tables.h is always considered out of date due to the forced rebuild 4648 of the makekeys util. This means the file is also rebuilt during 'make 4649 install', which is usually performed as root, which can to lead 4650 permission problems later on. 4651 4652 Signed-off-by: Martin Natano <natano@natano.net> 4653 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4654 4655commit 796f754cba6d75b676a0fc39b97802198fceda4f 4656Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4657Date: Sat May 5 14:43:30 2018 -0700 4658 4659 Change fall through comment in lcDB.c to match gcc's requirements 4660 4661 Needs to match one of the regexps shown under 4662 https://gcc.gnu.org/onlinedocs/gcc-7.3.0/gcc/Warning-Options.html#index-Wimplicit-fallthrough 4663 4664 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4665 4666commit 82ca6308757126fa7ffc6588f1e5d8e3be04251b 4667Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4668Date: Sat Mar 24 19:45:14 2018 -0700 4669 4670 Use size_t for buffer sizes in SetHints.c 4671 4672 These variables store values returned from strlen() as a size_t 4673 and are passed to Xmalloc, which expects a size_t, so lets stop 4674 converting back and forth to int along the way. 4675 4676 Reported by: Konstantin SKliarov 4677 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4678 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 4679 4680commit 50a5a98984097d614227d22a49265e12b172cac7 4681Author: Bhavi Dhingra <b.dhingra@samsung.com> 4682Date: Tue Jul 5 11:37:50 2016 +0530 4683 4684 Fix possible memory leak in cmsProp.c:140 4685 4686 https://bugs.freedesktop.org/show_bug.cgi?id=96814 4687 4688 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4689 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4690 4691commit a9dafdd57c71473fa3a2ec4887e973e4e9876d83 4692Author: Michal Srb <msrb@suse.com> 4693Date: Thu Mar 15 09:50:58 2018 +0100 4694 4695 Use flexible array member instead of fake size. 4696 4697 The _XimCacheStruct structure is followed in memory by two strings containing 4698 fname and encoding. The memory was accessed using the last member of the 4699 structure `char fname[1]`. That is a lie, prohibits us from using sizeof and 4700 confuses checkers. Lets declare it properly as a flexible array, so compilers 4701 don't complain about writing past that array. As bonus we can replace the 4702 XOffsetOf with regular sizeof. 4703 4704 Fixes GCC8 error: 4705 In function 'strcpy', 4706 inlined from '_XimWriteCachedDefaultTree' at imLcIm.c:479:5, 4707 inlined from '_XimCreateDefaultTree' at imLcIm.c:616:2, 4708 inlined from '_XimLocalOpenIM' at imLcIm.c:700:5: 4709 /usr/include/bits/string_fortified.h:90:10: error: '__builtin_strcpy' 4710 forming offset 2 is out of the bounds [0, 1] [-Werror=array-bounds] 4711 return __builtin___strcpy_chk (__dest, __src, __bos (__dest)); 4712 4713 Caused by this line seemingly writing past the fname[1] array: 4714 imLcIm.c:479: strcpy (m->fname+strlen(name)+1, encoding); 4715 4716 Reviewed-by: Keith Packard <keithp@keithp.com> 4717 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 4718 4719commit 34f4464f69a4d6ff0d1042500a62f9a5ed7f3647 4720Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4721Date: Tue Mar 6 11:42:27 2018 -0800 4722 4723 If XGetImage fails to create image, don't dereference it to bounds check 4724 4725 Reported by gcc 7.3: 4726 4727 GetImage.c:110:25: warning: potential null pointer dereference [-Wnull-dereference] 4728 if (planes < 1 || image->height < 1 || image->bytes_per_line < 1 || 4729 ~~~~~^~~~~~~~ 4730 4731 Introduced by 8ea762f94f4c942d898fdeb590a1630c83235c17 in Xlib 1.6.4 4732 4733 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4734 Reviewed-by: Emil Velikov <emil.velikov@collabora.com> 4735 4736commit e835a9dcc3362b5e92893be756dd7ae361e64ced 4737Author: wharms <wharms@bfs.de> 4738Date: Sun Sep 3 14:17:45 2017 +0200 4739 4740 silence gcc warning assignment discards 'const' qualifier from pointer target type 4741 4742commit 36a1ac0253fea82ff79cc52ba56c5691cfd07a3b 4743Author: wharms <wharms@bfs.de> 4744Date: Sun Aug 20 21:51:57 2017 +0200 4745 4746 remove empty line 4747 4748commit e02dfe54f32b4165351d2712a9d2e0584906a3ce 4749Author: wharms <wharms@bfs.de> 4750Date: Sun Aug 20 21:50:33 2017 +0200 4751 4752 add _X_UNUSED to avoid unused variable warnings 4753 4754commit 2911c39cecd63ed3747072a5eeeb9eedffc881e9 4755Author: walter harms <wharms@bfs.de> 4756Date: Sat Jun 4 17:19:59 2016 +0200 4757 4758 Fixes: warning: variable 'req' set but not,used 4759 4760 Fixes: warning: variable 'req' set but not used [-Wunused-but-set-variable] 4761 by marking req _X_UNUSED 4762 Solution was discussed on xorg-devel ML 4763 Peter Hutter, Alan Coopersmith 4764 Re: [PATCH libX11 3/5] fix: warning: pointer targets in passing argument 2 of '_XSend' differ in signedness [-Wpointer-sign] 4765 4766 Signed-off-by: harms wharms@bfs.de 4767 4768commit bf82ec0402479fd8399d69e7d62fc17d7956699a 4769Author: walter harms <wharms@bfs.de> 4770Date: Sat Jun 4 17:22:07 2016 +0200 4771 4772 mark _XDefaultIOError as no_return 4773 4774 mark _XDefaultIOError as no_return. No one comes back from exit() ... 4775 4776 Signed-off-by: harms wharms@bfs.de 4777 4778commit 9abe8380074edea3ac4e72466ec47e921ca05e47 4779Author: walter harms <wharms@bfs.de> 4780Date: Sat Jun 4 17:21:52 2016 +0200 4781 4782 no need to check XFree arguments 4783 4784 You can save a bit of code. The is no need to check XFree arguments bring free_fontdataOM in line with other free function and check for NULL arg 4785 4786 Signed-off-by: harms wharms@bfs.de 4787 4788commit 433477fcb7e07d0c26a22ba78aae88827ed1f440 4789Author: walter harms <wharms@bfs.de> 4790Date: Tue Apr 26 16:32:20 2016 +0200 4791 4792 fix memleak in error path 4793 4794 free all mem on error 4795 Signed-off-by: walter harms <wharms@bfs.de> 4796 4797commit ed9f0d34abc645eee56e21863f23acb4d0bb8e9a 4798Author: walter harms <wharms@bfs.de> 4799Date: Tue Apr 26 17:58:16 2016 +0200 4800 4801 fix memleak in error path 4802 4803 V2: remove unneeded NULL (reported by eric.engestrom@imgtec.com) 4804 4805 fix mem leak in error path 4806 Signed-off-by: walter harms <wharms@bfs.de> 4807 4808commit 7c78fc57693afa94cf26170f0f6276e3b7374ed0 4809Author: walter harms <wharms@bfs.de> 4810Date: Tue Apr 26 16:34:11 2016 +0200 4811 4812 no need to check args for Xfree() 4813 4814 simplify code 4815 4816 Signed-off-by: walter harms <wharms@bfs.de> 4817 4818commit c1c14af441ae73d1a8e67a971fafcf967e45ac48 4819Author: walter harms <wharms@bfs.de> 4820Date: Tue Apr 26 16:23:46 2016 +0200 4821 4822 remove stray extern 4823 4824 remove stray extern 4825 4826 Signed-off-by: walter harms <wharms@bfs.de> 4827 4828commit 714921f041a245dc5f37a689268b584226a2ccb9 4829Author: walter harms <wharms@bfs.de> 4830Date: Mon Apr 11 18:26:52 2016 +0200 4831 4832 no need to check argument for _XkbFree() 4833 4834 simplify code by removing unneeded checks 4835 4836 Signed-off-by: walter harms <wharms@bfs.de> 4837 4838commit d02c2466f65063a03c97dbcee05071c12a3676e6 4839Author: walter harms <wharms@bfs.de> 4840Date: Mon Apr 11 18:22:38 2016 +0200 4841 4842 fix more shadow warning 4843 4844 Signed-off-by: walter harms <wharms@bfs.de> 4845 4846commit 0355c3926d5372f9762f235071dbd94a89bbbdad 4847Author: walter harms <wharms@bfs.de> 4848Date: Thu Mar 31 19:16:33 2016 +0200 4849 4850 fix shadow char_size 4851 4852 Signed-off-by: walter harms <wharms@bfs.de> 4853 4854commit 916dffadf052135df3398651be873c353da629e1 4855Author: walter harms <wharms@bfs.de> 4856Date: Thu Mar 31 19:14:32 2016 +0200 4857 4858 remove argument check for free() adjust one inden 4859 4860 Signed-off-by: walter harms <wharms@bfs.de> 4861 4862commit 6ec901ebca3fea6a762e22090dc35b1b90911133 4863Author: walter harms <wharms@bfs.de> 4864Date: Thu Mar 31 19:12:17 2016 +0200 4865 4866 _XIOError(dpy); will never return so remore dead 4867 4868 Signed-off-by: walter harms <wharms@bfs.de> 4869 4870commit 83107a677b2ed458e4d62ea4a601e8181d3683d8 4871Author: walter harms <wharms@bfs.de> 4872Date: Thu Mar 31 19:10:49 2016 +0200 4873 4874 fix shadow warning 4875 4876 Signed-off-by: walter harms <wharms@bfs.de> 4877 4878commit 4fe66b1c5112b07bd09e28bbc021911d08a9621f 4879Author: Ryan C. Gordon <icculus@icculus.org> 4880Date: Wed Aug 2 02:41:03 2017 -0400 4881 4882 Valgrind fix for XStoreColor and XStoreColors. 4883 4884 If the "pad" field isn't set, Valgrind will report it as uninitialized 4885 memory accesses when the struct is copied into the Display's send buffer. 4886 4887 In practice, this is (probably) harmless, but Valgrind is correct in 4888 believing it's a bug. 4889 4890 https://bugs.freedesktop.org/attachment.cgi?id=133189 4891 4892 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4893 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4894 4895commit 7d2010fec25c2f52b873ad0572479eb43128b038 4896Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4897Date: Fri Apr 7 00:13:03 2017 -0700 4898 4899 Improve table formatting in XkbChangeControls & XkbKeyNumGroups man pages 4900 4901 Includes fix for Solaris Bug 24564279: "XkbKeyNumGroups.3x11 man page 4902 contains some malformed text" caused by extra whitespace after .TE macros 4903 4904 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4905 4906commit b856d5d929047d1ea169814d56e43784ea404c83 4907Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4908Date: Thu Mar 16 00:05:49 2017 -0700 4909 4910 Clarify state parameter to XkbSetNamedDeviceIndicator 4911 4912 Checking a Bool to see if it's NULL does not work well in C. 4913 Also reported in https://bugs.freedesktop.org/show_bug.cgi?id=251 4914 4915 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4916 Reviewed-by: Adam Jackson <ajax@redhat.com> 4917 4918commit c6dadd4cebd994aafb37a58b3adbaa82507c2d18 4919Author: Alan Coopersmith <alan.coopersmith@oracle.com> 4920Date: Wed Mar 15 23:50:26 2017 -0700 4921 4922 Make Xkb{Get,Set}NamedIndicator spec & manpages match code 4923 4924 The XKB Library spec and the man pages for XkbGetNamedIndicator & 4925 XkbSetNamedIndicator included a device_spec argument neither function 4926 takes, and do not include the XkbGetNamedDeviceIndicator & 4927 XkbSetNamedDeviceIndicator variants that do take it (along with two 4928 other arguments). 4929 4930 This updates them to match the interfaces the code has provided for 4931 decades. 4932 4933 This has been reported multiple times, so this fixes: 4934 https://bugs.freedesktop.org/show_bug.cgi?id=251 4935 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=729812 4936 Sun Bug 4528016 XkbSetNamedIndicator & XkbGetNamedIndicator man pages are wrong 4937 (filed: alan.coopersmith@sun.com 2001-11-15 - now aka Oracle bug 15087506) 4938 X.Org Group Defect Id #9418 4939 4940 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 4941 Reviewed-by: Adam Jackson <ajax@redhat.com> 4942 4943commit 2d20890e7ffd3ee88a9ceb25cdd2ac1fe7aaceb6 4944Author: Arthur Huillet <ahuillet@nvidia.com> 4945Date: Wed Feb 1 15:02:41 2017 +0100 4946 4947 _XDefaultError: set XlibDisplayIOError flag before calling exit 4948 4949 _XReply isn't reentrant, and it can lead to deadlocks when the default error 4950 handler is called: _XDefaultError calls exit(1). It is called indirectly by 4951 _XReply when a X protocol error comes in that isn't filtered/handled by an 4952 extension or the application. This means that if the application (or one of its 4953 loaded shared libraries such as the NVIDIA OpenGL driver) has registered any 4954 _fini destructor, _fini will get called while still on the call stack of 4955 _XReply. If the destructor interacts with the X server and calls _XReply, it 4956 will hit a deadlock, looping on the following in _XReply: 4957 4958 ConditionWait(dpy, dpy->xcb->reply_notify); 4959 4960 It is legal for an application to make Xlib calls during _fini, and that is 4961 useful for an OpenGL driver to avoid resource leaks on the X server side, for 4962 example in the dlopen/dlclose case. However, the driver can not readily tell 4963 whether its _fini is being called because Xlib called exit, or for another 4964 reason (dlclose), so it is hard to cleanly work around this issue in the driver. 4965 4966 This change makes it so _XReply effectively becomes a no-op when called after 4967 _XDefaultError was called, as though an XIOError had happened. The dpy 4968 connection isn't broken at that point, but any call to _XReply is going to hang. 4969 This is a bit of a kludge, because the more correct solution would be to make 4970 _XReply reentrant, maybe by broadcasting the reply_notify condition before 4971 calling the default error handler. However, such a change would carry a grater 4972 risk of introducing regressions in Xlib. 4973 4974 This change will drop some valid requests on the floor, but this should not 4975 matter, as it will only do so in the case where the application is dying: X will 4976 clean up after it once exit() is done running. There is the case of 4977 XSetCloseDownMode(RETAIN_PERMANENT), but an application using that and wishing 4978 to clean up resources in _fini would currently be hitting a deadlock, which is 4979 hardly a better situation. 4980 4981 Signed-off-by: Aaron Plattner <aplattner@nvidia.com> 4982 Reviewed-by: Jamey Sharp <jamey@minilop.net> 4983 4984commit 42f4d7af9cf6d1dbfa575552e057328b054a20c9 4985Author: Matt Turner <mattst88@gmail.com> 4986Date: Sat Feb 25 21:54:22 2017 -0800 4987 4988 libX11 1.6.5 4989 4990 Signed-off-by: Matt Turner <mattst88@gmail.com> 4991 4992commit a0da5835e8078445947c828fe2d86c1a31439012 4993Author: Adam Jackson <ajax@redhat.com> 4994Date: Tue Feb 14 15:33:29 2017 -0500 4995 4996 Revert "Compose sequences for rouble sign" 4997 4998 This reverts commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0. 4999 5000 Reported to break 'make check': 5001 5002 https://lists.freedesktop.org/archives/xorg-devel/2017-February/052720.html 5003 5004commit d9e34061307748cb7318ed6b5f83ee5ee9b81fd0 5005Author: Mihail Konev <k.mvc@ya.ru> 5006Date: Fri Feb 10 18:48:18 2017 +0500 5007 5008 Compose sequences for rouble sign 5009 5010 Cyrillic combinations mirror the Qwerty-Jcuken keyboard layout. 5011 Also add Cyrillic sequences for hryvnia sign. 5012 5013 Submitted-by: Victor V. Kustov <coyote@bks.tv> 5014 Reviewed-by: Victor V. Kustov <coyote@bks.tv> 5015 Signed-off-by: Mihail Konev <k.mvc@ya.ru> 5016 5017commit 23d9623c661694aba8cf1e8f277dffa7a86cf065 5018Author: Petr Písař <petr.pisar@atlas.cz> 5019Date: Sun Oct 30 12:49:11 2016 +0100 5020 5021 Revert cs_CZ.UTF-8 XLC_LOCALE to en_US.UTF-8 5022 5023 The cs_CZ.UTF-8/XLC_LOCALE is an empty file leading to unsupported cs_CZ.UTF-8 5024 locale and reporting this error: 5025 5026 Warning: locale not supported by Xlib, locale set to C 5027 5028 Therefore this patch reverts to the en_US.UTF-8 definition file that was used 5029 before. This patch also deduplicates the cs_CZ.UTF-8 entry. 5030 5031 <https://bugs.freedesktop.org/show_bug.cgi?id=98219> 5032 5033 This reverts commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51 5034 5035 Signed-off-by: Julien Cristau <jcristau@debian.org> 5036 5037commit 71b0929ebc1f0f877f63e3f6de260f529daa6c69 5038Author: Mihail Konev <k.mvc@ya.ru> 5039Date: Thu Jan 26 13:52:49 2017 +1000 5040 5041 autogen: add default patch prefix 5042 5043 Signed-off-by: Mihail Konev <k.mvc@ya.ru> 5044 5045commit 2979011bc170c55894b9185b26376f8efc6db7d4 5046Author: Emil Velikov <emil.l.velikov@gmail.com> 5047Date: Mon Mar 9 12:00:52 2015 +0000 5048 5049 autogen.sh: use quoted string variables 5050 5051 Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent 5052 fall-outs, when they contain space. 5053 5054 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 5055 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 5056 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5057 5058commit 4a0082a1b6affa65d38294f0e13511525cd8ad15 5059Author: Peter Hutterer <peter.hutterer@who-t.net> 5060Date: Tue Jan 24 10:32:07 2017 +1000 5061 5062 autogen.sh: use exec instead of waiting for configure to finish 5063 5064 Syncs the invocation of configure with the one from the server. 5065 5066 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5067 Reviewed-by: Emil Velikov <emil.velikov@collabora.com> 5068 5069commit c74b070f2712c95f0db7c320a10232b0e5c83049 5070Author: Julien Cristau <jcristau@debian.org> 5071Date: Sat Jan 7 16:20:31 2017 +0100 5072 5073 Fix wrong Xfree in XListFonts failure path 5074 5075 'ch' gets moved inside the allocated buffer as we're looping through 5076 fonts, so keep a reference to the start of the buffer so we can pass 5077 that to Xfree in the failure case. 5078 5079 Fixes: commit 20a3f99eba5001925b8b313da3accb7900eb1927 "Plug a memory leak" 5080 5081 Signed-off-by: Julien Cristau <jcristau@debian.org> 5082 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5083 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5084 5085commit 663f47075fe67bab4f99bc5d186c49175d4a4334 5086Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5087Date: Sun Jan 1 20:57:58 2017 -0800 5088 5089 specs/libX11: Update Portability Considerations for the 21st century 5090 5091 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5092 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5093 5094commit 28f4b989b52fcf45c7e75a878d8d9c2583cd6b3c 5095Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5096Date: Sun Jan 1 20:39:53 2017 -0800 5097 5098 specs/libX11: Fix broken synopsis for Data/Data16/Data32 5099 5100 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5101 5102commit 382561951e3460b09c21a1a23748cde0315fbb19 5103Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5104Date: Sun Jan 1 20:31:54 2017 -0800 5105 5106 specs/libX11: Add missing parameter types for XGetWindowProperty() 5107 5108 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5109 5110commit 2beaecdb66965b861d6b790d151ba947f65f0a22 5111Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5112Date: Sun Jan 1 20:16:08 2017 -0800 5113 5114 specs/libX11: Make paramdef spacing more consistent 5115 5116 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5117 5118commit 4c436c6c14cfd4c397b011563bf13c2872861133 5119Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5120Date: Sun Jan 1 20:02:53 2017 -0800 5121 5122 specs/libX11: Fix paramdef entries listing multiple parameters 5123 5124 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5125 5126commit 1728b1a8a4718315da8e5c9cbc2c04bb75c74107 5127Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5128Date: Sun Jan 1 19:38:40 2017 -0800 5129 5130 specs/libX11: More synopsis fixes 5131 5132 Mostly transforming macro definitions and functions taking void arguments 5133 from undecorated <para> tags to use <funcsynopsis> tags to get decorations. 5134 5135 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5136 5137commit f0dc83db7e3a3d4a76c0f9d24763b80f01c893a9 5138Author: Lucien Gentis <lucien.gentis@waika9.com> 5139Date: Wed Jun 1 17:02:47 2016 +0200 5140 5141 Typos in "Xlib - C Language X Interface" document - Chapter 02 5142 5143 This patch fixes typos and lack of tags in "Xlib - C Language X Interface" document - Chapter 02. 5144 5145 Signed-off-by: Lucien Gentis <lucien.gentis@waika9.com> 5146 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5147 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5148 5149commit 20a3f99eba5001925b8b313da3accb7900eb1927 5150Author: Emilio Pozuelo Monfort <pochu@debian.org> 5151Date: Tue Oct 25 21:30:15 2016 +0200 5152 5153 Plug a memory leak 5154 5155 This was introduced in 8ea762f. 5156 5157 Reported-by: Julien Cristau <jcristau@debian.org> 5158 Signed-off-by: Emilio Pozuelo Monfort <pochu@debian.org> 5159 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5160 5161commit 8f349feac24aacc958bd816afcc52380764e3d92 5162Author: Matthieu Herrb <matthieu.herrb@laas.fr> 5163Date: Tue Oct 4 21:01:39 2016 +0200 5164 5165 libX11 1.6.4 5166 5167 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr> 5168 5169commit 8ea762f94f4c942d898fdeb590a1630c83235c17 5170Author: Tobias Stoeckmann <tobias@stoeckmann.org> 5171Date: Sun Sep 25 21:25:25 2016 +0200 5172 5173 Validation of server responses in XGetImage() 5174 5175 Check if enough bytes were received for specified image type and 5176 geometry. Otherwise GetPixel and other functions could trigger an 5177 out of boundary read later on. 5178 5179 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 5180 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 5181 5182commit 8c29f1607a31dac0911e45a0dd3d74173822b3c9 5183Author: Tobias Stoeckmann <tobias@stoeckmann.org> 5184Date: Sun Sep 25 21:22:57 2016 +0200 5185 5186 The validation of server responses avoids out of boundary accesses. 5187 5188 v2: FontNames.c return a NULL list whenever a single 5189 length field from the server is incohent. 5190 5191 Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org> 5192 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 5193 5194commit 78851f6a03130e3c720b60c3cbf96f8eb216d741 5195Author: walter harms <wharms@bfs.de> 5196Date: Mon Aug 15 19:18:14 2016 +0200 5197 5198 XFree will accept NULL as argument 5199 5200 since Xfree is a define for free(): 5201 Xlibint.h:#define Xfree(ptr) free((ptr)) 5202 5203 Xfree will accept NULL and do nothing. 5204 5205 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 5206 5207commit 83adf3d1e3d0d6602244381334f75c216da4ab6e 5208Author: Matthew D. Fuller <fullermd@over-yonder.net> 5209Date: Sat Jun 4 11:24:01 2016 -0500 5210 5211 Fixup param specification for XChangeProperty() 5212 5213 Signed-off-by: Matthew D. Fuller <fullermd@over-yonder.net> 5214 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5215 5216commit 3129c757f9da8586ab8b8654a56c8f687cc9ef5c 5217Author: Mats Blakstad <mats.gbproject@gmail.com> 5218Date: Sun Feb 28 13:22:03 2016 -0500 5219 5220 New compose keys for local languages in Togo 5221 5222 Signed-off-by: James Cloos <cloos@jhcloos.com> 5223 5224commit e1011b9e2f6c82255959cf3cc1d8cda402ded0a9 5225Author: Daniel Albers <daniel@lbe.rs> 5226Date: Wed Mar 9 14:35:48 2016 +0100 5227 5228 Add Compose sequence for U+1F4A9. 5229 5230 Signed-off-by: Daniel Albers <daniel@lbe.rs> 5231 5232commit 6d7bb040c928485f2557c2c914b95cffb2354179 5233Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5234Date: Sat Feb 6 14:18:32 2016 -0800 5235 5236 xcms: use size_t for pointer offsets passed to strncmp 5237 5238 instead of converting to int and back 5239 5240 Fixes clang warnings of the form: 5241 HVC.c:190:43: warning: implicit conversion changes signedness: 'int' to 5242 'unsigned long' [-Wsign-conversion] 5243 if (strncmp(spec, _XcmsTekHVC_prefix, n) != 0) { 5244 ~~~~~~~ 5245 5246 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5247 5248commit a9266804eed38a83897ab5f0f9f8a8ab82a98882 5249Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5250Date: Sat Feb 6 13:32:44 2016 -0800 5251 5252 xcms: use unsigned indexes when looping through unsigned values 5253 5254 Clears many gcc warnings of the form: 5255 5256 uvY.c: In function ‘XcmsCIEuvYToCIEXYZ’: 5257 uvY.c:263:19: warning: comparison between signed and unsigned integer 5258 expressions [-Wsign-compare] 5259 for (i = 0; i < nColors; i++, pColor++) { 5260 ^ 5261 5262 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5263 5264commit 0ee0d383b4488b7b90d8bd50b75c371e0dc0d397 5265Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5266Date: Sat Feb 6 13:01:25 2016 -0800 5267 5268 xcms: use size_t for strlen/sizeof values instead of converting to int & back 5269 5270 Fixes gcc warnings of the form: 5271 5272 IdOfPr.c: In function ‘XcmsFormatOfPrefix’: 5273 IdOfPr.c:69:32: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] 5274 if ((len = strlen(prefix)) >= sizeof(string_buf)) { 5275 ^ 5276 IdOfPr.c:83:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] 5277 if (len >= sizeof(string_buf)) Xfree(string_lowered); 5278 ^ 5279 IdOfPr.c:97:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] 5280 if (len >= sizeof(string_buf)) Xfree(string_lowered); 5281 ^ 5282 IdOfPr.c:104:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] 5283 if (len >= sizeof(string_buf)) Xfree(string_lowered); 5284 ^ 5285 5286 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5287 5288commit 4de6ed3e7b1833c52c9d58ab74d59d57ca2a9f0d 5289Author: Dominik Muth <nxdomainuser-muth@yahoo.com> 5290Date: Thu Mar 26 07:52:58 2015 +0100 5291 5292 Xlib.h: Fix macros imitating C functions. 5293 5294 The basic rule "put parantheses around macro parameters" should be 5295 observed where possible. Otherwise code like 5296 5297 ConnectionNumber(foo = bar); 5298 5299 fails to compile. (It obviously passes if ConnectionNumber is a C 5300 function.) There are several other macros amended for the same reason. 5301 5302 This bug appeared while building http://ioccc.org/1993/cmills.c, so 5303 historically it was not present. 5304 5305 Signed-off-by: Dominik Muth <muth@nxdomain.no-ip.biz> 5306 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5307 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5308 5309commit 3706b0f2b14cc97578a6bee620266edca2722ebf 5310Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5311Date: Fri Nov 15 18:03:25 2013 -0800 5312 5313 Don't need to link libX11-xcb against libX11 5314 5315 libX11-xcb only accesses data structures defined in X11 headers, 5316 it doesn't call any functions or reference any global variables 5317 in libX11 itself. (Seems to have been left from previous XCL 5318 implementation.) 5319 5320 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5321 5322commit eddf1bbd18872b286a9f939140f0cd9ba4e93804 5323Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5324Date: Fri Jan 22 11:44:25 2016 -0800 5325 5326 Stop checking for preferred order of local transports 5327 5328 Removes --with-local-transport-order=... flag to configure. 5329 5330 Code which used this ordered list was removed in commit 15e5eaf6289 5331 which outsourced X11 connection handling & authentication to libxcb. 5332 5333 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5334 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 5335 5336commit 1a66c1e964ff8d11382313404f48b5a3d5ed8be8 5337Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5338Date: Fri Jan 22 09:39:28 2016 -0800 5339 5340 Stop checking XTRANS_SECURE_RPC_FLAGS since we no longer use them 5341 5342 Removes --enable-secure-rpc & --disable-secure-rpc flags to configure 5343 5344 Code that used SECURE_RPC definitions was removed in commit 15e5eaf6289 5345 which outsourced X11 connection handling & authentication to libxcb. 5346 5347 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5348 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 5349 5350commit 7eb724dc24505f1591ef32620fa63f079b540646 5351Author: Olivier Fourdan <ofourdan@redhat.com> 5352Date: Thu Jan 21 11:54:19 2016 +0100 5353 5354 XKB: fix XkbGetKeyboardByName with Xming server 5355 5356 XkbGetKeyboardByName relies on flags to read the data from the server. 5357 5358 If the X server sends us the wrong flags or if a subreply is smaller 5359 than it should be, XkbGetKeyboardByName will not read all the available 5360 data and leave data in the buffer, which will cause the next _XReply() 5361 to fail with: 5362 5363 [xcb] Extra reply data still left in queue 5364 [xcb] This is most likely caused by a broken X extension library 5365 [xcb] Aborting, sorry about that. 5366 xcb_io.c:576: _XReply: Assertion `!xcb_xlib_extra_reply_data_left' failed. 5367 Aborted 5368 5369 Check if there is some extra data left at the end of 5370 XkbGetKeyboardByName() and discard that data if any is found. 5371 5372 Many thanks to Peter Hutterer <peter.hutterer@who-t.net> for finding the 5373 root cause of the issue and Adam Jackson <ajax@redhat.com> for helping 5374 with the analysis! 5375 5376 Signed-off-by: Olivier Fourdan <ofourdan@redhat.com> 5377 Reviewed-by: Daniel Stone <daniels@collabora.com> 5378 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 5379 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5380 5381commit 43ba0a68d3d17b496ec1f48d44921122ddd7d7d9 5382Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5383Date: Sat Dec 19 18:03:41 2015 -0800 5384 5385 lcPubWrap: replace malloc(strlen) + strcpy with strdup 5386 5387 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5388 Reviewed-by: Adam Jackson <ajax@redhat.com> 5389 5390commit 6fc95cb12b70c5a67cb4fc5e5749f9f1ec741e2a 5391Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5392Date: Sat Dec 19 10:21:04 2015 -0800 5393 5394 XlcDL.c: reduce code duplication 5395 5396 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5397 Reviewed-by: Adam Jackson <ajax@redhat.com> 5398 5399commit f7ecc0856be58608881d2086954cb71857ad64e1 5400Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5401Date: Sat Dec 19 10:19:25 2015 -0800 5402 5403 XlcDL.c: replace strcpy+strcat sequences with snprintf 5404 5405 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5406 Reviewed-by: Adam Jackson <ajax@redhat.com> 5407 5408commit 522989b34398bd6a6ea144c4af0ba69d6dc4faea 5409Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5410Date: Sat Dec 19 10:05:42 2015 -0800 5411 5412 XDefaultOMIF: Remove comments referring to ancient Sun bug ids 5413 5414 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5415 Reviewed-by: Adam Jackson <ajax@redhat.com> 5416 5417commit b738a104ae80e4270dd1d215ad0c6a80016982c2 5418Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5419Date: Sat Dec 19 10:00:22 2015 -0800 5420 5421 XDefaultOMIF: additional code simplification 5422 5423 Don't need to test for a case that we already returned for, don't need 5424 to store a count that will only ever be 1 if we didn't return, don't 5425 need to increment pointers to allow storing more than one item when we 5426 can only ever possibly do one. 5427 5428 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5429 Reviewed-by: Adam Jackson <ajax@redhat.com> 5430 5431commit 31011cf100419269eae7409581c784638be503cf 5432Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5433Date: Sat Dec 19 09:46:31 2015 -0800 5434 5435 XDefaultOMIF: replace strlen+Xmalloc+strcpy with strdup 5436 5437 Code seems to have been originally written to handle appending multiple 5438 strings, but only ever operates on a single string. 5439 5440 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5441 Reviewed-by: Adam Jackson <ajax@redhat.com> 5442 5443commit c27c46d5e22bbf60fb5608eaabe584b7fdeb0b09 5444Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5445Date: Sat Dec 19 09:20:55 2015 -0800 5446 5447 Use strdup instead of Xmalloc+strcpy in _XDefaultOpenIM 5448 5449 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5450 Reviewed-by: Adam Jackson <ajax@redhat.com> 5451 5452commit 4359dfabc04af082872d2bc2d5b52e26d6d93290 5453Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5454Date: Fri Dec 4 22:20:53 2015 -0800 5455 5456 Delete #if 0 hunks of code 5457 5458 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5459 5460commit a2f9dfac286f37e54eb47d4736cc3f0150224a84 5461Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5462Date: Thu Dec 3 23:38:07 2015 -0800 5463 5464 Bug 93183: _XDefaultOpenIM memory leaks in out-of-memory error paths 5465 5466 Rework code to store allocations directly into XIM struct instead of 5467 temporary local variables, so we can use _XCloseIM to unwind instead 5468 of duplicating it, and consistently jump to error handler on failure, 5469 instead of sometimes leaking and sometimes freeing. 5470 5471 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93183 5472 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5473 5474commit 07a97b3944467dce085a1efd24706cc851d2caf2 5475Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5476Date: Thu Dec 3 23:19:48 2015 -0800 5477 5478 Bug 93184: read_EncodingInfo invalid free 5479 5480 Free the correct bits of memory if we run out and need to unwind 5481 5482 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93184 5483 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5484 5485commit 11118e9eb3705fcbe42b6a68d4a8aa86ab0211f1 5486Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5487Date: Sat Nov 28 13:18:11 2015 -0800 5488 5489 Remove unused definition of XCONN_CHECK_FREQ 5490 5491 The only use of XCONN_CHECK_FREQ was removed in commit 15e5eaf62897b3179 5492 when we dropped the old Xlib connection handling in favor of xcb's. 5493 5494 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5495 Reviewed-by: Mark Kettenis <kettenis@openbsd.org> 5496 5497commit 5f0da8311a61498edf073cc877f5b467bfd5f863 5498Author: James Cloos <cloos@jhcloos.com> 5499Date: Thu Dec 3 18:24:44 2015 -0500 5500 5501 Fix another missing update in cf4d5989383a 5502 5503 Reported in: 5504 5505 https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7 5506 5507 Signed-off-by: James Cloos <cloos@jhcloos.com> 5508 5509commit 33840a5465a2e5fecab520bfbdd2d1bd0a456f51 5510Author: James Cloos <cloos@jhcloos.com> 5511Date: Thu Dec 3 18:15:40 2015 -0500 5512 5513 Fix missing update in cf4d5989383a 5514 5515 Reported in: 5516 5517 https://bugs.freedesktop.org/show_bug.cgi?id=81875#c7 5518 5519 Signed-off-by: James Cloos <cloos@jhcloos.com> 5520 5521commit dbcb847a08c44d99e4e1de2ba777d63238fb0e03 5522Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5523Date: Sun Sep 27 18:38:32 2015 -0700 5524 5525 Get rid of some extraneous ; at the end of C source lines 5526 5527 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5528 Reviewed-by: Thomas Klausner <wiz@NetBSD.org> 5529 5530commit 121a1bad334459f66f78bfca6df53dc841cf97f8 5531Author: Gunnar Hjalmarsson <gunnarhj@ubuntu.com> 5532Date: Wed Sep 23 11:44:55 2015 -0400 5533 5534 Add compose file for pt_PT similar to pt_BR 5535 5536 This is a forward of the Ubuntu bug https://launchpad.net/bugs/518056 5537 5538 One of the conclusions from the discussion on that bug report, which 5539 basically is about typing the ccedilla character easily on a non- 5540 Portuguese keyboard, is that X11 should include a compose file for 5541 pt_PT.UTF-8 similar to the file for pt_BR.UTF-8. 5542 5543 FDO bug: https://bugs.freedesktop.org/show_bug.cgi?id=90300 5544 5545 Signed-off-by: Gunnar Hjalmarsson <gunnarhj@ubuntu.com> 5546 Signed-off-by: James Cloos <cloos@jhcloos.com> 5547 5548commit 3f41d8a7f82eb5ffbd5c5d36472cf7043186b904 5549Author: Julien Cristau <jcristau@debian.org> 5550Date: Fri May 1 13:50:15 2015 +0200 5551 5552 Mark _XNextRequest as hidden 5553 5554 It's only used inside XNextRequest(), so doesn't need to be exported. 5555 5556 Signed-off-by: Julien Cristau <jcristau@debian.org> 5557 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5558 5559commit a72d2d06c002b644b7040a0a9936c8525e092ba8 5560Author: Christian Linhart <chris@demorecorder.com> 5561Date: Mon Sep 7 17:17:32 2015 +0200 5562 5563 fix for Xlib 32-bit request number issues 5564 5565 Make use of the new 64-bit sequence number API in XCB 1.11.1 to avoid 5566 the 32-bit sequence number wrap in libX11. 5567 5568 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71338 5569 Signed-off-by: Christian Linhart <chris@demorecorder.com> 5570 Signed-off-by: Olivier Fourdan <ofourdan@redhat.com> 5571 Reviewed-by: Adam Jackson <ajax@redhat.com> 5572 5573commit 58af066a764305c506efea7065ef7679369a1a98 5574Author: Thomas Klausner <wiz@NetBSD.org> 5575Date: Sun Jul 19 10:23:21 2015 +0200 5576 5577 Ignore test-driver (used by newer autoconf). 5578 5579 Signed-off-by: Thomas Klausner <wiz@NetBSD.org> 5580 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5581 5582commit 80b9a346b9ba200fa4652560282e80d249519287 5583Author: Thomas Klausner <wiz@NetBSD.org> 5584Date: Sun Jul 19 10:22:45 2015 +0200 5585 5586 Do not return() after exit(). 5587 5588 Signed-off-by: Thomas Klausner <wiz@NetBSD.org> 5589 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5590 5591commit c827edcd1c4a7f920aa25208083b5b58d60d2b44 5592Author: Ross Burton <ross.burton@intel.com> 5593Date: Mon May 18 14:49:01 2015 +0100 5594 5595 Add missing NULL checks to ICWrap 5596 5597 ICWrap.c dereferences the xim parameter passed in from client code without a 5598 NULL check. I have seen mplayer trigger this resulting in a segfault. In this 5599 case mplayer had called XOpenIM and NULL was returned which was later passed 5600 into XCreateIC. 5601 5602 Patch originally by Drew Moseley <drew_moseley@mentor.com>. 5603 5604 Signed-off-by: Ross Burton <ross.burton@intel.com> 5605 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5606 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5607 5608commit 26e0d2de294f8adf1ce65f1dbff0b59af41a00b9 5609Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5610Date: Thu Jun 4 20:51:17 2015 -0700 5611 5612 Replace Xmalloc+memset pairs with Xcalloc calls 5613 5614 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5615 5616commit f0286b2770ece10aef5e2e8c004260217f12fd25 5617Author: Bhavi Dhingra <b.dhingra@samsung.com> 5618Date: Thu Jun 4 19:07:12 2015 -0700 5619 5620 omGeneric.c: Correct the parameter usage of sizeof 5621 5622 Incorrect parameter usage with sizeof. Earlier passed argument FontData 5623 will be 4 bytes always as its a pointer hence the change is needed and 5624 FontDataRec should be used for memset. 5625 5626 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5627 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5628 5629commit 47da70d75f9e48e800719c0db752f9ccd2d77aea 5630Author: Peter Hutterer <peter.hutterer@who-t.net> 5631Date: Tue May 19 12:30:22 2015 +1000 5632 5633 Fix three "use of uninitialized variable" coverity warnings 5634 5635 False positive, if rlen/nbytes are unset we quit early before using it. Still, 5636 initialize it so we don't have to deal with these warnings again. 5637 5638 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5639 Reviewed-by: Hans de Goede <hdegoede@redhat.com> 5640 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5641 5642commit 19a30f17f30e9ae9641a7c0634fc52134208b060 5643Author: Peter Hutterer <peter.hutterer@who-t.net> 5644Date: Mon May 18 07:56:22 2015 +1000 5645 5646 Fix an indentation issue 5647 5648 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5649 Reviewed-by: Hans de Goede <hdegoede@redhat.com> 5650 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5651 5652commit 013ccece124b990217ad3bcf2c41688e8fda1df8 5653Author: Peter Hutterer <peter.hutterer@who-t.net> 5654Date: Mon May 18 07:55:17 2015 +1000 5655 5656 Fix potential memory leak 5657 5658 If we hit the depth limit, filename leaks. Move the depth check up before we 5659 allocate filename. 5660 Introduced in 226622349a4b1e16064649d4444a34fb4be4f464. 5661 5662 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5663 Reviewed-by: Hans de Goede <hdegoede@redhat.com> 5664 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5665 5666commit d3415d1f052530760b4617db45affcb984cfe35c 5667Author: Mike FABIAN <mfabian@redhat.com> 5668Date: Mon Apr 20 17:59:30 2015 +0200 5669 5670 Fix spelling mistake introduced by 748d47e69f5c12d8557d56a8a8ec166588da7b93 5671 5672 Sorry, my patch to fix the spelling mistakes in the ks_IN and sd_IN 5673 locales fixed it only partly, I introduced a new spelling mistake 5674 in the sd_IN locales. This patch fixes this. 5675 5676 Signed-off-by: James Cloos <cloos@jhcloos.com> 5677 5678commit 748d47e69f5c12d8557d56a8a8ec166588da7b93 5679Author: Mike FABIAN <mfabian@redhat.com> 5680Date: Wed Feb 19 11:46:45 2014 +0100 5681 5682 fix spelling mistakes in ks_IN and sd_IN devanagari locales 5683 5684 The codeset must be *before* the modifier. 5685 5686 See also: http://pubs.opengroup.org/onlinepubs/7908799/xbd/envvar.html 5687 5688 opengroup> The syntax for these environment variables is thus defined as: 5689 opengroup> 5690 opengroup> [language[_territory][.codeset][@modifier]] 5691 5692 Signed-off-by: James Cloos <cloos@jhcloos.com> 5693 5694commit c64fe5553aa4738f9d1d74a795f5651fbb7b1b09 5695Author: Mike FABIAN <mfabian@redhat.com> 5696Date: Wed Feb 19 11:50:55 2014 +0100 5697 5698 add be_BY.UTF-8@latin and sr_RS.UTF-8@latin to locale.dir 5699 5700 See also: https://bugzilla.redhat.com/show_bug.cgi?id=1066910 5701 5702 If these are not in locale.dir, 5703 5704 $ LANG=sr_RS.UTF-8@latin xterm 5705 5706 and 5707 5708 $ LANG=sr_RS@latin xterm 5709 5710 give the warning: 5711 5712 Warning: locale not supported by Xlib, locale set to C 5713 5714 and some programs (like xmms) fail to find translations for Serbian 5715 in Latin because of this. 5716 5717 Signed-off-by: James Cloos <cloos@jhcloos.com> 5718 5719commit c85be01b006126c4407eebd1eb6e01a17312b7b4 5720Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5721Date: Sun Mar 22 16:46:45 2015 -0700 5722 5723 Move Compose \ o / to be with other emoji compose sequences 5724 5725 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5726 5727commit 5a499ca7b064bf7e6a4fcc169f22862dce0c60c5 5728Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5729Date: Mon Mar 9 15:28:29 2015 -0700 5730 5731 libX11 1.6.3 5732 5733 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5734 5735commit c8e19b393defd53f046ddc2da3a16881221b3c34 5736Author: Joonas Javanainen <joonas.javanainen@gmail.com> 5737Date: Thu Feb 5 17:31:04 2015 +0200 5738 5739 Fix XErrorEvent struct field order in man page 5740 5741 In the man page the field "resourceid" was in a different place than 5742 in the actual struct layout in Xlib.h 5743 5744 Signed-off-by: Joonas Javanainen <joonas.javanainen@gmail.com> 5745 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5746 5747commit ddf3b09bb262d01b56fbaade421ac85b0e60a69f 5748Author: Ran Benita <ran234@gmail.com> 5749Date: Tue Feb 3 13:23:50 2015 -0500 5750 5751 compose: fix the description of modifiers in compose sequences 5752 5753 The Compose format has a feature which allows specifying certain 5754 modifiers must (or must not) be present with a given keysym in the 5755 sequence. 5756 5757 The grammar in imLcPrs.c and the Compose man page both do not match what 5758 the code actually does (see the handling of the variables 5759 `modifier_mask` and `modifier` in parseline() in imLcPrs.c, which are 5760 eventually matched as `ev->state & modifier_mask == modifier`). 5761 5762 Also explicitly list the accepted modifier names, since they are 5763 not standard (e.g. "Ctrl" instead of "Control"). 5764 5765 Signed-off-by: Ran Benita <ran234@gmail.com> 5766 Signed-off-by: James Cloos <cloos@jhcloos.com> 5767 5768commit 129f13f385c50e3d8b53ea7441b17386b0f36aeb 5769Author: Ran Benita <ran234@gmail.com> 5770Date: Tue Feb 3 13:23:49 2015 -0500 5771 5772 xkb: fix misleading comment about consumed modifiers 5773 5774 In the spec and the man page the `mods_rtrn` argument is described as 5775 "backfilled with unconsumed modifiers" but actually it is backfilled 5776 with the *consumed* modifiers. This is also mentioned a few lines below 5777 in each case. 5778 5779 Signed-off-by: Ran Benita <ran234@gmail.com> 5780 Signed-off-by: James Cloos <cloos@jhcloos.com> 5781 5782commit 446f5f7f41317a85a0cd0efa5e6a1b37bc99fba2 5783Author: Ingo Schwarze <schwarze@usta.de> 5784Date: Tue Dec 9 10:44:13 2014 +0100 5785 5786 Fix pasto in XkbGetKeyBehaviors(3) manual 5787 5788 Reviewed-by: Thomas Klausner <wiz@NetBSD.org> 5789 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 5790 5791commit f3831dde6972e4da9e018c6a5f4013d8756a5e78 5792Author: Benno Schulenberg <bensberg@justemail.net> 5793Date: Sun Nov 23 21:35:36 2014 +0100 5794 5795 nls: Sorting compose sequences rigorously in mirroring pairs, as is custom. 5796 5797 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5798 5799commit a51681b60c84109fe19f5d449e13080522499324 5800Author: Benno Schulenberg <bensberg@justemail.net> 5801Date: Thu Sep 19 16:42:01 2013 +0200 5802 5803 nls: Remove a duplicate locale name, and sort some others more strictly. 5804 5805 Also improve the grammar of the initial comment. 5806 5807 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5808 Signed-off-by: James Cloos <cloos@jhcloos.com> 5809 5810commit 426b7f850f5376db96a4b12420ee141603fcc3cd 5811Author: Benno Schulenberg <bensberg@justemail.net> 5812Date: Fri Sep 20 12:37:29 2013 +0200 5813 5814 nls: Add a comment to the block of accented Hebrew letters. 5815 5816 And align them in a nicer manner. 5817 5818 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5819 Signed-off-by: James Cloos <cloos@jhcloos.com> 5820 5821commit 7474c6f1ee78dd097b1d0b4c7e3e4ea41317e335 5822Author: Benno Schulenberg <bensberg@justemail.net> 5823Date: Fri Sep 20 12:05:09 2013 +0200 5824 5825 nls: Add one lowercase compose variant for ®, to mirror those for ©. 5826 5827 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5828 Signed-off-by: James Cloos <cloos@jhcloos.com> 5829 5830commit 18dcd13514fa538afefa78c93523d9dbd4688e74 5831Author: Benno Schulenberg <bensberg@justemail.net> 5832Date: Sat Sep 7 20:10:43 2013 +0200 5833 5834 nls: Adding the visual composing characters to two comment lines. 5835 5836 The lines around them also contain the characters in the comments. 5837 5838 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5839 Signed-off-by: James Cloos <cloos@jhcloos.com> 5840 5841commit 18d8307575af748913d5da17e3de45da2a22ede4 5842Author: Benno Schulenberg <bensberg@justemail.net> 5843Date: Sat Sep 7 20:09:32 2013 +0200 5844 5845 nls: Grouping a lone superscript minus together with its mates. 5846 5847 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5848 Signed-off-by: James Cloos <cloos@jhcloos.com> 5849 5850commit 33301cc45e6a2b8aa841ed6325547af970f8c4db 5851Author: Benno Schulenberg <bensberg@justemail.net> 5852Date: Sat Sep 7 19:53:38 2013 +0200 5853 5854 nls: Grouping the compose sequences for Dstroke/dstroke together. 5855 5856 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 5857 Signed-off-by: James Cloos <cloos@jhcloos.com> 5858 5859commit 78fe1becb6c18fa33c0d5f04005b45d856f8952a 5860Author: Gioele Barabucci <gioele@svario.it> 5861Date: Sun Sep 21 23:05:55 2014 +0200 5862 5863 Add compose sequence for U+20B9 INDIAN RUPEE SIGN 5864 5865 The compose sequence for the new Indian Rupee sign is modelled after 5866 the sequence for the Euro sign. 5867 5868 Signed-off-by: Gioele Barabucci <gioele@svario.it> 5869 Signed-off-by: James Cloos <cloos@jhcloos.com> 5870 5871commit 6101b967b641355dd863fd1ce52c6a7d58bcbe68 5872Author: Gabriel Souza Franco <gabrielfrancosouza@gmail.com> 5873Date: Thu Jul 31 22:23:28 2014 -0300 5874 5875 Add double-arrow compose sequence 5876 5877 Signed-off-by: Gabriel Souza Franco <gabrielfrancosouza@gmail.com> 5878 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5879 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5880 5881commit 368a6401c6a3275d3497fec38a3dcbc38cd9df60 5882Author: James Cloos <cloos@jhcloos.com> 5883Date: Fri Aug 1 18:30:42 2014 -0400 5884 5885 Add cs_CZ.UTF-8 locale to configure.ac 5886 5887 Commit cf4d5989383a should have included this. 5888 5889 Reported-by: Colin Harrison <colin.harrison@virgin.net> 5890 Signed-off-by: James Cloos <cloos@jhcloos.com> 5891 5892commit cf4d5989383acc4ed1b7eebadde9f380f2129766 5893Author: James Cloos <cloos@jhcloos.com> 5894Date: Tue Jul 29 15:02:56 2014 -0400 5895 5896 Add nls for cs_CZ.UTF-8 5897 5898 Based on the iso8859-2 compose, and a bug report by Vladimír Marek, 5899 override the en_US.UTF-8 use of <dead_caron> <u> to enter »ǔ« instead 5900 to enter »ů«, and likewise for the majuscule, for the Czech locale. 5901 5902 This evidently is the norm for Czech keyboards. 5903 5904 Fixes bz#81875. 5905 5906 The XI18N_OBJS and XLC_LOCALE.pre files are empty, as they are for 5907 several other locales. That may require an update. 5908 5909 Reported-by: Vladimír Marek <vlmarek@volny.cz> 5910 Signed-off-by: James Cloos <cloos@jhcloos.com> 5911 5912commit d9f569572bd14db31921471e7b877523b5cf1e4c 5913Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5914Date: Sat Jul 26 12:17:47 2014 -0700 5915 5916 Fix source paths for out-of-tree lintlib builds 5917 5918 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5919 5920commit 0885cad1e4a9ed57266582be320be55259c881bf 5921Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5922Date: Sun Jul 20 09:21:20 2014 -0700 5923 5924 specs/XKB: Markup fractions as <{super,sub}script> instead of <emphasis> 5925 5926 Matches the way they were styled in original doc, before conversion 5927 to DocBook. 5928 5929 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5930 5931commit d0a9e9d56bb003315787201ee525b4d00fd54e06 5932Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5933Date: Sat Jul 19 23:55:47 2014 -0700 5934 5935 specs/XKB: acknowledge my contributions 5936 5937 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5938 5939commit 9c0be82017f513e2eb63d59b095f1cf1955f2e2b 5940Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5941Date: Sat Jul 19 23:53:48 2014 -0700 5942 5943 specs/XKB: Trim leading spaces off text lines 5944 5945 perl -i -p -e 's{/\*(\S)}{/* \1}g;' *.xml 5946 5947 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5948 5949commit 75b0b10990f38d966c6fcc821bf15e58c5a90c91 5950Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5951Date: Sat Jul 19 20:30:55 2014 -0700 5952 5953 specs/XKB: Add olinks to libX11 for "X Library Functions Affected by Xkb" 5954 5955 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5956 5957commit 5009621799444e9d1d284719f871d00be13e7330 5958Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5959Date: Sat Jul 19 12:34:28 2014 -0700 5960 5961 specs/XKB: Fix miscelleanous typos & spelling errors 5962 5963 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5964 5965commit ac219bf1522a592bd3e3283b1a6ea3dfd2c3c48c 5966Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5967Date: Sat Jul 19 00:46:41 2014 -0700 5968 5969 specs/XKB: add some more indexterms by hand 5970 5971 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5972 5973commit 5c3aa4c69e65ecf2e56d5e26f3833fb5d31973c2 5974Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5975Date: Sat Jul 19 00:48:02 2014 -0700 5976 5977 specs/XKB: fixup various formatting issues in <programlisting>s 5978 5979 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5980 5981commit 70c648ff852fd9bc784967cfc77ea70bd7f14c8d 5982Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5983Date: Fri Jul 18 23:56:29 2014 -0700 5984 5985 specs/XKB: fixup various formatting issues in <literallayout>s 5986 5987 Including translating some that are really just lists into 5988 <simplelist> markup. 5989 5990 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5991 5992commit 94b56774784ac00b9db02403aecea10bb0814c10 5993Author: Alan Coopersmith <alan.coopersmith@oracle.com> 5994Date: Fri Jul 18 23:18:52 2014 -0700 5995 5996 specs/XKB: add some more links by hand 5997 5998 random bits where a link looked handy 5999 6000 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6001 6002commit 7a15a934cdb07ed1b991bd0ef633f32ee00b1833 6003Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6004Date: Fri Jul 18 23:16:31 2014 -0700 6005 6006 specs/XKB: add links for terms in definition list under figure 1.1 6007 6008 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6009 6010commit 92b86fc3c50fbb7ab2e36af10a2fb3fe6284f58c 6011Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6012Date: Fri Jul 18 22:52:16 2014 -0700 6013 6014 specs/XKB: add links to more tables listing section references 6015 6016 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6017 6018commit 06a4483a05053b4f8d8c0d4cc0513c68ea912676 6019Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6020Date: Fri Jul 18 22:49:34 2014 -0700 6021 6022 specs/XKB: Table 4.1: remove page numbers & unnecessary para tags 6023 6024 Page numbers refer to old doc format, didn't translate to new one 6025 6026 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6027 6028commit 61bd55c6d1ab211b89d604bd05555dc417f6e53d 6029Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6030Date: Sat Jul 19 11:55:51 2014 -0700 6031 6032 specs/XKB: remove unwanted white space around C -> struct references 6033 6034 perl -i -0 -p -e 's{\s*->\s*}{->}g' *xml 6035 6036 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6037 6038commit 2be0cc0b2abbcc98cfd150210dea415a04787251 6039Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6040Date: Fri Jul 18 22:26:20 2014 -0700 6041 6042 specs/XKB: replace -> with → when used as arrow, not in C structs 6043 6044 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6045 6046commit bf1f3d6f6f995303624679ae546f507c70967dc0 6047Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6048Date: Fri Jul 18 22:21:10 2014 -0700 6049 6050 specs/XKB: replace => with ⇒ for double arrows 6051 6052 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6053 6054commit c7ee427fc0a72abd4a4f147ab16a5d1128a6a2ba 6055Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6056Date: Sat Jul 19 11:33:35 2014 -0700 6057 6058 specs/XKB: Markup some ranges with – instead of - 6059 6060commit ec4075303c6c0d1d64bfe378e585968f9a137da7 6061Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6062Date: Fri Jul 18 22:10:45 2014 -0700 6063 6064 specs/XKB: Markup some subtractions with − instead of - 6065 6066 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6067 6068commit 441a267e461132a38abed205245f028686526f1d 6069Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6070Date: Fri Jul 18 22:04:40 2014 -0700 6071 6072 specs/XKB: make sure all files have DOCTYPEs so standard entities work 6073 6074 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6075 6076commit 88fd70bee410f290b4f540405fdc7ecd85c26f25 6077Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6078Date: Fri Jul 18 21:55:41 2014 -0700 6079 6080 specs/XKB: Markup quoted terms as <quote> instead of with "" 6081 6082 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6083 6084commit 3576587ff10334a8f48c34b4fe5b7e829dec9a1b 6085Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6086Date: Sat Jul 19 09:01:26 2014 -0700 6087 6088 specs/XKB: fixup newlines between tags and punctuation 6089 6090 Get rid of unwanted whitespace before punctuation by moving them to the 6091 lines with the tags, instead of before & after. 6092 6093 perl -i -0 -p -e 's{\>\s*\n([\.,;:])}{>\1\n}g' *xml 6094 6095 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6096 6097commit 6d5ec492cd28c206423337f926503349702af5a6 6098Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6099Date: Fri Jul 18 21:29:33 2014 -0700 6100 6101 specs/XKB: fixup newlines between tags and parens 6102 6103 Get rid of unwanted whitespace inside parens by moving them to the 6104 lines with the tags, instead of before & after. 6105 6106 perl -i -0 -p \ 6107 -e 's{(?<!--) \(\s*\n\<}{\n(<}g;' \ 6108 -e 's{\>\s*\n\)([\.,;]?)(?! [^\n]*--)}{>)\1\n}g' *xml 6109 6110 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6111 6112commit 59d688f4c787250e0b401a92b1db0437d8c60f2d 6113Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6114Date: Fri Jul 18 21:09:24 2014 -0700 6115 6116 specs/XKB: Markup key terms as <firstterm> instead of <emphasis> 6117 6118 Also add <indexterm> entries for most of them, to make their definitions 6119 or introductions easy to find from the index. 6120 6121 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6122 6123commit 861f3087ee0f501362a67501f384c2ca4c7bfe73 6124Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6125Date: Thu Jul 10 20:00:53 2014 -0700 6126 6127 specs/XKB: Manual fixup of type markup 6128 6129 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6130 6131commit 6b96259dabe52701fd1bcaa0625b574180c4e769 6132Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6133Date: Thu Jul 10 15:00:30 2014 -0700 6134 6135 specs/XKB: Manual fixup of parameter markup 6136 6137 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6138 6139commit 5526dce6812a84102f556fdde8f2b52b21c8bcdc 6140Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6141Date: Thu Jul 10 13:40:40 2014 -0700 6142 6143 specs/XKB: Manual fixup of struct name/field markup 6144 6145 Handles typos that caused the scripts to miss matches, misnamed structs, etc. 6146 6147 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6148 6149commit c36ee1a4db4e7876526190b8ab6b0da5867f76f7 6150Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6151Date: Wed Jul 9 23:26:37 2014 -0700 6152 6153 specs/XKB: Manual fixup of symbol name markup 6154 6155 Handles typos that caused the scripts to miss matches, misnamed masks, etc. 6156 6157 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6158 6159commit eb1453a0c69606b8af96b90ddccf1b93a069fb35 6160Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6161Date: Wed Jul 9 00:03:23 2014 -0700 6162 6163 specs/XKB: Manual fixup of function name markup 6164 6165 Handles typos that caused the scripts to miss matches, functions like 6166 malloc & free from other libraries, function name patterns, etc. 6167 6168 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6169 6170commit 252d99c87b60ac6f3f2b36f292f9b3880daabe26 6171Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6172Date: Thu Jul 10 19:42:31 2014 -0700 6173 6174 specs/XKB: Markup protocol requests as <systemitem> instead of <emphasis> 6175 6176 No great fit in DocBook, so follow what we used in Xlib spec. 6177 6178 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6179 6180commit ed60df10aad15057577d5714c955d22d2a446e51 6181Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6182Date: Thu Jul 10 15:45:57 2014 -0700 6183 6184 specs/XKB: Markup keyboard keys as <keycap> instead of <emphasis> 6185 6186 Also uses <guilabel> for LED names/labels, for lack of a better fit 6187 in DocBook. 6188 6189 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6190 6191commit f57b91ee497414083cc1bf481d28eb9ad9f965fb 6192Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6193Date: Thu Jul 10 15:41:19 2014 -0700 6194 6195 specs/XKB: Markup characters & strings as <literal> instead of <emphasis> 6196 6197 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6198 6199commit cfd4279c9b59d7e87c9f7c67692c87973adb7667 6200Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6201Date: Thu Jul 10 11:08:14 2014 -0700 6202 6203 specs/XKB: Markup structs as <struct{name,field}> instead of <emphasis> 6204 6205 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6206 6207commit a014bb4cc013b0d1b76524b7868e860c7e7ebc79 6208Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6209Date: Wed Jul 9 23:30:54 2014 -0700 6210 6211 specs/XKB: Markup symbol names in table entries too 6212 6213 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6214 6215commit 907f7ad7a7a977bf4f19daa3143d47c4d07ca33d 6216Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6217Date: Tue Jul 8 23:35:48 2014 -0700 6218 6219 specs/XKB: Markup *Ptr as <type> instead of <emphasis> 6220 6221 perl -i -p -e 's{<emphasis>(\w*Ptr)</emphasis>}{<type>\1</type>}g' *xml 6222 6223 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6224 6225commit 9f6c00629fd4b713082cc11f9150f7aafd272c89 6226Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6227Date: Tue Jul 8 23:34:29 2014 -0700 6228 6229 specs/XKB: Markup *Rec as <structname> instead of <emphasis> 6230 6231 perl -i -p -e \ 6232 's{<emphasis>(\w*Rec)</emphasis>}{<structname>\1</structname>}g' *xml 6233 6234 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6235 6236commit 83839e37802b8b752c77859a95de60ad757feb79 6237Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6238Date: Tue Jul 8 23:29:49 2014 -0700 6239 6240 specs/XKB: Markup XKB macros as <symbol> instead of <emphasis> 6241 6242 Performed via: 6243 perl -n -e 'printf "s{<emphasis>\\s*%s\\s*</emphasis>}{<symbol>%s</symbol>};\n", $1, $1 if m{^#define\s+([^\s\(]*)}' \ 6244 /usr/include/X11/extensions/XK*h /usr/include/X11/XKBlib.h \ 6245 | sort -u > xkb-defines.pl 6246 perl -i -p -f xkb-defines.pl *.xml 6247 6248 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6249 6250commit 9e397ed37ce4cc70621de347de3a795df88d4506 6251Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6252Date: Tue Jul 8 22:53:12 2014 -0700 6253 6254 specs/XKB: Use ° instead of o for degrees. 6255 6256 Conversion from framemaker turned superscripted "o" into plain "o". 6257 6258 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6259 6260commit 3b8364c21f5119105a2c14ae8cc75a11494cb7a7 6261Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6262Date: Tue Jul 8 22:46:03 2014 -0700 6263 6264 specs/XKB: Markup keysyms as <keysym> instead of <emphasis> 6265 6266 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6267 6268commit 33bef065683c8f910f3722730503c0c0699ee8be 6269Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6270Date: Tue Jul 8 22:26:56 2014 -0700 6271 6272 specs/XKB: Markup NULL as <symbol> instead of <emphasis> 6273 6274 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6275 6276commit f10aa1e09468bd28454d85ac8ab55d9dc7178029 6277Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6278Date: Tue Jul 8 22:22:29 2014 -0700 6279 6280 specs/XKB: Markup function args as <parameter> instead of <emphasis> 6281 6282 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6283 6284commit 79ba58cceb69521fcb313c69233cf93a9fcb177f 6285Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6286Date: Mon Jul 7 23:56:13 2014 -0700 6287 6288 specs/XKB: Convert remaining error names to errorname tags 6289 6290 Most were caught by applying libX11 lists, but BadKeyboard & XKB*_Bad* 6291 are XKB-specific. (Plus some were badly split across tag boundaries.) 6292 6293 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6294 6295commit 42b2f5388c399949ece377f9cc9c479c06964972 6296Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6297Date: Mon Jul 7 23:16:56 2014 -0700 6298 6299 specs/XKB: manually fixup some more emphasis tagging mismatches 6300 6301 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6302 6303commit 3b2f47d44a55d93c65455ff183f3b47da04b1de1 6304Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6305Date: Mon Jul 7 23:02:31 2014 -0700 6306 6307 specs/XKB: re-normalize <emphasis> layout in xml files 6308 6309 Same script as before, just with <!-- xref --> comments out of the way 6310 now. 6311 6312 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6313 6314commit ce95f152eda509263874c53fb7c6b4a6bdab2c29 6315Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6316Date: Mon Jul 7 22:59:25 2014 -0700 6317 6318 specs/XKB: Remove remaining xref comments 6319 6320 All the places marked by these have been turned into the appropriate 6321 link, xref, or olink tags now. 6322 6323 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6324 6325commit 6590b66e19af8dff68888ac403ac82f5d585b4e2 6326Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6327Date: Mon Jul 7 22:56:36 2014 -0700 6328 6329 specs/XKB: make olinks to xkbproto for references in section 10.13 6330 6331 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6332 6333commit fcda446877a62e7443d7bc704ba3610e90d1e755 6334Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6335Date: Mon Jul 7 22:22:28 2014 -0700 6336 6337 specs/XKB: Add <figure> tags and make Figure references link to them 6338 6339 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6340 6341commit 087a2338476719e340dc3d5af0df6fdc4a26ce7a 6342Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6343Date: Mon Jul 7 21:30:01 2014 -0700 6344 6345 specs/XKB: Turn Table references into links 6346 6347 Adds id attributes to all table tags so we can link to them 6348 6349 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6350 6351commit 135fa07b74cb50172c6a75768d499cd87ddb336e 6352Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6353Date: Sun Jul 6 21:29:59 2014 -0700 6354 6355 specs/XKB: Turn section references into xref links 6356 6357 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6358 6359commit 53e931d79926af8a3996253efd8b5f6c21d9e5d7 6360Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6361Date: Sun Jul 6 21:17:18 2014 -0700 6362 6363 specs/XKB: Turn Chapter references into xref links 6364 6365 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6366 6367commit b7f00ce5bcb0c00696bb82503ab548e14f04d17d 6368Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6369Date: Sun Jul 6 20:40:18 2014 -0700 6370 6371 specs/XKB: Apply <emphasis> to semantic tag transformations from Xlib spec 6372 6373 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6374 6375commit b00a7ddff2744238fbfe31c2298b02028a45a6ff 6376Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6377Date: Sun Jul 6 20:34:51 2014 -0700 6378 6379 specs/XKB: Markup function names as <function> instead of <emphasis> 6380 6381 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6382 6383commit bfbb58b7679221cb5c9212665209ea9099ad079a 6384Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6385Date: Sun Jul 6 20:25:46 2014 -0700 6386 6387 specs/XKB: normalize <emphasis> layout in xml files 6388 6389 - Stop placing <emphasis> on empty space, commas, and periods. 6390 - Move periods & commas after closing </emphasis> tag 6391 - move <emphasis> open & close tags to same line, instead of mirroring 6392 nroff layout. 6393 6394 Simplifies automating further transformations of these tags. 6395 6396 Performed via: 6397 perl -i -0 -p \ 6398 -e 's{<emphasis>(\s*)</emphasis>}{}msg;' \ 6399 -e 's{<emphasis>([\s\.,]*)</emphasis>\s*}{\1}msg;' \ 6400 -e 's{\n([\.,])\s*}{\1\n}msg;' \ 6401 -e 's{([^\.])([\.,])\s*</emphasis>}{\1</emphasis>\2}msg;' \ 6402 -e 's{\s*<emphasis>\n\s*}{\n<emphasis>}msg;' *xml 6403 6404 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6405 6406commit b16ee69a0103109a661a88140a1765dcd7bda634 6407Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6408Date: Sun Jul 6 15:57:41 2014 -0700 6409 6410 specs/XKB: Convert to funcsynopsis+variablelist instead of informaltable 6411 6412 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6413 6414commit b41d43d4cf0c0a1a049a171ee8cf6fd8a3ee4335 6415Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6416Date: Sun Jul 6 19:29:38 2014 -0700 6417 6418 specs/XKB: Add index 6419 6420 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6421 6422commit 72ae1d793be078db521dda60af578ece71f364de 6423Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6424Date: Sun Jul 6 13:21:40 2014 -0700 6425 6426 specs/XKB: Fix various markup issues in functiondecl tables 6427 6428 - Merge some functionargdecl entries incorrectly split across rows 6429 - Add missing parameter name markup to some functionargdecls 6430 - Add missing function prototype markup to a functiondecl 6431 - Remove stray emphasis tags in a functiondecl 6432 6433 Allows them to correctly convert to funcsynopsis markup in next step. 6434 6435 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6436 6437commit 9fdb973012de80ac60dbc59c39162f4e839fc5a4 6438Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6439Date: Sun Jul 6 12:15:23 2014 -0700 6440 6441 specs/XKB: Convert header filenames to filename tags 6442 6443 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6444 6445commit 5525e8433f93bce464412f27cffa203ea628f368 6446Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6447Date: Tue Jul 8 00:05:32 2014 -0700 6448 6449 specs/libX11: disengender a user reference 6450 6451 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6452 6453commit d8679eae9317b389ad4acb0430360ee0663e2af8 6454Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6455Date: Fri Jul 11 18:41:42 2014 -0700 6456 6457 specs/libX11: Correct value of IconicState to match Xutil.h 6458 6459 Xutil.h has always had a value of 3 for IconicState, since 2 was 6460 previously used for the long-obsolete ZoomState, so make the spec 6461 match what programs have used for decades. 6462 6463 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6464 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 6465 6466commit 7ce2b0f12a48fb832f457cbafb0e1144ef557f9a 6467Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6468Date: Fri Jul 11 10:34:08 2014 -0700 6469 6470 Use C99 named initializers to fill in events passed to XSendEvent 6471 6472 Forces compiler to zero-fill unset fields in the struct (fixing bug 81236) 6473 and allows optimizer to order field initialization to best fit cache layout 6474 or other considerations. 6475 6476 Before & after output of gcc -S on AMD64 shows insertion of "rep stosq" 6477 instructions to rapidly zero-fill structs. 6478 6479 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6480 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 6481 6482commit 169805e1dc8743b37b00e24cf3a5eb8748f733ad 6483Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6484Date: Sun Jul 6 11:13:49 2014 -0700 6485 6486 Fix validation of ctrls parameter to XkbGetPerClientControls() 6487 6488 Nothing in the XKB spec states that the memory pointed to by ctrls has to 6489 be initialized to any given value when passed to the function, only that 6490 it is set by the function to the values returned by the X server: 6491 http://www.x.org/releases/X11R7.7/doc/libX11/XKB/xkblib.html#The_Miscellaneous_Per_client_Controls 6492 6493 The check for the incoming value seems to be copied from 6494 XkbSetPerClientControls without explanation. 6495 6496 Instead change it to checking if ctrls is non-NULL, since there's no 6497 point asking the X server to return a value the caller won't even see. 6498 6499 Found while investigating report from cppcheck-1.65: 6500 [src/xkb/XKB.c:699] -> [src/xkb/XKB.c:719]: (warning) Possible null pointer 6501 dereference: ctrls - otherwise it is redundant to check it against null. 6502 6503 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6504 6505commit 1e362fac92c6688fb42b195ccad16d7a337a34c1 6506Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6507Date: Sun Jul 6 10:54:57 2014 -0700 6508 6509 Fix map->num_types check in XkbAddKeyType() 6510 6511 Check is intended to ensure we allocate at least XkbNumRequiredTypes 6512 in map, but was accidentally marked with a ! causing the wrong check. 6513 6514 Reported-by: Harms <wharms@bfs,de> 6515 Reviewed-by: Matthieu Herrb <matthieu@herrb.eu> 6516 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6517 6518commit ff9a5c199251a84fa59d14fd48dadb3f8920b54b 6519Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6520Date: Sun Jul 6 15:08:21 2014 -0700 6521 6522 specs/libX11: Add missing spaces to 'unsignedint' & 'unsignedlong' types 6523 6524 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6525 6526commit a06ea86773568926c36ae650b188fc818d540db7 6527Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6528Date: Sun Jul 6 15:04:27 2014 -0700 6529 6530 specs/libX11: Fix height & width in parameter lists to be two separate entries 6531 6532 "unsigned int width, unsigned int height", not a single parameter "height" 6533 of type "unsignedintwidth,". 6534 6535 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6536 6537commit e4db5e503682b3304fe82e4b17b419a8e0f0a9f2 6538Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6539Date: Sun Jul 6 14:38:10 2014 -0700 6540 6541 specs/libX11: Fix x & y in parameter lists to be two separate parameters 6542 6543 "int x, int y" not a single parameter y of type "intx" 6544 6545 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6546 6547commit aa8bda0db2c6d82515b90ceb4a7d6403e38895e9 6548Author: walter harms <wharms@bfs.de> 6549Date: Sat Jun 7 12:03:17 2014 +0200 6550 6551 lcDefConv.c: fix use before check 6552 6553 * Do not use variables before checked for NULL. 6554 * remove some superfluid spaces (Mark Kettenis) 6555 6556 Signed-off-by: Harms <wharms@bfs,de> 6557 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6558 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6559 6560commit c0670e5d3ae330e611ecb05303d579a4f8a3d114 6561Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6562Date: Mon Jun 16 15:16:10 2014 -0700 6563 6564 Start adding Unicode 7.0 support to compose table 6565 6566 New characters defined in http://www.unicode.org/charts/PDF/U1F300.pdf 6567 6568 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6569 6570commit a4679baaa18142576d42d423afe816447f08336c 6571Author: walter harms <wharms@bfs.de> 6572Date: Sat Jun 7 11:54:34 2014 +0200 6573 6574 rm redundant null checks 6575 6576 remove more redundant NULL checks 6577 note that _XkbFree() is really Xfree() 6578 6579 Signed-off-by: Harms <wharms@bfs,de> 6580 Reviewed-by: Rémi Cardona <remi@gentoo.org> 6581 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 6582 6583commit 602d7f5030fe93b2fe7f29fb7310deb6f50cb6df 6584Author: walter harms <wharms@bfs.de> 6585Date: Sat Jun 7 15:17:27 2014 +0200 6586 6587 libX11: rm redundante NULL checks 6588 6589 This patch removes the last remaining NULL checks for Xfree() 6590 6591 Signed-off-by: Harms <wharms@bfs,de> 6592 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 6593 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 6594 6595commit d81fed46144d089bdfa1d916a28dffc9ebffe1e4 6596Author: walter harms <wharms@bfs.de> 6597Date: Fri Jun 6 22:53:05 2014 +0200 6598 6599 Remove more redundant null checks before Xfree() 6600 6601 Signed-off-by: Harms <wharms@bfs,de> 6602 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6603 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6604 6605commit 0b7fd7dbec136bae317bd9a329309eaa089beee3 6606Author: walter harms <wharms@bfs.de> 6607Date: Thu Jun 5 18:37:40 2014 +0200 6608 6609 Remove redundant null checks before free 6610 6611 This patch removes some redundant null checks before free. 6612 It should not change the code otherwise. Be aware that this 6613 is only the first series. 6614 6615 Signed-off-by: Harms <wharms@bfs,de> 6616 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6617 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6618 6619commit 7d452fad5068ba92b964e92bc46708046f4044aa 6620Author: walter harms <wharms@bfs.de> 6621Date: Wed Jun 4 17:12:31 2014 +0200 6622 6623 libX11/lcUTF8.c fix: dereferenced before check 6624 6625 * Do not use variables before checked for NULL. 6626 6627 Signed-off-by: Harms <wharms@bfs,de> 6628 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6629 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6630 6631commit a3808f51517a720e7ff738208af60865779dd6ef 6632Author: walter harms <wharms@bfs.de> 6633Date: Wed Jun 4 17:10:20 2014 +0200 6634 6635 libX11/XKBNames.c fix: dereferenced before check 6636 6637 * Do not use variables before checked for NULL. 6638 6639 Signed-off-by: Harms <wharms@bfs,de> 6640 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6641 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6642 6643commit b3c9f6a17e430aabe16aecbe097f7312a0f6ff78 6644Author: walter harms <wharms@bfs.de> 6645Date: Wed Jun 4 17:08:12 2014 +0200 6646 6647 libX11/lcGenConv.c fix: dereferenced before check 6648 6649 * Do not use variables before checked for NULL. 6650 6651 Signed-off-by: Harms <wharms@bfs,de> 6652 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6653 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6654 6655commit c6bc28d433243c32b3f74955f1478451b4fd27b5 6656Author: Andreas Schwab <schwab@linux-m68k.org> 6657Date: Sun Jan 19 16:59:13 2014 +0100 6658 6659 Restore lost tabs in sed commands 6660 6661 Signed-off-by: Andreas Schwab <schwab@linux-m68k.org> 6662 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6663 6664commit e3dc0d17339e61eaf0b51b8907510984e3bf23cb 6665Author: Benno Schulenberg <bensberg@justemail.net> 6666Date: Thu Sep 19 11:01:11 2013 +0200 6667 6668 nls: Transform Brazilian compose file to an include plus three overrides. 6669 6670 Signed-off-by: James Cloos <cloos@jhcloos.com> 6671 6672commit 16c87dda4da2271aaecc5d8b6fe6ecd072cc584c 6673Author: Benno Schulenberg <bensberg@justemail.net> 6674Date: Fri Sep 6 12:10:01 2013 +0200 6675 6676 nls: Adding the missing compose sequences with <comma> for O with ogonek. 6677 6678 All other vowels with ogoneks can already be composed using <comma>. 6679 6680 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6681 Signed-off-by: James Cloos <cloos@jhcloos.com> 6682 6683commit 655b60f48376069750b151c46da836fdd411c83b 6684Author: Benno Schulenberg <bensberg@justemail.net> 6685Date: Fri Sep 6 12:04:24 2013 +0200 6686 6687 nls: Adding the missing postfix sequences for composing vowels with ogoneks. 6688 6689 It existed for lowercase i, but not for uppercase I nor the other vowels. 6690 6691 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6692 Signed-off-by: James Cloos <cloos@jhcloos.com> 6693 6694commit 7f8f9a36ef901f31279c385caf960a22daeb33fe 6695Author: Owen W. Taylor <otaylor@fishsoup.net> 6696Date: Fri May 9 18:21:05 2014 -0400 6697 6698 Fix XNextRequest() after direct usage of XCB 6699 6700 When XCB owns the X socket, dpy->request is not updated, so 6701 NextRequest() and XNextRequest() return the wrong value. There's 6702 nothing we can do to fix NextRequest() while retaining ABI compat, 6703 but change XNextRequest() to grab the socket back from XCB, 6704 updating dpy->request. 6705 6706 Signed-off-by: Owen W. Taylor <otaylor@fishsoup.net> 6707 Reviewed-by: Uli Schlachter <psychon@znc.in> 6708 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6709 6710commit 0f9e734ea96556fe750a4baf354d42d5a87bcd14 6711Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6712Date: Sun May 4 11:54:59 2014 -0700 6713 6714 Add missing .TE tags to end tables in Xkb man pages 6715 6716 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6717 6718commit 280274e5292e013b43e552274111fab434f5ed4e 6719Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6720Date: Sun May 4 11:23:17 2014 -0700 6721 6722 XCreateGC.man: simplify table to work with Solaris tbl 6723 6724 Having every table cell be a text diversion (T{...T}) was too much for 6725 Solaris tbl to handle, and thus "man XCreateGC" would print the error 6726 /usr/man/man3x11/XCreateGC.3x11: line 402: Too many text block diversions 6727 tbl quits 6728 and not display the table of mask bits or any text in the man page after 6729 that table. Since the #define column doesn't need special handling, 6730 making it not use text diversions brings the table under the tbl limit. 6731 6732 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6733 6734commit 93bb325a54025dd15f4744abce54b358960420f9 6735Author: James Cloos <cloos@jhcloos.com> 6736Date: Tue May 20 17:30:10 2014 -0400 6737 6738 Revert "nls: Adding compose sequences (with <parenleft> first) that GTK also has." 6739 6740 Parenleft is already in use for sequences of the form <(> <letter> <)> 6741 to generate circled letters. 6742 6743 Eg, <Multikey> <parenleft> <a> <parenright> generates ⓐ. 6744 6745 This reverts commit f020235f4bd91fb6eade82f8c9f7b85a57981768. 6746 6747 Signed-off-by: James Cloos <cloos@jhcloos.com> 6748 6749commit 060707851be918f2f507a26d17b016f764ddf2b4 6750Author: Benno Schulenberg <bensberg@justemail.net> 6751Date: Fri Sep 6 11:42:59 2013 +0200 6752 6753 nls: Adding accessible compose sequences for Ș and Ț (with comma below). 6754 6755 Compose sequences with <dead_belowcomma> exist, but very few keyboard 6756 layouts contain that symbol. So a more usual character is needed to be 6757 able to easily compose Ș, ș, Ț and ț. The semicolon is normally only 6758 used for composing letters with ogoneks -- but only vowels take ogoneks, 6759 so the character is free for consonants, and thus <semicolon> is used 6760 here to compose commas below. It is somewhat fitting, because on most 6761 Romanian keyboards the Ș is placed on that key, and the Ț next to it. 6762 6763 (Oh -- the more obvious sequences with <comma> were already taken for 6764 composing S and T with cedillas.) 6765 6766 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6767 Signed-off-by: James Cloos <cloos@jhcloos.com> 6768 6769commit ca435c2f753aa2961fb35ac448cdb2cc77112755 6770Author: Benno Schulenberg <bensberg@justemail.net> 6771Date: Tue Sep 3 21:10:33 2013 +0200 6772 6773 nls: Ordering some compose sequences in a more customary way. 6774 6775 The custom seems to be: pairing the ones that have only the sequence of 6776 two keys reversed, and putting the one with the diacritic first first. 6777 6778 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6779 Signed-off-by: James Cloos <cloos@jhcloos.com> 6780 6781commit f020235f4bd91fb6eade82f8c9f7b85a57981768 6782Author: Benno Schulenberg <bensberg@justemail.net> 6783Date: Tue Sep 3 20:44:42 2013 +0200 6784 6785 nls: Adding compose sequences (with <parenleft> first) that GTK also has. 6786 6787 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6788 Signed-off-by: James Cloos <cloos@jhcloos.com> 6789 6790commit bda0b3b5bd19154206dc40166364e73d4b6b1374 6791Author: Benno Schulenberg <bensberg@justemail.net> 6792Date: Tue Sep 3 11:14:16 2013 +0200 6793 6794 nls: Allowing Romanian Ă and ă to be composed also with lowercase <u>. 6795 6796 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6797 Signed-off-by: James Cloos <cloos@jhcloos.com> 6798 6799commit 8be4610939b833587954957f5963eb4191b43d19 6800Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6801Date: Thu Mar 13 23:22:48 2014 -0700 6802 6803 Fix "follwing" typo in en_US.UTF-8/Compose comment 6804 6805 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6806 6807commit 6f30e9034f29c3ae6ad7e617b3d5e903aa107b6a 6808Author: Reuben Thomas <rrt@sc3d.org> 6809Date: Mon Jan 27 14:18:24 2014 +0000 6810 6811 en_US.UTF-8/Compose: Fix apparent copy-paste-o, changing capital to small A. 6812 6813 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6814 6815commit d6bd988bc00494914b38b95ee5df77ac4f32f19f 6816Author: Peter Hutterer <peter.hutterer@who-t.net> 6817Date: Mon Mar 3 12:38:48 2014 +1000 6818 6819 man: fix man page for XkbGetMap 6820 6821 Returned structure must be freed with XkbFreeKeyboard(). 6822 6823 Reported-by: Morten Bøgeskov <mb@dbc.dk> 6824 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 6825 6826commit a6dcf2201a05adbff54122df05a1e6325936abb6 6827Author: Ran Benita <ran234@gmail.com> 6828Date: Tue Feb 11 13:26:16 2014 +0200 6829 6830 Remove dead USE_OWN_COMPOSE-protected code 6831 6832 The build doesn't provide any way to define this option. It also refers 6833 to files (imComp.h) and functions (e.g. XimCompInitTables(), 6834 XimCompProcessSym()) which are not found anywhere, and the ordinary 6835 Compose implementation in xim doesn't use any of it. 6836 6837 Signed-off-by: Ran Benita <ran234@gmail.com> 6838 Reviewed-by: Julien Cristau <jcristau@debian.org> 6839 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6840 6841commit b64bee2ddb7b96f00713a8b8435f11ad9ac1c9e0 6842Author: Ran Benita <ran234@gmail.com> 6843Date: Sun Feb 16 15:24:58 2014 +0200 6844 6845 nls: always use XCOMM instead of # for comments in Compose.pre files 6846 6847 Lines starting with # are considered as preprocessor directives in the 6848 .pre files. 6849 6850 Fixes warnings like: 6851 <stdin>:3:0: error: invalid preprocessing directive #Khmer 6852 6853 Signed-off-by: Ran Benita <ran234@gmail.com> 6854 Signed-off-by: James Cloos <cloos@jhcloos.com> 6855 6856commit 470e2289a3ebc59c5a35e54e1adeb0f261d5bf88 6857Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6858Date: Thu Feb 6 13:48:08 2014 -0800 6859 6860 Fix typos in Xrm.c comments 6861 6862 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6863 6864commit aacf95dacc7c598e7297894580d4d655593813b2 6865Author: Marko Myllynen <myllynen@redhat.com> 6866Date: Mon Jan 13 16:43:18 2014 +0200 6867 6868 Annotate Finnish Compose map with Unicode code points 6869 6870 Scripted annotation, no functional changes. 6871 6872 Signed-off-by: James Cloos <cloos@jhcloos.com> 6873 6874commit 20fdccd81b54678376d49e00edfebbbe94951f07 6875Author: Teemu Likonen <tlikonen@iki.fi> 6876Date: Fri Jul 20 19:21:04 2012 +0300 6877 6878 Fix "RING ABOVE" key in the Finnish compose file 6879 6880 The Finnish keyboard standard defines that <dead_abovering> <space> must 6881 insert the character U+02DA RING ABOVE. Currently the Finnish Compose 6882 file inserts U+00B0 DEGREE SIGN even though the line's comment says 6883 "RING ABOVE". This commit changes the character to U+02DA RING ABOVE. 6884 6885 Signed-off-by: Teemu Likonen <tlikonen@iki.fi> 6886 Signed-off-by: James Cloos <cloos@jhcloos.com> 6887 6888commit 8757e2ac8e04f2932ff437127f3e2ae9ac20c1d7 6889Author: Ran Benita <ran234@gmail.com> 6890Date: Wed Jan 29 02:11:47 2014 +0200 6891 6892 nls: remove duplicate 'ohorn' and 'uhorn' compose sequences 6893 6894 Since <Ohorn> == <U01A0> and <ohorn> == <U01A1>, when translated to 6895 keysyms: 6896 6897 #define XK_Ohorn 0x10001a0 /* U+01A0 LATIN CAPITAL LETTER O WITH HORN */ 6898 #define XK_ohorn 0x10001a1 /* U+01A1 LATIN SMALL LETTER O WITH HORN */ 6899 6900 (and similarly for uhorn), there is no need to have both names. Remove 6901 the unicode literal ones. 6902 6903 Signed-off-by: Ran Benita <ran234@gmail.com> 6904 Signed-off-by: James Cloos <cloos@jhcloos.com> 6905 6906commit b98998cb3bea7cb3005f2e9d5bc5332d14b1d5d3 6907Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6908Date: Sat Jan 18 22:35:18 2014 -0800 6909 6910 Add RANDR 1.4 requests & events to XErrorDB 6911 6912 Checked against randrproto.txt & randr.h 6913 6914 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6915 6916commit 321392ded15a7ee9d177d4ebe8846336ba76741c 6917Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6918Date: Fri Jan 3 20:04:33 2014 -0800 6919 6920 Remove unused ETEST & ESZTEST macros from XlibInt.c 6921 6922 Left behind when 15e5eaf62897 removed support for building without XCB. 6923 6924 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6925 Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net> 6926 6927commit 1ffc0c5503d4f419fdbc765243832a53491bf5bc 6928Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6929Date: Thu Dec 26 09:26:13 2013 -0800 6930 6931 _XkbReadGeomOverlay: check for NULL first, then use pointer 6932 6933 Flagged by cppcheck 1.62: 6934 [lib/libX11/src/xkb/XKBGeom.c:479] -> [lib/libX11/src/xkb/XKBGeom.c:480]: 6935 (warning) Possible null pointer dereference: row - otherwise it is 6936 redundant to check it against null. 6937 6938 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6939 6940commit ddf5f130cc29bb3bf8b9c757dcbac31bc56e9379 6941Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6942Date: Thu Dec 26 09:22:49 2013 -0800 6943 6944 XkbSelectEventDetails: remove unnecessary assignments 6945 6946 clear & selectAll are set to 0 already a few lines earlier, 6947 affectWhich is set to XkbMapNotifyMask a few lines later. 6948 None are used between the other assignments and the removed ones. 6949 6950 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6951 6952commit 3d69b0a83e62f8f6fbdd952fc49cdbdf8825e1e6 6953Author: Jon TURNEY <jon.turney@dronecode.org.uk> 6954Date: Thu Oct 7 18:46:08 2010 +0100 6955 6956 Don't try so hard to find a matching font with the given encoding 6957 6958 See http://sourceware.org/bugzilla/show_bug.cgi?id=10948 6959 6960 Currently, if the locale is UTF-8, no CJK fonts are installed, and someone 6961 does XCreateFontSet() with a font name of "*", we end up asking the server 6962 to list the (non-existent) fonts 11 times for each CJK encoding, which can 6963 take a while. 6964 6965 A * wildcard can match multiple components in a XLFD name in XListFonts(), 6966 so there's no need to try adding more than one to get a match. 6967 6968 We do try once with a leading '*-' in case the fontname isn't a full 6969 well-formed XLFD name, maybe even that isn't needed? 6970 6971 (See also http://invisible-island.net/xterm/xterm.faq.html#slow_menus) 6972 6973 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> 6974 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6975 6976commit 7e163300735d4bcd3386b86eec112acdad139c59 6977Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6978Date: Mon Dec 2 21:51:27 2013 -0800 6979 6980 unifdef -UISC 6981 6982 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6983 6984commit 1e43c262d13cab2b759665f9f13bdedbc7afbfd4 6985Author: Benno Schulenberg <bensberg@justemail.net> 6986Date: Thu Sep 19 13:20:05 2013 +0200 6987 6988 nls: Fix transposed locale identifier for Khmer. 6989 6990 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 6991 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6992 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 6993 6994commit 0e45f64766c0557c8e99a979c70ca6f55664dae7 6995Author: Alan Coopersmith <alan.coopersmith@oracle.com> 6996Date: Sat Nov 16 20:21:54 2013 -0800 6997 6998 Drop X_LOCALE fallback for OS'es without setlocale() 6999 7000 C89 or bust! This was documented as being needed for "only Lynx, 7001 Linux-libc5, OS/2" and has never been enabled in modular builds, 7002 since none of those platforms have had anyone step up to add support 7003 since the X11R7 conversion to autotools. 7004 7005 Mostly performed with unifdef -UX_LOCALE, followed by removal of files 7006 left without any purpose, and manual cleanup of remaining references. 7007 7008 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7009 Reviewed-by: Adam Jackson <ajax@redhat.com> 7010 7011commit 6cb02b166361200da35ba14f52cd9aaa493eb0ea 7012Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7013Date: Wed Oct 23 10:37:53 2013 -0700 7014 7015 Xcms file parsing should not require the impossible to succeed 7016 7017 The field2 helper function, to split lines from Xcms.txt files into 7018 two tab delimited fields, contained a check: 7019 7020 if ((*pBuf != '\n') || (*pBuf != '\0')) { 7021 return(XcmsFailure); 7022 7023 which would cause it to return failure unless *pBuf had a value that 7024 was simultaneously equal to both \n & \0, and no one wants to live in 7025 a world where that could ever be true. 7026 7027 This has gone unnoticed since 1991, since this only caused lines 7028 in Xcms.txt that started with whitespace to be rejected, but now 7029 gcc -Wlogicalop has brought it to our attention, and 7030 https://bugs.freedesktop.org/show_bug.cgi?id=70803 was filed. 7031 7032 Now that we see it, and cannot unsee it, we change it to use the 7033 same logic as the check at other points in this function, to return 7034 failure only if we hit \n or \0 before we find the first non-whitespace 7035 character, so that lines starting with whitespace will have the space 7036 skipped over to get to the color name to be defined. 7037 7038 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7039 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 7040 7041commit 18a5278b008e9faa59b346fcab18a8d74b875fda 7042Author: Gaetan Nadon <memsize@videotron.ca> 7043Date: Sat Sep 28 17:33:52 2013 -0400 7044 7045 makekeys: don't need to use target-specific CFLAGS 7046 7047 It's the only thing built in that directory, so we can use AM_CFLAGS 7048 and AM_CPPFLAGS as usual. 7049 7050 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7051 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7052 7053commit 5dcb40f28d59587597d2ff6e6ac64c71cfe6ff7b 7054Author: James Cloos <cloos@jhcloos.com> 7055Date: Tue Sep 17 12:50:42 2013 -0400 7056 7057 nls/en_US.UTF-8/Compose.pre: Fix typo. 7058 7059 Fix typo added in 215ce6a67863, s/actute/acute/. 7060 7061 Fixes bug #69476. Reported by Jean Krohn. 7062 7063 Signed-off-by: James Cloos <cloos@jhcloos.com> 7064 7065commit cb107760df33ffc8630677e66e2e50aa37950a5c 7066Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7067Date: Sun Sep 8 18:37:01 2013 -0700 7068 7069 libX11 1.6.2 7070 7071commit 215ce6a67863de7acfd6dd3562b4fd97ef87b411 7072Author: Benno Schulenberg <bensberg@justemail.net> 7073Date: Sun Sep 1 12:38:30 2013 +0200 7074 7075 nls: Adding more accessible compose sequences for J́ and j́. 7076 7077 Few keyboards have an <acute> key, so this adds the much more 7078 accessible and usual compose sequences with <apostrophe>, ánd 7079 the most comfortable ones with <dead_acute>. 7080 7081 Signed-off-by: Benno Schulenberg <bensberg@justemail.net> 7082 Signed-off-by: James Cloos <cloos@jhcloos.com> 7083 7084commit e9b14d10d0258bfcc273ff8bc84cd349dccda62c 7085Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7086Date: Sat Aug 24 17:27:43 2013 -0700 7087 7088 Bug 68413 - [Bisected]Error in `xterm': realloc(): invalid next size 7089 7090 Pass *new* size to realloc, not old size. 7091 7092 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7093 7094commit c2b8e30790c21d6386767265263b3294ce1b1f9a 7095Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7096Date: Fri Aug 16 21:04:02 2013 -0700 7097 7098 Stop checking for HAVE_DIX_CONFIG_H on the client side 7099 7100 Leftover from when these XKB files were shared with the server sources 7101 and could be compiled in either the client or server, with the different 7102 autoconf config files in each. 7103 7104 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7105 7106commit 84276609b2f0aec74fb464c428c7db5714b0fcfc 7107Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7108Date: Fri Aug 16 18:27:28 2013 -0700 7109 7110 Rearrange some variable declarations & initializations in XKB 7111 7112 Little things noticed during XKB restyling that seemed to make the 7113 code easier to read. 7114 7115 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7116 7117commit b90b7e859cf45ec76921fa21bbfc1f3840d6e8d1 7118Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7119Date: Sun Aug 11 13:29:33 2013 -0700 7120 7121 Reindent XKB code to X.Org standard style 7122 7123 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7124 7125commit c0a0f78eb49c2e4ad956209de77475c85b9314ea 7126Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7127Date: Fri Aug 16 18:14:14 2013 -0700 7128 7129 Fix overflow checks in _XkbReadKeySyms when key_sym_map is already created 7130 7131 We were checking to make sure that the largest keysym value was within 7132 the range of the allocated buffer, but checking against different limits 7133 in the not-yet-allocated vs. the already-allocated branches. 7134 7135 The check should be the same in both, and reflect the size used for the 7136 allocation, which is based on the maximum key code value, so we move it 7137 to be a common check, before we branch, instead of duplicating in each 7138 branch. 7139 7140 map->key_sym_map is an array of XkbSymMapRec structs, [0..max_key_code] 7141 map->syms is the array for which num_syms is recorded, hence is not the 7142 right value to check for ensuring our key_sym_map accesses are in range. 7143 7144 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7145 Reported-by: Barry Kauler <bkauler@gmail.com> 7146 Tested-by: Barry Kauler <bkauler@gmail.com> 7147 7148commit bea6cbd027973142fc64532274e1d16861b47190 7149Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7150Date: Sun Aug 11 17:02:21 2013 -0700 7151 7152 Remove long unused src/udcInf.c 7153 7154 I can find no record of what this file was for. Neither the X11R6.8.2 7155 monolith Imakefile nor any modular release Makefile.am have ever built 7156 it and nothing else references it. 7157 7158 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7159 7160commit bf3501e0395abe890acfea98fdd9f50a6966f118 7161Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7162Date: Sun Aug 11 00:07:33 2013 -0700 7163 7164 Remove unnecessary casts of pointers to (char *) in calls to Xfree() 7165 7166 Left one cast behind that is necessary to change from const char * 7167 to char * in src/xlibi18n/lcCharSet.c. 7168 7169 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7170 7171commit 6ead9dd92ab90aabd9f0e328d59597e6b5bc09d3 7172Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7173Date: Sat Aug 10 23:57:55 2013 -0700 7174 7175 Don't cast sizeof() results to unsigned when passing to Xmalloc/Xcalloc 7176 7177 sizeof() returns size_t, malloc() & calloc() expect sizes in size_t, 7178 don't strip down to unsigned int and re-expand unnecessarily. 7179 7180 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7181 7182commit 25a7a329def672fc8d26078538173777850c6390 7183Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7184Date: Sat Aug 10 23:51:08 2013 -0700 7185 7186 Remove even more casts of return values from Xmalloc/Xrealloc 7187 7188 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7189 7190commit f8fa16092a148b74ca35b4beb182053352606f2f 7191Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7192Date: Sat Aug 10 23:05:13 2013 -0700 7193 7194 xlibi18n: fix argsize argument to _XlcParsePath 7195 7196 The array is defined as having NUM_LOCALEDIR entries, so use that 7197 instead of hardcoded 256 value (the other two calls already did this). 7198 7199 Reported by parfait: 7200 Buffer overflow (CWE 120): In pointer dereference of argv[argc] with index argc 7201 Pointer size is 64 elements (of 8 bytes each), index is 255 7202 at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'. 7203 called at line 178 in function '_XlcParsePath' with argv = argv. 7204 called at line 722 in function '_XlcLocaleLibDirName' with argv = args, argsize = 256. 7205 at line 82 of src/xlibi18n/lcFile.c in function 'parse_line'. 7206 called at line 178 in function '_XlcParsePath' with argv = argv. 7207 called at line 638 in function '_XlcLocaleDirName' with argv = args, argsize = 256. 7208 7209 [ This bug was found by the Parfait 1.2.0 bug checking tool. 7210 http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ] 7211 7212 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7213 7214commit e7d46c6452c0b90fd66ae9f538546b968e0dd608 7215Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7216Date: Sat Aug 10 22:32:42 2013 -0700 7217 7218 i18n modules: Fix some const cast warnings 7219 7220 imRm.c: In function '_XimSetICMode': 7221 imRm.c:2419:37: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7222 imRm.c:2420:30: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7223 7224 lcGenConv.c: In function 'byteM_parse_codeset': 7225 lcGenConv.c:345:13: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7226 7227 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7228 7229commit cbd86eccf175dc82a5cbcea54c8bd21ce18b70c0 7230Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7231Date: Sat Aug 10 22:21:54 2013 -0700 7232 7233 xlibi18n: Fix a bunch of const cast warnings 7234 7235 Add const qualifiers to casts where needed, remove other casts that 7236 are no longer needed. 7237 7238 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7239 7240commit eb3676113fc2dd0f34d92b89beb81b3f61569aa1 7241Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7242Date: Sat Aug 10 22:18:00 2013 -0700 7243 7244 Fix const handling in XSetLocaleModifiers 7245 7246 Instead of reusing the input parameter to store the output, make a 7247 result variable instead, so that there's less const confusion. 7248 7249 Fixes gcc warnings: 7250 lcWrap.c: In function 'XSetLocaleModifiers': 7251 lcWrap.c:87:18: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7252 lcWrap.c:91:25: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7253 lcWrap.c:93:12: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 7254 7255 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7256 7257commit 8ebbffa98563960910152e4f2e31cb032375d871 7258Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7259Date: Sat Aug 10 21:46:37 2013 -0700 7260 7261 Constify lc_name argument to _XlcLocaleDirName() & _XlcLocaleLibDirName() 7262 7263 Makes code considerably less crufty and clears gcc warnings: 7264 XlcDL.c: In function '_XlcDynamicLoad': 7265 XlcDL.c:384:44: warning: cast discards '__attribute__((const))' qualifier 7266 from pointer target type [-Wcast-qual] 7267 XlcDL.c:386:51: warning: cast discards '__attribute__((const))' qualifier 7268 from pointer target type [-Wcast-qual] 7269 7270 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7271 7272commit 07e4e864107b38c2f393564fdacc90f4e858f23f 7273Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7274Date: Sat Aug 10 13:37:53 2013 -0700 7275 7276 init_om: remove unneeded extra copy of string to local buffer 7277 7278 Strings from the supported_charset_list[] were being copied one by 7279 one to a stack buffer, and then strdup called on that buffer. 7280 7281 Instead, just strdup the original string, without the local copy, 7282 and use a more traditional for loop, so it's easier to figure out 7283 what the code is doing (cleaning up a gcc const-cast warning in 7284 the process). 7285 7286 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7287 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7288 7289commit 1cec14dad904ba21a861f4af131be5982ecb83dd 7290Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7291Date: Sat Aug 10 12:34:53 2013 -0700 7292 7293 Delete unused XKB_INSURE_SIZE macro from XKBlibint.h 7294 7295 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7296 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7297 7298commit 5f32182c7c4045540ff3833c48ee24a3a25726e2 7299Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7300Date: Sat Aug 10 12:19:17 2013 -0700 7301 7302 miRegionOp(): ensure region size is not updated if realloc fails 7303 7304 This function performs operations on a region, and when finished, 7305 checks to see if it should compact the rectangle list. If the 7306 number of rectangles for which memory is allocated in the list is 7307 more than twice the number used, it tries to shrink. realloc() 7308 should not fail in this case, but if it does, might as well keep 7309 the correct value for the number of allocated rectangles, so we 7310 don't try to grow it unnecessarily later if adding to the region. 7311 7312 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7313 7314commit bd2a0b5a187798bb2e2f05dc5062ca79e37075dd 7315Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7316Date: Sat Aug 10 12:19:17 2013 -0700 7317 7318 miRegionCopy(): handle realloc failure better 7319 7320 Zero out the region size when freeing the region so callers don't think 7321 there's anything there. (Pointer is already set to NULL from the realloc 7322 result itself.) Return 0 to the callers, and have them cascade that back 7323 to their callers to indicate failure, instead of their usual return value 7324 of 1 on success. 7325 7326 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7327 7328commit 5dc8b5385d513bbda88697c2372db750d23f46d4 7329Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7330Date: Sat Aug 10 11:27:22 2013 -0700 7331 7332 Avoid memory leak/corruption if realloc fails in Xregion.h:MEMCHECK macro 7333 7334 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7335 7336commit 453c4ee436ef32d91501d7736d7a91c1aeafc565 7337Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7338Date: Sat Aug 10 12:07:51 2013 -0700 7339 7340 Avoid memory leak/corruption if realloc fails in imLcPrs.c:parseline() 7341 7342 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7343 7344commit b3fea74ec5b7d4f83755a52a8d49c564b71c6d12 7345Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7346Date: Sat Aug 10 12:30:39 2013 -0700 7347 7348 lcDB.c: ensure buffer size is updated correctly if realloc fails 7349 7350 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7351 7352commit 43bb822c714a73c3b2d15e621ffb3333cd10da8c 7353Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7354Date: Sat Aug 10 11:07:47 2013 -0700 7355 7356 Avoid memory leak/corruption if realloc fails in XlcDL.c:resolve_object() 7357 7358 Previously, if realloc failed to increase the size, we'd still 7359 record that we had allocated the larger size, but the pointer 7360 to it would be NULL, causing future calls to be broken, and the 7361 previous allocation to be lost/leaked. 7362 7363 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7364 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7365 7366commit 5d47a39978e92bb34ec928b1b15d71c0c2434870 7367Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7368Date: Fri Aug 9 23:33:03 2013 -0700 7369 7370 omGeneric.c: convert sprintf calls to snprintf 7371 7372 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7373 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7374 7375commit 88a27a2aa9b7d35cb79b16334ea3413e572b724a 7376Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7377Date: Fri Aug 9 23:30:30 2013 -0700 7378 7379 ximcp/imRm.c: convert sprintf calls to snprintf 7380 7381 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7382 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7383 7384commit 4fadae243fb485628c9a137f5da3489ed6214b21 7385Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7386Date: Fri Aug 9 23:02:12 2013 -0700 7387 7388 xlibi18n: convert sprintf calls to snprintf 7389 7390 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7391 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7392 7393commit 36a7edf0e5edfc5ef4ff2c3a8b4fa3dc4796e854 7394Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7395Date: Fri Aug 9 23:02:12 2013 -0700 7396 7397 lcfile: skip over any null entries in args list 7398 7399 Previous code seemed to assume that printf("%s", NULL) would result 7400 in a 0-length string, not "(null)" or similar, but since there's no 7401 point looking for files in "(null)/filepath...", instead we just 7402 skip over NULL entries in search paths when generating file names. 7403 7404 In the *DirName() functions, this effectively just moves the "bail on 7405 NULL in arg[i]" check up from the later code that assigned it to targetdir 7406 and then bailed if that was NULL. 7407 7408 Not sure how there ever could be a NULL in arg[i], given the current 7409 implementation of XlcParsePath, but it's easy enough to check once and 7410 reject up front instead of on every reference. 7411 7412 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7413 7414commit ee0824f24392d5ca3d5fd5f5ed8d78c0d892f7c0 7415Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7416Date: Fri Aug 9 22:00:09 2013 -0700 7417 7418 Fix file leak on malloc error in XlcDL.c:resolve_object() 7419 7420 File Leak: Leaked File fp 7421 at line 219 of lib/libX11/src/xlibi18n/XlcDL.c in function 'resolve_object'. 7422 fp initialized at line 198 with fopen 7423 7424 [ This bug was found by the Parfait 1.2.0 bug checking tool. 7425 http://labs.oracle.com/pls/apex/f?p=labs:49:::::P49_PROJECT_ID:13 ] 7426 7427 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7428 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7429 7430commit 9b291044a240e5b9b031ed814e0c84e53a1c3084 7431Author: Julien Cristau <jcristau@debian.org> 7432Date: Sat Jun 15 18:02:21 2013 +0200 7433 7434 Add missing locales to configure.ac 7435 7436 Commits 40761898692e5063957bfa2518cca3d35b2e354a and 7437 f198c6aa98f88ff285d903175a3c4c0fd33a4575 added two new locales 7438 (sr_CS.UTF-8 and km_KH.UTF-8), but didn't list them in configure.ac, 7439 meaning they're not included in tarballs. 7440 7441 Signed-off-by: Julien Cristau <jcristau@debian.org> 7442 Reviewed-by: James Cloos <cloos@jhcloos.com> 7443 7444commit 8f58e54a5f46c3cd4897a23b89950f4800ae38d4 7445Author: ISHIKAWA,chiaki <ishikawa@yk.rim.or.jp> 7446Date: Tue Dec 18 15:28:05 2012 +0000 7447 7448 Fix bogus timestamp generated by XIM 7449 7450 Fix bogus timestamp generted by XIM due to uninitialized 7451 data field. Also set appropriate serial, too. 7452 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39367 7453 7454 Signed-off-by: Chiaki ISHIKAWA <ishikawa@yk.rim.or.jp> 7455 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7456 7457commit e7fd6f0eda57300df4d6b695b7064610ca5dec57 7458Author: Egbert Eich <eich@freedesktop.org> 7459Date: Thu Jun 16 18:47:49 2011 +0200 7460 7461 XIM: Fix sync problem on focus change. 7462 7463 XSetICFocus() and XUnsetICFocus() are both asynchronous events. 7464 This is a pretty stupid idea: those functions may undo certain 7465 settings on the client side for which requests from the server 7466 may still be in the queue unprocessed. Thus things may be set 7467 in the wrong order ie instead of set -> unest it will be unset -> set. 7468 Moreover there is no way for either the client or the server to 7469 cause the event queue to be flushed - which is pretty bad as 7470 XIM is bidirectional. 7471 The scenario is as follows: 7472 Two ICs are created: 7473 ic1 = XCreateIC(im, 7474 XNInputStyle, XIMPreeditCallbacks | XIMStatusCallbacks, 7475 XNClientWindow, window, 7476 XNPreeditAttributes, preedit_attr, 7477 XNStatusAttributes, status_attr, 7478 NULL); 7479 ic2 = XCreateIC(im, XNInputStyle, 7480 XIMPreeditNothing | XIMStatusNothing, 7481 XNClientWindow, window, NULL); 7482 Then the focus is removed from ic2: 7483 XUnsetICFocus(ic2); 7484 If SCIM is used as the input server it will send a bunch of requests 7485 following an XCreateIC(). One of the requests registers a key release 7486 filter. XUnsetICFocus() unsets both key press and release filters. 7487 Since it is asynchronous, the input server requests to register key 7488 press and release filters may not have been processed, when XUnsetICFocus() 7489 is called. Since there is no explicite way for client programs to enforce 7490 the request queue to be flushed explicitely before an X[Set/Unset]ICFocus() 7491 call it would be safest to make those two calls synchronous in the sense 7492 that they ensure the request queue has been handled before they execute. 7493 The easiest way to do this from Xlib is thru a call to XGetICValues() 7494 which sends a request to the server and subsequently reads the queue 7495 from the server to the client. This will cause all outstanding requests 7496 in the queue to be read and handled. 7497 This is an ugly hack and this could be fixed directly in the client, 7498 however it seems to be easier to fix Xlib than to fix numerous clients. 7499 This problem arose since there is no well documented way how to handle 7500 and synchronize XIM requests and not all input servers send requests 7501 when an IC is created. 7502 This has been discussed extensively in: 7503 https://bugzilla.novell.com/show_bug.cgi?id=221326 7504 7505 Signed-off-by: Egbert Eich <eich@freedesktop.org> 7506 7507commit 26ec7d3821bc19debc73c8c3e42e6e33ef6f856e 7508Author: Egbert Eich <eich@freedesktop.org> 7509Date: Thu Jun 16 17:28:39 2011 +0200 7510 7511 XIM: Fix race on focus change: set 'FABRICATED' only when keyev filters in place. 7512 7513 When synthesized key events are sent on commit XIM sets the 'fabricated' 7514 flag so that the keypress handler knows that these were not real events. 7515 This also happens when committing due to the loss of focus. However in this 7516 case the keypress/release filters which consume and unset this flag are no 7517 longer in the filter chain. 7518 So the flag is erronously set when a real keyboard event is received after 7519 focus has been regained. So the first event is wrongly treated as a 7520 fabricated key in the keypress handler which will at the same time reset 7521 the flag so the second key event is treated correctly. 7522 This fix only sets the flag when at least one of the keyboard filters is in 7523 place. 7524 How to reproduce this bug: run scim, choose a Japanese input method start 7525 two instances of xterm: start typing in one xterm (this should pop up an 7526 IM window). Without comitting (hitting 'enter') move focus to the other 7527 xterm, then move focus back. Start typing again. The first character will 7528 be committed immediately without popping up an input window. 7529 With this fix this behavior is gone. 7530 7531 See also: https://bugzilla.novell.com/show_bug.cgi?id=239698 7532 7533 Signed-off-by: Egbert Eich <eich@freedesktop.org> 7534 7535commit 44f84223f5e2dd46883fcbd352af2798bfa9aeb6 7536Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7537Date: Mon Jul 29 21:29:49 2013 -0700 7538 7539 libX11 1.6.1 7540 7541 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7542 7543commit d19cfaca15826943d3c637ef7fa5db0a23d5feed 7544Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7545Date: Sat Jul 27 12:19:00 2013 -0700 7546 7547 Fix undefined XCMSDIR error when building lint library 7548 7549 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7550 7551commit 3083cd43d7dcd59da587975e7cadda681cd8a103 7552Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7553Date: Sat Jul 27 00:36:08 2013 -0700 7554 7555 Add ku_TR.UTF-8 (Kurdish language, Turkey region) to compose/locale.dir 7556 7557 Upstreaming from changes originally integrated into OpenSolaris 7558 under Sun bug id 6882572. 7559 7560 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7561 Reviewed-by: Thomas Klausner <wiz@NetBSD.org> 7562 7563commit 208e586c808e88a2ee819e4450dc27f557afc2bf 7564Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7565Date: Sat Jul 27 01:03:18 2013 -0700 7566 7567 omGeneric: remove space between struct name & member name 7568 7569 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7570 7571commit 7db74514e454d3fc4ff70aa08ddac66bfffda4dd 7572Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7573Date: Tue Jul 23 22:18:46 2013 -0700 7574 7575 Refactor common code from XAddHost & XRemoveHost into single function 7576 7577 On the Xlib side, the only real difference is the mode flag we send 7578 to the server with the address, so just make that an argument to the 7579 function with the common code for packing the address into the request. 7580 7581 (Aside from labels, gcc 4.7.2 generates identical code before & after 7582 this change due to inlining, verified via diff of gcc -S output.) 7583 7584 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7585 7586commit 3292195a64a9ce4f0d27134cd544651ec647e728 7587Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7588Date: Tue Jul 23 22:11:34 2013 -0700 7589 7590 XSetModifierMapping: Use Data instead of GetReqExtra 7591 7592 Handle arbitrary length data in the same fashion as other calls, 7593 avoiding need to ensure it fits all in the request buffer. 7594 7595 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7596 7597commit feb131b18aee31c2c125dc3275b0260940245882 7598Author: Kees Cook <kees@outflux.net> 7599Date: Sun Jun 9 11:13:43 2013 -0700 7600 7601 libX11: check "req" when calling GetReqExtra 7602 7603 This fixes the two callers of GetReqExtra to check that "req" is non-NULL 7604 to avoid crashing now that GetReqExtra does internal bounds-checking on 7605 the resulting buffer sizes. 7606 7607 Additionally updates comment describing return values to use names 7608 instead of only literal values. 7609 7610 Signed-off-by: Kees Cook <kees@outflux.net> 7611 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7612 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7613 7614commit 54540d7cba0c2bfe9176221c7bca910058d304df 7615Author: Kees Cook <kees@outflux.net> 7616Date: Sun Jun 9 11:13:42 2013 -0700 7617 7618 libX11: check size of GetReqExtra after XFlush 7619 7620 Two users of GetReqExtra pass arbitrarily sized allocations from the 7621 caller (ModMap and Host). Adjust _XGetRequest() (called by the GetReqExtra 7622 macro) to double-check the requested length and invalidate "req" when 7623 this happens. Users of GetReqExtra passing lengths greater than the Xlib 7624 buffer size (normally 16K) must check "req" and fail gracefully instead 7625 of crashing. 7626 7627 Any callers of GetReqExtra that do not check "req" for NULL 7628 will experience this change, in the pathological case, as a NULL 7629 dereference instead of a buffer overflow. This is an improvement, but 7630 the documentation for GetReqExtra has been updated to reflect the need 7631 to check the value of "req" after the call. 7632 7633 Bug that manifested the problem: 7634 https://bugs.launchpad.net/ubuntu/+source/x11-xserver-utils/+bug/792628 7635 7636 Signed-off-by: Kees Cook <kees@outflux.net> 7637 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7638 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7639 7640commit 24d3ee0d08f24e23c91d55702f010f73d7b908e5 7641Author: Thomas Klausner <wiz@NetBSD.org> 7642Date: Tue Jun 25 22:35:29 2013 +0200 7643 7644 Tighten out-of-range comparisons. 7645 7646 [For all of these, LONG_MAX was the correct value to prevent overflows 7647 for the recent CVEs. Lowering to INT_MAX catches buggy replies from 7648 the server that 32-bit clients would reject but 64-bit would accept, 7649 so we catch bugs sooner, and really, no sane & working server should 7650 ever report more than 2gb of extension names, font path entries, 7651 key modifier maps, etc. -alan- ] 7652 7653 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7654 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7655 7656commit 6d926088d80a08e13e6d6c4ff207b81ad52e667f 7657Author: Thomas Klausner <wiz@NetBSD.org> 7658Date: Tue Jun 25 18:34:32 2013 +0200 7659 7660 Fix out-of-range comparison in _XF86BigfontQueryFont 7661 7662 clang complained (correctly): 7663 warning: comparison of constant 768614336404564650 with expression 7664 of type 'CARD32' (aka 'unsigned int') is always true 7665 [-Wtautological-constant-out-of-range-compare] 7666 7667 [While LONG_MAX is correct, since it's used in size_t math, the 7668 numbers have to be limited to 32-bit range to be usable by 32-bit 7669 clients, and values beyond that range are far more likely to be 7670 bugs in the data from the server than valid numbers of characters 7671 in a font. -alan- ] 7672 7673 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7674 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7675 7676commit 383e2b0d029482a0f4c39fe00e15397538576fc1 7677Author: Thomas Klausner <wiz@NetBSD.org> 7678Date: Tue Jun 25 18:33:56 2013 +0200 7679 7680 Check for symbol existence with #ifdef, not #if 7681 7682 Reviewed-by: Jamey Sharp <jamey@minilop.net> 7683 7684 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7685 7686commit 9955d1c8de994a90fe7f2e3187e7362611d7d265 7687Author: Thomas Klausner <wiz@NetBSD.org> 7688Date: Tue Jun 25 18:33:07 2013 +0200 7689 7690 Use newer callback-based API for XIM. 7691 7692 Let libX11 load and make available the newer (X11R6) callback-based 7693 API for XIM (expected by emacs). 7694 7695 This patch updates the files to match the other nls/ files. 7696 7697 Patch from Ian D. Leroux <idleroux@fastmail.fm> on pkgsrc-users@NetBSD.org 7698 following a hint by Nhat Minh Lê <nhat.minh.le@gmail.com>. 7699 7700 Reviewed-by: James Cloos <cloos@jhcloos.com> 7701 7702 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7703 7704commit a17ceb7100bd36c2db210473ee701deb5d515731 7705Author: Thomas Klausner <wiz@NetBSD.org> 7706Date: Tue Jun 25 18:31:32 2013 +0200 7707 7708 Stop truncating source to destination length if it is larger. 7709 7710 It seems useless to do that since the code tests for both source 7711 length and destination to be non-zero. This fixes a cut'n'paste 7712 problem in xterm where the paste length was limited to 1024 (BUFSIZ) 7713 in button.c. 7714 7715 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7716 7717commit a336db9a0add3ae0783dda6e52459236622a12af 7718Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7719Date: Mon Jun 24 23:02:05 2013 -0700 7720 7721 Require ANSI C89 pre-processor, drop pre-C89 token pasting support 7722 7723 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7724 7725commit 9dfb0f3c0a761590bcdc1f3396b1e064da4e18e8 7726Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7727Date: Fri Jun 7 11:30:11 2013 -0700 7728 7729 troff macro expansion in specs/libX11 7730 7731 Many of the custom nroff macros (.ds <macro> <contents>) were left 7732 unsubstituted in the nroff->docbook conversion. This substitution 7733 is now performed, via the following perl script: 7734 7735 #! /usr/bin/perl -w -i 7736 7737 use Text::Wrap; 7738 7739 while ($_ = <>) { 7740 while ($_ =~ m/\((\w+)\b/g) { 7741 my $m = $1; 7742 if (exists $macro{$m}) { 7743 $_ =~ s/\($m/$macro{$m}/; 7744 $_ = wrap('', '', $_); 7745 $_ =~ s/[ \t]+$//; 7746 } 7747 } 7748 7749 if ($_ =~ /\<!-- .ds (\w+) (.*) -->/) { 7750 my ($m, $s) = ($1, $2); 7751 $macro{$m} = $s; 7752 while ($macro{$m} =~ /\\\s*$/) { 7753 $macro{$m} =~ s/\\\s*$//ms; 7754 $macro{$m} .= <>; 7755 chomp($macro{$m}); 7756 } 7757 $macro{$m} =~ s/\\ / /g; 7758 } else { 7759 print $_; 7760 } 7761 } 7762 7763 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7764 7765commit 20c17bd9ebf767a24643279e45866dddcb57b5ce 7766Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7767Date: Fri Jun 7 09:27:26 2013 -0700 7768 7769 specs/libX11: correct prototype for XListPixmapFormats/XImageByteOrder 7770 7771 The XListPixmapFormats arguments was being shown with XImageByteOrder's 7772 name and return types. Appears to have been a glitch in the nroff -> 7773 docbook conversion. 7774 7775 Reported-by: ZHANG Zhaolong <zhangzl2013@126.com> 7776 Reviewed-by: Jamey Sharp <jamey@minilop.net> 7777 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7778 7779commit ed3d830243c8a0eefaf24e15b11823272ffe5049 7780Author: Thomas Klausner <wiz@NetBSD.org> 7781Date: Sun Jun 2 20:49:55 2013 +0200 7782 7783 Deal with the limited range of VAX floating point numbers when compiling for VAX. 7784 7785 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7786 7787commit 0a48235d63639fb917c44d27c86e928e79fbac66 7788Author: Eric S. Raymond <esr@thyrsus.com> 7789Date: Thu Jun 6 16:43:56 2013 -0400 7790 7791 Remove call to undefined macro. 7792 7793commit 9e4719b9b719f2f8d255f6778e2e8c1809e32599 7794Author: Eric S. Raymond <esr@thyrsus.com> 7795Date: Thu Jun 6 16:42:20 2013 -0400 7796 7797 Remove call to undefined macro. 7798 7799commit 8496122eb00ce6cd5d2308ee54f64b68c378e455 7800Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7801Date: Mon Jun 3 20:06:43 2013 -0700 7802 7803 Update README to reflect where to find the Xlib specs now 7804 7805 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7806 7807commit 655d631e86c95b14888758b27ed2836ca3e3ce86 7808Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7809Date: Mon Jun 3 19:21:06 2013 -0700 7810 7811 libX11 1.6.0 7812 7813 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7814 7815commit 4a89b7ea908554628f374537a79931c8006a2de3 7816Author: Thomas Klausner <wiz@NetBSD.org> 7817Date: Sun Jun 2 11:49:54 2013 -0700 7818 7819 cmsMath.c: Add missing stdio header for printf(3) in DEBUG build. 7820 7821 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7822 7823commit 96dcf747f13b26b8e4d17b1bc8605d933c3e1dc6 7824Author: Thomas Klausner <wiz@NetBSD.org> 7825Date: Sun Jun 2 20:49:48 2013 +0200 7826 7827 XCreateGC man page: Avoid .TS H and .TH macros 7828 7829 Avoid .TS H and .TH for now as it doesn't alter the output in this case, 7830 and improve the output with mandoc(1). 7831 7832 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7833 7834commit 95a388158c9d73df7d24016d6a3d61506d7d53a4 7835Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7836Date: Thu May 23 19:43:35 2013 -0700 7837 7838 libX11 1.5.99.902 (1.6 RC2) 7839 7840 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7841 7842commit a3bdd2b090915fe0163b062f0e6576fe05dd332e 7843Author: Julien Cristau <jcristau@debian.org> 7844Date: Thu May 23 20:39:46 2013 +0200 7845 7846 xkb: fix off-by-one in _XkbReadGetNamesReply and _XkbReadVirtualModMap 7847 7848 The size of the arrays is max_key_code + 1. This makes these functions 7849 consistent with the other checks added for CVE-2013-1997. 7850 7851 Also check the XkbGetNames reply when names->keys was just allocated. 7852 7853 Signed-off-by: Julien Cristau <jcristau@debian.org> 7854 Tested-by: Colin Walters <walters@verbum.org> 7855 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7856 7857commit 7e30056e78e4b7979ff47f102e00327617266019 7858Author: Niveditha Rau <Niveditha.Rau@Oracle.COM> 7859Date: Fri May 17 15:26:21 2013 -0700 7860 7861 Make sure internal headers include required headers 7862 7863 Fixes builds with Solaris Studio 12.3 when lint is enabled, since it no 7864 longer ignores *.h files, but complains when they reference undefined 7865 typedefs or macros. 7866 7867 Signed-off-by: Niveditha Rau <Niveditha.Rau@Oracle.COM> 7868 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7869 7870commit 2820100bf8ba130b94253f415e7fa5ac28bb2037 7871Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7872Date: Thu May 16 23:05:36 2013 -0700 7873 7874 Free fs->properties in _XF86BigfontQueryFont overflow error path 7875 7876 Fixes small memory leak introduced in commit 5669a22081 7877 7878 Reported-by: Julien Cristau <jcristau@debian.org> 7879 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7880 7881commit 3131740513133a9ff7cb12123d29ceb18584fc38 7882Author: Matthieu Herrb <matthieu.herrb@laas.fr> 7883Date: Wed May 8 19:33:09 2013 +0200 7884 7885 XListFontsWithInfo: Re-decrement flist[0] before calling free() on it. 7886 7887 Freeing a pointer that wasn't returned by malloc() is undefined 7888 behavior and produces an error with OpenBSD's implementation. 7889 7890 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7891 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7892 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7893 7894commit 3fe4bea086149f06a142a8f1d575f627ec1e22c7 7895Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7896Date: Fri Apr 19 14:30:40 2013 -0700 7897 7898 Give GNU & Solaris Studio compilers hints about XEatData branches 7899 7900 Try to offset the cost of all the recent checks we've added by giving 7901 the compiler a hint that the branches that involve us eating data 7902 are less likely to be used than the ones that process it. 7903 7904 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7905 7906commit e1b457beb8d4e831ef44279dada6c475cb955738 7907Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7908Date: Sun Mar 31 12:22:35 2013 -0700 7909 7910 _XkbReadGetMapReply: reject maxKeyCodes smaller than the minKeyCode 7911 7912 Various other bounds checks in the code assume this is true, so 7913 enforce it when we first get the data from the X server. 7914 7915 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7916 7917commit 12ad4c6432496897ff000eb7cfecd0fb4b290331 7918Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7919Date: Sat Mar 16 10:03:13 2013 -0700 7920 7921 Use calloc in XOpenDisplay to initialize structs containing pointers 7922 7923 Prevents trying to free uninitialized pointers if we have to bail out 7924 partway through setup, such as if we receive a corrupted or incomplete 7925 connection setup block from the server. 7926 7927 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7928 7929commit d38527e25f8b6e2f1174ecc21260c5c5416f972e 7930Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7931Date: Thu Mar 7 23:46:05 2013 -0800 7932 7933 Remove more unnecessary casts from Xmalloc/calloc calls 7934 7935 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7936 7937commit b2c86b582c58f50c7b14da01cf7ebd20ef12a6b2 7938Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7939Date: Sat Mar 2 16:56:16 2013 -0800 7940 7941 Convert more _XEatData callers to _XEatDataWords 7942 7943 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7944 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7945 7946commit 192bbb9e2fc45df4e17b35b6d14ea0eb418dbd39 7947Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7948Date: Sat Mar 9 11:04:37 2013 -0800 7949 7950 Make XGetWindowProperty() always initialize returned values 7951 7952 Avoids memory corruption and other errors when callers access them 7953 without checking to see if XGetWindowProperty() returned an error value. 7954 7955 Callers are still required to check for errors, this just reduces the 7956 damage when they don't. 7957 7958 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 7959 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7960 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7961 7962commit db1b1c871da29aa0545182bf888df81627f165a5 7963Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7964Date: Sat Mar 2 15:08:21 2013 -0800 7965 7966 Avoid overflows in XListExtensions() [CVE-2013-1997 15/15] 7967 7968 Ensure that when breaking the returned list into individual strings, 7969 we don't walk past the end of allocated memory to write the '\0' bytes 7970 7971 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7972 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7973 7974commit 8d5936594993921acdfec778dd8f41b555e2543a 7975Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7976Date: Sat Mar 2 15:08:21 2013 -0800 7977 7978 Avoid overflows in XGetFontPath() [CVE-2013-1997 14/15] 7979 7980 Ensure that when breaking the returned list into individual strings, 7981 we don't walk past the end of allocated memory to write the '\0' bytes 7982 7983 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7984 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7985 7986commit 0c404db6a92dc2c198328bf586c02d8abbe02013 7987Author: Alan Coopersmith <alan.coopersmith@oracle.com> 7988Date: Sat Mar 2 15:08:21 2013 -0800 7989 7990 Avoid overflows in XListFonts() [CVE-2013-1997 13/15] 7991 7992 Ensure that when breaking the returned list into individual strings, 7993 we don't walk past the end of allocated memory to write the '\0' bytes 7994 7995 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7996 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 7997 7998commit 0b0f5d4358c3de7563d6af03f0d2ce454702a06a 7999Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8000Date: Sat Mar 2 15:08:21 2013 -0800 8001 8002 integer overflow in XGetModifierMapping() [CVE-2013-1981 13/13] 8003 8004 Ensure that we don't underallocate when the server claims a very large reply 8005 8006 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8007 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8008 8009commit a351b8103b2ba78882e1c309e85893ca3abe2073 8010Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8011Date: Sat Mar 2 15:08:21 2013 -0800 8012 8013 integer overflow in XGetPointerMapping() & XGetKeyboardMapping() [CVE-2013-1981 12/13] 8014 8015 Ensure that we don't underallocate when the server claims a very large reply 8016 8017 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8018 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8019 8020commit 833f6b70bc789d33607f6dbfee9e0a4178ec4b59 8021Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8022Date: Sat Mar 2 15:08:21 2013 -0800 8023 8024 integer overflow in XGetImage() [CVE-2013-1981 11/13] 8025 8026 Ensure that we don't underallocate when the server claims to have sent a 8027 very large reply. 8028 8029 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8030 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8031 8032commit 79d8dc08eb98842173ce239b9dd60df0e9e9ae72 8033Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8034Date: Fri Mar 8 22:25:35 2013 -0800 8035 8036 integer overflow in XGetWindowProperty() [CVE-2013-1981 10/13] 8037 8038 If the reported number of properties is too large, the calculations 8039 to allocate memory for them may overflow, leaving us returning less 8040 memory to the caller than implied by the value written to *nitems. 8041 8042 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8043 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8044 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8045 8046commit 164bf4dfe839b1cc75cdeee378a243d04a8200e4 8047Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8048Date: Sat Mar 2 13:18:48 2013 -0800 8049 8050 integer overflows in TransFileName() [CVE-2013-1981 9/13] 8051 8052 When trying to process file paths the tokens %H, %L, & %S are expanded 8053 to $HOME, the standard compose file path & the xlocaledir path. 8054 If enough of these tokens are repeated and values like $HOME are set to 8055 very large values, the calculation of the total string size required to 8056 hold the expanded path can overflow, resulting in allocating a smaller 8057 string than the amount of data we'll write to it. 8058 8059 Simply restrict all of these values, and the total path size to PATH_MAX, 8060 because really, that's all you should need for a filename path. 8061 8062 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8063 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8064 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8065 8066commit 460e8a223b87d4fa0ea1e97823e998a770e0f2a2 8067Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8068Date: Fri Mar 1 18:37:37 2013 -0800 8069 8070 integer truncation in _XimParseStringFile() [CVE-2013-1981 8/13] 8071 8072 Called from _XimCreateDefaultTree() which uses getenv("XCOMPOSEFILE") 8073 to specify filename. 8074 8075 If the size of off_t is larger than the size of unsigned long (as in 8076 32-bit builds with large file flags), a file larger than 4 gigs could 8077 have its size truncated, leading to data from that file being written 8078 past the end of the undersized buffer allocated for it. 8079 8080 While configure.ac does not use AC_SYS_LARGEFILE to set large file mode, 8081 builders may have added the large file compilation flags to CFLAGS on 8082 their own. 8083 8084 size is left limited to an int, because if your Xim file is 8085 larger than 2gb, you're doing it wrong. 8086 8087 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8088 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8089 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8090 8091commit 226622349a4b1e16064649d4444a34fb4be4f464 8092Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8093Date: Sat Mar 2 12:39:58 2013 -0800 8094 8095 Unbounded recursion in _XimParseStringFile() when parsing include files [CVE-2013-2004 2/2] 8096 8097 parseline() can call _XimParseStringFile() which can call parseline() 8098 which can call _XimParseStringFile() which can call parseline() .... 8099 eventually causing recursive stack overflow and crash. 8100 8101 Limit is set to a include depth of 100 files, which should be enough 8102 for all known use cases, but could be adjusted later if necessary. 8103 8104 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8105 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8106 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8107 8108commit 236b603d235dc264d1c6250dca09c745458a9088 8109Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8110Date: Sat Mar 2 12:01:39 2013 -0800 8111 8112 Unbounded recursion in GetDatabase() when parsing include files [CVE-2013-2004 1/2] 8113 8114 GetIncludeFile() can call GetDatabase() which can call GetIncludeFile() 8115 which can call GetDatabase() which can call GetIncludeFile() .... 8116 eventually causing recursive stack overflow and crash. 8117 8118 Easily reproduced with a resource file that #includes itself. 8119 8120 Limit is set to a include depth of 100 files, which should be enough 8121 for all known use cases, but could be adjusted later if necessary. 8122 8123 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8124 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8125 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8126 8127commit 076428918e6c35f66b9b55c3fa097ff06496d155 8128Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8129Date: Fri Mar 1 18:37:37 2013 -0800 8130 8131 integer overflow in ReadInFile() in Xrm.c [CVE-2013-1981 7/13] 8132 8133 Called from XrmGetFileDatabase() which gets called from InitDefaults() 8134 which gets the filename from getenv ("XENVIRONMENT") 8135 8136 If file is exactly 0xffffffff bytes long (or longer and truncates to 8137 0xffffffff, on implementations where off_t is larger than an int), 8138 then size may be set to a value which overflows causing less memory 8139 to be allocated than is written to by the following read() call. 8140 8141 size is left limited to an int, because if your Xresources file is 8142 larger than 2gb, you're very definitely doing it wrong. 8143 8144 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8145 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8146 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8147 8148commit 90fd5abac2faca86f9f100353a3c9c7b89f31484 8149Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8150Date: Sat Mar 2 11:44:19 2013 -0800 8151 8152 Integer overflows in stringSectionSize() cause buffer overflow in ReadColornameDB() [CVE-2013-1981 6/13] 8153 8154 LoadColornameDB() calls stringSectionSize() to do a first pass over the 8155 file (which may be provided by the user via XCMSDB environment variable) 8156 to determine how much memory needs to be allocated to read in the file, 8157 then allocates the returned sizes and calls ReadColornameDB() to load the 8158 data from the file into that newly allocated memory. 8159 8160 If stringSectionSize() overflows the signed ints used to calculate the 8161 file size (say if you have an xcmsdb with ~4 billion lines in or a 8162 combined string length of ~4 gig - which while it may have been 8163 inconceivable when Xlib was written, is quite possible today), then 8164 LoadColornameDB() may allocate a memory buffer much smaller than the 8165 amount of data ReadColornameDB() will write to it. 8166 8167 The total size is left limited to an int, because if your xcmsdb file 8168 is larger than 2gb, you're doing it wrong. 8169 8170 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8171 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8172 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8173 8174commit b9ba832401734e1cbd30a930c0d11d850293f3f9 8175Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8176Date: Sat Mar 2 11:25:25 2013 -0800 8177 8178 unvalidated length in _XimXGetReadData() [CVE-2013-1997 12/15] 8179 8180 Check the provided buffer size against the amount of data we're going to 8181 write into it, not against the reported length from the ClientMessage. 8182 8183 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8184 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8185 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8186 8187commit de2e6c322c4aca22856b380f67f8e488e7510015 8188Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8189Date: Sat Mar 2 11:11:08 2013 -0800 8190 8191 unvalidated index/length in _XkbReadGetNamesReply() [CVE-2013-1997 11/15] 8192 8193 If the X server returns key name indexes outside the range of the number 8194 of keys it told us to allocate, out of bounds memory writes could occur. 8195 8196 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8197 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8198 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8199 8200commit 2df882eeb3a70256170127a746a9ba26376599a1 8201Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8202Date: Sat Mar 2 11:01:04 2013 -0800 8203 8204 unvalidated index in _XkbReadVirtualModMap() [CVE-2013-1997 10/15] 8205 8206 If the X server returns modifier map indexes outside the range of the number 8207 of keys it told us to allocate, out of bounds memory writes could occur. 8208 8209 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8210 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8211 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8212 8213commit 4d7c422a37eb9617fb22f8e37527c2b34b105665 8214Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8215Date: Sat Mar 2 11:04:44 2013 -0800 8216 8217 unvalidated index in _XkbReadExplicitComponents() [CVE-2013-1997 9/15] 8218 8219 If the X server returns key indexes outside the range of the number of 8220 keys it told us to allocate, out of bounds memory writes could occur. 8221 8222 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8223 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8224 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8225 8226commit e56a2ada719c5cfac5ed61a52a80ade86c0f5957 8227Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8228Date: Sat Mar 2 10:51:51 2013 -0800 8229 8230 unvalidated index in _XkbReadModifierMap() [CVE-2013-1997 8/15] 8231 8232 If the X server returns modifier map indexes outside the range of the number 8233 of keys it told us to allocate, out of bounds memory writes could occur. 8234 8235 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8236 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8237 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8238 8239commit 06c086e8a1d8374ea9a95ff989f053c96bb1bdca 8240Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8241Date: Sat Mar 2 10:39:21 2013 -0800 8242 8243 unvalidated index in _XkbReadKeyBehaviors() [CVE-2013-1997 7/15] 8244 8245 If the X server returns key behavior indexes outside the range of the number 8246 of keys it told us to allocate, out of bounds memory writes could occur. 8247 8248 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8249 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8250 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8251 8252commit 00626c3830b869259098985afa38933d77ccec72 8253Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8254Date: Sat Mar 2 09:40:22 2013 -0800 8255 8256 unvalidated index in _XkbReadKeyActions() [CVE-2013-1997 6/15] 8257 8258 If the X server returns key action indexes outside the range of the number 8259 of keys it told us to allocate, out of bounds memory access could occur. 8260 8261 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8262 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8263 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8264 8265commit fd7d4956bc7a1c4b5c38661b12777ebee4d685d9 8266Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8267Date: Sat Mar 2 09:28:33 2013 -0800 8268 8269 unvalidated index in _XkbReadKeySyms() [CVE-2013-1997 5/15] 8270 8271 If the X server returns keymap indexes outside the range of the number of 8272 keys it told us to allocate, out of bounds memory access could occur. 8273 8274 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8275 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8276 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8277 8278commit 59ae16a00d18588e98af57d26e442af8ea42b7aa 8279Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8280Date: Sat Mar 2 09:18:26 2013 -0800 8281 8282 unvalidated indexes in _XkbReadGetGeometryReply() [CVE-2013-1997 4/15] 8283 8284 If the X server returns color indexes outside the range of the number of 8285 colors it told us to allocate, out of bounds memory access could occur. 8286 8287 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8288 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8289 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8290 8291commit bff938b9fe1629cbacb726509edfa2a3840b7207 8292Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8293Date: Sat Mar 2 09:12:47 2013 -0800 8294 8295 unvalidated indexes in _XkbReadGeomShapes() [CVE-2013-1997 3/15] 8296 8297 If the X server returns shape indexes outside the range of the number 8298 of shapes it told us to allocate, out of bounds memory access could occur. 8299 8300 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8301 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8302 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8303 8304commit f293659d5a4024bda386305bb7ebeb4647c40934 8305Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8306Date: Fri Mar 1 22:49:01 2013 -0800 8307 8308 unvalidated index in _XkbReadGetDeviceInfoReply() [CVE-2013-1997 2/15] 8309 8310 If the X server returns more buttons than are allocated in the XKB 8311 device info structures, out of bounds writes could occur. 8312 8313 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8314 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8315 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8316 8317commit cddc4e7e3cb4b9b7ad25f8591971a86901c249f2 8318Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8319Date: Fri Mar 1 19:30:09 2013 -0800 8320 8321 unvalidated lengths in XAllocColorCells() [CVE-2013-1997 1/15] 8322 8323 If a broken server returned larger than requested values for nPixels or 8324 nMasks, XAllocColorCells would happily overflow the buffers provided by 8325 the caller to write the results into. 8326 8327 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8328 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8329 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8330 8331commit 2cd62b5eb99ffbb2fce99f3c459455e630b35bf7 8332Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8333Date: Fri Mar 1 22:49:01 2013 -0800 8334 8335 integer overflow in XListHosts() [CVE-2013-1981 5/13] 8336 8337 If the reported number of host entries is too large, the calculations 8338 to allocate memory for them may overflow, leaving us writing beyond the 8339 bounds of the allocation. 8340 8341 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8342 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8343 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8344 8345commit 1f6a3dbf699b85c0ea715ef21de7e7095a714e12 8346Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8347Date: Fri Mar 1 22:49:01 2013 -0800 8348 8349 integer overflow in XGetMotionEvents() [CVE-2013-1981 4/13] 8350 8351 If the reported number of motion events is too large, the calculations 8352 to allocate memory for them may overflow, leaving us writing beyond the 8353 bounds of the allocation. 8354 8355 v2: Ensure nEvents is set to 0 when returning NULL events pointer 8356 8357 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8358 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8359 8360commit 39515b7c3ba8cae9021bf6695e378ae19487082f 8361Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8362Date: Fri Mar 1 22:49:01 2013 -0800 8363 8364 integer overflow in XListFontsWithInfo() [CVE-2013-1981 3/13] 8365 8366 If the reported number of remaining fonts is too large, the calculations 8367 to allocate memory for them may overflow, leaving us writing beyond the 8368 bounds of the allocation. 8369 8370 v2: Fix reply_left calculations, check calculated sizes fit in reply_left 8371 v3: On error cases, also set values to be returned in pointer args to 0/NULL 8372 8373 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8374 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8375 8376commit 5669a220816b7d58fcaf0c302ead16fbe5c87817 8377Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8378Date: Fri Mar 1 21:05:27 2013 -0800 8379 8380 integer overflow in _XF86BigfontQueryFont() [CVE-2013-1981 2/13] 8381 8382 Similar to _XQueryFont, but with more ways to go wrong and overflow. 8383 Only compiled if libX11 is built with XF86BigFont support. 8384 8385 v2: Fix reply_left calculations, check calculated sizes fit in reply_left 8386 8387 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8388 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8389 8390commit 6df8a63d34b7514077188e2062a13774f920c085 8391Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8392Date: Fri Mar 1 21:05:27 2013 -0800 8393 8394 integer overflow in _XQueryFont() on 32-bit platforms [CVE-2013-1981 1/13] 8395 8396 If the CARD32 reply.nCharInfos * sizeof(XCharStruct) overflows an 8397 unsigned long, then too small of a buffer will be allocated for the 8398 data copied in from the reply. 8399 8400 v2: Fix reply_left calculations, check calculated sizes fit in reply_left 8401 8402 Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> 8403 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8404 8405commit 9f5d83706543696fc944c1835a403938c06f2cc5 8406Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8407Date: Fri Mar 1 20:54:24 2013 -0800 8408 8409 Add _XEatDataWords to discard a given number of 32-bit words of reply data 8410 8411 Matches the units of the length field in X protocol replies, and provides 8412 a single implementation of overflow checking to avoid having to replicate 8413 those checks in every caller. 8414 8415 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8416 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8417 8418commit d7f04c340ade3834e603c23d543132e1ee4e0c63 8419Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8420Date: Sat Mar 2 13:03:55 2013 -0800 8421 8422 Move repeated #ifdef magic to find PATH_MAX into a common header 8423 8424 Lets stop duplicating the mess all over 8425 8426 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8427 Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr> 8428 8429commit f3a553a4e4a55d9d19deda1ea01883e1d5d682b1 8430Author: Pander <pander@users.sourceforge.net> 8431Date: Tue May 7 18:38:14 2013 -0400 8432 8433 Add compose sequences for J́ and j́. 8434 8435 The resulting sequences are: 8436 8437 U+004A LATIN CAPITAL LETTER J U+0301 COMBINING ACUTE ACCENT 8438 U+006A LATIN SMALL LETTER J U+0301 COMBINING ACUTE ACCENT 8439 8440 Used in Dutch, per: 8441 8442 http://lists.x.org/archives/xorg-devel/2013-February/035514.html 8443 https://nl.wikipedia.org/wiki/Accenttekens_in_de_Nederlandse_spelling 8444 8445 Signed-off-by: Pander <pander@users.sourceforge.net> 8446 Signed-off-by: James Cloos <cloos@jhcloos.com> 8447 8448commit f49bb2dd6d4ea45c55bd21acc0efe2b764441020 8449Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8450Date: Sat Mar 16 18:30:56 2013 -0700 8451 8452 Move big request comment in XOpenDisplay to the right place 8453 8454 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8455 8456commit 3996543c1b2919e97d61a5d70fe1ebd7cd76fc83 8457Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8458Date: Sat Mar 9 19:16:03 2013 -0800 8459 8460 libX11 1.5.99.901 (1.6 RC1) 8461 8462 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8463 8464commit f9cd175a471116a616e681fb0ca1a61b3d84a6a0 8465Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8466Date: Fri Mar 8 22:33:28 2013 -0800 8467 8468 Fix very weird indenting in src/GetFProp.c 8469 8470 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8471 8472commit e9bd757630368afb374c5d1bcc5d4d85ad3c6c4c 8473Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8474Date: Fri Mar 8 15:37:33 2013 -0800 8475 8476 XAllocClassHint: Assume calloc sets pointers in allocated memory to NULL 8477 8478 While the C standard technically allows for the compiler to translate 8479 pointer = 0 or pointer = NULL into something other than filling the 8480 pointer address with 0 bytes, the rest of the Xlib code already assumes 8481 that calloc initializes any pointers in the struct to NULL, and there 8482 are no known systems supported by X.Org where this is not true. 8483 8484 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8485 8486commit 39547d600a13713e15429f49768e54c3173c828d 8487Author: Karl Tomlinson <xmail@karlt.net> 8488Date: Mon Feb 18 01:25:34 2013 +0000 8489 8490 MakeBigReq: don't move the last word, already handled by Data32 8491 8492 MakeBigReq inserts a length field after the first 4 bytes of the request 8493 (after req->length), pushing everything else back by 4 bytes. 8494 8495 The current memmove moves everything but the first 4 bytes back. 8496 If a request aligns to the end of the buffer pointer when MakeBigReq is 8497 invoked for that request, this runs over the buffer. 8498 Instead, we need to memmove minus the first 4 bytes (which aren't moved), 8499 minus the last 4 bytes (so we still align to the previous tail). 8500 8501 The 4 bytes that fell out are already handled with Data32, which will 8502 handle the buffermax correctly. 8503 8504 The case where req->length = 1 was already not functional. 8505 8506 Reported by Abhishek Arya <inferno@chromium.org>. 8507 8508 https://bugzilla.mozilla.org/show_bug.cgi?id=803762 8509 8510 Reviewed-by: Jeff Muizelaar <jmuizelaar@mozilla.com> 8511 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 8512 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8513 8514Notes: 8515 CVE-2013-7439 8516 8517commit 3cdb6c3a1646f670afa03d424ec12ac418181d1e 8518Author: Quentin Glidic <sardemff7+git@sardemff7.net> 8519Date: Tue Jan 15 21:07:17 2013 +0000 8520 8521 nls/Makefile.am: Use LOG_COMPILER 8522 8523 TESTS_ENVIRONMENT is deprecated 8524 8525 Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net> 8526 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8527 8528commit df66d7a98e2bc7f44fb5583b645df87d525f07f1 8529Author: Quentin Glidic <sardemff7+git@sardemff7.net> 8530Date: Tue Jan 15 21:07:16 2013 +0000 8531 8532 nls/Makefile.am: Remove unneeded $(srcdir) 8533 8534 Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net> 8535 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 8536 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8537 8538commit 3facbe5c0df1b5597571b7b00d5f7bdbc92fb278 8539Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8540Date: Sat Mar 2 12:01:39 2013 -0800 8541 8542 Add <X11/Xresource.h> hint to all Xrm* man pages 8543 8544 Help users figure out which header file they need to #include 8545 8546 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8547 8548commit 466404007f2c8f7166e4faddfea1454c5bfe1e9a 8549Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8550Date: Fri Mar 8 17:13:09 2013 -0800 8551 8552 _xudc_code_to_glyph: check for NULL pointer *before* writing to it, not after 8553 8554 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8555 8556commit 47d2eff64d167b3245d346d7f38ca703be645e26 8557Author: Ken Moffat <ken@linuxfromscratch.org> 8558Date: Thu Mar 7 18:27:54 2013 -0500 8559 8560 dead_double_grave and dead_inverted_breve should only have one underscore. 8561 8562 Correct instances of dead_double_grave and dead_inverted_breve to 8563 dead_doublegrave and dead_invertedbreve. 8564 8565 Signed-off-by: Ken Moffat <ken@linuxfromscratch.org> 8566 Signed-off-by: James Cloos <cloos@jhcloos.com> 8567 8568commit c23d61d1b84dca3740bf4786978c7908d0065fb9 8569Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8570Date: Fri Mar 1 18:10:27 2013 -0800 8571 8572 Assume size_t is always available, since it was defined in C89 8573 8574 Don't provide a fallback definition #ifdef X_NOT_POSIX anymore. 8575 We already use size_t throughout the rest of Xlib, just had this 8576 one instance left in XKBGAlloc.c of a fallback definition. 8577 8578 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8579 8580commit 9bcfd84aa1410387bc8cf002a5f90f44705aa0d1 8581Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8582Date: Fri Mar 1 18:09:07 2013 -0800 8583 8584 unifdef XKB_IN_SERVER 8585 8586 Leftovers from XKB files that were previously shared between the client 8587 and server code, but aren't any more. 8588 8589 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8590 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 8591 8592commit 769a0efa2298040fe8316a89fc9e75fb61e288e5 8593Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8594Date: Thu Feb 28 20:04:25 2013 -0800 8595 8596 unifdef CRAY & _CRAY 8597 8598 (mostly performed with unifdef, followed by some manual cleanup of 8599 the remaining code) 8600 8601 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8602 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 8603 8604commit ca106eb03e5f5468df8033300c5caae3d3c6936b 8605Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8606Date: Thu Feb 28 20:04:25 2013 -0800 8607 8608 unifdef WORD64 8609 8610 WORD64 seems to have only been defined in <X11/Xmd.h> when building for 8611 CRAY, to handle int being a 64-bit value (ILP64, not LP64) and having 8612 64-bit alignment requirements. 8613 8614 It hadn't been fully supported even before autotooling, as can be 8615 seen by removed code such as: 8616 8617 #ifdef WORD64 8618 _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8 8619 #endif 8620 8621 (mostly performed with unifdef, followed by some manual cleanup of 8622 the remaining code) 8623 8624 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8625 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 8626 8627commit 9399caf2c12cbe1ed56f4f6b368c5811cb5d0458 8628Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8629Date: Thu Feb 28 20:04:25 2013 -0800 8630 8631 unifdef MUSTCOPY 8632 8633 MUSTCOPY seems to have only been defined in <X11/Xmd.h> when building for 8634 CRAY, to handle missing some sizes of integer type. 8635 8636 (mostly performed with unifdef, followed by some manual cleanup of 8637 spacing/indenting in the remaining code) 8638 8639 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8640 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 8641 8642commit b687440c28c7da6ee0ae44514d20248db5161606 8643Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8644Date: Sat Feb 16 10:42:23 2013 -0800 8645 8646 Convert more sprintf calls to snprintf 8647 8648 You could analyze most of these and quickly recognize that there was no 8649 chance of buffer overflow already, but why make everyone spend time doing 8650 that when we can just make it obviously safe? 8651 8652 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8653 8654commit b092dabbd712d7b656abcf572d253b9b206c0237 8655Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8656Date: Fri Feb 15 23:43:12 2013 -0800 8657 8658 XKeysymToString: move variable declarations to the scope of their usage 8659 8660 Makes it easier for readers to understand scope of variable usage, and 8661 clears up gcc warning: 8662 8663 KeysymStr.c: In function 'XKeysymToString': 8664 KeysymStr.c:128:13: warning: declaration of 'i' shadows a previous local [-Wshadow] 8665 KeysymStr.c:73:18: warning: shadowed declaration is here [-Wshadow] 8666 8667 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8668 8669commit f0b171c8ea7b055ba520272ea9a2604e18841ac7 8670Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8671Date: Fri Feb 15 23:34:40 2013 -0800 8672 8673 Preserve constness in casting arguments through the Data*() routines 8674 8675 Casts were annoying gcc by dropping constness when changing types, 8676 when routines simply either copy data into the request buffer or 8677 send it directly to the X server, and never modify the input. 8678 8679 Fixes gcc warnings including: 8680 ChProp.c: In function 'XChangeProperty': 8681 ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8682 ChProp.c:65:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8683 ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8684 ChProp.c:74:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8685 ChProp.c:83:6: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8686 SetHints.c: In function 'XSetStandardProperties': 8687 SetHints.c:262:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8688 SetPntMap.c: In function 'XSetPointerMapping': 8689 SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8690 SetPntMap.c:46:5: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8691 StBytes.c: In function 'XStoreBuffer': 8692 StBytes.c:97:33: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8693 StName.c: In function 'XStoreName': 8694 StName.c:40:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8695 StName.c: In function 'XSetIconName': 8696 StName.c:51:27: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8697 8698 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8699 8700commit 6c558ee357292dd9dfc6d9006f4525f625327c52 8701Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8702Date: Fri Feb 15 22:58:54 2013 -0800 8703 8704 Fix comment typo & confusing indentation levels in Data() macro definition 8705 8706 The final } matches the one on the #define line, not one that doesn't 8707 appear after the else statement it was lined up with 8708 8709 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8710 8711commit afd6593da90e51234d59f8921c411317f91ab48b 8712Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8713Date: Fri Feb 15 23:25:38 2013 -0800 8714 8715 XStringToKeysym: preserve constness when casting off unsignedness for strcmp 8716 8717 Fixes gcc warning: 8718 StrKeysym.c:97:17: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8719 8720 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8721 8722commit 7e3bf4dd83fec22bd568146de75e6d59eff74e21 8723Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8724Date: Fri Feb 15 23:14:40 2013 -0800 8725 8726 XRebindKeysym: Drop unnecessary const-removing cast 8727 8728 C89 defines memcpy as taking a const void *, so casting from 8729 const unsigned char * to char * simply angers gcc for no benefit: 8730 8731 KeyBind.c:1017:24: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8732 8733 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8734 8735commit 54527eab93d46055cf11eb6c18abb353a03ae544 8736Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8737Date: Fri Feb 15 22:45:19 2013 -0800 8738 8739 cmsColNm.c: maintain constness of arguments to qsort helper function 8740 8741 Fixes gcc warning: 8742 8743 cmsColNm.c: In function 'FirstCmp': 8744 cmsColNm.c:257:20: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8745 cmsColNm.c:257:45: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual] 8746 8747 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8748 8749commit deedeada53676ee529d700bf96fde0b29a3a1def 8750Author: Nickolai Zeldovich <nickolai@csail.mit.edu> 8751Date: Tue Jan 22 10:03:00 2013 -0500 8752 8753 XListFontsWithInfo: avoid accessing realloc'ed memory 8754 8755 If exactly one of the two reallocs in XListFontsWithInfo() fails, the 8756 subsequent code accesses memory freed by the other realloc. 8757 8758 Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu> 8759 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8760 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8761 8762commit f57fd760cba92ad846917f21e94e73e9c846185f 8763Author: Colin Walters <walters@verbum.org> 8764Date: Wed Jan 4 17:37:06 2012 -0500 8765 8766 autogen.sh: Implement GNOME Build API 8767 8768 http://people.gnome.org/~walters/docs/build-api.txt 8769 8770 Signed-off-by: Adam Jackson <ajax@redhat.com> 8771 8772commit 51c102d39e855cf1704d9eb3afba76a2e73c6b81 8773Author: Adam Jackson <ajax@redhat.com> 8774Date: Tue Jan 15 14:28:48 2013 -0500 8775 8776 configure: Remove AM_MAINTAINER_MODE 8777 8778 Signed-off-by: Adam Jackson <ajax@redhat.com> 8779 8780commit 3cd974b1d4d1fa6389d3695fa9fcc0c22a51d50c 8781Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8782Date: Wed Dec 26 22:57:39 2012 -0800 8783 8784 Remove unused DECnet ("DNETCONN") code from Xlib 8785 8786 Has never been converted to build in modular builds, so has been unusable 8787 since X11R7.0 release in 2005. DNETCONN support was removed from xtrans 8788 back in 2008. 8789 8790 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8791 8792commit a6e5b36a3e6d4a7a9fb4bad905ed127e67b1957e 8793Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8794Date: Wed Dec 26 22:56:38 2012 -0800 8795 8796 Remove unused TLI ("STREAMSCONN") code from Xlib 8797 8798 Has never been converted to build in modular builds, so has been unusable 8799 since X11R7.0 release in 2005. All known platforms with TLI/XTI support 8800 that X11R7 & later releases run on also have (and mostly prefer) BSD 8801 socket support for their networking API. 8802 8803 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8804 8805commit 59c9ee8cd58857c5769b643611cbe526005a9e45 8806Author: Alan Coopersmith <alan.coopersmith@oracle.com> 8807Date: Sun Dec 16 17:44:42 2012 -0800 8808 8809 Tell clang not to report -Wpadded warnings on public headers we can't fix 8810 8811 Better to silence the compiler warning than break ABI. 8812 8813 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8814 8815commit 0b148750027fd0557c5ed93afda861ddf4b92e0f 8816Author: Jon TURNEY <jon.turney@dronecode.org.uk> 8817Date: Mon Nov 12 17:27:52 2012 +0000 8818 8819 Fix config check for loadable modules 8820 8821 The config check of the results of testing for dlfcn.h or dl.h just tests the 8822 value of the ac_cv_ variables, which will be 'yes' or 'no', rather than checking 8823 it is 'yes', so loadable module support would always be detected. 8824 8825 This is necessary for successful compilation for the MinGW target without the 8826 optional dlfcn-win32 library. 8827 8828 v2: Also, fixed typoed name of ac_cv_header_dlfcn_h, so check still works 8829 correctly when dlfcn.h is available 8830 8831 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> 8832 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 8833 8834commit cd25cab4b5b957641183ce72dd1ae0424aff1663 8835Author: Egbert Eich <eich@freedesktop.org> 8836Date: Fri May 20 18:27:02 2011 +0200 8837 8838 Install Xcms.txt in $(datadir) rather than $(libdir). 8839 8840 This file is an architecture independent data and should be where 8841 other databases are. 8842 This is the Xlib provided sample file, applications are free to 8843 use a different one specifying its location in the XCMSDB env 8844 variable. 8845 8846 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8847 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8848 8849commit 51fef7e273b5c6256e4c9604e3e1afe5dc8f6a1a 8850Author: Egbert Eich <eich@freedesktop.org> 8851Date: Fri May 20 18:25:24 2011 +0200 8852 8853 Don't hard code path to Xcms.txt file. 8854 8855 The path to this file is configurable at build time. The source 8856 however contains a hard coded path. 8857 8858 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8859 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8860 8861commit 9833489e6c3829a1e835bc0a11f028fc180809e4 8862Author: Colin Walters <walters@verbum.org> 8863Date: Fri Dec 7 08:51:21 2012 -0500 8864 8865 Fix build after dropping locales 8866 8867 They also needed to be removed from configure.ac 8868 8869 Signed-off-by: Colin Walters <walters@verbum.org> 8870 Signed-off-by: Julien Cristau <jcristau@debian.org> 8871 8872commit 0a740a574aaf0c0eec78859b773a532cff3b74c3 8873Author: Egbert Eich <eich@freedesktop.org> 8874Date: Fri May 20 13:04:11 2011 +0200 8875 8876 i18n: Uppercased all occurances if 'iso8859' in the full locale name. 8877 8878 Making all occurances of iso8859 upper case in the full local name 8879 makes the alias entries more consistent and match the entries on 8880 locale.dir. 8881 8882 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8883 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8884 8885commit 5e7d589697755a70fb22d85c6a1ae82b39843e53 8886Author: Egbert Eich <eich@freedesktop.org> 8887Date: Fri May 20 17:55:49 2011 +0200 8888 8889 i18n: Remove ja.S90 and ja.U90 locales. 8890 8891 Both locales carry a copyright notice and a prorietary statement: 8892 8893 Copyright 1995 by FUJITSU LIMITED 8894 This is source code modified by FUJITSU LIMITED under the Joint 8895 Development Agreement for the CDEnext PST. 8896 This is unpublished proprietary source code of FUJITSU LIMITED 8897 8898 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8899 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8900 8901commit 99eae3dcb7bf6764e3b243d2a2934a4d1ecce90e 8902Author: Egbert Eich <eich@freedesktop.org> 8903Date: Fri May 20 16:46:15 2011 +0200 8904 8905 i18n: Bring locale.dir and compose.dir in sync. 8906 8907 Some entries for locale/encoding combinations were missing from 8908 either file or just misspelled, some entries were wrong or just 8909 aliases. 8910 8911 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8912 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8913 8914commit d1e6baa4e290b758e430077cb74e7c03ad850771 8915Author: Egbert Eich <eich@freedesktop.org> 8916Date: Fri May 20 14:02:39 2011 +0200 8917 8918 i18n: Treat 'a3_AZ' as an alias for 'az_AZ'. 8919 8920 locale.alias contains a comment: 8921 XCOMM a3 is not an ISO 639 language code, but in Cyrillic, "Z" looks like "3". 8922 Thus lets treat 'a3' as an alias for 'az'. 8923 8924 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8925 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8926 Reviewed-by: Magnus Kessler <Magnus.Kessler@gmx.net> 8927 8928commit fa2aab0bea18efa26a56977d3166277582ab7b07 8929Author: Egbert Eich <eich@freedesktop.org> 8930Date: Fri May 20 13:09:38 2011 +0200 8931 8932 i18n: Fixed typos in full locale names. 8933 8934 Fixing those typos those names actually match entries in 8935 locale.dir. 8936 8937 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8938 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 8939 8940commit b88dd95005dce4c40f9b4d5f938f945e7955fd04 8941Author: Egbert Eich <eich@freedesktop.org> 8942Date: Sat May 21 07:29:38 2011 +0200 8943 8944 i18n: Remove duplicates and aliases to oneself. 8945 8946 Aliases to itself don't make much sense. 8947 This changes occurances of: 8948 xy_UV.UTF-8: xy_UV.UTF-8 8949 to 8950 xy_UV: xy_UV.UTF-8 8951 where applicable. 8952 8953 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8954 Reviewed-by: James Cloos <cloos@jhcloos.com> 8955 8956commit f198c6aa98f88ff285d903175a3c4c0fd33a4575 8957Author: Jens Herden <jens.herden@email.de> 8958Date: Fri May 20 17:29:03 2011 +0200 8959 8960 i18n: Add support for Khmer locale and compose table. 8961 8962 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8963 Reviewed-by: James Cloos <cloos@jhcloos.com> 8964 8965commit 40761898692e5063957bfa2518cca3d35b2e354a 8966Author: Kalman Kemenczy <kkemenczy@novell.com> 8967Date: Fri May 20 17:03:24 2011 +0200 8968 8969 i18n: Add support for Serbian specific compose table entries. 8970 8971 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8972 Reviewed-by: James Cloos <cloos@jhcloos.com> 8973 8974commit 7c14aacc9f01d7a975f8d9d033b9b13cbd777a61 8975Author: Egbert Eich <eich@freedesktop.org> 8976Date: Fri May 20 14:00:35 2011 +0200 8977 8978 i18n: Adding and removing comments. 8979 8980 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8981 Reviewed-by: James Cloos <cloos@jhcloos.com> 8982 8983commit 7754d68976106183751243c2c35a84134be17b34 8984Author: Egbert Eich <eich@freedesktop.org> 8985Date: Fri May 20 13:45:31 2011 +0200 8986 8987 i18n: Add missing locales which existed in locale.alias. 8988 8989 Signed-off-by: Egbert Eich <eich@freedesktop.org> 8990 Reviewed-by: James Cloos <cloos@jhcloos.com> 8991 8992commit 952eccd0d25ba66023acfd31873eee2e71c38c42 8993Author: Egbert Eich <eich@freedesktop.org> 8994Date: Fri May 20 10:57:57 2011 +0200 8995 8996 i18n: Consolidate compose handling for locales with UTF-8 encoding. 8997 8998 - add an entry to include the default en_US compose file 8999 for the ja_JP, ko_KR, th_TH, zh_CN, zh_HK and zh_TW locales. 9000 - add missing entries for zh_CN. and zh_HK and am_ET. 9001 - change entries for the UTF-8 encoding for ru_RU, ja_JP, 9002 ko_KR, th_TH and zh_TW to point to their native directory 9003 entries. 9004 9005 Signed-off-by: Egbert Eich <eich@freedesktop.org> 9006 Reviewed-by: James Cloos <cloos@jhcloos.com> 9007 9008commit d14b6a250f004fa405179db7020f6953001d17b9 9009Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 9010Date: Mon Oct 22 13:54:11 2012 -0500 9011 9012 XIM: remove Private and Public macros 9013 9014 Private is a struct member name in mingw-w64 <winioctl.h>, causing this 9015 useless define in a private header to break the build. 9016 9017 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 9018 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk> 9019 9020commit f2a8def423a46d52e834cf7ea49fa0079427663a 9021Author: Adam Jackson <ajax@redhat.com> 9022Date: Wed Oct 17 14:40:43 2012 -0400 9023 9024 XErrorDB: Add GLXBadProfileARB 9025 9026 Signed-off-by: Adam Jackson <ajax@redhat.com> 9027 9028commit d45b3fc19fbe95c41afc4e51d768df6d42332010 9029Author: Ross Burton <ross.burton@intel.com> 9030Date: Wed Sep 12 14:39:40 2012 +0100 9031 9032 Allow overriding location of keysymdef.h 9033 9034 Currently keysymdef.h is found by using the includedir of xproto. This doesn't 9035 work when cross-compiling with a sysroot as that ends up being /usr/include/X11, 9036 not a path into the cross-build environment. 9037 9038 So, add an option to allow explicitly specifying the location of keysymdef.h, 9039 and verify that the specified or found path exists. 9040 9041 (original patch by Martin Jansa <martin.jansa@gmail.com>, revised by myself) 9042 9043 Signed-off-by: Ross Burton <ross.burton@intel.com> 9044 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 9045 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 9046 9047commit 44cdc0dc2c68d67654023ec707b807145d3a38c0 9048Author: Eric S. Raymond <esr@thyrsus.com> 9049Date: Thu Aug 23 19:15:07 2012 -0400 9050 9051 Renove some unnecessary low-level markup. 9052 9053 Also, SYNTAX -> SYNOPSIS so function prototypes get parsed by doclifter. 9054 This appears to have been somebody's thinko, it's only in a few of the files. 9055 9056 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9057 9058commit b686600ab5ca93b5750f827786e79c329ab2db4d 9059Author: Eric S. Raymond <esr@thyrsus.com> 9060Date: Thu Aug 23 10:53:33 2012 -0400 9061 9062 The .NT/.NE macro pair is no longer used. Remove it. 9063 9064 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9065 9066commit b83f2898528c5dd683acd2a2143879a760a8dad5 9067Author: Eric S. Raymond <esr@thyrsus.com> 9068Date: Thu Aug 23 10:42:46 2012 -0400 9069 9070 Remove the one and only use of the .NT/.NE pair. 9071 9072 The problem with these macros is that they rely on being able to 9073 center the note label. That doesn't play well with modern HTML, 9074 not anyway without coomplications like CSS. This use was just a cute 9075 trick, not adding enough value to be kept. 9076 9077commit e9509fa6745d25eee01ec6f1c34edf8a806d66b3 9078Author: Eric S. Raymond <esr@thyrsus.com> 9079Date: Thu Aug 23 10:23:45 2012 -0400 9080 9081 The .C{ and .C} macros are never used. Remove them. 9082 9083 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9084 9085commit 6b2f7ddeea6c87dbec4ddfc19b1fed33f1bb8575 9086Author: Eric S. Raymond <esr@thyrsus.com> 9087Date: Thu Aug 23 10:08:08 2012 -0400 9088 9089 The .FN macro, paired with .FD, is also never used. Remove it. 9090 9091 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9092 9093commit cac6572701c976542da562b2b277b07ffb892a6f 9094Author: Eric S. Raymond <esr@thyrsus.com> 9095Date: Thu Aug 23 10:03:00 2012 -0400 9096 9097 The ".FD" macro is never used. Remove it. 9098 9099 It was a temptation to presentation-level klugery and is best gone. 9100 9101 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9102 9103commit 6e27a828f39f7028bc7f4a8736e7262fca250632 9104Author: Eric S. Raymond <esr@thyrsus.com> 9105Date: Thu Aug 23 09:54:25 2012 -0400 9106 9107 Clean up, my last commit missed four cases. 9108 9109 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9110 9111commit a7fb575957ff4d9dd3671994a005ac3be8bb10fe 9112Author: Eric S. Raymond <esr@thyrsus.com> 9113Date: Thu Aug 23 08:24:17 2012 -0400 9114 9115 Eliminate all uses of tab stops in the libX11 man pages. 9116 9117 Two steps: First, expand tabs to 8 spaces in code and structure 9118 listings. Second, make the .Ds used to wrap code listings switch to 9119 constant-width font (CW) rather than numeric font position 1, which 9120 maps to R on most systems. 9121 9122 It is possible some archaic systems won't know what CW is, but the 9123 only risk is that code listings won't look quite right on troff 9124 devices; the PostScript and DVI drivers definitely grok it, so those 9125 important cases are OK. 9126 9127 The purpose of these changes is to get rid of presentation-level 9128 markup so these pages will lift clean to DocBook. 9129 9130 Signed-off-by: Eric S. Raymond <esr@thyrsus.com> 9131 9132commit 8042f88ace33573f9d0dfaa21ed54ac7cef266d5 9133Author: Will Thompson <will@willthompson.co.uk> 9134Date: Mon Jul 9 18:00:27 2012 +0100 9135 9136 Add compose sequences for "therefore" and "because". 9137 9138 These sequences look sensible to me. I added them to the APL-related 9139 section of Mathematical Operators—they're in that section of Unicode 9140 anyway. 9141 9142 https://bugs.freedesktop.org/show_bug.cgi?id=51922 9143 9144 Signed-off-by: Will Thompson <will@willthompson.co.uk> 9145 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 9146 9147commit 65358ea5079236b2508f787ac2fb2024a477e36d 9148Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9149Date: Fri Jun 29 23:08:04 2012 -0700 9150 9151 Convert XCreate{Pix,Bit}map...Data to use C99 designated initializers 9152 9153 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9154 9155commit 0dc93f7e43deb102b1f8fb7c4c4844cdce7ffd1e 9156Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9157Date: Fri Jun 29 22:57:13 2012 -0700 9158 9159 XCreate{Pix,Bit}map...Data: Free pixmap in error path if XCreateGC fails 9160 9161 Fixes leaks in error paths found by Parfait 1.0.0: 9162 9163 Error: X Resource Leak 9164 Leaked X Resource pix 9165 at line 62 of CrBFData.c in function 'XCreateBitmapFromData'. 9166 pix initialized at line 60 with XCreatePixmap 9167 Error: X Resource Leak 9168 Leaked X Resource pix 9169 at line 70 of CrPFBData.c in function 'XCreatePixmapFromBitmapData'. 9170 pix initialized at line 66 with XCreatePixmap 9171 9172 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9173 Reviewed-by: Aaron Plattner <aplattner@nvidia.com> 9174 9175commit dce84b8c39ad5a8908c29bb6de25b6c3004c1ab7 9176Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9177Date: Wed Jun 6 13:31:16 2012 -0700 9178 9179 libX11 spec: Correct prototype for XConvertSelection 9180 9181 selection & target parameters were accidentally run together 9182 9183 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9184 9185commit 6c5cb2a90a6479f56855e5167039c37c234cdfe7 9186Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9187Date: Fri Jun 1 23:37:09 2012 -0700 9188 9189 libX11 1.5.0 9190 9191 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9192 9193commit 05c587ec3be880721131a17c1dd4366e458fdd8b 9194Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9195Date: Sat May 26 14:37:28 2012 -0700 9196 9197 libX11 1.4.99.902 (1.5 RC2) 9198 9199 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9200 9201commit d817834d9772e3500d5102e2eae868b04a33c31f 9202Author: Pander <pander@users.sourceforge.net> 9203Date: Sat May 5 19:02:10 2012 +0200 9204 9205 Compose: Reassigned squences with minus and a or o 9206 9207 Reassigned squences with minus and a or o (vice versa and lower and 9208 upper case) to conform existing series and not resulting in tilde. 9209 Also added noe missing underscore sequence. 9210 9211 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9212 9213commit 6bfd1dca6e7cb2046ee6bf9dbbddc0af5ef7cc00 9214Author: Pander <pander@users.sourceforge.net> 9215Date: Sat May 5 16:56:05 2012 +0200 9216 9217 Compose: Removed <slash> <U> and vice versa 9218 9219 Combination with lower case u suffices 9220 9221 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9222 9223commit 3a8b1637132d1e36eb5e11f52dfb284081772d14 9224Author: Pander <pander@users.sourceforge.net> 9225Date: Sat May 5 16:44:31 2012 +0200 9226 9227 Compose: Removed <n> <minus> for n with tilde, also vice versa & for upper case 9228 9229 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9230 9231commit fd514d7a35119dd4413b51c84f2a536f2ca501a6 9232Author: Pander <pander@users.sourceforge.net> 9233Date: Sat May 5 16:40:20 2012 +0200 9234 9235 Compose: Removed <S> in combination with a numeral for superscript 9236 9237 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9238 9239commit 1b5cad3ca54410c4edbca79c23c463e9e088bc0d 9240Author: Pander <pander@users.sourceforge.net> 9241Date: Sat May 5 16:28:01 2012 +0200 9242 9243 Compose: Reassigned <o> <apostrophe> to oacute, also for upper case. 9244 9245 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9246 9247commit 0bbc0d5e605e2a4a3eb00a229b651d4546e8aef4 9248Author: James Cloos <cloos@jhcloos.com> 9249Date: Thu May 10 12:27:28 2012 -0400 9250 9251 Remove duplicate compose sequences from commit e51e37c118ae. 9252 9253 The compose-check script only handles compiled Compose files, not 9254 the Compose.pre files. One must remember to use: 9255 9256 ./autogen.sh; make; make check 9257 9258 when reviewing patches to the Compose.pre files.... 9259 9260 Signed-off-by: James Cloos <cloos@jhcloos.com> 9261 9262commit e51e37c118ae6cb9ced8244ce1c410677e0279ce 9263Author: Geoff Streeter <geoff@dyalog.com> 9264Date: Thu Mar 22 15:02:00 2012 +0000 9265 9266 Add APL support to compose 9267 9268 Signed-off-by: Geoff Streeter <geoff@dyalog.com> 9269 Signed-off-by: James Cloos <cloos@jhcloos.com> 9270 9271commit dac90324cee224df977a428afe80d960dceca769 9272Author: Julien Cristau <jcristau@debian.org> 9273Date: Sat May 5 16:05:07 2012 +0200 9274 9275 configure: make previous change work with older autoconf 9276 9277 autoconf 2.63 doesn't seem to like the nested AC_CHECK_DECL/FUNC. So do 9278 the tests separately. 9279 9280 Reported-by: Dave Airlie 9281 Signed-off-by: Julien Cristau <jcristau@debian.org> 9282 9283commit f5b50af4324186962e258ffe9be78d5ee4681982 9284Author: Julien Cristau <jcristau@debian.org> 9285Date: Sun Apr 29 16:43:09 2012 +0200 9286 9287 configure: check if issetugid is declared 9288 9289 GNU/kFreeBSD has issetugid in libc (for legacy apps?), but doesn't 9290 declare it anywhere, causing gcc to error out with 9291 -Werror=implicit-function-declaration. Use AC_CHECK_DECL in addition to 9292 AC_CHECK_FUNC so we disable this code instead of failing to build it. 9293 9294 Debian bug#669670 <http://bugs.debian.org/669670> 9295 9296 Signed-off-by: Julien Cristau <jcristau@debian.org> 9297 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9298 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9299 9300commit 52e1b5cc3b6de76ccf4285b55652474a522ed9a8 9301Author: Peter Hutterer <peter.hutterer@who-t.net> 9302Date: Mon Apr 30 16:36:47 2012 +1000 9303 9304 Typo fix 9305 9306 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9307 9308commit b64969f0e510d5d3300cf968741a3726a6409577 9309Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9310Date: Tue Apr 17 18:12:02 2012 -0700 9311 9312 Add X11R7 sections to the libX11 & XKBlib credits to cover Docbook conversion 9313 9314 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9315 9316commit 9ea611696f317ac3b3fb67893f1d6d87d49e3b5e 9317Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9318Date: Tue Apr 17 18:01:36 2012 -0700 9319 9320 Add olinks from libX11 & localedb specs to ICCCM spec 9321 9322 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9323 9324commit b3c1b8cdab7d14220426c9b997ac362dc16318fc 9325Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9326Date: Tue Apr 17 17:49:44 2012 -0700 9327 9328 Add olinks from libX11 spec to ICCCM spec 9329 9330 Also convert ICCCM title mentions from <emphasis> to <citetitle> 9331 9332 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9333 9334commit ebebb65e753007ad01966dccc90bd6ca9a826488 9335Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9336Date: Sat Apr 14 23:40:01 2012 -0700 9337 9338 libX11 AppC: Fix section headers that didn't translate from nroff properly 9339 9340 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9341 Acked-by: Peter Hutterer <peter.hutterer@who-t.net> 9342 9343commit d5ab4ae0e74ae1fb30fb72add0751effe2759bf2 9344Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9345Date: Sat Apr 14 23:13:05 2012 -0700 9346 9347 Add olinks from libX11 spec to x11protocol spec 9348 9349 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9350 Acked-by: Peter Hutterer <peter.hutterer@who-t.net> 9351 9352commit 83878a0e34fffd255597300dd3e6cd43fcd645b0 9353Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9354Date: Sat Apr 14 22:28:53 2012 -0700 9355 9356 libX11 spec: Remove .br nroff macro left behind in XGetWindowProperty prototype 9357 9358 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9359 9360commit abc523fce31fcf2687229697a8eb656e343ecb0c 9361Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9362Date: Thu Mar 15 22:14:45 2012 -0700 9363 9364 libX11 1.4.99.901 (1.5 RC1) 9365 9366 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9367 9368commit b2cc5905a4a6d519957223e8ba2caef71520040c 9369Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9370Date: Thu Mar 15 22:03:21 2012 -0700 9371 9372 Remove "register" qualifier that annoys Solaris Studio compiler 9373 9374 Fixes warning: 9375 "Xrm.c", line 1094: warning: storage class after type is obsolescent 9376 9377 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9378 9379commit 62d42953893f93a98db0504eaf06d650ceaf5811 9380Author: James Cloos <cloos@jhcloos.com> 9381Date: Wed Mar 14 17:25:46 2012 -0400 9382 9383 Fix the gtk+ additions 9384 9385 (Some of) the Dstroke and dstroke entries already were present as U011[01], 9386 even though XK_Dstroke and XK_dstroke are part of the latin2 set in keysymdef.h. 9387 9388 The addition of <Multi_key> <o> <apostrophe> as a postfix version of 9389 <Multi_key> <apostrophe> <o> blocks the existing entries for ǻ and Ǻ. 9390 That prevents its and <Multi_key> <O> <apostrophe>’s addition. 9391 9392 Signed-off-by: James Cloos <cloos@jhcloos.com> 9393 9394commit 91bcce48d94792f78333d2aea73961cc2e739d2e 9395Author: Pander <pander@users.sourceforge.net> 9396Date: Wed Mar 14 12:54:53 2012 -0400 9397 9398 Complete compose key sequences for musical symbols 9399 9400 Signed-off-by: Pander <pander@users.sourceforge.net> 9401 Signed-off-by: James Cloos <cloos@jhcloos.com> 9402 9403commit cf040016d455bc37f7665d6714337c5eafd8ea94 9404Author: Pander <pander@users.sourceforge.net> 9405Date: Wed Mar 14 12:46:25 2012 -0400 9406 9407 Add compose sequences from gtk+ to X.Org 9408 9409 Signed-off-by: Pander <pander@users.sourceforge.net> 9410 Signed-off-by: James Cloos <cloos@jhcloos.com> 9411 9412commit a4c591ced5cac9301b9abfa0e521be2d0b267882 9413Author: Keith Packard <keithp@keithp.com> 9414Date: Sun Mar 4 02:00:13 2012 -0800 9415 9416 Block for other threads in _XUserLockDisplay 9417 9418 Wait for all other threads to release the user-level lock when 9419 acquiring it. This ensures that only one thread at a time holds the 9420 user-level lock, necessary as it is a nesting lock and a single 9421 variable is used to determine when the lock is nesting and when it is 9422 contended. 9423 9424 Signed-off-by: Keith Packard <keithp@keithp.com> 9425 Reviewed-by: Jamey Sharp <jamey@minilop.net> 9426 9427commit ed00b460acb08787b695f27b864e96102dfd4867 9428Author: Jon TURNEY <jon.turney@dronecode.org.uk> 9429Date: Fri Oct 28 11:09:20 2011 -0500 9430 9431 Don't use caddr_t casts 9432 9433 (caddr_t) isn't used anywhere else in xcb or libX11. 9434 Cast to (char *) for consistency. 9435 9436 Removing this cast allows building for MinGW without patching. 9437 9438 v2: Cast to (char *) rather than just dropping the cast 9439 9440 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> 9441 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9442 9443commit 20adca02c2a1d0b7c95ecbe387d68f881fd57754 9444Author: Julien Cristau <jcristau@debian.org> 9445Date: Sun Feb 19 13:27:38 2012 +0100 9446 9447 Revert "xcb: Add TCP fallback" 9448 9449 This reverts commit f09c5299a381e2729e800a0ac43f1c0e371f65f6. 9450 9451 The TCP fallback ended up falling back to UNIX socket connection if 9452 $DISPLAY was set to e.g. some.host:0 and the initial attempt failed. 9453 9454 Debian bug#659558 <http://bugs.debian.org/659558> 9455 9456 Signed-off-by: Julien Cristau <jcristau@debian.org> 9457 9458 Conflicts: 9459 9460 src/OpenDis.c 9461 9462commit f4378193619baa9bb973c1b5b718721bbcbe92c7 9463Author: James Cloos <cloos@jhcloos.com> 9464Date: Wed Feb 22 14:13:20 2012 -0500 9465 9466 Make the compose sequence for ẞ work. 9467 9468 There is no XK_Ssharp symbol for U+1E9E LATIN CAPITAL LETTER SHARP S, 9469 so use the U1e9e symbol in the Compose sequence. 9470 9471 (Compose sequences do not work when the target symbol is unknown.) 9472 9473 Signed-off-by: James Cloos <cloos@jhcloos.com> 9474 9475commit 2ca641c3a506dcbee97e279b67990d5387389f36 9476Author: Marko Myllynen <myllynen@redhat.com> 9477Date: Mon Feb 20 17:04:59 2012 +0200 9478 9479 Use ezh/EZH in compose maps 9480 9481 Related: https://bugs.freedesktop.org/show_bug.cgi?id=19687 9482 9483 Signed-off-by: Marko Myllynen <myllynen@redhat.com> 9484 Reviewed-by: Matt Dew <marcoz@osource.org> 9485 Signed-off-by: James Cloos <cloos@jhcloos.com> 9486 9487commit 61725822f20f47684a545c1797183ee7075243ac 9488Author: Marko Myllynen <myllynen@redhat.com> 9489Date: Sun Feb 20 17:09:43 2011 +0200 9490 9491 Provide translation from XK_permille (ad5) to Unicode (U2030) 9492 9493 https://bugs.freedesktop.org/show_bug.cgi?id=19687 9494 9495 Signed-off-by: Marko Myllynen <myllynen@redhat.com> 9496 Reviewed-by: Matt Dew <marcoz@osource.org> 9497 Signed-off-by: James Cloos <cloos@jhcloos.com> 9498 9499commit d2cce0abba0fa0143f49026442c8cab5ed721625 9500Author: Jeremy Huddleston <jeremyhu@apple.com> 9501Date: Sun Feb 12 19:01:43 2012 -0800 9502 9503 nls: Use LC_CTYPE=C for sed magic 9504 9505 Stricter versions of sed can trip up if the input does not match 9506 LC_CTYPE 9507 9508 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9509 9510commit f180e043f55531933bedfa6e0ff36a00a9ec07f3 9511Author: Frédéric Boiteux <fboiteux@calistel.com> 9512Date: Wed Nov 30 12:47:31 2011 +0000 9513 9514 Compositions with the dead greek symbol 9515 9516 FreeDesktop Bug 21475 <https://bugs.freedesktop.org/show_bug.cgi?id=21475> 9517 9518 Signed-off-by: Frédéric Boiteux <fboiteux@calistel.com> 9519 Signed-off-by: James Cloos <cloos@jhcloos.com> 9520 9521commit d58e8f8e27790017fcfdeca0843b7318d541c189 9522Author: Ryan Pavlik <rpavlik@iastate.edu> 9523Date: Sun Jan 1 21:04:52 2012 +0000 9524 9525 Use pthreads on MinGW also 9526 9527 Use pthreads (provided by the pthreads-win32 compatability library which implements 9528 them using native Win32 threading) on MinGW 9529 9530 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk> 9531 9532commit bf15ccb6821664746ec23d769d757edf8059007e 9533Author: Ryan Pavlik <rpavlik@iastate.edu> 9534Date: Mon Jan 23 14:18:02 2012 +0000 9535 9536 Add XWindows.h include to Xxcbint.h 9537 9538 This avoids some conflicting type re-definition errors which occur if 9539 we attempt to include Windows headers after Xmd.h 9540 9541 Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk> 9542 9543commit cadcbd376f0a5d17a71a2fe2f8ced5d93232921a 9544Author: Matt Dew <marcoz@osource.org> 9545Date: Sat Jan 21 17:59:51 2012 -0700 9546 9547 informaltable & table cleanup 9548 9549 On certain tables, add top and bottom borders to table header 9550 and a bottom border to the table. This matches what those 9551 tables in the old pdfs looked like. 9552 9553 the <?dbfo keep-together='always'> prevents tables from 9554 splitting across pages. Useful for tiny tables. 9555 9556 Converting the colwidth to a floating point, IE, 1* -> 1.0* 9557 cleans up these build errors: 9558 WARNING: table-layout="fixed" and column-width unspecified => 9559 falling back to proportional-column-width(1) 9560 9561 Signed-off-by: Matt Dew <marcoz@osource.org> 9562 9563commit bb551f654df8f647c867f79252241964521e689e 9564Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9565Date: Wed Dec 28 21:22:41 2011 -0800 9566 9567 Add more Xkb man pages to the See Also lists for core keyboard functions 9568 9569 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9570 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 9571 9572commit b7bb23fe7c8b8a17128e5bc98a05f68595190438 9573Author: Gaetan Nadon <nadon@memsize.(none)> 9574Date: Fri Dec 30 17:08:14 2011 -0500 9575 9576 docbook.am: embed css styles inside the HTML HEAD element 9577 9578 Rather than referring to the external xorg.css stylesheet, embed the content 9579 of the file in the html output produced. This is accomplished by using 9580 version 1.10 of xorg-xhtml.xsl. 9581 9582 This makes the whole html docs tree much more relocatable. 9583 In addition, it eliminates xorg.css as a runtime file which makes 9584 xorg-sgml-doctools a build time only package. 9585 9586 Signed-off-by: Gaetan Nadon <nadon@memsize.(none)> 9587 9588commit 70505468b7c4a7068cc39be42e421dcee34ec595 9589Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9590Date: Tue Dec 13 19:58:30 2011 -0800 9591 9592 makekeys: move buf declaration from global to main to silence gcc -Wshadow 9593 9594 The global was only referenced in the main() function, which passes it 9595 as an argument of the same name to the parse_line() function, leading 9596 to gcc -Wshadow warnings: 9597 9598 makekeys.c: In function ‘parse_line’: 9599 makekeys.c:58:24: warning: declaration of ‘buf’ shadows a global declaration 9600 makekeys.c:54:13: warning: shadowed declaration is here 9601 9602 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9603 9604commit 87e10a7b9a97c951ab4d477f61177779ac0a6a66 9605Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 9606Date: Wed Dec 14 02:17:55 2011 -0500 9607 9608 XQueryColors: Split a request into multiple requests if necessary 9609 9610 https://bugs.freedesktop.org/show_bug.cgi?id=9160 9611 9612 Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp> 9613 Signed-off-by: James Cloos <cloos@jhcloos.com> 9614 9615commit d3b3570592e9b9e57f270a0bd86762fd205a2833 9616Author: Andreas Wettstein <wettstein509@solnet.ch> 9617Date: Tue Nov 8 20:18:09 2011 +0000 9618 9619 XIM: Allow modifier releases in compose sequences (#26705) 9620 9621 Currently, only non-modifier keys (actually, keysyms) can be part of a compose 9622 sequence, and they are matched against the defined compose sequences at the 9623 time the key is pressed. The patch allows to use modifier keys an well, but 9624 matches them on key release, and only if no other key has been pressed after 9625 the modifier. 9626 9627 Releasing a non-matched modifier during an ongoing compose sequence only aborts 9628 the sequence if any modifier release would have matched. In particular, if no 9629 compose sequences with modifiers are specified, the compose mechanism works 9630 exactly as without this patch. 9631 9632 Even if modifiers are part of a compose sequence, they are not filtered. This 9633 is because modifiers affect the keyboard state no matter what we do here and, 9634 therefore, filtering them only could confuse clients. 9635 9636 The purpose is this extension to the compose mechanism is to allow to make 9637 better use of keys in convenient reach for touch typing. 9638 9639 Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch> 9640 Signed-off-by: James Cloos <cloos@jhcloos.com> 9641 9642commit 56448a626fc90bcf75a1fa2f4c294b0eb1f23bd6 9643Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9644Date: Tue Nov 29 23:56:57 2011 -0800 9645 9646 Reject negative string counts in copy_string_list 9647 9648 Silences parfait warning of a potential memory leak: 9649 Memory leak of pointer 'dst' allocated with malloc(length) 9650 at line 160 of FSWrap.c in function 'copy_string_list'. 9651 'dst' allocated at line 145 with malloc(length). 9652 dst leaks when count <= 0 at line 154. 9653 9654 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9655 9656commit bf2d7c8c6d70539c72560b1921e18df2610acf29 9657Author: Peter Hutterer <peter.hutterer@who-t.net> 9658Date: Fri Nov 11 14:56:36 2011 +1000 9659 9660 libX11 1.4.99.1 9661 9662 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9663 9664commit 24283d40b1e4314c6647dda49d2a159833341a8b 9665Author: Derek Buitenhuis <derek.buitenhuis@gmail.com> 9666Date: Thu Sep 22 18:44:13 2011 -0400 9667 9668 makekeys: Fix build/target word size mismatch when cross-compiling 9669 9670 Since makekeys is built using build environment's gcc and 9671 runs natively, we have to make sure that the size of the 9672 Signature type is the same on both the native environment 9673 and the target, otherwise we get mismatches upon running X, 9674 and some LSB test failures (xts5). 9675 9676 Use an unsigned 32-bit integer on all platforms. Also, 9677 eliminate the redundant multiple typedefs for the 9678 Signature type. 9679 9680 Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com> 9681 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 9682 9683commit 1c41f3b9b86b5eeedfa3bff92e519d45aa097587 9684Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9685Date: Wed Nov 9 21:00:36 2011 -0800 9686 9687 XlcSL.c: convert old-style function definitions to ANSI C89 style 9688 9689 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9690 9691commit f6dad6aaa384bf836baa28ccb72b476a85c40eff 9692Author: Peter Hutterer <peter.hutterer@who-t.net> 9693Date: Thu Oct 27 13:53:22 2011 +1000 9694 9695 Use GetReqSized for GetReq and GetReqExtra 9696 9697 GetEmptyReq and GetResReq cannot do this due to the final typecast - 9698 typically requests that need either of those do not have their own typedef 9699 in the protocol headers. 9700 9701 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9702 Reviewed-by: Jamey Sharp <jamey@minilop.net> 9703 9704commit c9c99058b9d98789c0b2d7e78a23443c2b57a047 9705Author: Peter Hutterer <peter.hutterer@who-t.net> 9706Date: Fri Oct 14 14:51:06 2011 +1000 9707 9708 include: Add GetReqSized() for request buffers of specific size 9709 9710 Some XI2 requests change in size over different versions and libXi would 9711 need to hack around GetReq and GetReqExtra. Add a new GetReqSized so the 9712 library can explicitly specify the size of the request in 4-byte units. 9713 9714 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9715 Reviewed-by: Jamey Sharp <jamey@minilop.net> 9716 9717commit ba8a7a19165e30d14bc165f43f67c19b6a115585 9718Author: Peter Hutterer <peter.hutterer@who-t.net> 9719Date: Thu Oct 27 13:24:10 2011 +1000 9720 9721 Switch GetEmptyReq and GetResReq to call _XGetRequest 9722 9723 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9724 Reviewed-by: Jamey Sharp <jamey@minilop.net> 9725 9726commit 4a060f993bf676cf21ad9784e010f54134da7b40 9727Author: Peter Hutterer <peter.hutterer@who-t.net> 9728Date: Mon Oct 17 09:45:15 2011 +1000 9729 9730 Add _XGetRequest as substitute for GetReq/GetReqExtra 9731 9732 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9733 Reviewed-by: Jamey Sharp <jamey@minilop.net> 9734 9735commit d8956520deb79c1cbb5e974c175bf8493859b22b 9736Author: Alan Coopersmith <alan.coopersmith@oracle.com> 9737Date: Fri Oct 14 17:53:00 2011 -0700 9738 9739 Fix "nomal" -> "normal" typo in several comments 9740 9741 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9742 9743commit 002b36e308a26a152504f9b40aa08a0dce9a7991 9744Author: Xue Wei <Wei.Xue@Sun.COM> 9745Date: Fri Oct 14 17:39:21 2011 -0700 9746 9747 mbtocs should not truncate input 9748 9749 Fixes pasting more than 1024 bytes into xterm, as described in 9750 https://bugs.freedesktop.org/show_bug.cgi?id=25209 9751 9752 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9753 9754commit 9b8d8c9e5b27273e8856a3851ba9b68022bed3cd 9755Author: Marko Myllynen <myllynen@redhat.com> 9756Date: Mon Oct 10 09:41:13 2011 +0300 9757 9758 Add new compose sequences 9759 9760 This patch adds few new compose sequences to the en_US map, ligatures 9761 common in typesetting and sequences already in use in the fi_FI map. 9762 9763 https://bugs.freedesktop.org/show_bug.cgi?id=30621 9764 https://bugs.freedesktop.org/show_bug.cgi?id=34523 9765 9766 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 9767 9768commit 738f7b867341c1da87ee667d48815822715c3e75 9769Author: Choe Hwanjin <choe.hwanjin@gmail.com> 9770Date: Thu Oct 13 07:58:02 2011 +0900 9771 9772 XIM: Make Xim handle NEED_SYNC_REPLY flag 9773 9774 NEED_SYNC_REPLY flag should be in Xim not in Xic. 9775 Because the focused Xic can be changed before sending sync reply. 9776 After focused Xic changed, the new Xic doesn't have NEED_SYNC_REPLY 9777 flag enabled, so libX11 doesn't send XIM_SYNC_REPLY packet. 9778 9779 This patch adds sync reply flag to Xim and removes sync reply 9780 from Xic. 9781 9782 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=7869 9783 9784 Signed-off-by: Choe Hwanjin <choe.hwanjin@gmail.com> 9785 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 9786 9787commit 50e15379331c436851abb194251cde96999e67b3 9788Author: Bodo Graumann <mail@bodograumann.de> 9789Date: Mon Oct 10 16:40:52 2011 +0200 9790 9791 libX11: Fixing modifier key range in Xutil.h (Bug #21910) 9792 9793 IsModifierKey, defined in include/X11/Xutil.h, is a macro determining, 9794 which keys are regarded as modifiers. The constants ISO_Level5_Shift, 9795 ISO_Level5_Latch and ISO_Level5_Lock where excluded previously, leaving 9796 some Neo2 modifiers functionless in combination with compose. 9797 This patch adjusts the range to include the correct, full range of 9798 modifier constants. 9799 9800 Neo2 Bug 277 <http://wiki.neo-layout.org/ticket/277> 9801 9802 X.Org Bug 21910 <http://bugs.freedesktop.org/show_bug.cgi?id=21910> 9803 9804 Signed-off-by: Bodo Graumann <mail@bodograumann.de> 9805 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 9806 9807commit b5a108624331fabf393223c0891914cc54d4caf4 9808Author: Yann Droneaud <yann@droneaud.fr> 9809Date: Tue Oct 11 17:27:59 2011 +0200 9810 9811 Return name instead of False in XSetICValues() 9812 9813 In case of error, XSetICValues() must return the first argument 9814 that failed to be set. 9815 9816 But in some error paths, it returns False, which is converted to NULL, 9817 so the function returns OK in case of error. 9818 9819 Signed-off-by: Yann Droneaud <yann@droneaud.fr> 9820 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 9821 9822commit 10992cb8dc5e4d938e7e5a633b68a81b5875f3ba 9823Author: Yann Droneaud <yann@droneaud.fr> 9824Date: Sun Oct 9 17:56:45 2011 +0200 9825 9826 Return name instead of value in XGetIMValues() and XSetIMValues() 9827 9828 As stated in man page (XOpenIM) and Xlib documentation (chapter 13.5.3), 9829 XGetIMValues() and XSetImValues() "returns the name of the first argument 9830 that could not be obtained." 9831 9832 But currently, 9833 9834 err = XGetIMValues(im, "invalid", &arg, NULL); 9835 9836 returns &arg instead of "invalid". 9837 9838 This patch fixes https://bugs.freedesktop.org/show_bug.cgi?id=12897 9839 9840 Signed-off-by: Yann Droneaud <yann@droneaud.fr> 9841 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 9842 9843commit f2651e03f3295a453a2965c3749bc8b6e66f1c09 9844Author: Jeremy Huddleston <jeremyhu@apple.com> 9845Date: Mon Oct 10 14:09:17 2011 -0700 9846 9847 Mark XKeycodeToKeysym as _X_DEPRECATED 9848 9849 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9850 9851commit 25d7bb9c705623ebbb3afba3b86d0ded5b6bb4fb 9852Author: Jeremy Huddleston <jeremyhu@apple.com> 9853Date: Sun Oct 9 02:38:28 2011 -0700 9854 9855 Fix nobreakspace for pt_BR.UTF-8 9856 9857 https://bugs.freedesktop.org/show_bug.cgi?id=31334 9858 9859 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9860 9861commit 913603660c17ec59a4d39d3e3b9d60469ee5c060 9862Author: Jeremy Huddleston <jeremyhu@apple.com> 9863Date: Sun Oct 9 02:26:45 2011 -0700 9864 9865 Fix potential uninitialized variable access in _XimMakeICAttrIDList 9866 9867 Found by clang static analysis 9868 9869 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9870 9871commit 566ceaf5a92c721ac7155528e4d0d2e5cbef023f 9872Author: Jeremy Huddleston <jeremyhu@apple.com> 9873Date: Sun Oct 9 02:25:50 2011 -0700 9874 9875 Remove self-resolving aliases 9876 9877 https://bugs.freedesktop.org/show_bug.cgi?id=30112 9878 9879 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9880 9881commit 22ba43d198dcca86c88eb15a56fc7d8fc47c422e 9882Author: Matt Dew <marcoz@osource.org> 9883Date: Fri Oct 7 22:52:30 2011 -0600 9884 9885 Cleanup IDs and links in doc 9886 9887 1 - fix the capitalization of the ID attributes to match either the 9888 <title> or <funcdef> string it goes with. 9889 2 - fix any <linkend>'s that were affected by 1. 9890 3 - any <function> in the docs that has an actual funcdef, 9891 will become an olink. 9892 9893 Signed-off-by: Matt Dew <marcoz@osource.org> 9894 9895commit f858f3326adbc0c5711669b92a64a84a9083a055 9896Author: James Cloos <cloos@jhcloos.com> 9897Date: Tue Oct 4 17:11:11 2011 -0400 9898 9899 [nls] Fix typo/synco. 9900 9901 The iso8859-1 Compose table includes an optional section which uses 9902 Ctrl<T> as a substitute for <Multi_key>. In that section the sequence 9903 to generate an @ (at) either was incorrectly copied from the Multi_key 9904 section or was not kept in sync with the Multi_key section. 9905 9906 Fixing this eliminates the warning from compose-check.pl: 9907 9908 ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9909 ┃ Clash with existing sequence in iso8859-1/Compose.pre 9910 ┃ on line 661: Ctrl<T> <A> <A> 9911 ┃ line #661: Ctrl<T> <A> <A> : "\305" Aring 9912 ┃ line #480: Ctrl<T> <A> <A> : "@" at 9913 ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9914 9915 Signed-off-by: James Cloos <cloos@jhcloos.com> 9916 9917commit 7f35f7efc23234b3f8529fee31a802c21846d9ab 9918Author: Jeremy Huddleston <jeremyhu@apple.com> 9919Date: Mon Sep 26 11:32:56 2011 -0700 9920 9921 Remove conflicting compose sequences for cent and colon 9922 9923 Regression from 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4 9924 Found by tinderbox 9925 9926 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9927 9928commit 4d78ad4bf6dcabca9bb5f84c770abfbb02d3f7a4 9929Author: Jeremy Huddleston <jeremyhu@apple.com> 9930Date: Sun Sep 25 16:29:17 2011 -0700 9931 9932 Add additional compose sequences for pound sterling, yen, and cent (mixed case) 9933 9934 https://bugs.freedesktop.org/show_bug.cgi?id=1013 9935 9936 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9937 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 9938 9939commit 0ca4153a373e2af7a9eb3cef003393ab332bc79e 9940Author: Jeremy Huddleston <jeremyhu@apple.com> 9941Date: Sat Sep 24 16:44:04 2011 -0700 9942 9943 Use a configure check for seteuid 9944 9945 HP-UX doesn't have seteuid 9946 9947 https://bugs.freedesktop.org/show_bug.cgi?id=1497 9948 9949 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 9950 9951commit 79594b4d66344f248eb4314ecb78eb81f632c3ab 9952Author: Gaetan Nadon <memsize@videotron.ca> 9953Date: Tue Sep 20 16:06:10 2011 -0400 9954 9955 localedb specs: use <copyright> for first holder of multi license 9956 9957 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9958 9959commit 154430268cf004b06920edde2f1812dc16d71fae 9960Author: Gaetan Nadon <memsize@videotron.ca> 9961Date: Tue Sep 20 16:01:52 2011 -0400 9962 9963 libX11 specs: use <copyright> for first holder of multi license 9964 9965 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9966 9967commit 66e621f58dee804acb795ff2c79ce6dd38b34a3f 9968Author: Tollef Fog Heen <tfheen@err.no> 9969Date: Tue Sep 20 13:07:37 2011 -0400 9970 9971 NLS: Add more vulgar fractions 9972 9973 Add 1/7, 1/9, 1/10 and 0/3 vulgar fractions. 9974 9975 Signed-off-by: Tollef Fog Heen <tfheen@err.no> 9976 Signed-off-by: James Cloos <cloos@jhcloos.com> 9977 9978commit e37a6da814b5653be46000a9a76902729660a2e6 9979Author: Alexander Polakov <polachok@gmail.com> 9980Date: Sat Sep 17 20:54:58 2011 +0400 9981 9982 XGrabKey manual page: change XAllowAccess to XAllowEvents in See Also 9983 9984 There is no XAllowAccess man page to see. 9985 9986 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9987 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 9988 9989commit a23f3323f269b33a43d16dce01395cd28bc41b45 9990Author: Gaetan Nadon <memsize@videotron.ca> 9991Date: Mon Sep 19 15:27:44 2011 -0400 9992 9993 XKB: provide adequate quotes for the license text 9994 9995 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9996 9997commit a9c7a5cad91d968f5536ef8fc735036e921832a3 9998Author: Gaetan Nadon <memsize@videotron.ca> 9999Date: Mon Sep 19 15:18:53 2011 -0400 10000 10001 XIM: refactor the multi licensing legal text 10002 10003 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10004 10005commit c46f934ed89e7de79746a0387c9f998d91994ea6 10006Author: Gaetan Nadon <memsize@videotron.ca> 10007Date: Mon Sep 19 10:33:30 2011 -0400 10008 10009 xim trans: restore Fujitsu copyright legal text 10010 10011 Somehow lost during docbook conversion. text from x.org ftp R7.5. 10012 10013 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10014 10015commit 3d75f993382bfdc89d31668d7dfc71c91222e0d7 10016Author: Gaetan Nadon <memsize@videotron.ca> 10017Date: Mon Sep 19 10:11:46 2011 -0400 10018 10019 xtrans: restore X Consortium original legal text 10020 10021 Asking X Consortium permission to use The Open Group name makes no sense. 10022 Even more so in 1994 before X Window System was passed on to the Open Group. 10023 10024 Using original text from xorg-docs/general/License 10025 10026 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10027 10028commit 136a381585a4a9686c11bad1a6130837978e677f 10029Author: Gaetan Nadon <memsize@videotron.ca> 10030Date: Mon Sep 19 10:03:21 2011 -0400 10031 10032 Framework: restore X Consortium copyright 10033 10034 Somewhat dammaged during docbook conversion. 10035 Also restore pasrt of the original license text 10036 10037 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10038 10039commit 33f346878406762704a0c13cdc018111c666f3c1 10040Author: Gaetan Nadon <memsize@videotron.ca> 10041Date: Mon Sep 19 09:50:00 2011 -0400 10042 10043 localedb: restore X Consortium original legal text 10044 10045 Asking X Consortium permission to use The Open Group name makes no sense. 10046 Even more so in 1994 before X Window System was passed on to the Open Group. 10047 10048 Using original text from xorg-docs/general/License 10049 10050 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10051 10052commit e99c9338e413efca026210b9d830efb5c74d34e5 10053Author: Gaetan Nadon <memsize@videotron.ca> 10054Date: Mon Sep 19 08:23:18 2011 -0400 10055 10056 specs: support multi licensed copyright notice and license text 10057 10058 For documentation having multiple licenses, the copyright and legalnotice 10059 elements sequence cannot instantiated multiple times. 10060 The copyright notice and license text are therefore coded inside a 10061 legalnotice element. The role attribute on the paragraph is used to allow 10062 styling of the copyright notice text which should not be italicized. 10063 10064 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10065 10066commit b9dedc757e667333ce899115d618f25cdaa4dd5e 10067Author: Gaetan Nadon <memsize@videotron.ca> 10068Date: Mon Sep 19 08:17:50 2011 -0400 10069 10070 localedb: add release info to spec 10071 10072 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10073 10074commit 4519c89a87ddce01e599542737a27c26030fe4ff 10075Author: Gaetan Nadon <memsize@videotron.ca> 10076Date: Mon Sep 19 08:05:32 2011 -0400 10077 10078 specs: fix The Open Group license text 10079 10080 The warranty referred to the X Consortium 10081 10082 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10083 10084commit 08ac378423e7be72e340197ba5559a1a36f1783c 10085Author: Gaetan Nadon <memsize@videotron.ca> 10086Date: Mon Sep 19 08:03:39 2011 -0400 10087 10088 specs: The strandard name is still "X Consortium Standard" 10089 10090 This spec, and fsproto spec, are the only two docs with a different 10091 standard name. 10092 10093 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10094 10095commit 8dfbeb1b1b1e8aa17f69d0a373155419a5f6a172 10096Author: Gaetan Nadon <memsize@videotron.ca> 10097Date: Sun Sep 18 13:22:34 2011 -0400 10098 10099 specs: support multi licensed copyright notice and license text 10100 10101 For documentation having multiple licenses, the copyright and legalnotice 10102 elements sequence cannot instantiated multiple times. 10103 The copyright notice and license text are therefore coded inside a legalnotice 10104 element. The role attribute on the paragraph is used to allow styling of the 10105 copyright notice text which should not be italicized. 10106 10107 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10108 10109commit 278ca8947c1dabb2b819527dca0fa6190c034f67 10110Author: Gaetan Nadon <memsize@videotron.ca> 10111Date: Fri Sep 16 11:45:56 2011 -0400 10112 10113 docs: merge copyright holder under the same copyright notice 10114 10115 As per the docbook markup dtd. 10116 10117 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10118 10119commit 7ff012bb433109a622ac122ade71669e842ebdcf 10120Author: Gaetan Nadon <memsize@videotron.ca> 10121Date: Fri Sep 16 10:09:57 2011 -0400 10122 10123 specs: handle multiple sets of copyright notice/license/warranty 10124 10125 Docbook groups all the <copyright> elements together and all the 10126 <legalnotice> elements together. 10127 10128 We cannot have a sequence: 10129 <copyright> <legalnotice> <copyright> <legalnotice> [...] 10130 10131 A workaround, which was done in some documents, is to put the copyright 10132 notice inside the legalnotice in plain text without the <copyright> element. 10133 A formal paragraph title is added here which makes the copyright notice bold, 10134 and makes it much easier to locate. 10135 10136 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10137 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10138 10139commit 4a550c71b8221c37b1a9378d5a170da9eaa03405 10140Author: Gaetan Nadon <memsize@videotron.ca> 10141Date: Wed Sep 14 15:55:42 2011 -0400 10142 10143 specs: remove orphan affiliation. 10144 10145 Authors affiliation are correct. 10146 10147 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10148 10149commit 0cc02a6df65214560b7575e89cebee741d6d2469 10150Author: Gaetan Nadon <memsize@videotron.ca> 10151Date: Tue Sep 13 17:39:02 2011 -0400 10152 10153 specs: use appropriate markup for Copyright statements 10154 10155 Also move <releaseinfo> to match order of appearance 10156 10157 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10158 10159commit afe13e19ebf0fc4e3460644164433af016f0add7 10160Author: Gaetan Nadon <memsize@videotron.ca> 10161Date: Mon Sep 12 16:54:45 2011 -0400 10162 10163 docs: use the &fullrelvers; entity to set X11 release information 10164 10165 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10166 10167commit 22a2153282db13e9d94881022b8c979539841a2c 10168Author: Gaetan Nadon <memsize@videotron.ca> 10169Date: Sun Sep 11 19:49:53 2011 -0400 10170 10171 docs: remove <productnumber> which is not used by default 10172 10173 This element is not rendered by default on the title. A template 10174 customization is required to display it. 10175 X Window System does not have a product number. 10176 10177 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10178 10179commit 719f16570d9fcfd15247813ee51fa51ac8a6ff4c 10180Author: Gaetan Nadon <memsize@videotron.ca> 10181Date: Sun Sep 11 17:40:21 2011 -0400 10182 10183 docs: use the &fullrelvers; entity to set X11 release information 10184 10185 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10186 10187commit a6b2992f50b571d612ea9ade631c432c3099bc12 10188Author: Gaetan Nadon <memsize@videotron.ca> 10189Date: Sun Sep 11 17:38:42 2011 -0400 10190 10191 docs: remove orphan <affiliation> 10192 10193 Somehow created during the conversion from roff. Unable to locate 10194 the author to which it belongs. 10195 10196 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10197 10198commit c7420060b6e47b8467ba50c796ec3c1bab090bc7 10199Author: Gaetan Nadon <memsize@videotron.ca> 10200Date: Sun Sep 11 17:38:05 2011 -0400 10201 10202 docs: remove <productnumber> which is not used by default 10203 10204 This element is not rendered by default on the title. A template 10205 customization is required to display it. 10206 X Window System does not have a product number. 10207 10208 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10209 10210commit 7d5b718c1edbc43c54b7001be548e515a49540ea 10211Author: Gaetan Nadon <memsize@videotron.ca> 10212Date: Fri Sep 9 14:46:39 2011 -0400 10213 10214 docbook.am: embed css styles inside the HTML HEAD element 10215 10216 Rather than referring to the external xorg.css stylesheet, embed the content 10217 of the file in the html output produced. This is accomplished by using 10218 version 1.10 of xorg-xhtml.xsl. 10219 10220 This makes the whole html docs tree much more relocatable. 10221 In addition, it eliminates xorg.css as a runtime file which makes 10222 xorg-sgml-doctools a build time only package. 10223 10224 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10225 10226commit 24632d280491c465dfe208ef788815890892e326 10227Author: Gaetan Nadon <memsize@videotron.ca> 10228Date: Mon Sep 5 12:53:12 2011 -0400 10229 10230 compose: upgrade makefile to support olinking on chunked html 10231 10232 The essential differences over the regular docbook.am are: 10233 Adding root.filename parameter for naming of chapters html files. 10234 Using xhtml xmlto format and xorg-chunk.xsl stylesheet 10235 Set olink.base.uri for pdf but not for chunked html 10236 Olink is not applicable to ps and txt formats. 10237 10238 Html chapters are added to shelf_DATA as they are also installed. 10239 The xml is generated from a perl script and not distributed. 10240 10241 Requires version 1.10 of xorg-sgml-doctools. 10242 10243 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10244 10245commit 93ab0a1c9a4f9094cf0c8e1c686130673e681798 10246Author: Gaetan Nadon <memsize@videotron.ca> 10247Date: Wed Aug 31 19:46:20 2011 -0400 10248 10249 docbook.am: refactor common flags for xmlto and xsltproc 10250 10251 Maximize reuse and reduce risk of setting the wrong flag 10252 at the wrong place. 10253 10254 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10255 10256commit 85b725aa9ef76aeb5e94d42e0cd60f896da7ab97 10257Author: Gaetan Nadon <memsize@videotron.ca> 10258Date: Wed Aug 31 10:46:59 2011 -0400 10259 10260 docbook.am: add search path for local entities 10261 10262 Currently, only $(XORG_SGML_PATH)/X11 is searched for xml entities. 10263 A module may want to add entities that apply only to itself and 10264 not to all modules, like the xserver does. 10265 10266 This feature may or may not be used in this module, but all modules 10267 share a copy of docbook.am. 10268 10269 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10270 10271commit b5ed2d368d47f244d43d2513329fa2758bc4ef54 10272Author: Gaetan Nadon <memsize@videotron.ca> 10273Date: Thu Aug 25 14:39:20 2011 -0400 10274 10275 docbook.am: explicitly list xmlto flags for each target 10276 10277 Normal evolution of make targets make it impractical to factor out 10278 common command flags for xmlto. 10279 10280 The targets now list each command option as its presence or absence 10281 needs to be justified. 10282 10283 xorg.ss is only needed by xmlto for html. 10284 masterdb is only needed by xmlto. 10285 img.src.path must not be used by html. 10286 xsltproc need to use customization layer xorg-*.xsl. 10287 txt format is not required to search masterdb. 10288 10289 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10290 10291commit d5b3605f3b0639452e892e9ff3afb18bb7442069 10292Author: Gaetan Nadon <memsize@videotron.ca> 10293Date: Thu Aug 25 13:12:49 2011 -0400 10294 10295 docbook.am: do not generate docs if docbook customization layer is missing 10296 10297 The stylesheets used to be only about style, fonts or colors. 10298 Complex features are now used like olink and chunked html which 10299 may cause a build break when stylesheets are missing or lead to hard 10300 to find problem. 10301 10302 Some modules may be built when stylesheets are present while others 10303 are built without. There is no requirement to build crippled docs. 10304 10305 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10306 10307commit 7f23c72c94d9d14122426b0d8c66054f1402769a 10308Author: Gaetan Nadon <memsize@videotron.ca> 10309Date: Fri Aug 5 14:33:32 2011 -0400 10310 10311 libX11 specs: review doclifter generated tables 10312 10313 Many tables had a questionnable layout and some had information dropped. 10314 Each table was cross-referenced with a pre-docbook version 10315 to ensure semantic integrity. 10316 10317 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10318 10319commit 1efdbeb8cd08f443188ac783cbab6ba0654a66d6 10320Author: Gaetan Nadon <memsize@videotron.ca> 10321Date: Wed Jul 20 15:28:47 2011 -0400 10322 10323 credits.xml: remove toc from Acknowledgments 10324 10325 There should be no toc for a simple preface with only one 10326 Acknowledgments section. 10327 10328 Use <simplesect> markup rather than sect1. 10329 10330 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10331 10332commit 77cc17ed0817c1df3ac5feb9dc83a5ba5d021a4b 10333Author: Gaetan Nadon <memsize@videotron.ca> 10334Date: Thu Aug 4 19:59:37 2011 -0400 10335 10336 libX11 specs: move </para> above <varaiablelist> 10337 10338 Many, but not all, function synopsis have a paragraph that 10339 nests the variable list. The code was generated by doclifter 10340 and there is no apprent reasons as to why it was done this way. 10341 Found while investigating a spacing issue. 10342 10343 Before the patch: 10344 <para>Some blurb about the function 10345 <funcsynopsis> 10346 </funcsynopsis> 10347 <variablelist> 10348 </variablelist> 10349 </para> 10350 10351 After the patch: 10352 <para>Some blurb about the function</para> 10353 <funcsynopsis> 10354 </funcsynopsis> 10355 <variablelist> 10356 </variablelist> 10357 10358 There are no noticable differences when reading the doc 10359 other than than the removed few pixels of spacing between 10360 the function sysnopsis and the variable list block. 10361 10362 In some cases, there are no "blurb about the fucntion" 10363 and the empty paragraph is removed. 10364 10365 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10366 10367commit 18595f0d8f95ff1c12e7b9f73bd4b30dda5c7168 10368Author: Gaetan Nadon <memsize@videotron.ca> 10369Date: Thu Jul 28 19:46:14 2011 -0400 10370 10371 compose specs: generate chunked html 10372 10373 For large 600 page documents such as this one, chunked html 10374 provide faster browser load time and better navigation. 10375 10376 Simply click on the locale of your choice in the toc and 10377 the browser loads just that one file. Being a DocBook, 10378 it benefits from all of the usual features and can be 10379 integrated with the rest of the documentation. 10380 10381 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10382 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10383 10384commit 5cc8815b8a117cf0b6af37e7d55c5a420d9fff57 10385Author: Gaetan Nadon <memsize@videotron.ca> 10386Date: Thu Jun 23 15:16:47 2011 -0400 10387 10388 specs: build compose keys tables in specs/i18n/compose 10389 10390 Build the tables article together with the rest of the specs. 10391 Make it transparent that the source in generated in nls. 10392 Reuse docbook.am and get all the features such as olink. 10393 10394 The docbook article file stem and id must be the same. The new name 10395 for the main article is libX11-keys. The new installation location 10396 is $docdir/libX11/i18n/compose. 10397 10398 The nls dir retains the role of generating the DocBook/XML source 10399 but does not build neither installs output formats or generated source. 10400 10401 The tables article now has to specify that each included locale 10402 section is a dependency. It did not matter before as they were web links. 10403 10404 The xorg-sgml-doctools masterdb shall be updated to include this new doc. 10405 Install location moved from doc/libX11/Compose to doc/libX11/i18n/compose. 10406 10407 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10408 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10409 10410commit adebbe3856b4aee39c0c7af902c1dea847dd6d7c 10411Author: Gaetan Nadon <memsize@videotron.ca> 10412Date: Thu Jun 23 14:43:31 2011 -0400 10413 10414 nls: restructure charts as a single article with sections 10415 10416 Looks more like a real article with a toc rather than individual 10417 web pages. Looks nicer in pdf. 10418 10419 Each locale is a "section" rather than an "article". 10420 Using XInclude to aggregate xml source files gets you the toc for free. 10421 10422 The single document is over 600 pages while there were 62 separate 10423 documents previously. FOP version 1.0 is required to handle missing 10424 character like capital sharp s. 10425 10426 Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10427 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10428 10429commit fc74dc12b1ff3c43e240e1a713316ce1bf525d61 10430Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10431Date: Thu Jul 28 19:32:40 2011 -0700 10432 10433 libX11 1.4.4 10434 10435 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10436 10437commit ab1f30231755d99ed123af6873b981834640f01d 10438Author: Matt Dew <marcoz@osource.org> 10439Date: Wed Jul 13 12:33:40 2011 -0600 10440 10441 Add id attributes to error codes to allow linking from other docs. 10442 10443 Signed-off-by: Matt Dew <marcoz@osource.org> 10444 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10445 Acked-by: Gaetan Nadon <memsize@videotron.ca> 10446 10447commit 5c831fef402914ccf2ec14005c25be48852f119b 10448Author: David Coppa <dcoppa@gmail.com> 10449Date: Tue Jul 12 10:05:47 2011 +0200 10450 10451 Fix libpthread linkage on OpenBSD. 10452 10453 OpenBSD prefers to use the -pthread to fetch pthread libs when needed. 10454 10455 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr> 10456 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10457 10458commit 445b0b3efc04a74fc77cd3e483b25cfb5da82f12 10459Author: Gaetan Nadon <memsize@videotron.ca> 10460Date: Wed Jun 22 20:46:26 2011 -0400 10461 10462 config: use XORG_WITH_PERL macro to replace custom program check 10463 10464 New in version util-macros 1.15, the macro does the usual checking 10465 for program path and provides an interface for makefile variables. 10466 10467 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10468 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10469 10470commit 47d99a324db9232029a22dd523e3bc3ed0936aca 10471Author: Gaetan Nadon <memsize@videotron.ca> 10472Date: Wed Jun 22 20:41:18 2011 -0400 10473 10474 config: update XORG_WITH_FOP macro usage 10475 10476 A new feature to test for a minimum version has been added. 10477 10478 The parameter position changed. Due to limited usage 10479 and requiring 1.15, it turns out to be backward compatible. 10480 10481 There is no functional change to how docs are build. 10482 10483 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10484 10485commit e659683ad51ecc26c989fbdefdfe20f83af2c5a5 10486Author: Gaetan Nadon <memsize@videotron.ca> 10487Date: Wed Jun 22 11:48:49 2011 -0400 10488 10489 Remove unused xmlrules.in in specs, now in /nls 10490 10491 The specs dir has been converted to docbook.am 10492 10493 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10494 10495commit 9992092609bd83a0db778e9a9ac1e879d445e637 10496Author: Gaetan Nadon <memsize@videotron.ca> 10497Date: Thu Mar 31 10:23:22 2011 -0400 10498 10499 nls: move xmlrules.in from specs to nls 10500 10501 It is no longer used in the specs dir. 10502 Remove html from CLEANFILES as it breaks. Make do not clean directories 10503 Remove SUFFIXES as pattern rules are used rather than suffix rules. 10504 10505 Other improvements are possible, the first priority is to move it out 10506 of the specs dir. 10507 10508 Acked-by: Peter Hutterer <peter.hutterer@who-t.net> 10509 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10510 10511commit b370eeb0191b0e2a8bde2f03a442a8ca62e35d58 10512Author: Gaetan Nadon <memsize@videotron.ca> 10513Date: Sun Jun 5 16:27:36 2011 -0400 10514 10515 Install target dbs alongside generated documents 10516 10517 This matches a change in xorg-sgml-docs whereby the masterdb will look for 10518 the target dbs into the same location as the generated documents. 10519 10520 The target dbs are now installed alongside the generated documents. 10521 Previously they are installed in $prefix/sgml/X11/dbs alongside masterdb which 10522 has the potential of installing outside the package prefix and cause 10523 distcheck to fail when user does not have write permission in this package. 10524 10525 Requires XORG_CHECK_SGML_DOCTOOLS(1.8) which was released 2011-06-11 10526 10527commit 4e741654b7e0a283779b8b4af5bb32e05b7fc678 10528Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10529Date: Tue Jun 7 18:41:30 2011 -0700 10530 10531 specs/libX11: Turn appendix references into links 10532 10533 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10534 10535commit a3d1152bc4b300f59ea2f08925739aab1ebfbc4e 10536Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10537Date: Tue Jun 7 18:37:41 2011 -0700 10538 10539 specs/libX11: Turn chapter references into links 10540 10541 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10542 10543commit d0cc949dfe44a41a71f19e4fe6c7547bb3d9bdbe 10544Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10545Date: Tue Jun 7 00:52:32 2011 -0700 10546 10547 specs/libX11: Turn many "see section ..." into links 10548 10549 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10550 10551commit 02d4c08dead2f266809b3f93ec72377783423ac7 10552Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10553Date: Sun Jun 5 22:50:37 2011 -0700 10554 10555 specs/libX11: Convert some header filenames to filename tags 10556 10557 perl -i -p -e 's{^<(.*\.h)>\ *}{<filename class="headerfile"><\1></filename>}' *.xml 10558 10559 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10560 10561commit c51a011329afa6e5b9d2b285349c132683ecf9fb 10562Author: Matt Dew <marcoz@osource.org> 10563Date: Tue May 31 20:03:23 2011 -0600 10564 10565 Add id attributes to funcsynopsis to allow other docs to olink to them. 10566 10567 Signed-off-by: Matt Dew <marcoz@osource.org> 10568 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10569 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 10570 10571commit 6841260c8bb15404a0b4805bee3b0bdfec7176b3 10572Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10573Date: Fri May 27 20:14:36 2011 -0700 10574 10575 Bug 37624 - parameter x is missing from synopsis of XDrawString in libX11.html 10576 10577 https://bugs.freedesktop.org/show_bug.cgi?id=37624 10578 10579 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10580 10581commit 205af6a4e557c62a6395feadc1c89f4a9fe1713b 10582Author: Matt Dew <marcoz@osource.org> 10583Date: Sat May 21 00:23:06 2011 -0600 10584 10585 Add id's to functions, to make clickable links. 10586 10587 Signed-off-by: Matt Dew <marcoz@osource.org> 10588 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 10589 10590commit 5c06bc594473f6ab234724cd90db32e7b57fe811 10591Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10592Date: Fri May 20 15:38:08 2011 -0700 10593 10594 libX11 spec: fix monospaced column alignment after deligaturization 10595 10596 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10597 10598commit ef0019c714c273cb9b4ad3eba52d0b905109826b 10599Author: James Cloos <cloos@jhcloos.com> 10600Date: Fri May 20 15:51:46 2011 -0400 10601 10602 Expand latin ligatures out to their NFKC equivalents. 10603 10604 Ligatures should be done by the typesetting system. 10605 Manuscripts should not bother with them. 10606 10607 Signed-off-by: James Cloos <cloos@jhcloos.com> 10608 Reviewed-by: Keith Packard <keithp@keithp.com> 10609 10610commit 761b8aa0c9b3c58c478ac5ea1b3aaafadcfc1325 10611Author: Daniel Stone <daniel@fooishbar.org> 10612Date: Thu May 12 16:21:50 2011 +0200 10613 10614 XCB: Add more friendly error messages for common asserts 10615 10616 This patch adds more friendly error messages for three common classes of 10617 assertion: 10618 - missed sequence numbers due to being griefed by another thread 10619 - unknown requests in queue due to being griefed by another thread 10620 - extensions dequeuing too much or too little reply data 10621 10622 It adds error messages offering advice (e.g. call XInitThreads() first) 10623 on stderr, but still generates actual assertions. Hopefully this means 10624 it's a little more Googleable and a little less frightening. 10625 10626 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 10627 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 10628 10629commit 159bf292477048b9a2f074735afc516f52c93d80 10630Author: Gaetan Nadon <memsize@videotron.ca> 10631Date: Thu May 19 20:22:11 2011 -0400 10632 10633 Remove misplaced hyphens in libX11 DocBook/XML #37364 10634 10635 Reported-by: Christopher Yeleighton <giecrilj@stegny.2a.pl> 10636 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 10637 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10638 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 10639 10640commit 0ee6d8247d397500ae183ef180b6ff21bceefa1c 10641Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10642Date: Tue May 17 20:49:59 2011 -0700 10643 10644 Fix man page and comment references to use XFreeModifiermap (lowercase map) 10645 10646 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10647 10648commit 3275ec4ca8790d571b6a1902367ca5f68a2dc5fd 10649Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10650Date: Tue May 17 20:30:54 2011 -0700 10651 10652 XKB man pages: Fix coordinantes typo in multiple pages 10653 10654 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10655 10656commit ca33a80606cd68e4572fa49ed6aaa1d1fc31a47f 10657Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10658Date: Tue May 17 16:27:32 2011 -0700 10659 10660 DisplayOfCCC.man: Fix typo "ClientWhitePointOfCC" -> "ClientWhitePointOfCCC" 10661 10662 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10663 10664commit ae39d82b01359b07dd1fe092d867f79ea95a5438 10665Author: Jeremy Huddleston <jeremyhu@apple.com> 10666Date: Sun May 8 09:08:07 2011 -0700 10667 10668 Silence clang static analysis warnings for SetReqLen 10669 10670 This provides a simplified version of the SetReqLen macro when using clang for 10671 static analysis. Prior to this change, we would see many Idempotent operation 10672 warnings inside this macro due to the common case of calling with arg2 and 10673 arg3 being the same variable. This has no effect on code produced during 10674 compilation, but it silences a number of false positives in static analysis. 10675 10676 XIPassiveGrab.c:170:5: warning: Assigned value is always the same as the existing value 10677 SetReqLen(req, num_modifiers, num_modifiers); 10678 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 10679 In file included from XIPassiveGrab.c:26: 10680 .../include/X11/Xlibint.h:580:8: note: instantiated from: 10681 n = badlen; \ 10682 ^ 10683 10684 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10685 10686commit 130af2bc02090ee2526adb2f9803fc07019b8f1f 10687Author: Jeremy Huddleston <jeremyhu@apple.com> 10688Date: Wed May 4 15:34:31 2011 -0700 10689 10690 XKB: XkbComputeSectionBounds: Check correct bounds in default switch-case 10691 10692 XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds') 10693 _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1); 10694 ^~~~~~~ 10695 10696 Found-by: clang static analyzer 10697 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10698 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 10699 10700commit c987ef04bea345d5810e60cf6c53b747ddbe910e 10701Author: Jeremy Huddleston <jeremyhu@apple.com> 10702Date: Wed May 4 15:31:17 2011 -0700 10703 10704 Revert "XKB: Avoid a possible NULL dereference" 10705 10706 Sorry for the noise. I accidentally pushed and didn't mean to. =( 10707 10708 This reverts commit 4024091678ea07e0d898b798df9b29f3bf68eb08. 10709 10710commit 46d04bc4e7ff6c86385002c929d8eb0310f737a8 10711Author: Jeremy Huddleston <jeremyhu@apple.com> 10712Date: Wed May 4 11:35:31 2011 -0700 10713 10714 Dead code removal 10715 10716 XKBGeom.c:118:27: warning: Assigned value is always the same as the existing value 10717 for (key=row->keys,pos=k=0;k<row->num_keys;k++,key++) { 10718 ~~~^~~~ 10719 XKBGeom.c:115:5: warning: Value stored to 'pos' is never read 10720 pos= 0; 10721 ^ ~ 10722 10723 Found-by: clang static analyzer 10724 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10725 10726commit 4024091678ea07e0d898b798df9b29f3bf68eb08 10727Author: Jeremy Huddleston <jeremyhu@apple.com> 10728Date: Wed May 4 11:34:53 2011 -0700 10729 10730 XKB: Avoid a possible NULL dereference 10731 10732 XKBGeom.c:191:25: warning: Access to field 'x1' results in a dereference of a null pointer (loaded from variable 'rbounds') 10733 _XkbCheckBounds(bounds,rbounds->x1,rbounds->y1); 10734 ^~~~~~~ 10735 10736 Found-by: clang static analyzer 10737 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10738 10739commit 626e5e34be55b7cf734d745bd40a7ee3359029f8 10740Author: Jeremy Huddleston <jeremyhu@apple.com> 10741Date: Wed May 4 11:21:41 2011 -0700 10742 10743 Annotate _XIOError as _X_NORETURN 10744 10745 Found-by: clang static analyzer 10746 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10747 10748commit d3d7896408b435ab42656398839ff4351a37724d 10749Author: Jeremy Huddleston <jeremyhu@apple.com> 10750Date: Tue May 3 09:32:53 2011 -0700 10751 10752 clang analyzer: Don't warn about Xmalloc(0) 10753 10754 This will prevent a number of false positives in where clang's 10755 static analysis reports about calls to malloc(0). 10756 10757 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 10758 10759commit 393921cf2188b2b0713cc157effaf17d0abab783 10760Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10761Date: Thu Apr 21 16:21:04 2011 -0700 10762 10763 Delete special case code to append "/sparcv9" to i18n module path 10764 10765 Was triggered by defined(__sparcv9) so only built on Solaris SPARC 64-bit. 10766 Inconsistent with all other platforms, and a bit overcomplicated. 10767 10768 Should anyone need to continue using that path, simply build with 10769 a #define POSTLOCALELIBDIR "sparcv9" to get the same result. 10770 10771 Fixes Solaris bug 7038737: 10772 sparcv9 Xlib looking in wrong path for i18n loadable modules 10773 10774 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10775 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 10776 10777commit b9ff9cfd5c56ae6d52c6503f7019756d9de29557 10778Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10779Date: Thu Apr 14 13:47:12 2011 -0700 10780 10781 specs/libX11: Fix XOpenDisplay() prototype in chapter 2 [bug 36244] 10782 10783 Was incorrectly showing AllPlanes() instead when describing XOpenDisplay() 10784 10785 https://bugs.freedesktop.org/show_bug.cgi?id=36244 10786 10787 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10788 10789commit b850adbdebcf500c659f85285d4d7374e15857f5 10790Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10791Date: Tue Apr 12 22:30:45 2011 -0700 10792 10793 Convert malloc(strlen()); strcpy() sets to strdup 10794 10795 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10796 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 10797 10798commit 09194042d3dc44a463add1f7c122a68ffd5ef0bf 10799Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10800Date: Tue Apr 12 21:27:45 2011 -0700 10801 10802 Replace Xmalloc+bzero pairs with Xcalloc calls 10803 10804 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10805 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 10806 10807commit ac1e2bff7121987fd768500a11d428d9fb9447c5 10808Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10809Date: Fri Mar 25 14:47:35 2011 +0200 10810 10811 om: Fix memory leaks on get_font_name error paths. 10812 10813 While at it, remove unneeded check for NULL before Xfree. 10814 10815 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10816 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10817 10818commit 7c362e275c93c92b4e68fe862e73ee36665de703 10819Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10820Date: Wed Mar 30 15:06:10 2011 +0300 10821 10822 xcms: Fix memory leaks on LINEAR_RGB_InitSCCData error path. 10823 10824 pScreenData is replaced when building per visual intensity tables. If 10825 malloc failed the old value of pScreenData (stored also in 10826 pScreenDefaultData) was being leaked. Also, property_return wasn't 10827 free'd in that case. 10828 10829 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10830 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10831 10832commit 5c810e2ac233e00f361549bafb9b59e8a9e05eff 10833Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10834Date: Wed Mar 30 15:19:28 2011 +0300 10835 10836 xcms: Fix error on LINEAR_RGB_InitSCCData error path. 10837 10838 Due to what looks like a copy & paste error, pScreenData->pBlueTbl would 10839 be accessed after being free'd. 10840 10841 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10842 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10843 10844commit d749948f9492fd9b61c74655a08e32c595e0e3a5 10845Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10846Date: Fri Mar 25 16:06:15 2011 +0200 10847 10848 om: Fix potential memory leak in init_om. 10849 10850 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10851 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10852 10853commit d0749d6abdf0fd4d8b4e59b02dad8ccda3f10995 10854Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10855Date: Thu Mar 24 20:40:41 2011 +0200 10856 10857 om: Fix memory leak on read_EncodingInfo error path. 10858 10859 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10860 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10861 10862commit 4b2e8d00f5b6969c14003ee8eb258b9f0e4dd7c3 10863Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10864Date: Thu Mar 24 19:55:33 2011 +0200 10865 10866 Fix memory leaks on _XimCbDispatch error path. 10867 10868 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10869 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10870 10871commit 46e6c78b1a89e4774e0f7e0f4d6d0fd060c3897e 10872Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10873Date: Thu Mar 24 19:52:05 2011 +0200 10874 10875 Fix memory leak on _XimCommitRecv error path. 10876 10877 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10878 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10879 10880commit 0ace642a2d47265f01450bfa2b2fd48eb6956a83 10881Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10882Date: Thu Mar 24 19:48:44 2011 +0200 10883 10884 Fix memory leaks on _XimWriteCachedDefaultTree error paths. 10885 10886 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10887 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10888 10889commit e29be94edbb58b3b8dab545377a710d1f73b61e0 10890Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10891Date: Thu Mar 24 19:42:36 2011 +0200 10892 10893 Fix memory leaks on _XimGetAttributeID error paths. 10894 10895 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10896 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10897 10898commit d144a50512466766f55ce61e3884925334b08f0d 10899Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10900Date: Thu Mar 24 19:36:56 2011 +0200 10901 10902 Fix memory leaks on _XimProtoCreateIC error paths. 10903 10904 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10905 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10906 10907commit 6a452f7a98499508f753cb8a7c3f08bcbec736b9 10908Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10909Date: Thu Mar 24 16:24:29 2011 +0200 10910 10911 Fix leaks in _XimEncodingNegotiation error paths. 10912 10913 name_ptr and detail_ptr weren't free'd in some cases before returning 10914 False. 10915 10916 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10917 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10918 10919commit 396e5a452a59c1f121220ba72167b720a863b30f 10920Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10921Date: Thu Mar 24 14:17:44 2011 +0200 10922 10923 FSWrap: fix potential leak in copy_string_list 10924 10925 If list_count is 0, dst would be allocated and leaked. 10926 10927 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10928 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10929 10930commit 1a944260182bb552b954d69f6355c2760d4415df 10931Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10932Date: Wed Mar 23 14:36:22 2011 +0200 10933 10934 Fix memory leak with broken bitmap files in XReadBitmapFileData 10935 10936 Bitmap file data is read looping through the lines in the input file. If 10937 there is extra data after the bitmap, these lines will be processed and 10938 if this data represents another bitmap it will replace the one read 10939 before causing the memory allocated for bits to leak. 10940 10941 This changes the code to stop processing the file once a bitmap was 10942 read. 10943 10944 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 10945 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10946 10947commit c1c91e9a221a0e762d96cc12a3f189d6e59f4865 10948Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10949Date: Tue Apr 12 20:50:42 2011 -0700 10950 10951 Fix "attrinute" typo in comments in ximcp 10952 10953 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10954 10955commit 7103b83a88edb9c93acfa68f3b556595b087baff 10956Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10957Date: Fri Apr 8 21:36:11 2011 -0700 10958 10959 Make doc install subdirectories more like the pre-docbook.am paths 10960 10961 Haven't restored the "specs" path in the install path, but have restored 10962 libX11 & i18n subdirectories in the path for better logical grouping. 10963 10964 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10965 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 10966 10967commit 9a1ae7a90f815a737a942e42d5eb1ffff400977b 10968Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10969Date: Fri Apr 8 13:24:28 2011 -0700 10970 10971 Install xml versions of specs even if HAVE_XMLTO is false 10972 10973 Moves HAVE_XMLTO check into docbook.am, more closely matches behaviour 10974 from before docbook.am changes (commit e8c76407d2f6e3) 10975 10976 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10977 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 10978 10979commit a0ad0d5c99023bb9a8ce3944dbc3267f5265721e 10980Author: Alan Coopersmith <alan.coopersmith@oracle.com> 10981Date: Tue Apr 5 13:29:04 2011 -0700 10982 10983 libX11 1.4.3 10984 10985 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 10986 10987commit e8c76407d2f6e3b11babdb84426e43e780c859c3 10988Author: Gaetan Nadon <memsize@videotron.ca> 10989Date: Fri Feb 25 09:23:54 2011 -0500 10990 10991 Documentation: add Docbook external references support 10992 10993 When writing technical documentation, it is often necessary to cross 10994 reference to other information. When that other information is not in the 10995 current document, additional support is needed, namely <olink>. 10996 10997 A new feature with version 1.7 of xorg-sgml-doctools adds references to 10998 other documents within or outside this package. 10999 11000 This patch adds technical support for this feature but does not change 11001 the content of the documentation as seen by the end user. 11002 11003 Each book or article must generate a database containing the href 11004 of sections that can be referred to from another document. This database 11005 is installed in DATAROOTDIR/sgml/X11/dbs. There is a requirement that 11006 the value of DATAROOTDIR for xorg-sgml-doctools and for the package 11007 documentation is the same. This forms a virtual document tree. 11008 11009 This database is consulted by other documents while they are being generated 11010 in order to fulfill the missing information for linking. 11011 Refer to the xorg-sgml-doctools for further technical information. 11012 11013 Co-authored-by: Matt Dew <marcoz@osource.org> 11014 11015 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11016 11017commit 25eb76b3d2726f2f1de759901971ae53b2539dc4 11018Author: Harshula Jayasuriya <harshula@gmail.com> 11019Date: Mon Mar 21 14:49:37 2011 +0000 11020 11021 Add #define XK_SINHALA 11022 11023 Add #define XK_SINHALA so that the Sinhala keysyms can be used by 11024 the lk xkb keymap. 11025 11026 Signed-off-by: Harshula Jayasuriya <harshula@gmail.com> 11027 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 11028 11029commit db8b20b789112717ac0590b40f0b4dc2171797d0 11030Author: Jeremy Huddleston <jeremyhu@apple.com> 11031Date: Thu Mar 17 16:15:00 2011 -0700 11032 11033 configure.ac: Bump version to 1.4.2 11034 11035 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 11036 11037commit 83e1ba59c48c79f8b0a7e7aa0b9c9cfd84fa403d 11038Author: Jamey Sharp <jamey@minilop.net> 11039Date: Tue Mar 15 16:48:07 2011 -0700 11040 11041 Call _XErrorFunction without holding the Display lock. 11042 11043 Historically, Xlib dropped the Display lock around the upcall to any 11044 user-supplied _XErrorFunction, but somewhere along the way I quit doing 11045 that if you built with XCB. The reasons are lost somewhere in the 11046 pre-git history of Xlib/XCB, and I can't now see any reason to hold the 11047 lock. 11048 11049 The documentation for XSetErrorHandler still applies though: 11050 11051 Because this condition is not assumed to be fatal, it is acceptable 11052 for your error handler to return; the returned value is ignored. 11053 However, the error handler should not call any functions (directly 11054 or indirectly) on the display that will generate protocol requests 11055 or that will look for input events. 11056 11057 So while you are now once again permitted to re-enter Xlib from the 11058 error handler, you're only allowed to call non-protocol functions. 11059 11060 Signed-off-by: Jamey Sharp <jamey@minilop.net> 11061 11062commit fd85aca7a616c595fc17b2520f84316a11e8906f 11063Author: Jamey Sharp <jamey@minilop.net> 11064Date: Mon Mar 14 14:45:35 2011 -0700 11065 11066 Ignore user locks after sleeping in _XReply and _XReadEvents. 11067 11068 This bug appears as a hang in applications that wait for replies from 11069 multiple threads, where one such thread has taken a user lock using 11070 XLockDisplay. 11071 11072 Prior to this fix, the code could deadlock in this way: If thread 1 goes 11073 to sleep waiting for a reply, and then thread 2 takes a user lock and 11074 waits for a reply, then thread 2 will wait for thread 1 to process its 11075 reply (because responses must be processed in order), but thread 1 will 11076 wait for thread 2 to drop its user lock. 11077 11078 Fixed by making thread 1 not wait for thread 2 to drop its user lock. 11079 This makes the semantics of user locks hard to define, but they were 11080 already hard to define. The new behavior appears to be consistent with 11081 the way Xlib worked historically, anyway. 11082 11083 Fixes: http://lists.freedesktop.org/archives/xcb/2011-March/006802.html 11084 11085 There was a similar potential for deadlock in _XReadEvents, fixed the 11086 same way, with the same caveats about user-lock semantics. 11087 11088 Signed-off-by: Jamey Sharp <jamey@minilop.net> 11089 11090commit 690f8bffd48a4e7e74298360ddd0431dc95dcd3f 11091Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11092Date: Tue Jan 18 15:58:20 2011 +0200 11093 11094 xkb: XkbPropertyPtr determined allocation success from wrong variables 11095 11096 Cannot reach dead statement "return NULL;" 11097 11098 Check for the NULLness of prop->name and prop->value instead of 11099 name and value, which was checked earlier anyway. Decided against 11100 using strdup due to curious memory allocation functions and the 11101 rest of the xkb not using it either. 11102 11103 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11104 Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com> 11105 11106commit 6a4d027284e7bb5dd458157947bbb1ff580ad071 11107Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11108Date: Mon Jan 10 16:37:22 2011 +0200 11109 11110 keyBind: Use Xcalloc to initialize allocated _XKeytrans 11111 11112 Using uninitialized value "p->modifiers" 11113 11114 Small fix by using Xcalloc instead of Xmalloc 11115 11116 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11117 Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com> 11118 11119commit b993d73bb3214ecc24646f5427c8003b816c6921 11120Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11121Date: Mon Jan 10 16:22:45 2011 +0200 11122 11123 im/ximcp: release modifiermap before returning 11124 11125 Variable "map" goes out of scope 11126 11127 Release modifiermap before returning. Reordered code to call 11128 XGetModifierMapping after the first return from the function. 11129 11130 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11131 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de> 11132 11133commit 807a7fc0354f2212dfa5ff1f9e4ede56d8e69ef4 11134Author: Gaetan Nadon <memsize@videotron.ca> 11135Date: Fri Feb 25 09:23:54 2011 -0500 11136 11137 Docbook: change the book id to match the xml file basename 11138 11139 This is required for the up-coming external references support. 11140 11141 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11142 11143commit 40812b53ff5fe548f6eaf43ba4c8781cb43dab43 11144Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 11145Date: Mon Feb 21 21:54:17 2011 +0100 11146 11147 Make the Local XIM request key releases for braille 11148 11149 Braille chords management needs key release events. We need to explicitly 11150 request then, else GTK would not pass them throught XFilterEvent and braille 11151 wouldn't work. 11152 11153 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 11154 11155commit c97c42c49cd5095462abecdf908b416fb0b540b6 11156Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 11157Date: Mon Feb 21 17:27:38 2011 +0100 11158 11159 Match braille patterns with compose tree 11160 11161 Braille patterns should also be usable in Compose. This combines the 11162 implementation of braille chords and compose tree: only emit the braille 11163 pattern if it can not be found in the compose tree, if any. 11164 11165 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 11166 11167commit 0c6ca565d7c8a47ef3ea823569a9ca5298a5307d 11168Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 11169Date: Mon Feb 21 15:56:54 2011 +0100 11170 11171 Fix status reporting for braille patterns 11172 11173 _XimLocalMbLookupString can return a braille keysym even if _Xlcwctomb can't 11174 convert to the current MB charset. 11175 _XimLocalUtf8LookupString needs to set the braille keysym and status too. 11176 11177 Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> 11178 11179commit 993abe751f4141f54d8d28b8b73588a1c9085970 11180Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11181Date: Fri Feb 11 14:49:17 2011 -0800 11182 11183 Clean up memory when first XCreateRegion succeeds and second fails 11184 11185 Error: Memory leak (CWE 401) 11186 Memory leak of pointer 's' allocated with XCreateRegion() 11187 at line 387 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XShrinkRegion'. 11188 's' allocated at line 387 with XCreateRegion(). 11189 s leaks when s != 0 at line 387. 11190 Error: Memory leak (CWE 401) 11191 Memory leak of pointer 'tra' allocated with XCreateRegion() 11192 at line 1452 of /export/alanc/X.Org/sx86-gcc/lib/libX11/src/Region.c in function 'XXorRegion'. 11193 'tra' allocated at line 1451 with XCreateRegion(). 11194 tra leaks when tra != 0 at line 1451. 11195 11196 [ This bug was found by the Parfait 0.3.6 bug checking tool. 11197 For more information see http://labs.oracle.com/projects/parfait/ ] 11198 11199 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11200 11201commit 6ac417cea1136a3617f5e40f4b106aaa3f48d6c2 11202Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11203Date: Fri Feb 11 14:20:24 2011 -0800 11204 11205 ximcp: Prevent memory leak & double free if multiple %L in string 11206 11207 In the highly unlikely event that TransFileName was passed a path 11208 containing multiple %L entries, for each entry it would call 11209 _XlcFileName, leaking the previous results, and then for each entry it 11210 would copy from that pointer and free it, resulting in invalid pointers 11211 & possible double frees for each use after the first one freed it. 11212 11213 Error: Use after free (CWE 416) 11214 Use after free of pointer 'lcCompose' 11215 at line 358 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'. 11216 Previously freed at line 360 with free. 11217 Error: Use after free (CWE 416) 11218 Use after free of pointer 'lcCompose' 11219 at line 359 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'. 11220 Previously freed at line 360 with free. 11221 Error: Double free (CWE 415) 11222 Double free of pointer 'lcCompose' 11223 at line 360 of modules/im/ximcp/imLcPrs.c in function 'TransFileName'. 11224 Previously freed at line 360 with free. 11225 11226 [ This bug was found by the Parfait 0.3.6 bug checking tool. 11227 For more information see http://labs.oracle.com/projects/parfait/ ] 11228 11229 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11230 11231commit 502d414118c97d35a44f8e295709682022876331 11232Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11233Date: Thu Feb 3 17:08:57 2011 +0200 11234 11235 xcms/cmsProp: don't deal with uninitialized values, fail instead 11236 11237 Properly handle the return value of XGetWindowProperty by considering 11238 if after the loop as well. 11239 11240 Using freed pointer "prop_ret" 11241 11242 There were numerous things wrong in how this function interacted with 11243 XGetWindowProperty. 11244 11245 None of the local variables were initialized and remained that way if 11246 the call to XGetWindowProperty returned 1 (not Succeed). That doesn't 11247 result in after_ret being initialized in which case if it happens to 11248 be 0, the loop was exited. In that case format_ret and nitems_ret were 11249 uninitialized and the function might return with success (but with 11250 uninitialized pointer in prop_ret) or XcmsFailure. 11251 11252 As the buffer enlarging code was called only when XGetWindowProperty 11253 failed (returned not Success), after_ret would not have been 11254 initialized. It would have been initialized only if the 11255 XGetWindowProperty has returned Success earlier, but in that case the 11256 code fragment would not have been reached. 11257 11258 This patch alters the function to return XcmsFailure if the call to 11259 XGetWindowProperty fails. 11260 11261 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11262 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11263 Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi> 11264 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11265 11266commit c37e278993b9e5b3d7025ef4c434373a011996ec 11267Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11268Date: Mon Jan 31 14:02:07 2011 +0200 11269 11270 xcms/LRGB: don't double-free property_return 11271 11272 property_return was free'd before and in the case the conditional is true, 11273 the call to XcmsGetProperty failed which means that property_return wasn't 11274 set so there is no need to free it again. 11275 11276 Double free of pointer "property_return" in call to "free" 11277 11278 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11279 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11280 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11281 11282commit 50f4107811249806718a100f9d34f996c58e5e25 11283Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11284Date: Wed Feb 2 08:58:45 2011 -0800 11285 11286 Xrm.c: ReadInFile: refactor fstat error handling 11287 11288 We can simplify the fstat failure case now that the GetFileSize macro 11289 has been expanded inline. 11290 11291 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11292 Reviewed-by: Julien Cristau <jcristau@debian.org> 11293 11294commit 5e9c40fcb5da43c9fdacf12967d090bf202daf2a 11295Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11296Date: Wed Feb 2 08:56:00 2011 -0800 11297 11298 Expand GetSizeOfFile() macro at the one place it's called 11299 11300 Removes XrmI.h header that only contained this single macro 11301 11302 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11303 Reviewed-by: Julien Cristau <jcristau@debian.org> 11304 11305commit 450e17422c0e374d25c643f343ea268cec68da38 11306Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11307Date: Mon Jan 31 14:01:57 2011 +0200 11308 11309 XlibInt: Use strncpy+zero termination instead of strcpy to enforce buffer size 11310 11311 Possible overrun of 8192 byte fixed size buffer "buffer" by copying 11312 "ext->name" without length checking 11313 11314 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11315 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11316 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11317 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11318 11319commit e2566e43b02d2d7b7c1c3bb7db7c5ae81c1245fa 11320Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11321Date: Mon Jan 31 14:02:13 2011 +0200 11322 11323 lc/def/lcDefConv: Use Xcalloc to avoid use of uninitialized memory 11324 11325 Fixed by zero'ing conv on allocation with Xcalloc. Then 11326 close_converter works properly. 11327 11328 Using uninitialized value "conv->state" in call to function "close_converter" 11329 11330 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11331 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11332 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11333 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11334 11335commit 03f81ad8fb9783986cf9b17661dd31e95c396681 11336Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11337Date: Mon Jan 31 14:02:16 2011 +0200 11338 11339 xlibi18n/lcFile: Removed superfluous check for NULL target_dir 11340 11341 The situation is already handled before this code. 11342 11343 Cannot reach dead expression "0U" inside statement "if (1U + (target_dir ? strl..." 11344 11345 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11346 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11347 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11348 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11349 11350commit 1346b9ea3b3882201ff8c3ee462ff4b0d4edf639 11351Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11352Date: Mon Jan 31 14:02:15 2011 +0200 11353 11354 ximcp/imLckup: Handle negative return value from _Xlcwctomb 11355 11356 Fixed by negative value to memcpy by checking for the negative return 11357 value of _Xlcwctomb and returning 0/XLookupNone in that case. 11358 11359 a negative value was passed to memcpy 11360 11361 Unfortunately the other return values for *status don't fit into the 11362 error (which appears to indicate some internal error or running out of 11363 memory). The other valid status codes are XBufferOverflow, 11364 XLookupNone, XLookupChars, XLookupKeySym, and XLookupBoth. Each of 11365 these has a specific meaning attached. 11366 11367 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11368 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11369 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11370 11371commit 79a5c86e020f08cc108184298a72e2777036de39 11372Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11373Date: Mon Jan 31 14:02:14 2011 +0200 11374 11375 ximcp/imTrX: Handle failing XGetWindowProperty 11376 11377 Checked return value of XGetWindowProperty and return false if it fails. 11378 11379 Return value of "XGetWindowProperty(im->core.display, spec->lib_connect_wid, prop, 0L, (length + bytes_after_ret + 3UL) / 4UL, 1, 0UL, &type_ret, &format_ret, &nitems, &bytes_after_ret, &prop_ret)" is not checked 11380 11381 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11382 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11383 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11384 11385commit 2ace8d5c89c8f6d9f42b4068f4b508ca28f0ced1 11386Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11387Date: Mon Jan 31 14:02:12 2011 +0200 11388 11389 XlibInt: info_list->watch_data was reallocated, but result was discarded 11390 11391 info_list->watch_data was being reallocated, but the return value of 11392 the reallocation was stored only into a local variable. This might 11393 cause some funky behavior and crashes. 11394 11395 Variable "wd_array" goes out of scope 11396 Value "wd_array" is overwritten in "wd_array = (XPointer*)realloc((char*)info_list->watch_data, (((dpy->watcher_count + 1) * 4U == 0U) ? 1U : ((dpy->watcher_count + 1) * 4U)))" 11397 11398 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11399 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11400 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11401 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11402 11403commit 2b16a7e683e355c9746290b2cee2fd0dd2bf342a 11404Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11405Date: Mon Jan 31 14:02:10 2011 +0200 11406 11407 GetProp: Zero-initialized error so its resourceID field is initialized 11408 11409 Using uninitialized value "error.resourceID" in call to function "_XError" 11410 11411 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11412 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11413 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11414 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11415 11416commit 85e9f38e016137f0ff2791eb0d092ab027382d2c 11417Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11418Date: Tue Feb 1 11:07:25 2011 +0200 11419 11420 xcms/LRGB: Add a label for freeing property_return. 11421 11422 The rest of the code uses goto's to free memory allocated later 11423 and prevent memory leaks, but there were several paths were 11424 property_return was free'd just before a goto. 11425 11426 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11427 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11428 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11429 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11430 11431commit 3161dc57d4e9b70f852f05e5e474455e121b06ab 11432Author: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11433Date: Mon Jan 31 14:02:08 2011 +0200 11434 11435 xcms/LRGB: Fix potential resource leak. 11436 11437 property_return was not free'd if the allocation of pRedTbl failed. 11438 11439 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11440 Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11441 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11442 11443commit be3e6c205d94dedc1cdebf5d17b987f0f828377a 11444Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11445Date: Mon Jan 31 14:02:06 2011 +0200 11446 11447 Xrm: Handle the extremely unlikely situation of fstat failing 11448 11449 Tracked variable "size" was passed to a negative sink. 11450 11451 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11452 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11453 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11454 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11455 11456commit 3fae16c64d6ef76fd4a25a54c7f7de76596457db 11457Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11458Date: Mon Jan 31 14:02:05 2011 +0200 11459 11460 Xrm: NEWTABLE had a memory leak after a memory allocation error 11461 11462 The NEWTABLE macro missed freeing its allocated memory on subsequent 11463 memory allocation errors. Added call to Xfree. 11464 11465 Variable "table" goes out of scope 11466 11467 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11468 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11469 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11470 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11471 11472commit 7110ac653349a23c80c365f11f6270dc27c8975a 11473Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11474Date: Mon Jan 31 14:02:03 2011 +0200 11475 11476 ImUtil: Handle a memory leak in one early return branch 11477 11478 Fixed memory leak by adding Xfree for image 11479 11480 Variable "image" goes out of scope 11481 11482 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11483 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11484 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11485 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11486 11487commit d695f5da9f5b778e54b6987d9177048e32818c4d 11488Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11489Date: Mon Jan 31 14:02:01 2011 +0200 11490 11491 ximcp/imRm: Handle leaking colormap_ret 11492 11493 Fixed memory leak by adding Xfree for colormap_ret 11494 11495 Variable "colormap_ret" goes out of scope 11496 11497 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11498 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11499 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11500 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11501 11502commit f9eda19d08c1ea0973dfe0bc10a2519d6fd26cc7 11503Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11504Date: Mon Jan 31 14:01:59 2011 +0200 11505 11506 ximcp/imRmAttr: Handle leaking missing_list 11507 11508 Fixed memory leak by adding Xfree and initializing missing_list with NULL 11509 11510 Variable "missing_list" goes out of scope 11511 11512 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11513 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11514 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11515 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11516 11517commit 3183269e0979c9dbce9d55d9e03937897dc9fb3b 11518Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11519Date: Mon Jan 31 14:01:58 2011 +0200 11520 11521 ximcp/imRmAttr: Handle leaking colormap_ret 11522 11523 XFree colormap_ret and initialize it when appropriate. 11524 11525 Variable "colormap_ret" goes out of scope 11526 11527 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11528 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11529 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11530 11531commit f66a032a937030f2d9baa81744d36dc585bb085c 11532Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11533Date: Mon Jan 31 14:01:56 2011 +0200 11534 11535 xlibi18n/lcGeneric: Initialize uninitialized local variable 11536 11537 Using uninitialized value "new" 11538 11539 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11540 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11541 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11542 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11543 11544commit d0266e06d38110ec908ca28379014eff743630b7 11545Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11546Date: Mon Jan 31 14:01:54 2011 +0200 11547 11548 xcmx/cmxColNm: Removed unused assignments to pBuf (in two functions) 11549 11550 Pointer "pBuf" returned from "fgets(buf, 256, stream)" is never used 11551 11552 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11553 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11554 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11555 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11556 11557commit cc686655d7bfdeab8b67e01a24bd452a2e9e3fcf 11558Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11559Date: Tue Feb 1 12:30:56 2011 +0200 11560 11561 XDefaultOMIF.c: Fix memory leaks in get_font_name 11562 11563 Instead of copying the value returned by get_prop_name and then releasing it, 11564 directly use the return value of get_prop_name, which allocates memory for the 11565 name. 11566 11567 If get_prop_name returns NULL, continue on to XFreeFont to release the font 11568 before returning the NULL via the normal function return. 11569 11570 Reviewed-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11571 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11572 11573commit 95796351435d81eaef8166d49ba3a6e7b633d70a 11574Author: Gaetan Nadon <memsize@videotron.ca> 11575Date: Mon Jan 31 14:50:19 2011 -0500 11576 11577 config: comment, minor upgrade, quote and layout configure.ac 11578 11579 Group statements per section as per Autoconf standard layout 11580 Quote statements where appropriate. 11581 Autoconf recommends not using dnl instead of # for comments 11582 11583 Use AC_CONFIG_FILES to replace the deprecated AC_OUTPUT with parameters. 11584 Remove redundant AC_CANONICAL_HOST included in XORG_DEFAULT_OPTIONS 11585 11586 This helps automated maintenance and release activities. 11587 Details can be found in http://wiki.x.org/wiki/NewModuleGuidelines 11588 11589 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11590 11591commit e994ab227ed28655565c3193e63733630105e7f9 11592Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11593Date: Tue Jan 18 12:49:48 2011 +0200 11594 11595 Comparing array against NULL is not useful "&xkb->server->vmods != NULL" 11596 11597 Removed superfluous comparison. 11598 11599 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de> 11600 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11601 11602commit 4115f051e62f9b098efce691e070d44f09f30f1c 11603Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11604Date: Wed Jan 12 15:51:11 2011 +0200 11605 11606 Variable "entry" tracked as NULL was dereferenced. 11607 11608 Check entry for non-nullness before dereferencing it 11609 11610 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de> 11611 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11612 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11613 11614commit 0edb76d1d5fdca5a2543332699be2e72386dab24 11615Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11616Date: Thu Jan 27 09:54:00 2011 +0200 11617 11618 Dereferencing possibly NULL "str" in call to function "memcpy" (Deref assumed on the basis of 'nonnull' parameter attribute.) 11619 11620 If _XkbGetReadBufferPtr returns NULL, goto BAILOUT 11621 11622 Reviewed-by: Dirk Wallenstein <halsmit@t-online.de> 11623 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11624 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11625 11626commit 59da8a211ef723909d0530c0331d541db8e63378 11627Author: Erkki Seppälä <erkki.seppala@vincit.fi> 11628Date: Mon Jan 10 16:17:47 2011 +0200 11629 11630 Using freed pointer "e" 11631 11632 Reordered code to first to do the comparison and then to release data 11633 11634 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11635 Reviewed-by: Ander Conselvan de Oliveira <ander.conselvan-de-oliveira@nokia.com> 11636 Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi> 11637 11638commit 0f11c229f7099f7c5aeed4691b358dca151dac7d 11639Author: Matt Dew <marcoz@osource.org> 11640Date: Thu Jan 27 00:31:23 2011 -0700 11641 11642 Remove <literal> tags from compose-chart.pl so pdf building doesn't die on soft-hyphens. 11643 11644 Signed-off-by: Matt Dew <marcoz@osource.org> 11645 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11646 11647commit ca00bb202f3afcdbbcb8c4cb50308c5dd03f2322 11648Author: Gaetan Nadon <memsize@videotron.ca> 11649Date: Fri Jan 28 16:07:07 2011 -0500 11650 11651 config: replace deprecated AC_HELP_STRING with AS_HELP_STRING 11652 11653 This silences an Automake warning. 11654 11655 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11656 11657commit 8dac08c7515cb6f8a8f4d0a721713697eae755d8 11658Author: Gaetan Nadon <memsize@videotron.ca> 11659Date: Fri Jan 28 14:59:04 2011 -0500 11660 11661 config: remove unrequired AC_HEADER_STDC 11662 11663 Autoconf says: 11664 "This macro is obsolescent, as current systems have conforming 11665 header files. New programs need not use this macro". 11666 11667commit d4483375e37a72631d5821413cb75a423e990ffe 11668Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11669Date: Tue Jan 18 19:37:02 2011 -0800 11670 11671 config: Use correct AC_CONFIG_HEADERS macro 11672 11673 Replaces obsolete AM_CONFIG_HEADER and undocumented AC_CONFIG_HEADER 11674 11675 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11676 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 11677 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 11678 11679commit c6405fe931f313a0d159bdad00244fc8aae12468 11680Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11681Date: Sun Jan 16 11:55:23 2011 -0800 11682 11683 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 11684 11685 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 11686 AC_PROG_C_C99. This sets gcc with -std=gnu99. 11687 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 11688 11689 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11690 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 11691 11692commit 17aa1f2305da0952ad594c7329d08aba21f13ac8 11693Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11694Date: Sun Jan 16 11:40:59 2011 -0800 11695 11696 Update to xf86bigfontproto >= 1.2.0 header name 11697 11698 Clears compile-time warning of: 11699 "X11/extensions/xf86bigfstr.h", line 1: #warning: "xf86bigfstr.h is obsolete and may be removed in the future." 11700 "X11/extensions/xf86bigfstr.h", line 2: #warning: "include <X11/extensions/xf86bigfproto.h> for the protocol defines." 11701 11702 Requires xf86bigfontproto >= 1.2.0 if --disable-xf86bigfont is not 11703 passed to configure. 11704 11705 Also removes unnecessary AC_SUBST of BIGFONT_CFLAGS & BIGFONT_LIBS 11706 that PKG_CHECK_MODULES does automatically 11707 11708 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11709 Reviewed-by: Julien Cristau <jcristau@debian.org> 11710 11711commit a9228fcc676aacf9a760dd94891c89f9bc82b20d 11712Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11713Date: Tue Jan 11 17:55:22 2011 -0800 11714 11715 libX11 1.4.1 11716 11717 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11718 11719commit 92fa96451af37a7a0f2592de07643fdaed7a9efd 11720Author: Pauli Nieminen <ext-pauli.nieminen@nokia.com> 11721Date: Mon Jan 3 12:25:28 2011 -0500 11722 11723 Initialize event type 11724 11725 If we receive unsupported event closing connection triggers valgrind 11726 error. 11727 11728 ==12017== Conditional jump or move depends on uninitialised value(s) 11729 ==12017== at 0x487D454: _XFreeDisplayStructure (OpenDis.c:607) 11730 ==12017== by 0x486857B: XCloseDisplay (ClDisplay.c:72) 11731 *snip* 11732 ==12017== Uninitialised value was created by a heap allocation 11733 ==12017== at 0x4834C48: malloc (vg_replace_malloc.c:236) 11734 ==12017== by 0x4894147: _XEnq (XlibInt.c:877) 11735 ==12017== by 0x4891BF3: handle_response (xcb_io.c:335) 11736 ==12017== by 0x4892263: _XReply (xcb_io.c:626) 11737 *snip* 11738 11739 Problem is that XFreeDisplaySturture is checking for qelt->event.type == 11740 GenericEvent while _XUnknownWireEvent doesn't store the type. 11741 11742 Reviewed-by: Adam Jackson <ajax@redhat.com> 11743 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 11744 Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com> 11745 11746commit fa9747b9d4443eef9b1687b8b1208801f5c585da 11747Author: Fernando Carrijo <fcarrijo.lists@gmail.com> 11748Date: Tue Dec 28 16:18:57 2010 -0500 11749 11750 specs: convert images from gif to svg format. 11751 11752 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11753 11754commit 0012e2a4eaddc0840617b4758931ad976ca2eb7c 11755Author: Philipp Reh <sefi@s-e-f-i.de> 11756Date: Wed Dec 22 15:14:05 2010 +0000 11757 11758 Events: Store event cookie when dequeuing event 11759 11760 When we dequeue an event in XCheckTypedEvent or XCheckTypedWindowEvent, 11761 make sure to store the corresponding cookie too. 11762 11763 Signed-off-by: Philipp Reh <sefi@s-e-f-i.de> 11764 Reviewed-by: Daniel Stone <daniel@fooishbar.org> 11765 11766commit d2714d65e85b44abedf5f82e1a31506dba397ef2 11767Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11768Date: Tue Dec 21 18:47:16 2010 -0800 11769 11770 Revert "Mark atom names argument to XInternAtoms as const" 11771 11772 This reverts commit c8701115462b482d99ecff24d9de0f2806084ba5. 11773 11774 The constification of a pointer to a pointer caused unexpected issues, 11775 and xorg-devel was unable to come up with a clean, safe, reasonable way 11776 to handle them, so we're chalking this up for now as yet another mistake 11777 in the Xlib API definition we'll be living with. 11778 11779 See https://bugs.freedesktop.org/show_bug.cgi?id=32098 for details. 11780 11781 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11782 11783commit 6459f9a48a82de3bc0235bd36acab2df11e436d5 11784Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11785Date: Mon Dec 20 20:54:07 2010 -0800 11786 11787 specs/XKB: make acknowlegement section <preface> instead of <chapter> 11788 11789 Fixes numbering of actual chapters to match their filenames and to 11790 make the table/figure numbering match the references in the text. 11791 11792 Reported-by: jelmd on freenode irc 11793 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11794 11795commit 42c653d5413fc5870486a9cf51dd28d1e88f7c69 11796Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11797Date: Sun Dec 19 23:53:02 2010 -0800 11798 11799 Add XKBlib spec images to dist_spec_DATA for distribution & installation 11800 11801 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11802 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 11803 11804commit 8f8a3c8c42af82976d261ac2447cfbb0f90fa77d 11805Author: Matt Dew <matt@osource.org> 11806Date: Tue Nov 30 10:52:07 2010 -0500 11807 11808 specs: convert XKBlib spec from Framemaker to DocBook.xml 11809 11810 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11811 11812commit c8701115462b482d99ecff24d9de0f2806084ba5 11813Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11814Date: Sun Nov 21 20:49:05 2010 -0800 11815 11816 Mark atom names argument to XInternAtoms as const 11817 11818 Updates code & docs for XInternAtoms. 11819 11820 The single atom name argument to XInternAtom was already const char * 11821 in the code, but not the docs, so updated it in the docs too. 11822 11823 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11824 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 11825 11826commit 24db5b5ff0d51b0ed9d9701a792c5f14f23b638d 11827Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11828Date: Sun Nov 21 14:04:32 2010 -0800 11829 11830 libX11 1.4.0 11831 11832 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11833 11834commit 18d3c3e9de19e936281b111d5a82acfb3a3f552e 11835Author: Dan Nicholson <dbn.lists@gmail.com> 11836Date: Sat Nov 6 21:58:10 2010 +0000 11837 11838 config: Check host string when deciding architecture to build for 11839 11840 When checking for the OS/2 platform, $target_os is used. However, unless 11841 building a cross compiler, the $host* strings contain the platform 11842 details for the build system. See: 11843 11844 http://www.gnu.org/software/automake/manual/automake.html#Cross_002dCompilation 11845 11846 $host_os is already being used to determine the transport and threading 11847 options. 11848 11849 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com> 11850 Reviewed-by: Julien Cristau <jcristau@debian.org> 11851 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11852 11853commit affc2488a7f2660a74dc8354fc3e0bff2c4f879c 11854Author: Dan Nicholson <dbn.lists@gmail.com> 11855Date: Sat Nov 6 21:58:09 2010 +0000 11856 11857 config: Fix architecture check for OS/2 to skip nios2 cpu 11858 11859 The OS/2 platform requires some utility functions as well as having a 11860 non-32 bit wchar_t. Fix the configure check so that it doesn't also 11861 affect the nios2 cpu, which wouldn't influence these operating system 11862 issues. 11863 11864 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com> 11865 Tested-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com> 11866 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11867 Reviewed-by: Julien Cristau <jcristau@debian.org> 11868 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11869 11870commit 8cbca8a10761d1ea75a75bafa647632d6c0dac71 11871Author: Gaetan Nadon <memsize@videotron.ca> 11872Date: Tue Nov 9 13:04:44 2010 -0500 11873 11874 config: HTML file generation: use the installed copy of xorg.css 11875 11876 Currenlty the xorg.css file is copied in each location 11877 where a DocBook/XML file resides. This produces about 11878 70 copies in the $(docdir) install tree. 11879 11880 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11881 11882commit 3e8907305e1818369aef5a5c0da61f09e20de4f8 11883Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11884Date: Fri Oct 29 22:02:10 2010 -0700 11885 11886 libX11 1.3.99.903 (1.4.0 RC3) 11887 11888 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11889 11890commit 5d245d8013289b13e0c42100951b26166c7fada4 11891Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11892Date: Fri Oct 29 22:01:39 2010 -0700 11893 11894 Require xorg-macros 1.11 now that it is released 11895 11896 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11897 11898commit 1ec89689fc771f116a6165226b9e076f54254a40 11899Author: Dan Nicholson <dbn.lists@gmail.com> 11900Date: Fri Oct 22 19:09:07 2010 -0700 11901 11902 docs: Disable fop documentation by default 11903 11904 fop is used to generate the pdf and ps formats of the documentation. 11905 This can significantly slow down the build, especially when creating all 11906 the compose key charts. Since few people probably want the full set of 11907 doc formats, set the default to 'no'. 11908 11909 The default parameter for XORG_WITH_FOP is only available in recent 11910 macros. Users generating configure from older macros will just get 11911 'auto' as the default. 11912 11913 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com> 11914 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 11915 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11916 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11917 11918commit 03877bda911b72ac60a7144d6eced7d5be4b4d72 11919Author: Jon TURNEY <jon.turney@dronecode.org.uk> 11920Date: Wed Apr 14 13:38:18 2010 +0100 11921 11922 Add C.UTF-8 and C.ASCII locale aliases 11923 11924 Add C.UTF-8 locale as an alias for en_US.UTF-8 11925 Add C.ASCII locale as an alias for C 11926 11927 (C.UTF-8 is the default locale for cygwin. It also exists in debian, 11928 although I don't think it's intended for use by humans.) 11929 11930 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> 11931 Acked-by: Julien Cristau <jcristau@debian.org> 11932 Reviewed-by: James Cloos <cloos@jhcloos.com> 11933 11934commit 81d1d9a6bdca779a44f931730b866eea75777c00 11935Author: Dan Nicholson <dbn.lists@gmail.com> 11936Date: Sat Oct 9 16:00:47 2010 -0700 11937 11938 docs: Remove directory prerequisites from make rules 11939 11940 Make expects prerequisites to be files with valid timestamps, and 11941 directories are treated as always being out of date. Thus, any targets 11942 depending on directories will always be rebuilt. 11943 11944 Instead, the doc rules are changed to always create the target's leading 11945 directory. This should prevent the documentation from being rebuilt when 11946 "make install" is run. 11947 11948 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com> 11949 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 11950 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11951 11952commit d0cbf388919364fe6b5b9127f36426eb744090a0 11953Author: Jeremy Huddleston <jeremyhu@apple.com> 11954Date: Sun Sep 26 21:21:16 2010 -0700 11955 11956 Add an "X11_" string to header guards to avoid possible collision 11957 11958 This addresses a build failure which can result from <X11/Xlocale.h> and 11959 <xlocale.h> being included in the same code since they both used the same 11960 _XLOCALE_H_ protection. 11961 11962 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 11963 11964commit b51ac675f44f5ce1a28c5734bf3c26983f8b4192 11965Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11966Date: Tue Sep 21 19:53:03 2010 -0700 11967 11968 libX11 1.3.99.902 (1.4.0 RC2) 11969 11970 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11971 11972commit 986bb6d1d54368fe91e3ea24f518d43ce6179782 11973Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11974Date: Tue Sep 14 00:10:31 2010 -0700 11975 11976 Bug 19379 - Provide docs with overview of all compose key combinations 11977 11978 Adds compose-chart.pl to generate DocBook/XML documents listing compose 11979 keys, and Makefile rules to generate HTML & PDF output from them if xmlto 11980 is present. 11981 11982 https://bugs.freedesktop.org/show_bug.cgi?id=19379 11983 11984 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11985 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> 11986 Reviewed-by: James Cloos <cloos@jhcloos.com> 11987 Tested-by: Gaetan Nadon <memsize@videotron.ca> 11988 11989commit 3eb064071695ebf0f371163ed818a428dfeba8e6 11990Author: Alan Coopersmith <alan.coopersmith@oracle.com> 11991Date: Sat Sep 11 00:49:21 2010 -0700 11992 11993 Make locale data build non-recursive / parallelizable 11994 11995 On a 4 core CPU with gmake -j 16 the nls subdir builds in half the time, 11996 plus this simplifies the next set of changes. 11997 11998 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 11999 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> 12000 Reviewed-by: James Cloos <cloos@jhcloos.com> 12001 Tested-by: Gaetan Nadon <memsize@videotron.ca> 12002 12003commit a3fc78ac352e5a70a958996ef6aec50f653974d1 12004Author: Gaetan Nadon <memsize@videotron.ca> 12005Date: Fri Sep 17 17:38:11 2010 -0400 12006 12007 config: remove man page suffix from bottom summary 12008 12009 The man page suffix is the same for all libraries on a given 12010 platform and is not configurable. It should have been removed 12011 in commit 09edc6de6. 12012 12013 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12014 12015commit a52b0068f36c739eb7e426214cd72a8d9bb776ca 12016Author: Gaetan Nadon <memsize@videotron.ca> 12017Date: Fri Sep 17 17:29:30 2010 -0400 12018 12019 config: remove unhelpful comment in .gitignore 12020 12021 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12022 12023commit 10dd881e7818be9074da4ab3a68577adcdaf71d2 12024Author: Adam Jackson <ajax@redhat.com> 12025Date: Wed Sep 8 10:44:23 2010 -0400 12026 12027 Zero buffer data in BufAlloc() 12028 12029 Inspired by a pattern in NoMachine's NX. Consistently zeroed buffers 12030 compress better with ssh and friends. Note that you'll need to rebuild 12031 all your protocol libraries to take advantage of this. 12032 12033 Signed-off-by: Adam Jackson <ajax@redhat.com> 12034 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 12035 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12036 12037commit da099f0d2015831ea5f4a5d0740ed962aa4d5d8b 12038Author: James Cloos <cloos@jhcloos.com> 12039Date: Sun Sep 12 17:00:54 2010 -0400 12040 12041 Revert “Dolt-ify" 12042 12043 Libtool’s is now sufficiently fast that DOLT is no longer 12044 worth the bother, even on those few systems where is works. 12045 12046 This reverts commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60 12047 and part of commit d31e644c65c52828ea3e7abd94a8cf9aee12265c. 12048 12049 It conflicted with commit f6a4fd0c7615684d08e848245849dea4017a5214 12050 which moved dolt from configure.ac to m4/dolt.m4. 12051 12052 And it addresses: http://bugs.freedesktop.org/show_bug.cgi?id=28188 12053 12054 Signed-off-by: James Cloos <cloos@jhcloos.com> 12055 12056commit 2661fbe6b809c937a60c2c1ca5ddb8280e399bd8 12057Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12058Date: Thu Sep 9 19:24:29 2010 -0700 12059 12060 XIM spec title page minor formatting cleanup 12061 12062 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12063 12064commit f99f1380bbf7bb2a0c491acad4a3d8db44bfd752 12065Author: Jens Petersen <petersen@redhat.com> 12066Date: Mon Aug 23 18:08:10 2010 -0700 12067 12068 Bug 29773: aliases for nb_NO.utf8 and nn_NO.utf8 12069 12070 <https://bugs.freedesktop.org/show_bug.cgi?id=29773> 12071 12072 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12073 12074commit b15b8a558ec64c834cc8f6a52d7f3c4f530c8c4b 12075Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12076Date: Fri Sep 3 23:11:53 2010 -0700 12077 12078 Sun's copyrights are now owned by Oracle 12079 12080 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12081 12082commit 7742bf62b1fa652da4270587e280249945367cae 12083Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12084Date: Fri Sep 3 18:31:44 2010 -0700 12085 12086 libX11 1.3.99.901 (1.4.0 RC1) 12087 12088 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12089 12090commit 5a02d6d74b5451d48d8f55709dfd4ecd3c068586 12091Author: Jeroen Hoek <mail@jeroenhoek.nl> 12092Date: Sun Aug 29 09:47:06 2010 -0400 12093 12094 libX11/nls compose tables, add LATIN CAPITAL LETTER SHARP S 12095 12096 Add LATIN CAPITAL LETTER SHARP S 12097 (See https://bugs.freedesktop.org/show_bug.cgi?id=29448 for rationale) 12098 12099 Signed-off-by: James Cloos <cloos@jhcloos.com> 12100 12101commit 09edc6de6619a2eabda3b808ebff4165550664f1 12102Author: Gaetan Nadon <memsize@videotron.ca> 12103Date: Fri Aug 13 14:21:08 2010 -0400 12104 12105 config: remove man-pages configuration option 12106 12107 This option was added in commit 6e752ea120 with no explanation. 12108 The section number is provoded by XORG_MANPAGE_SECTIONS 12109 There is no case where libX11 should be different than other libs 12110 The option was also used to disable building of the man pages, 12111 which build in 14 secs. No indication this is required. 12112 12113 If there is a requirement from system builders to disable building 12114 of man pages, it could be done consistently for all modules. 12115 12116 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12117 12118commit f92e754297ec5fdb81068b56a4435026666224fa 12119Author: Aaron Plattner <aplattner@nvidia.com> 12120Date: Sun Aug 15 21:51:38 2010 -0700 12121 12122 XOpenDisplay: save the correct display_name value 12123 12124 The X Test Suite's XDisplayString test checks the invariant 12125 XDisplayString(XOpenDisplay(str)) == str. The Xlib XOpenDisplay violates this 12126 invariant by expanding str to the canonical form "host:display.scrn" (unless 12127 HAVE_LAUNCHD is set and it starts with "/tmp/launch"). E.g., this expands ":1" 12128 to ":1.0": 12129 12130 400|26 1 1 19:26:41|IC Start 12131 200|26 1 19:26:41|TP Start 12132 520|26 1 00032625 1 1|VSW5TESTSUITE PURPOSE 1 12133 520|26 1 00032625 1 2|Assertion XDisplayString-1.(A) 12134 520|26 1 00032625 1 3|A call to XDisplayString returns the string that was used 12135 520|26 1 00032625 1 4|as the argument to the XOpenDisplay call that returned the 12136 520|26 1 00032625 1 5|value used as the display argument. 12137 520|26 1 00032625 1 6|METH: Open a connection using XOpenDisplay. 12138 520|26 1 00032625 1 7|METH: Obtain the display string using XDisplayString. 12139 520|26 1 00032625 1 8|METH: Verify that the value of the string is the parameter used in XOpenDisplay. 12140 520|26 1 00032625 1 9|METH: Close the display using XCloseDisplay. 12141 520|26 1 00032625 1 10|REPORT: XDisplayString() returned ":1.0" instead of ":1". 12142 220|26 1 1 19:26:41|FAIL 12143 410|26 1 1 19:26:41|IC End 12144 12145 Fix this by deleting all of the code to construct the canonical path and just 12146 stashing a copy of the original display_name in dpy->display_name. 12147 12148 Signed-off-by: Aaron Plattner <aplattner@nvidia.com> 12149 Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> 12150 12151commit 1dc401f678469d0235e6d0b28eb4356f03327c9f 12152Author: Gaetan Nadon <memsize@videotron.ca> 12153Date: Fri Aug 13 13:06:51 2010 -0400 12154 12155 man: xkb: remove unused variable LIB_MAN_DIR_SUFFIX 12156 12157 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12158 12159commit cbe9eebd11e5ebab4280d25b57b4cd4001241619 12160Author: Gaetan Nadon <memsize@videotron.ca> 12161Date: Fri Aug 13 11:42:29 2010 -0400 12162 12163 man: simplify building of shadow man pages 12164 12165 Store the shadow files in git as any other man page. 12166 Move man pages to man dir and use the common makefile 12167 12168 Local fix in CVS for bug 5628 is not required 12169 as the problem has been fixed in 12170 util-macros d9062e4077ebfd0985baf8418f3d0f111b9ddbba 12171 12172 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12173 12174commit bfd899c9af2318ecbe84cb69007dba4223fdb502 12175Author: Gaetan Nadon <memsize@videotron.ca> 12176Date: Mon Aug 9 09:39:26 2010 -0400 12177 12178 config: reinstate XORG_PROG_RAWCPP erroneoulsy removed 12179 12180 in commit 76e07ef6911734eac418e399b114f1b544512736. 12181 12182 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12183 12184commit 76e07ef6911734eac418e399b114f1b544512736 12185Author: Gaetan Nadon <memsize@videotron.ca> 12186Date: Fri Aug 6 14:01:51 2010 -0400 12187 12188 man: using the C preprocessor is not required for man pages. 12189 12190 There were no special symbols needing cpp. 12191 Everything can be handled by the default MAN_SUBSTS in util-macros. 12192 12193 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12194 12195commit 4b8ff7db39f2fe7ef12968d462aaf3f9054b6c18 12196Author: Jamey Sharp <jamey@minilop.net> 12197Date: Fri Aug 6 15:51:56 2010 -0700 12198 12199 Fix use-after-free in _XReply on X errors. 12200 12201 _XReply would always call dequeue_pending_request on errors. When it 12202 got an error for the current request, it would call dequeue, then break 12203 out of the loop; then, if it had an error in the event queue, it would 12204 compare it with the sequence number of the now-freed pending request. 12205 _XReply already stored that sequence number in dpy->last_request_read 12206 before freeing it, so look at that instead. 12207 12208 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29412 12209 12210 Signed-off-by: Jamey Sharp <jamey@minilop.net> 12211 Signed-off-by: Josh Triplett <josh@joshtriplett.org> 12212 12213commit 9fa146b30046396b70d64986e50d6617b3a8ac48 12214Author: Gaetan Nadon <memsize@videotron.ca> 12215Date: Sat Jul 31 16:15:35 2010 -0400 12216 12217 specs: xsl stylesheet requires abs path to images 12218 12219 Using abs_builddir requires automake 1.10 or later. 12220 12221 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12222 12223commit a8727d4bc39015e303b4128d0ad8aaf6d5fc9f0a 12224Author: Gaetan Nadon <memsize@videotron.ca> 12225Date: Sun Aug 1 14:13:07 2010 -0400 12226 12227 specs: update .gitignore now that all groff generated files are gone 12228 12229 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12230 12231commit f70527f05677aaf6a493ba42d2d95f738aaebafa 12232Author: Adam Jackson <ajax@redhat.com> 12233Date: Thu Jul 29 11:11:21 2010 -0400 12234 12235 nls: Switch one of the interrobang sequences to gnaborretni 12236 12237 Since gnaborretni is primarily used in LTR locales, the ? part of the 12238 sequence reflects the position of the bulge in the ?. When scanning an 12239 inverted interrobang left-to-right, you see the bulge first, so the 12240 compose sequence ?! is used; upright interrobang shows the bulge 12241 last, so it's !?. 12242 12243 This is a change in behaviour, but I was unable to come up with anything 12244 better. ??!! or !!?? would match the patterns for ¿ and ¡, but they 12245 would delay evaluation of them; "¿Qué?" wouldn't show the ¿ until you hit 12246 Q. Likewise ?!?! and !?!? would delay showing the interrobang itself. 12247 ~!? and ~?! were considered but are arguably less intuitive. 12248 12249 Reviewed-by: James Cloos <cloos@jhcloos.com> 12250 Signed-off-by: Adam Jackson <ajax@redhat.com> 12251 12252commit 511c4f6d29b2da4f71093feabcbb3913cb5d12a7 12253Author: Gaetan Nadon <memsize@videotron.ca> 12254Date: Wed Jul 28 10:27:39 2010 -0400 12255 12256 specs: move indexterm from glossdef to glossterm 12257 12258 This move fixes a Java class cast exception in the glossary. 12259 The problem was introduced in commit 12260 26f4f0d50840fe5ba4c46aae0a8e68db0059434b 12261 12262 It may not happen on all versions of the doc toolchain. 12263 There is no reason why indexterm cannot appear in glossdef, 12264 this is a workaround to an implementation problem found by 12265 trial and error. 12266 12267 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12268 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12269 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12270 12271commit 87a108f3aa565aa803dacb3a53b811f45025b609 12272Author: Matt Dew <matt@osource.org> 12273Date: Sat Jul 24 14:46:57 2010 -0400 12274 12275 specs: replace troff source with docbook-xml source 12276 12277 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12278 12279commit 8e603413586ff42b1351688f6f99318cbe243bca 12280Author: Julien Cristau <jcristau@debian.org> 12281Date: Sat Jul 24 22:13:07 2010 +0200 12282 12283 Compose.man: default user compose file is .XCompose, not .Xcompose 12284 12285 The path was correct in the DESCRIPTION section, but not in FILES. 12286 12287 Signed-off-by: Julien Cristau <jcristau@debian.org> 12288 12289commit 554da76ece85d0fc0cada45a86860e69c2107e9a 12290Author: Daniel Stone <daniel@fooishbar.org> 12291Date: Tue Jul 20 12:34:48 2010 +0100 12292 12293 NLS: Add \o/ Compose sequence 12294 12295 Unicode is even more pointlessly awesome! Add a binding for Compose-\o/ 12296 to U+1F64C PERSON RAISING BOTH HANDS IN CELEBRATION. 12297 12298 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12299 12300commit 2e7a18b6a617b9b4bfcea2d36f2bd2d7e0c4a3dd 12301Author: Daniel Stone <daniel@fooishbar.org> 12302Date: Fri Jul 9 18:13:13 2010 +0100 12303 12304 XStringToKeysym: Cope with 0x1234cafe-style input 12305 12306 If we get input in the style of 0xdeadbeef, just return that exact 12307 keysym. Introduces a dependency on strtoul, which I'm told is OK on all 12308 the systems we care about. 12309 12310 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12311 12312commit 3df45ed0c29b98ff468a0ff0ba24830bb664fd5a 12313Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12314Date: Mon Jul 19 17:58:27 2010 -0700 12315 12316 specs/libX11: Fix column count of Gravity Attributes table 12317 12318 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12319 12320commit 53bcba0d1d2dc3fd5e0de4bae3da30a1aa31a0c4 12321Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12322Date: Mon Jul 19 14:43:38 2010 -0700 12323 12324 specs/libX11: Convert \- to − and \^ to either   or removed 12325 12326 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12327 12328commit 5a0b45275638281e1bb2ae7d3e16b98e6470dae9 12329Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12330Date: Mon Jul 19 11:28:50 2010 -0700 12331 12332 specs/libX11: Manual cleanup pass over Ch. 1 12333 12334 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12335 12336commit 21567992958a6f12b3d0186f2f27a5b2a41b8448 12337Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12338Date: Mon Jul 19 14:51:10 2010 -0700 12339 12340 specs/libX11: make sure all files have DOCTYPEs so standard entities work 12341 12342 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12343 12344commit c944a8521f5760b485192658b921145159cdb439 12345Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12346Date: Sun Jul 18 13:12:20 2010 -0700 12347 12348 specs/libX11: Fix up list of header files in Ch.1 & add index entries 12349 12350 Combination of manual editing and automatic substitution via: 12351 12352 perl -i -p -00 -e 's{<varlistentry>(\s+)<term><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>$4<para>}g' CH01.xml 12353 12354 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12355 12356commit bb66e8f1ce330043278d3e8cbc7d5d3bc56ee030 12357Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12358Date: Sun Jul 18 12:47:27 2010 -0700 12359 12360 specs/libX11: Add indexterms for headerfiles that .hN used to provide 12361 12362 Automatic substitution performed via: 12363 perl -i -p -e 's{^<<filename class="headerfile">([^<]+)</filename>>(.*)$}{<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"><$1></filename></secondary></indexterm>}' *.xml 12364 12365 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12366 12367commit d0b0d215cb85da48b323b1392149ce181e7ef5ec 12368Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12369Date: Fri Jul 16 21:15:13 2010 -0700 12370 12371 specs/libX11: convert multicolumn lists from tables to <simplelist> 12372 12373 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12374 12375commit e679f80c6210c0aefe3669f8f536353619097a4b 12376Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12377Date: Fri Jul 16 20:51:52 2010 -0700 12378 12379 specs/libX11: Tag WM_* as <property> 12380 12381 Mass substitution done by: 12382 perl -i -p -e 'if ($_ !~ m{^\<}) { $_ =~ s{(WM_\w+)}{<property>$1</property>}g; }' *.xml 12383 12384 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12385 12386commit c7b7e59b3b22221d0be6286c540001c360308f69 12387Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12388Date: Fri Jul 16 20:41:33 2010 -0700 12389 12390 specs/libX11: Convert simpler eqn markup to docbook tags 12391 12392 Mostly "sup" to <superscript> 12393 12394 There's several more complicated equations that will probably need 12395 MathML or SVG to solve. 12396 12397 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12398 12399commit 897486c54c6a54771867d667441aaf9a4b9c35ca 12400Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12401Date: Fri Jul 16 00:50:39 2010 -0700 12402 12403 specs/libX11: App. D: convert literallayouts to synopsis or programlisting 12404 12405 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12406 12407commit 692906c3b3d8aa8b8927fbc230c3050d633785d2 12408Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12409Date: Fri Jul 16 00:36:43 2010 -0700 12410 12411 specs/libX11: Fix section headers in Appendix D 12412 12413 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12414 12415commit bcc41baa02e137884d847aec0f3ff8ca5c85c32f 12416Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12417Date: Wed Jul 14 07:22:26 2010 -0700 12418 12419 specs/libX11: Explicitly tag document as English 12420 12421 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12422 12423commit 6783544706ff370e900c137f951e90230586d6dc 12424Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12425Date: Mon Jul 12 22:03:18 2010 -0700 12426 12427 specs/libX11: Appendix C: convert literallayouts to synopsis or programlisting 12428 12429 Fixes display of a bunch of function prototypes and sample code 12430 12431 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12432 12433commit 8834cd9285e6aaf437aee56292c508d99a090fa9 12434Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12435Date: Mon Jul 12 18:37:16 2010 -0700 12436 12437 xmlrules.in: Add chunked-html rule to manually generate smaller files 12438 12439 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12440 12441commit 53f78680c34ebc94ce26b5e14c2a0003435cf10d 12442Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12443Date: Mon Jul 12 18:35:59 2010 -0700 12444 12445 specs/libX11: Convert Appendix A tables to real tables 12446 12447 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12448 12449commit 705a1257a8a47f4a1f94979aee09a837500ede7d 12450Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12451Date: Mon Jul 12 18:38:09 2010 -0700 12452 12453 specs/libX11: Fix out-of-place text in Chapter 1 12454 12455 DTD violation error introduced in 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4 12456 12457 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12458 12459commit a151346a2b7810e988f1de4b1e00b37672dc587a 12460Author: Jeremy Huddleston <jeremyhu@apple.com> 12461Date: Mon Jul 12 16:52:12 2010 -0700 12462 12463 launchd: Explicitly search /sbin 12464 12465 Previously, launchd wasn't found if /sbin wasn't in the user's PATH. 12466 https://bugs.freedesktop.org/show_bug.cgi?id=29028 12467 12468 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 12469 12470commit 89cc2e02e7c29ef9a02f0cfdf2090a2934f539e4 12471Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12472Date: Mon Jul 12 13:38:04 2010 -0700 12473 12474 Replace untranslated .hN macros with <filename> tags to show header names 12475 12476 Translation performed by: 12477 perl -i -p -e 's{\<\!-- .hN (\S+) (\S+)?\s*-->} 12478 {<<filename class="headerfile">$1</filename>>$2}' *.xml 12479 12480 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12481 12482commit 75480440946603b8efdbbf78b88d59c641b6d2c8 12483Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12484Date: Sat Jul 10 00:06:13 2010 -0700 12485 12486 specs/libX11: Mass substitution of <function> tags with more specific tags 12487 12488 Matched names from X headers & "nm libX11.so" output to names in spec to 12489 map to more specific tags. Tags used: 12490 12491 <code> code fragments 12492 <constant> enum values 12493 <errorname> X protocol errors 12494 <filename> filenames 12495 <function> functions 12496 <function> function-like macros (#define foo(a,b)...) 12497 <returnvalue> function return codes 12498 <structfield> struct members 12499 <structname> struct names (even when typedefed) 12500 <symbol> simple value #defines (#define NAME value) 12501 <systemitem> X protocol requests 12502 <systemitem class="event"> X protocol events 12503 <type> non-struct typedefs 12504 <varname> global variables 12505 12506 (Also fixed a couple typos detected by failures of this matching, such as 12507 XESSetPrintErrorValues -> XESetPrintErrorValues.) 12508 12509 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12510 12511commit 26f4f0d50840fe5ba4c46aae0a8e68db0059434b 12512Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12513Date: Fri Jul 9 00:37:23 2010 -0700 12514 12515 specs/libX11: Glossary cross-reference links 12516 12517 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12518 12519commit 930b52d84031de0e15e0a11cf4ecfd3b2f59e073 12520Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12521Date: Thu Jul 8 21:19:08 2010 -0700 12522 12523 specs/libX11: Make whitespace around <function> tags more uniform 12524 12525 Simplifies regular expressions for further mass substitutions. 12526 12527 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12528 12529commit 730ce6b8aca77b77385aa894d234cbde5fcbce37 12530Author: Gaetan Nadon <memsize@videotron.ca> 12531Date: Fri Jul 9 20:10:37 2010 -0400 12532 12533 specs: use pattern rules rather than suffix rules 12534 12535 This allows target to rebuild when included .xml files are changed. 12536 12537 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12538 12539commit 5527b4bc8ce7de60123eb28789ddef54aa48a378 12540Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12541Date: Thu Jul 8 20:56:41 2010 -0700 12542 12543 specs/libX11: Move punctuation outside of the <function> tags 12544 12545 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12546 12547commit d66d2134dfc38cd866c7c2d9ea45ed4b4dad23f0 12548Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12549Date: Thu Jul 8 20:42:50 2010 -0700 12550 12551 specs/libX11: Mark a number of <acronym>s 12552 12553 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12554 12555commit d5bbb12f55ac9f691eab6242cedc53207275131b 12556Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12557Date: Thu Jul 8 19:31:36 2010 -0700 12558 12559 specs/libX11: Glossary terms should not be marked as functions 12560 12561 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12562 12563commit 5decf7bc5dae0ae0c45c47eaaa9c4fdf9515ca1c 12564Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12565Date: Thu Jul 8 19:28:18 2010 -0700 12566 12567 specs/libX11: Clean up author list & acknowledgements a little 12568 12569 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12570 12571commit 0cd29bdb332c333e5123dce65b25ad1d97fbdae3 12572Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12573Date: Thu Jul 8 17:30:34 2010 -0700 12574 12575 specs/libX11: Convert troff .IN macros to docbook <indexterm> tags 12576 12577 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12578 12579commit a67d99ccbe22c8ce5f7a12f13b8991d9e0cf4491 12580Author: Gaetan Nadon <memsize@videotron.ca> 12581Date: Thu Jul 8 17:29:21 2010 -0400 12582 12583 specs: specdir is required to install xml files 12584 12585 The source is installed as well as the targets. 12586 This failed when configuring with --without-xmlto 12587 12588 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12589 12590commit e14273f44c1501ab51db4adcb83b18a1073787d8 12591Author: Matt Dew <matt@osource.org> 12592Date: Thu Jul 8 14:42:32 2010 -0400 12593 12594 specs: replace troff source with docbook-xml source 12595 12596 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 12597 12598commit bea0873caf50e9ed1b89255775d9ab912cbecd45 12599Author: Daniel Stone <daniel@fooishbar.org> 12600Date: Thu Jul 8 16:49:51 2010 +0100 12601 12602 XStringToKeysym: Check strdup() return value 12603 12604 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12605 Reviewed-by: Keith Packard <keithp@keithp.com> 12606 12607commit cffa71b4a5743e2b6675b9a917d15aef4177d513 12608Author: Marko Myllynen <myllynen@redhat.com> 12609Date: Mon Jun 28 15:08:05 2010 +0300 12610 12611 Fix two typos in SFS 5966 Annex 3 12612 12613 X.Org bug#28792 <https://bugs.freedesktop.org/show_bug.cgi?id=28792> 12614 12615 Signed-off-by: Julien Cristau <jcristau@debian.org> 12616 12617commit eb023c0f8919e809b8b609e1467b14d20a290aa7 12618Author: Daniel Stone <daniel@fooishbar.org> 12619Date: Tue Jun 15 18:49:43 2010 +0100 12620 12621 Delete now-redundant XKeysymDB 12622 12623 Since XStringToKeysym now supports all the vendor keysyms, just delete 12624 our XKeysymDB, which was incomplete at best, misleading at worst, and 12625 always an annoyance. 12626 12627 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12628 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12629 12630commit 00175397480b76d32bf82b0c7c94c91a2a95954e 12631Author: Daniel Stone <daniel@fooishbar.org> 12632Date: Tue Jun 15 18:48:48 2010 +0100 12633 12634 makekeys: Scan vendor keysyms as well as core 12635 12636 Since we can't really live without vendor keysyms, scan them all in to 12637 generate ks_tables.h, rather than only doing the core ones, and leaving 12638 the vendor syms to be manually synchronised with XKeysymDB. 12639 12640 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12641 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12642 12643commit ebd6ef0a4db0ddef0ae17ad14571518ccdeea5ba 12644Author: Daniel Stone <daniel@fooishbar.org> 12645Date: Tue Jun 15 18:47:37 2010 +0100 12646 12647 XStringToKeysym: Special case for XF86 keysyms 12648 12649 Some XFree86 keysyms were in XKeysymDB as XF86_foo, despite really being 12650 XF86foo. So, if we get to the bottom of XStringToKeysym and haven't 12651 found our XF86_foo, try it again as XF86foo. 12652 12653 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12654 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12655 12656commit 8c2ffce9e5c6eef8a04f7e2732db46b3b62e13bb 12657Author: Daniel Stone <daniel@fooishbar.org> 12658Date: Tue Jun 15 17:20:48 2010 +0100 12659 12660 configure.ac: Change from deprecated AC_DEFINE_DIR to AX_ 12661 12662 AC_DEFINE_DIR is deprecated as it's squatting on the autoconf-builtin 12663 namespace, so start using the more proper AX_DEFINE_DIR instead. 12664 12665 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 12666 Reviewed-by: Gaetan Nadon <memsize@videotron.ca> 12667 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12668 12669commit 6de368c9aa7ccd2fcd62fca5a2b278913db4d03d 12670Author: Fernando Carrijo <fcarrijo@yahoo.com.br> 12671Date: Thu Jul 1 06:50:47 2010 -0300 12672 12673 Purge macros NEED_EVENTS and NEED_REPLIES 12674 12675 Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br> 12676 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com> 12677 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12678 12679commit 0b724231be17f19538ee115a03b82b221b1f83c0 12680Author: Adam Jackson <ajax@redhat.com> 12681Date: Wed Jun 30 16:30:49 2010 -0400 12682 12683 Use -version-number consistently instead of -version-info 12684 12685 The latter is libtool braindamage. 12686 12687 Signed-off-by: Adam Jackson <ajax@redhat.com> 12688 12689commit 241a990afcf9c967c587ad6fd245df21b68b1ad9 12690Author: Juliusz Chroboczek <jch@pps.jussieu.fr> 12691Date: Tue Jun 29 18:32:42 2010 +0200 12692 12693 Make Compose-comma map to ogonek for I and U in UTF-8 locales. 12694 12695 With the preceding patch, it makes the UTF-8 compose map consistent with 12696 the ISO 8859-4 and -13 maps. 12697 12698commit af55e582f485a668c2bf43129be972bc65f03c60 12699Author: Juliusz Chroboczek <jch@pps.jussieu.fr> 12700Date: Tue Jun 29 18:20:18 2010 +0200 12701 12702 Make Compose-comma map to Ogonek for A and E in UTF-8 locales. 12703 12704 This makes the UTF-8 compose file consistent with the ISO 8859-2 compose file 12705 as far as Polish is concerned. We only sacrifice one pair of characters, 12706 e-cedilla, which is not used in any language. 12707 12708commit d6ba13009fc9ef876a104b907ffef73c6e405d4b 12709Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12710Date: Thu Jun 24 13:13:11 2010 -0700 12711 12712 Define FILE_MAN_DIR_SUFFIX so XCompose shadow page has correct path 12713 12714 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12715 12716commit 978c7238789037de917b9423ea9adfb006da1260 12717Author: Jamey Sharp <jamey@minilop.net> 12718Date: Sat Jun 19 10:44:55 2010 -0700 12719 12720 poll_for_response: Really handle xcb_poll_for_reply getting a reply. 12721 12722 Don't lose async replies. That's bad. 12723 12724 `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the 12725 _XReply path worked; that path waited for replies, rather than polling. 12726 12727 However, XRecordProcessReplies, which does nothing but call XPending, 12728 relied on the event-handling path to process async replies, and that was 12729 busted. 12730 12731 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595 12732 12733 Signed-off-by: Jamey Sharp <jamey@minilop.net> 12734 12735commit ef67486c5a8aada24fd95b2991a3c4979f53805f 12736Author: Jamey Sharp <jamey@minilop.net> 12737Date: Mon Jun 21 13:34:15 2010 -0700 12738 12739 Revert "xcb_io.c: poll_for_response doesn't guarantee there's a pending request." 12740 12741 This reverts commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e, because as 12742 a matter of fact, if poll_for_response returns NULL when we know 12743 dpy->xcb->next_event is non-NULL, there *is* guaranteed to be a pending 12744 request. 12745 12746commit 301ec5b41e9d253a446db656e1789ac7345dc433 12747Author: Jamey Sharp <jamey@minilop.net> 12748Date: Mon Jun 21 13:24:24 2010 -0700 12749 12750 Revert "poll_for_response: Really handle xcb_poll_for_reply getting a reply." 12751 12752 This reverts commit c115095d7f2bc4f5a4fb26380e3698fefdad7611. We *do* 12753 need to check poll_for_event every time through the loop in 12754 poll_for_response, so the commit did too much. 12755 12756commit 4a8b6528ff69f6feb8c0e119939b4ce6c088f29e 12757Author: Jamey Sharp <jamey@minilop.net> 12758Date: Mon Jun 21 10:54:02 2010 -0700 12759 12760 xcb_io.c: poll_for_response doesn't guarantee there's a pending request. 12761 12762 Fixes the second bug reported in: 12763 https://bugs.freedesktop.org/show_bug.cgi?id=28595 12764 12765 Signed-off-by: Jamey Sharp <jamey@minilop.net> 12766 12767commit c115095d7f2bc4f5a4fb26380e3698fefdad7611 12768Author: Jamey Sharp <jamey@minilop.net> 12769Date: Sat Jun 19 10:44:55 2010 -0700 12770 12771 poll_for_response: Really handle xcb_poll_for_reply getting a reply. 12772 12773 Don't lose async replies. That's bad. 12774 12775 `xlsfonts -l`, which uses XListFontsWithInfo, worked fine, because the 12776 _XReply path worked; that path waited for replies, rather than polling. 12777 12778 However, XRecordProcessReplies, which does nothing but call XPending, 12779 relied on the event-handling path to process async replies, and that was 12780 busted. 12781 12782 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=28595 12783 12784 Signed-off-by: Jamey Sharp <jamey@minilop.net> 12785 Signed-off-by: Josh Triplett <josh@joshtriplett.org> 12786 12787commit a25ae169862ab9b76daf259613b37c6b07bc2ef2 12788Author: Marko Myllynen <myllynen@redhat.com> 12789Date: Sun Jun 13 19:23:05 2010 +0300 12790 12791 Implement SFS 5966 Annex 4 for Finland 12792 12793 This patch adds Annex 4 of SFS 5966 for Finland and fixes two typos in 12794 Annex 3 (which were actually copied verbatim from the standard text). 12795 12796 The implementation of the standard is now complete. 12797 12798 X.Org bug#28498 <https://bugs.freedesktop.org/show_bug.cgi?id=28498> 12799 12800 Signed-off-by: Julien Cristau <jcristau@debian.org> 12801 12802commit 47b04195d8a31c8f9e6dd804196162c6cfca3ac6 12803Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12804Date: Fri Jun 4 13:20:17 2010 -0700 12805 12806 LINEAR_RGB_InitSCCData: When malloc fails, don't try to free unallocated bits 12807 12808 One of the malloc failure checks had a goto to the wrong spot in the 12809 list of cleanup free() calls to unwind at the end, and was freeing 12810 bits that hadn't been initialized/allocated yet, since they would be 12811 stored in the struct that just failed to be allocated. 12812 12813 Error: Null pointer dereference (CWE 476) 12814 Read from pointer that could be constant 'NULL' 12815 at line 805 of /export/alanc/X.Org/sx86/lib/libX11/src/xcms/LRGB.c in function 'LINEAR_RGB_InitSCCData'. 12816 Pointer checked against constant 'NULL' at line 754 but does not protect the dereference. 12817 12818 [ This bug was found by the Parfait bug checking tool. 12819 For more information see http://research.sun.com/projects/parfait ] 12820 12821 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12822 12823commit aebbf3623888119b43893b253195d93b613a6e1e 12824Author: Josh Triplett <josh@joshtriplett.org> 12825Date: Sun Jun 6 19:11:55 2010 -0700 12826 12827 Mark the rest of Xprivate.h as _X_HIDDEN. 12828 12829 None of the functions in Xprivate.h should have any callers outside of 12830 Xlib, by definition. 12831 12832 Signed-off-by: Josh Triplett <josh@joshtriplett.org> 12833 Suggested-by: Jamey Sharp <jamey@minilop.net> 12834 Reviewed-by: Jamey Sharp <jamey@minilop.net> 12835 12836commit abcd1b67c8135ad652833e5a60d818837c39fce1 12837Author: Julien Cristau <jcristau@debian.org> 12838Date: Fri Jun 4 13:57:45 2010 +0200 12839 12840 Hide _XSeqSyncFunction 12841 12842 Commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43 made _XSeqSyncFunction 12843 non-static, but we don't need to export it. 12844 12845 Signed-off-by: Julien Cristau <jcristau@debian.org> 12846 Reviewed-by: Adam Jackson <ajax@redhat.com> 12847 Reviewed-by: Jamey Sharp <jamey@minilop.net> 12848 12849commit 15e5eaf62897b3179d1fbe457cb19f886f0449f8 12850Author: Josh Triplett <josh@joshtriplett.org> 12851Date: Thu Jun 3 09:41:01 2010 -0700 12852 12853 Remove support for building without XCB 12854 12855 And there was much rejoicing. 12856 12857 Signed-off-by: Josh Triplett <josh@joshtriplett.org> 12858 Reviewed-by: Jamey Sharp <jamey@minilop.net> 12859 12860 Consensus on #xorg-devel agrees with removing --without-xcb; in 12861 particular, acks from Adam Jackson, Daniel Stone, Kristian Høgsberg, 12862 Julien Cristau, and Rémi Cardona. 12863 12864commit fb19eb767a32fd4ff74021c51bc6f60eb8bdff74 12865Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12866Date: Thu Jun 3 15:21:40 2010 -0700 12867 12868 libX11 1.3.4 12869 12870 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12871 12872commit 8f3db40ca5108a919244f3fff7466d01a14b3ce2 12873Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12874Date: Thu Jun 3 13:08:17 2010 -0700 12875 12876 Workaround bug in groff flag processing that breaks distcheck 12877 12878 At least with the groff 1.19.2 package I have installed, groff passes 12879 on the -I flags for the include path to grohtml, which if they come 12880 after the -P-I... flag we pass to grohtml to specify the image file 12881 name pattern cause it to override that flag and put the images in 12882 the wrong place, breaking "make distcheck" - changing the flag order 12883 works around this. 12884 12885 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12886 12887commit bdb31a1fb707cebccc9efbbf68cb55c10fa4ea3e 12888Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12889Date: Thu Jun 3 11:25:33 2010 -0700 12890 12891 Fix typo that made configure always report "none" for man page suffix 12892 12893 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12894 12895commit 6bb7d2d21d6b58019fb37b724b59c9744f30b9ce 12896Author: Paul Bender <pebender@gmail.com> 12897Date: Thu Jun 3 11:20:26 2010 -0700 12898 12899 Bug 22591 - configure does not obey the --enable-*-transport options 12900 12901 https://bugs.freedesktop.org/show_bug.cgi?id=22591 12902 12903 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12904 12905commit d7f35946d673bb00f2768e025b6e1017422b2174 12906Author: Paul Bender <pebender@gmail.com> 12907Date: Thu Jun 3 11:17:45 2010 -0700 12908 12909 Bug 22590 - libX11 1.2.1 has broken abstract namespace support 12910 12911 https://bugs.freedesktop.org/show_bug.cgi?id=22590 12912 12913 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12914 12915commit b2487d07f7b355f230a56e32c763acd08862735c 12916Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12917Date: Thu Jun 3 11:06:15 2010 -0700 12918 12919 Bug 22584 - libX11 does not cross compile 12920 12921 Adaptation of patch submitted by Paul Bender in attachment 27301 to 12922 https://bugs.freedesktop.org/show_bug.cgi?id=22584 12923 12924 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12925 Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net> 12926 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 12927 12928commit 4378219ff8c1418418c70086085358b69d574e74 12929Author: Alan Coopersmith <alan.coopersmith@oracle.com> 12930Date: Tue Jun 1 18:55:48 2010 -0700 12931 12932 Clarify requirements in XRestackWindows man page 12933 12934 The required common parent window is not specified in the arguments, 12935 just implied as the parent of the first window in the list. 12936 12937 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 12938 12939commit f09c5299a381e2729e800a0ac43f1c0e371f65f6 12940Author: Jeremy Huddleston <jeremyhu@apple.com> 12941Date: Wed May 12 16:42:18 2010 -0700 12942 12943 xcb: Add TCP fallback 12944 12945 If we fail to connect to a UNIX socket and the transport isn't specified, 12946 fallback on TCP. This matches behavior with the xtrans codepath and the 12947 Xlib spec. 12948 12949 http://lists.x.org/archives/xorg-devel/2010-April/007915.html 12950 12951 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 12952 Reviewed-by: Jamey Sharp <jamey@minilop.net> 12953 12954commit fd82552d5c0ce1931f29006a0c36f5e03cf8577e 12955Merge: aae2a4a7 933aee1d 12956Author: Jamey Sharp <jamey@minilop.net> 12957Date: Mon May 10 16:51:24 2010 -0700 12958 12959 Merge branch 'xlib-xcb-thread-fixes' 12960 12961commit aae2a4a7aab26de3fa715d6ecd0a0e0926b37fc9 12962Author: Jeremy Huddleston <jeremyhu@apple.com> 12963Date: Fri Apr 23 21:50:29 2010 -0700 12964 12965 Don't append the screen number when using a launchd socket 12966 12967 ssh gets confused by this. XQuartz is the only DDX using this 12968 functionality, and it doesn't support different screens, so 12969 let's just not include this until most ssh know how to handle 12970 this. 12971 12972 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 12973 12974commit adcd0ec209abf832a769d52db660fb37eaad6e0c 12975Author: Jeremy Huddleston <jeremyhu@apple.com> 12976Date: Fri Apr 23 16:33:44 2010 -0700 12977 12978 Remove launchd logic from _XConnectXCB as it's handled in XCB 12979 12980 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 12981 12982commit d232b259c36fdde1f4179822809fec1480867dc5 12983Author: Jeremy Huddleston <jeremyhu@apple.com> 12984Date: Tue Feb 2 17:01:28 2010 -0800 12985 12986 Fix various build warnings 12987 12988 imLcIm.c: In function '_XimCachedFileName': 12989 imLcIm.c:361: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int' 12990 imLcIm.c:364: warning: format '%03x' expects type 'unsigned int', but argument 8 has type 'long unsigned int' 12991 12992 imRm.c: In function '_XimDefaultArea': 12993 imRm.c:597: warning: cast from pointer to integer of different size 12994 imRm.c: In function '_XimDefaultColormap': 12995 imRm.c:626: warning: cast from pointer to integer of different size 12996 12997 lcFile.c:224: warning: no previous prototype for 'xlocaledir' 12998 12999 lcUTF8.c: In function 'iconv_cstombs': 13000 lcUTF8.c:1841: warning: assignment discards qualifiers from pointer target type 13001 lcUTF8.c:1869: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness 13002 lcUTF8.c:1873: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness 13003 lcUTF8.c: In function 'iconv_mbstocs': 13004 lcUTF8.c:1935: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness 13005 lcUTF8.c: In function 'iconv_mbtocs': 13006 lcUTF8.c:2031: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness 13007 lcUTF8.c: In function 'iconv_mbstostr': 13008 lcUTF8.c:2121: warning: pointer targets in passing argument 2 of 'mbtowc' differ in signedness 13009 lcUTF8.c: In function 'iconv_strtombs': 13010 lcUTF8.c:2180: warning: pointer targets in passing argument 1 of 'wctomb' differ in signedness 13011 lcUTF8.c: In function '_XlcAddGB18030LocaleConverters': 13012 lcUTF8.c:2367: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13013 lcUTF8.c:2368: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13014 lcUTF8.c:2373: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13015 lcUTF8.c:2374: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13016 lcUTF8.c:2375: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13017 lcUTF8.c:2376: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13018 lcUTF8.c:2377: warning: passing argument 5 of '_XlcSetConverter' from incompatible pointer type 13019 13020 XlibInt.c: In function '_XGetHostname': 13021 XlibInt.c:3441: warning: implicit declaration of function 'gethostname' 13022 XlibInt.c:3441: warning: nested extern declaration of 'gethostname' 13023 13024 ConnDis.c: In function '_XDisconnectDisplay': 13025 ConnDis.c:540: warning: old-style function definition 13026 ConnDis.c: In function '_XSendClientPrefix': 13027 ConnDis.c:554: warning: old-style function definition 13028 ConnDis.c: In function 'XSetAuthorization': 13029 ConnDis.c:677: warning: old-style function definition 13030 13031 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 13032 13033commit 3e11c73187acb2e2be9a812840bbbea947527ccb 13034Author: Daniel Kahn Gillmor <dkg@fifthhorseman.net> 13035Date: Fri Apr 23 00:01:54 2010 +0200 13036 13037 Allow X11 users to compose anarchism 13038 13039 I can easily compose the hammer and sickle (the symbol representative 13040 of communism) by pressing the key sequence: 13041 13042 <Compose> <C> <C> <C> <P> -> ☭ 13043 13044 But i can't easily make the circled A (the symbol representative of 13045 anarchism). 13046 13047 I'd like to be able to use <Compose> <O> <A> (this is a 13048 currently unused key sequence) to generate the symbol: Ⓐ 13049 13050 Debian bug#555938 <http://bugs.debian.org/555938> 13051 13052 Signed-off-by: Julien Cristau <jcristau@debian.org> 13053 13054commit 933aee1d5c53b0cc7d608011a29188b594c8d70b 13055Author: Jamey Sharp <jamey@minilop.net> 13056Date: Fri Apr 16 20:18:28 2010 -0700 13057 13058 Fix Xlib/XCB for multi-threaded applications (with caveats). 13059 13060 Rather than trying to group all response processing in one monolithic 13061 process_responses function, let _XEventsQueued, _XReadEvents, and 13062 _XReply each do their own thing with a minimum of code that can all be 13063 reasoned about independently. 13064 13065 Tested with `ico -threads 20`, which seems to be able to make many 13066 icosahedrons dance at once quite nicely now. 13067 13068 Caveats: 13069 13070 - Anything that was not thread-safe in Xlib before XCB probably still 13071 isn't. XListFontsWithInfo, for instance. 13072 13073 - If one thread is waiting for events and another thread tries to read a 13074 reply, both will hang until an event arrives. Previously, if this 13075 happened it might work sometimes, but otherwise would trigger either 13076 an assertion failure or a permanent hang. 13077 13078 - Versions of libxcb up to and including 1.6 have a bug that can cause 13079 xcb_wait_for_event or xcb_wait_for_reply to hang if they run 13080 concurrently with xcb_writev or other writers. So you'll want that fix 13081 as well. 13082 13083 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13084 Reviewed-by: Josh Triplett <josh@freedesktop.org> 13085 13086commit aab43278ae619eb57d2dd9c7396f460f078588fc 13087Author: Jamey Sharp <jamey@minilop.net> 13088Date: Fri Apr 16 22:24:16 2010 -0700 13089 13090 Use InternalLockDisplay on code paths called from LockDisplay. 13091 13092 It's easier to reason about the code when we can't re-enter the 13093 Xlib-private sync-handlers while they're already running. 13094 13095 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13096 Reviewed-by: Josh Triplett <josh@freedesktop.org> 13097 13098commit 660b7d05f4ca4ab4661c9fe7ce655a4909b4e556 13099Author: Jamey Sharp <jamey@minilop.net> 13100Date: Fri Apr 16 19:45:11 2010 -0700 13101 13102 Fix _XSend to enqueue the right range of pending requests. 13103 13104 _XSend was off-by-one on both ends. It should not re-enqueue the last 13105 request that was already flushed, but it should enqueue the last request 13106 currently being flushed. 13107 13108 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13109 Reviewed-by: Josh Triplett <josh@freedesktop.org> 13110 13111commit f2735889908d6e5a7f8dbee42f00c54a52665191 13112Author: Jamey Sharp <jamey@minilop.net> 13113Date: Fri Apr 16 19:45:11 2010 -0700 13114 13115 Pending requests are always added in-order. 13116 13117 Replace insert_pending_request, which did an in-order search of the 13118 queue to find the right insertion point, with a simpler 13119 append_pending_request, and use that in _XSend as well. 13120 13121 Includes assertions to check that the list of pending requests is in 13122 order by sequence number and does not have duplicates. 13123 13124 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13125 Reviewed-by: Josh Triplett <josh@freedesktop.org> 13126 13127commit d9cf5885b0f97942fbbd2a7cc50118132ece50f6 13128Author: Markus Duft <markus.duft@salomon.at> 13129Date: Fri Apr 16 08:39:34 2010 -0700 13130 13131 Bug 26839: Fix build problem on Interix (POSIX layer on Windows) 13132 13133 https://bugs.freedesktop.org/show_bug.cgi?id=26839 13134 13135 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13136 13137commit a3f5f1b90936d23e9894e3261b2d77fb7b32a51a 13138Author: Josh Triplett <josh@freedesktop.org> 13139Date: Thu Apr 15 14:24:21 2010 -0700 13140 13141 Stop returning an int from _XIDHandler and _XSeqSyncFunction 13142 13143 _XIDHandler and _XSeqSyncFunction originally ran from dpy->synchandler, and 13144 thus had to return an int. Now, they only run from _XPrivSyncHandler or 13145 LockDisplay, neither of which needs to check their return value since they 13146 always returned 0. Make them both void. 13147 13148 Signed-off-by: Josh Triplett <josh@freedesktop.org> 13149 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13150 13151commit a6d974dc59f2722b36e2df9d4f07aeee4f83ce43 13152Author: Jamey Sharp <jamey@minilop.net> 13153Date: Thu Apr 15 13:05:08 2010 -0700 13154 13155 Move XID and sync handling from SyncHandle to LockDisplay to fix races. 13156 13157 XID and sync handling happened via _XPrivSyncHandler, assigned to 13158 dpy->synchandler and called from SyncHandle. _XPrivSyncHandler thus ran 13159 without the Display lock, so manipulating the Display caused races, and 13160 these races led to assertions in multithreaded code (demonstrated via 13161 ico). 13162 13163 In the XTHREADS case, after you've called XInitThreads, we can hook 13164 LockDisplay and UnlockDisplay. Use that to run _XIDHandler and 13165 _XSeqSyncHandler from LockDisplay rather than SyncHandle; we then know 13166 that we hold the lock, and thus we can avoid races. We think it makes 13167 sense to do these both from LockDisplay rather than UnlockDisplay, so 13168 that you know you have valid sync and a valid XID before you start 13169 setting up the request you locked to prepare. 13170 13171 In the !XTHREADS case, or if you haven't called XInitThreads, you don't 13172 get to use Xlib from multiple threads, so we can use the logic we have 13173 now (with synchandler and savedsynchandler) without any concern about 13174 races. 13175 13176 This approach gets a bit exciting when the XID and sequence sync 13177 handlers drop and re-acquire the Display lock. Reacquisition will re-run 13178 the handlers, but they return immediately unless they have work to do, 13179 so they can't recurse more than once. In the worst case, if both of 13180 them have work to do, we can nest the Display lock three deep. In the 13181 case of the _XIDHandler, we drop the lock to call xcb_generate_id, which 13182 takes the socket back if it needs to request more XIDs, and taking the 13183 socket back will reacquire the lock; we take care to avoid letting 13184 _XIDHandler run again and re-enter XCB from the return_socket callback 13185 (which causes Very Bad Things, and is Not Allowed). 13186 13187 Tested with ico (with 1 and 20 threads), and with several test programs 13188 for XID and sequence sync. Tested with and without XInitThreads(), and 13189 with and without XCB. 13190 13191 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=23192 13192 13193 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13194 Signed-off-by: Josh Triplett <josh@freedesktop.org> 13195 13196commit b089b53b697c2851db2985d32af3b29f1da5e31e 13197Author: Jamey Sharp <jamey@minilop.net> 13198Date: Wed Apr 14 12:59:11 2010 -0700 13199 13200 Honest. Extensions get to filter async errors too. 13201 13202 Under some circumstances, errors are picked up by calling 13203 xcb_poll_for_reply, rather than xcb_poll_for_event, because Xlib issued 13204 the requests with the XCB_REQUEST_CHECKED flag. That happens when either 13205 an async handler is queued at the time the requests are flushed, or when 13206 XSetEventQueueOwner has been used to prevent Xlib from processing XCB's 13207 event queue. 13208 13209 This bugfix extends 405132dab64bf2375f8e57d02b1b53da2311933a to cover 13210 those cases. 13211 13212 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545 13213 13214 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13215 13216commit a15c31274650e391bc6de5d0951eb4464c228139 13217Author: Jamey Sharp <jamey@minilop.net> 13218Date: Wed Apr 14 12:10:34 2010 -0700 13219 13220 _XError already runs async handlers; only call them directly for replies. 13221 13222 The previous behavior probably would have triggered bug reports someday. 13223 13224 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13225 13226commit 405132dab64bf2375f8e57d02b1b53da2311933a 13227Author: Jamey Sharp <jamey@minilop.net> 13228Date: Tue Apr 13 12:12:36 2010 -0700 13229 13230 Extensions get to filter async errors too. 13231 13232 Apparently I misread XlibInt.c:_XReply and thought that handlers set 13233 with XESetError should be consulted only for the sequence number that 13234 _XReply is currently looking for. In fact, the error handlers are also 13235 consulted when an error arrives for a request that was not expected to 13236 have a reply. 13237 13238 However, in an odd twist, the error handlers are *not* consulted outside 13239 of _XReply--that is, when looking for events, or waiting to be able to 13240 flush the output queue. So this patch takes some small pains to preserve 13241 that behavior, though it seems likely to have been unintentional. 13242 13243 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=26545 13244 13245 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13246 13247commit d3eab4a38f3e53ef21eb3b6fa66ead7afecf6227 13248Author: Jamey Sharp <jamey@minilop.net> 13249Date: Tue Apr 13 12:49:59 2010 -0700 13250 13251 Prefer the xcb_generic_error_t we already have over casting to xEvent. 13252 13253 Just a minor cleanup. 13254 13255 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13256 13257commit 75ea8c37935ccc911557d16a303ba595b8ab106b 13258Author: Jamey Sharp <jamey@minilop.net> 13259Date: Mon Apr 12 11:30:20 2010 -0700 13260 13261 Run the user's synchandler as well as any internal synchandlers. 13262 13263 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=27595 13264 13265 Signed-off-by: Jamey Sharp <jamey@minilop.net> 13266 13267commit c3f3e4a9e531d010312c97e753d6e543e607094d 13268Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 13269Date: Fri Feb 12 17:16:10 2010 +0900 13270 13271 Fix manual typos. 13272 13273 Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp> 13274 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13275 13276commit 75c6c5595408caba9ef8233839d77021d02f86f2 13277Author: Dirk Wallenstein <dirkwallenstein@t-online.de> 13278Date: Fri Feb 5 20:42:46 2010 +0100 13279 13280 man: Correct the XkbAllAccessXEventsMask mask name 13281 13282 This mask has probably been renamed but not been updated in the manuals. 13283 13284 Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de> 13285 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13286 13287commit 9262aceaff24c954cab8c252690032bc2ec25e7d 13288Author: Dirk Wallenstein <dirkwallenstein@t-online.de> 13289Date: Fri Feb 5 20:42:45 2010 +0100 13290 13291 man: Add missing geometry component flag 13292 13293 Signed-off-by: Dirk Wallenstein <dirkwallenstein@t-online.de> 13294 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13295 13296commit 50cf134465f6fd263f37acecac94518468ef1c6d 13297Author: Dirk Wallenstein <halsmit@t-online.de> 13298Date: Tue Feb 2 21:24:32 2010 +0100 13299 13300 man: Return value of XkbGetState is Status and not Bool 13301 13302 Signed-off-by: Dirk Wallenstein <halsmit@t-online.de> 13303 Reviewed-by: Julien Cristau <jcristau@debian.org> 13304 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13305 13306commit 7a93ae2bb5292a97fdd27ed818bc81248b37f641 13307Author: Dirk Wallenstein <halsmit@t-online.de> 13308Date: Tue Feb 2 21:24:30 2010 +0100 13309 13310 man: Fix return value specification of XkbKeyActionEntry 13311 13312 The XkbKeyActionEntry macro expands to a pointer. 13313 13314 Signed-off-by: Dirk Wallenstein <halsmit@t-online.de> 13315 Reviewed-by: Julien Cristau <jcristau@debian.org> 13316 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13317 13318commit e9884d4a05a5661ec343ea8a2aa0562b6419e086 13319Author: Dirk Wallenstein <halsmit@t-online.de> 13320Date: Tue Feb 2 21:24:29 2010 +0100 13321 13322 man: Redirect users from XKeycodeToKeysym to XkbKeycodeToKeysym #25732 13323 13324 XKeycodeToKeysym keeps compatibility with pre-XKB and thus only sees 2 13325 groups with 2 levels each. It wraps the index into the next group. 13326 This behavior confuses the unaware user, and therefore this will add a 13327 reference to XkbKeycodeToKeysym in the corresponding man paragraph. 13328 13329 Another bug had that issue, too. #5349 13330 13331 Signed-off-by: Dirk Wallenstein <halsmit@t-online.de> 13332 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13333 13334commit 6474cdf97cd648afaeaa872155f3fa21cfb43f80 13335Author: Alan Coopersmith <alan.coopersmith@oracle.com> 13336Date: Thu Apr 8 19:59:55 2010 -0700 13337 13338 Fix typo in new fi_FI.UTF-8 that was reported by "make check" 13339 13340 Unrecognized pattern in Compose on line #154: 13341 <dead_diaeresis> <space> " "¨" 13342 13343 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 13344 13345commit 91bd5e7af2d3a9526aefb73dc10f417896f0e4dd 13346Author: Marko Myllynen <myllynen@redhat.com> 13347Date: Thu Apr 8 19:52:42 2010 -0700 13348 13349 Bug 27465 - Rewritten fi_FI.UTF-8 Compose file 13350 13351 I've rewritten the fi_FI.UTF-8 Compose file so that it 13352 13353 - includes en_US.UTF-8/Compose for base compose definitions (and thus gets any 13354 possible additions to en_US.UTF-8/Compose automatically included) 13355 - overrides any en_US.UTF-8/Compose definitions with fi_FI.UTF-8/Compose 13356 definitions (thus making it safe to include en_US.UTF-8/Compose) 13357 - lists all the definitions specified in the SFS 5966 standard (thus making it 13358 easy to verify that the implementation is valid and complete) 13359 - adds a header about the file itself as required by the standard 13360 13361 The rewritten version is ~320 lines compared to over 5000 lines of the previous 13362 version. 13363 13364 https://bugs.freedesktop.org/show_bug.cgi?id=27465 13365 13366commit f6a4fd0c7615684d08e848245849dea4017a5214 13367Author: Gaetan Nadon <memsize@videotron.ca> 13368Date: Thu Apr 1 21:46:12 2010 -0400 13369 13370 config: update and relocate AC_DEFINE_DIR macro 13371 13372 Remove deprecated acinclude.m4 macro container file 13373 Use separate macro files as per autoconf recommendation 13374 Use the latest macro from GNU (ax) which replaces 13375 the non-gnu version (ac) 13376 This preserves the Autoconf macro AC namespace. 13377 13378 Also moved out of acinclude.m4 is the DOLT macro 13379 13380 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13381 13382commit 61d5231db418cec51bd718633c3bba39b18689d6 13383Author: Gaetan Nadon <memsize@videotron.ca> 13384Date: Mon Mar 29 14:53:48 2010 -0400 13385 13386 config: remove the pkgconfig pc.in file from EXTRA_DIST 13387 13388 Automake always includes it in the tarball. 13389 13390 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13391 13392commit 69fc76a0321c6b25cc4286173c53435c69ab8e0e 13393Author: Fernando Carrijo <fcarrijo@yahoo.com.br> 13394Date: Wed Mar 17 23:53:16 2010 -0300 13395 13396 No need for req->firstKeyCode to be set twice. 13397 13398 There is no reason to set req->firstKeyCode twice when a client, wishful 13399 for changing keyboard mappings, calls XChangeKeyboardMapping. This patch 13400 fixes the mistake by making no functional changes to libX11 whatsoever. 13401 13402 Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br> 13403 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 13404 13405commit b5797597f83ee62228abfb88f026ef92163e07a5 13406Author: Gaetan Nadon <memsize@videotron.ca> 13407Date: Tue Feb 16 10:37:21 2010 -0500 13408 13409 config: move CWARNFLAGS from configure.ac to Makefile.am 13410 13411 Compiler warning flags should be explicitly set in the makefile 13412 rather than being merged with other packages compiler flags. 13413 13414 Reviewed-by: Dan Nicholson <dbn.lists@gmail.com> 13415 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13416 13417commit 052c4d7995b0e52a038933beb85cd544571c6b7e 13418Author: Gaetan Nadon <memsize@videotron.ca> 13419Date: Tue Feb 16 14:28:21 2010 -0500 13420 13421 specs: change install cmd due to automake 1.11 13422 13423 specData_INSTALL is defined in 1.9 and 1.10 but not 1.11 13424 13425 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13426 13427commit 7277a18cda9f5a6807d11cf4ceb9449b1c302c18 13428Author: Gaetan Nadon <memsize@videotron.ca> 13429Date: Tue Feb 9 17:27:08 2010 -0500 13430 13431 doc: use $(mkdir_p) rather than $(MKDIR_P) due to automake 1.9.6 13432 13433 $(MKDIR_P) is not defined in automake 1.9. 13434 13435 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13436 13437commit e66eda95b520727969d721416d4ea49b54ef2b03 13438Author: Gaetan Nadon <memsize@videotron.ca> 13439Date: Tue Feb 2 16:29:26 2010 -0500 13440 13441 specs: install html images in $docdir with html files 13442 13443 The images required by the html files have been omitted. 13444 13445 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13446 13447commit 137d1ff6be99d040e77c8c4c52029c5bdf1f4808 13448Author: Gaetan Nadon <memsize@videotron.ca> 13449Date: Sun Jan 31 14:16:20 2010 -0500 13450 13451 doc: use new macros to control doc generation 13452 13453 Namely XORG_WITH_GROFF for the groff generation tool 13454 XORG_WITH_PS2PDF for the conversion of PS docs to PDF 13455 XORG_ENABLE_SPECS for the generation of specs 13456 13457 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13458 13459commit 9ca583c5d6fcf9cd8151ee7b019630b141d32fab 13460Author: Gaetan Nadon <memsize@videotron.ca> 13461Date: Tue Jan 26 19:41:08 2010 -0500 13462 13463 doc: clean-up generated html images 13464 13465 Generate images in /images as is the convention 13466 Provide a base file name for images rather than process ID 13467 Remove images directory when running make clean 13468 13469 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 13470 13471commit eb289b34bfc9978eef724e0f268975938334c06c 13472Author: Kristian Høgsberg <krh@bitplanet.net> 13473Date: Thu Jan 21 13:39:17 2010 -0500 13474 13475 XErrorDB: Add new DRI2 request names 13476 13477commit 6babf2123d05adb9349394c49cc2d81d1f66cf9e 13478Author: Julien Cristau <jcristau@debian.org> 13479Date: Sat Jan 16 22:47:03 2010 +0000 13480 13481 man: Fix typo in Makefile 13482 13483 Missing line continuation was preventing the XCompose alias from being 13484 generated. 13485 13486 Signed-off-by: Julien Cristau <jcristau@debian.org> 13487 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13488 13489commit 68530599e61cf9fef3fcc91650e7cef21837fce7 13490Author: Alan Coopersmith <alan.coopersmith@sun.com> 13491Date: Sat Jan 16 11:41:32 2010 -0800 13492 13493 libX11 1.3.3 13494 13495 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13496 13497commit 1e5af224b9fbdb154620f898558704f2bc4bf4e9 13498Author: Alan Coopersmith <alan.coopersmith@sun.com> 13499Date: Thu Jan 14 18:33:39 2010 -0800 13500 13501 Update COPYING file with actual licenses from libX11 code & docs 13502 13503 COPYING was previously a years out-of-date copy of the LICENSES 13504 doc summarizing all licenses for all XFree86 components, many of 13505 which did not apply to libX11. 13506 13507 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13508 13509commit ddb1786720d6136b1b28be599c75c10ad1b76a2a 13510Author: Alan Coopersmith <alan.coopersmith@sun.com> 13511Date: Thu Jan 14 17:38:26 2010 -0800 13512 13513 Purge CVS/RCS id tags 13514 13515 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13516 13517commit c1db9ddcfa7091f173478dca45bf720badedfca4 13518Author: Alan Coopersmith <alan.coopersmith@sun.com> 13519Date: Thu Jan 14 15:39:14 2010 -0800 13520 13521 Update Sun license notices to current X.Org standard form 13522 13523 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13524 13525commit 1aaa7c0358c5c035b99625f5715cb722bc2d2a2f 13526Author: Alan Coopersmith <alan.coopersmith@sun.com> 13527Date: Thu Jan 14 08:28:22 2010 -0800 13528 13529 Add %S substitutions to Compose man page 13530 13531 Commit 9df349a7894725f9469b106af645f57f7f3f9af3 added support for %S 13532 in Compose file include statements - this documents it in the Compose 13533 file man page. 13534 13535 Also changes the existing substitution documentation to list format 13536 instead of a paragraph format to allow easier adding of %S. 13537 13538 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13539 13540commit 94b45db93058cffa25598fe27dd3e385ace9d305 13541Author: Richard Purdie <rpurdie@linux.intel.com> 13542Date: Thu Jan 14 08:04:28 2010 -0800 13543 13544 Fix configure.ac PKG_ macro calls 13545 13546 If the first call to PKG_CHECK_MODULES() may not be called during execution 13547 which is the case here since its inside a case statement, call 13548 PKG_PROG_PKG_CONFIG() to ensure things worked correctly. 13549 13550 Fixes a configure failure introduced by commit 13551 87529c039050ce3336ff9ce00f1b5a21d15690da when xcb is enabled. 13552 13553 http://bugs.freedesktop.org/show_bug.cgi?id=26041 13554 13555 Signed-off-by: Richard Purdie <rpurdie@linux.intel.com> 13556 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13557 13558commit df559d8bc2238a2c369152fcb06e5b5ab132a5dc 13559Author: Alan Coopersmith <alan.coopersmith@sun.com> 13560Date: Tue Jan 12 09:18:15 2010 -0800 13561 13562 Raise xorg-macros requirement to 1.4 13563 13564 Needed since the changes in 464390f16d7ed4aa5bf80f89863ba92273075ec2 13565 depend on the INSTALL file delivered in xorg-macros 1.4 13566 13567 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13568 13569commit 9df349a7894725f9469b106af645f57f7f3f9af3 13570Author: James Cloos <cloos@jhcloos.com> 13571Date: Wed Jan 13 11:04:38 2010 -0500 13572 13573 Allow inclusion of system level compose files. 13574 13575 With the release of XFree86 4.4 an inclusion system of compose files was 13576 added to Xlib to allow inclusion of the default compose file (with %L), 13577 any compose files from user’s home directory (with %H), or a compose 13578 file with a hard coded path¹. However, even today including system level 13579 compose files is not possible in a platform independent manner although 13580 the machinery for including compose files and overriding previously 13581 defined compositions is already in place. 13582 13583 With the ability to include system level compose files one could greatly 13584 reduce the need for compose file rule duplication and the work needed to 13585 propagate changes in one compose file to others. For example, currently 13586 the Finnish compose file fi_FI.UTF-8 weights over 5000 lines² but it 13587 is almost identical with en_US.UTF-8 except for perhaps half a dozen 13588 compositions. 13589 13590 This commit allows one to include system level compose files with the 13591 following kind of syntax: 13592 13593 include "%S/en_US.UTF-8/Compose" 13594 13595 1] http://www.xfree86.org/4.4.0/RELNOTES5.html#42 13596 2] http://cgit.freedesktop.org/xorg/lib/libX11/tree/nls/fi_FI.UTF-8/Compose.pre 13597 13598 Signed-off-by: Marko Myllynen <myllynen@redhat.com> 13599 Signed-off-by: James Cloos <cloos@jhcloos.com> 13600 13601commit ccf21a8877f40136e25f2f62d0668b0c76b3cfbd 13602Author: Thien-Thi Nguyen <ttn@gnuvola.org> 13603Date: Tue Dec 1 10:31:47 2009 +0100 13604 13605 libX11: Fix comment: Invert polarity (direction) of reformat description. 13606 13607 Signed-off-by: Thien-Thi Nguyen <ttn@gnuvola.org> 13608 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13609 13610commit 4d4676c0d6324d2ff9d00d15ab485fa5d848a369 13611Author: Osamu Sayama <osamu.sayama@sun.com> 13612Date: Tue Jan 5 17:26:40 2010 -0800 13613 13614 set_fontset_extents crash after get_rotate_fontname fix in 2bef065b70f7 13615 13616 - In set_fontset_extents, check font_data is not NULL before running 13617 loop that may increment it to a non-NULL invalid pointer. 13618 - Make sure get_rotate_fontname counts the final field 13619 13620 Fixes OpenSolaris bug 6913809: X application dumps core in ja_JP.UTF-8 locale 13621 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6913809> 13622 13623 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13624 Reviewed-by: Adam Jackson <ajax@redhat.com> 13625 13626commit 87529c039050ce3336ff9ce00f1b5a21d15690da 13627Author: Alan Coopersmith <alan.coopersmith@sun.com> 13628Date: Tue Jan 5 18:02:37 2010 -0800 13629 13630 Merge X11, XKBPROTO, & XPROTO pkg-config lists 13631 13632 Since the XPROTO_CFLAGS & XKBPROTO_CFLAGS are just merged into X11_CFLAGS 13633 in configure.ac anyway, might as well combine the lists passed to 13634 PKG_CHECK_MODULES to reduce duplication in the flags. 13635 13636 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13637 Reviewed-by: Rémi Cardona <remi@gentoo.org> 13638 Acked-by: Peter Hutterer <peter.hutterer@who-t.net> 13639 13640commit 2dd053667b44c55e2bf601aec7b75fc6fee1cf44 13641Author: Alan Coopersmith <alan.coopersmith@sun.com> 13642Date: Tue Jan 5 17:56:58 2010 -0800 13643 13644 Add XCompose man page shadow for Compose man page 13645 13646 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13647 13648commit 02543bba816e065c02cd36e0c9e21519a9ee9c96 13649Author: Alan Coopersmith <alan.coopersmith@sun.com> 13650Date: Tue Jan 5 17:47:44 2010 -0800 13651 13652 Add $(AM_V_GEN) to silence lint rules too 13653 13654 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13655 13656commit db7c6fdeeaef9475458498e4cf09d6b1329e9aa3 13657Author: Alan Coopersmith <alan.coopersmith@sun.com> 13658Date: Tue Jan 5 17:42:42 2010 -0800 13659 13660 Remove GCC_WARNINGS now that XORG_CWARNFLAGS sets them for us 13661 13662 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13663 13664commit 6b4e526e7ba105622f6b143e8e2639b44cb6cc19 13665Author: Gaetan Nadon <memsize@videotron.ca> 13666Date: Sun Nov 29 10:44:24 2009 -0500 13667 13668 configure.ac: restore correct order for XORG_DEFAULT_OPTIONS 13669 13670 It must be following AC_USE_SYSTEM_EXTENSION 13671 13672commit 3dbaa11d1c8705e1b309c6686f354766b651ad20 13673Author: Gaetan Nadon <memsize@videotron.ca> 13674Date: Sun Nov 29 10:01:44 2009 -0500 13675 13676 Add .gitignore in /specs for generated files 13677 13678 Being in /specs, it will apply for all 3 subdirs 13679 13680commit f45d39d37aab04742e44cd8c3c993aad7587e40c 13681Author: Gaetan Nadon <memsize@videotron.ca> 13682Date: Fri Nov 27 20:56:03 2009 -0500 13683 13684 Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES 13685 13686 Now that the INSTALL file is generated. 13687 Allows running make maintainer-clean. 13688 13689commit 464390f16d7ed4aa5bf80f89863ba92273075ec2 13690Author: Gaetan Nadon <memsize@videotron.ca> 13691Date: Wed Oct 28 14:09:10 2009 -0400 13692 13693 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 13694 13695 Add missing INSTALL file. Use standard GNU file on building tarball 13696 README may have been updated 13697 Remove AUTHORS file as it is empty and no content available yet. 13698 Remove NEWS file as it is empty and no content available yet. 13699 13700commit 6360e7f0ce6a75da2bed33ede7fea783a1fb80e1 13701Author: Gaetan Nadon <memsize@videotron.ca> 13702Date: Tue Oct 27 15:07:25 2009 -0400 13703 13704 Deploy the new XORG_DEFAULT_OPTIONS #24242 13705 13706 This macro aggregate a number of existing macros that sets commmon 13707 X.Org components configuration options. It shields the configuration file from 13708 future changes. 13709 13710commit d02f943c98fde2f14319bc57fd9ad77eb9a2a572 13711Author: Gaetan Nadon <memsize@videotron.ca> 13712Date: Mon Oct 26 22:08:42 2009 -0400 13713 13714 Makefile.am: ChangeLog not required: EXTRA_DIST or *CLEANFILES #24432 13715 13716 ChangeLog filename is known to Automake and requires no further 13717 coding in the makefile. 13718 13719commit f77c89c751a6a63c8ef11ecdddca2aed11ff6e29 13720Author: Gaetan Nadon <memsize@videotron.ca> 13721Date: Thu Oct 22 12:34:19 2009 -0400 13722 13723 .gitignore: use common defaults with custom section # 24239 13724 13725 Using common defaults will reduce errors and maintenance. 13726 Only the very small or inexistent custom section need periodic maintenance 13727 when the structure of the component changes. Do not edit defaults. 13728 13729commit 46c7b0e9d0d85cf5ccc0d40d19821fcc3335503b 13730Author: Alan Coopersmith <alan.coopersmith@sun.com> 13731Date: Thu Nov 12 23:05:40 2009 -0800 13732 13733 Add compose-check.pl to EXTRA_DIST 13734 13735 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13736 13737commit 54c64267cc8bc98641cc39a22cb7bd71673e89e0 13738Author: Julien Cristau <jcristau@debian.org> 13739Date: Thu Oct 29 17:17:24 2009 +0100 13740 13741 man: fix XCopyGC argument order 13742 13743 Ubuntu bug#408337 13744 13745commit aad10032651cdc2a53b359035954454a28d6db67 13746Author: Alan Coopersmith <alan.coopersmith@sun.com> 13747Date: Fri Oct 23 13:55:14 2009 -0700 13748 13749 libX11 1.3.2 13750 13751 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13752 13753commit 6303ada89cb228c0f9656036f798703afb42fc29 13754Author: Peter Hutterer <peter.hutterer@who-t.net> 13755Date: Wed Oct 21 12:42:07 2009 +1000 13756 13757 Add smiley faces to compose sequences. 13758 13759 I wonder how we could have lasted that long without them. 13760 13761 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 13762 Acked-by: Daniel Stone <daniel@fooishbar.org> 13763 Acked-By: James Cloos <cloos@jhcloos.com> 13764 13765commit 0e104ebd8628803c27e36b16922ad1edd891325a 13766Author: Alan Coopersmith <alan.coopersmith@sun.com> 13767Date: Thu Oct 22 23:12:30 2009 -0700 13768 13769 Add man page for Compose file format 13770 13771 Based on grammar description in modules/im/ximcp/imLcPrs.c and 13772 note on XFree86 changes formerly found in xorg-docs RELNOTES.sgml 13773 13774 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13775 13776commit 9c95f2af7c442b3a59b1a30cf804f1ef4e7fc5b5 13777Author: Jeremy Huddleston <jeremyhu@freedesktop.org> 13778Date: Tue Oct 20 12:46:03 2009 -0700 13779 13780 Add extra configuration and sanity checks for groff and ps2pdf 13781 13782 1) Add AC_ARG_VAR for GROFF and PS2PDF to inform users of these 13783 environment variables. 13784 2) Check that groff -ms works 13785 13786 Some distributions ship the ms macros as a separate package which may 13787 not be installed together with groff, so we need to make sure that groff 13788 works and the required macros are actually installed before attempting 13789 to build the specs. 13790 13791 Signed-off-by: Jeremy Huddleston <jeremyhu@freedesktop.org> 13792 Signed-off-by: Dan Nicholson <dbn.lists@gmail.com> 13793 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13794 13795commit d3f801fd2f9198eaad6797414dba652f9c006c6d 13796Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13797Date: Sun Oct 18 17:34:53 2009 -0500 13798 13799 Fix VPATH build of libX11 specs 13800 13801 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13802 13803commit ad15e1a89d30ccc11d80745897b83def1448e8c0 13804Author: Alan Coopersmith <alan.coopersmith@sun.com> 13805Date: Sat Oct 17 16:14:34 2009 -0700 13806 13807 libX11 1.3.1 13808 13809 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13810 13811commit 0cbf98c17a7484bb58b2464b98d63bb3b4ea2594 13812Author: Alan Coopersmith <alan.coopersmith@sun.com> 13813Date: Sat Oct 17 16:28:25 2009 -0700 13814 13815 Fix make distcheck 13816 13817 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13818 13819commit 082e62ad268ef16a3bebc5a3c9fa008dbdc483ed 13820Author: Alan Coopersmith <alan.coopersmith@sun.com> 13821Date: Fri Oct 16 19:25:35 2009 -0700 13822 13823 Use $(AM_V_GEN) to silence echo commands for generating shadow man pages 13824 13825 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13826 13827commit 4e66da0783b2e5e3b288aaecd3c89396ed425c20 13828Author: Alan Coopersmith <alan.coopersmith@sun.com> 13829Date: Wed Oct 14 16:18:24 2009 -0700 13830 13831 Move libX11 & XIM/locale specs from xorg-docs 13832 13833 If groff is found, and --disable-specs is not passed to configure, 13834 specs will be converted to text, html and ps (or pdf if ps2pdf is 13835 found) and installed to $(docdir) 13836 13837 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13838 13839commit 5d3d817a42ddcc8d0c6efd33efd1442fe14f5c6b 13840Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13841Date: Tue Oct 13 19:15:51 2009 -0500 13842 13843 Provide _Xsetlocale compat wrappers on Cygwin 13844 13845 Previous versions of Cygwin did not have proper locale support, so Cygwin/X 13846 defined X_LOCALE, using _Xsetlocale instead. Cygwin 1.7 has added locale 13847 support, but we can't remove the _Xsetlocale entry point without breaking 13848 ABI. 13849 13850 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13851 13852commit 2c8b3a877a713bb66a6316a7051b43c46af6e1a0 13853Author: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13854Date: Tue Oct 13 19:15:49 2009 -0500 13855 13856 dolt: add Cygwin to supported platforms 13857 13858 Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net> 13859 13860commit a2c8e3e34b15b57ff881a52101fc961a602f35e4 13861Author: Alan Coopersmith <alan.coopersmith@sun.com> 13862Date: Wed Oct 14 13:23:30 2009 -0700 13863 13864 Recognize XSUNBUFFERSIZE alias for XLIBBUFFERSIZE on Solaris 13865 13866 Also fix indenting of the XLIBBUFFERSIZE code to match surrounding code 13867 13868 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13869 13870commit 34ddfca7b7d58240b0fe74bec6b2e0652d15c051 13871Author: Jon TURNEY <jon.turney@dronecode.org.uk> 13872Date: Tue Feb 10 17:47:25 2009 +0000 13873 13874 Include sys/select.h for select() and struct timeval, if it exists 13875 13876 This is a cygwin build fix 13877 13878commit 383165916ddac91740d4c780174d4c0d07cdb994 13879Author: Xake <xake@rymdraket.net> 13880Date: Sun Sep 27 11:16:36 2009 +0200 13881 13882 Use AM_V_GEN instead of customized macros for AM_SILENT_RULES 13883 13884 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13885 13886commit 854269d55cdda8caf425515bfed2855e211a5ada 13887Author: Julien Cristau <jcristau@debian.org> 13888Date: Tue Oct 6 16:11:24 2009 +0200 13889 13890 configure: quote argument to m4_pattern_forbid 13891 13892 Without this, configure spits out 13893 ../configure: line 12364: ac_fn_c_check_member: command not found 13894 ../configure: line 12378: ac_fn_c_check_type: command not found 13895 13896 Also anchor the pattern to make it stricter. 13897 13898 Signed-off-by: Julien Cristau <jcristau@debian.org> 13899 13900commit 3bb020587ce74e0737ec7aceb20041f1e77d3b87 13901Author: Jeremy Huddleston <jeremyhu@freedesktop.org> 13902Date: Thu Oct 1 22:20:38 2009 -0700 13903 13904 Split CFLAGS into CPPFLAGS and CFLAGS 13905 13906 On some build systems, CPPFLAGS is set to "-I/some/prefix/include". If older 13907 X11 headers are in /some/prefix/include, they will be preferred over the 13908 shipped headers. This corrects that problem. 13909 13910commit d54caf1c9c55af8247621b7ba6afb20b23699839 13911Author: Peter Hutterer <peter.hutterer@who-t.net> 13912Date: Fri Oct 2 10:59:08 2009 +1000 13913 13914 libX11 1.3 13915 13916 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 13917 13918commit dd201bcf9e7f8863b7ef32273a5ef021678133c1 13919Author: Peter Hutterer <peter.hutterer@who-t.net> 13920Date: Fri Oct 2 12:10:29 2009 +1000 13921 13922 nls: remove duplicate Compose sequences from pt_BR.UTF-8 13923 13924 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 13925 13926commit a293ae9e83739067fc92db1a39c262857bc2283e 13927Author: Peter Hutterer <peter.hutterer@who-t.net> 13928Date: Fri Sep 25 11:19:41 2009 +1000 13929 13930 Add XF86TouchpadToggle to XKeysymDB 13931 13932 Lenovo laptops provide a key to enable or disable the touchpad and the 13933 trackstick. This key is usually located on Fn + F8. 13934 13935 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 13936 Acked-by: Adam Jackson <ajax@redhat.com> 13937 13938commit 69839f8903a24eab08f17a781b3797fb64dce9cf 13939Author: Alan Coopersmith <alan.coopersmith@sun.com> 13940Date: Sun Sep 27 10:34:16 2009 -0700 13941 13942 Bug 24173: libX11 from git fails to build with automake older then 1.11 13943 13944 AM_CONDITIONAL must come *before* the AC_OUTPUT that creates the 13945 Makefiles, instead of after. 13946 <http://bugs.freedesktop.org/show_bug.cgi?id=24173> 13947 13948 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13949 13950commit f5effd041f58ef07703cca2b4f396758811e1eec 13951Author: Alan Coopersmith <alan.coopersmith@sun.com> 13952Date: Wed Apr 15 10:59:23 2009 -0700 13953 13954 Resolve conflicting Compose sequences in iso8859-2, el_GR.UTF-8 & pt_BR.UTF-8 13955 13956 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13957 13958commit 3843778358d3a0cd6a2d07dba5dd061248053ac9 13959Author: Alan Coopersmith <alan.coopersmith@sun.com> 13960Date: Wed Apr 15 10:56:09 2009 -0700 13961 13962 Add perl script to check for duplicate or conflicting compose file entries 13963 13964 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13965 13966commit 19cc5e1fa17a285045662820a8b4de2a0f9a194d 13967Author: Alan Coopersmith <alan.coopersmith@sun.com> 13968Date: Fri Sep 18 17:10:04 2009 -0700 13969 13970 Use make rules instead of shell for loops to generate shadow man pages 13971 13972 Allows parallel make and simpler build logs/error reporting 13973 13974 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13975 13976commit 7dabcac973d0b6692a3cd62bd6d8e0467b24200b 13977Author: Alan Coopersmith <alan.coopersmith@sun.com> 13978Date: Fri Sep 18 16:58:53 2009 -0700 13979 13980 Add AM_SILENT_RULES support for cpp rules for man & nls files 13981 13982 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13983 13984commit bfa19cddd8546b4930a773f3bbd81137c362d82b 13985Author: Alan Coopersmith <alan.coopersmith@sun.com> 13986Date: Fri Sep 18 16:58:16 2009 -0700 13987 13988 Update to using xorg-macros 1.3 & XORG_DEFAULT_OPTIONS 13989 13990 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 13991 13992commit 51396066c8003a96a6399c9a4bed2a81e512b582 13993Author: Mikko Niskanen <mikko.niskanen@iki.fi> 13994Date: Fri Sep 4 10:11:15 2009 +1000 13995 13996 Fix wrong typedef on HP-UX (#18998) 13997 13998 shl_dt doesn't exist, the type is shl_t. 13999 14000 X.Org Bug 18998 <http://bugs.freedesktop.org/show_bug.cgi?id=18998> 14001 14002 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14003 14004commit 615220a312b9430a580fe6dcf51703c6ef244f66 14005Author: Paul Bender <pebender@gmail.com> 14006Date: Fri Sep 4 09:57:10 2009 +1000 14007 14008 Don't require xdmcp in configure.ac (#22583) 14009 14010 X.Org Bug 22583 <http://bugs.freedesktop.org/show_bug.cgi?id=22583> 14011 14012 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14013 14014commit 20f9ecd86ad2a9ac6966f7eee32234cb5ef77c29 14015Author: Peter Hutterer <peter.hutterer@who-t.net> 14016Date: Thu Sep 3 14:02:44 2009 +1000 14017 14018 man: fix parameters to XkbAllocGeomOverlay{Rows|Keys} (#23499) 14019 14020 X.Org Bug 23499 <http://bugs.freedesktop.org/show_bug.cgi?id=23499> 14021 14022 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14023 14024commit c2814a614dc4d9409bfa0f44c18bfd69ad7f7b85 14025Author: Peter Hutterer <peter.hutterer@who-t.net> 14026Date: Thu Sep 3 12:05:25 2009 +1000 14027 14028 man: XQueryTree may return BadWindow. (#23416) 14029 14030 X.Org Bug 23416 <http://bugs.freedesktop.org/show_bug.cgi?id=XXX> 14031 14032 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14033 14034commit dbe98d456ccc6eeca9fa3e241a3db0a4d83c5a65 14035Author: Alan Coopersmith <alan.coopersmith@sun.com> 14036Date: Fri Aug 28 23:07:58 2009 +0800 14037 14038 Fix version tag in .TH line of several XKB man pages 14039 14040 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14041 14042commit bf24400936c10af6f5aa0c75cfe2207ab9b680b4 14043Author: Alan Coopersmith <alan.coopersmith@sun.com> 14044Date: Fri Aug 28 23:04:38 2009 +0800 14045 14046 XkbSetDeviceButtonActions.man: remove non-existent actions argument 14047 14048 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14049 14050commit 53affa9335116f2d367f041e6502a411d4619e47 14051Author: Alan Coopersmith <alan.coopersmith@sun.com> 14052Date: Fri Aug 28 23:00:17 2009 +0800 14053 14054 XkbQueryExtension.man: Arguments should be pointers 14055 14056 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14057 14058commit 6233948885acc5873a5abddfff235afec555f3c2 14059Author: Alan Coopersmith <alan.coopersmith@sun.com> 14060Date: Fri Aug 28 22:53:03 2009 +0800 14061 14062 XkbSAActionSetCtrls.man: Fix typo in formatting macro 14063 14064 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14065 14066commit 28a9ca57cdec4aa9ca92322f963b01f0c2daf47a 14067Author: Alan Coopersmith <alan.coopersmith@sun.com> 14068Date: Fri Aug 28 22:49:31 2009 +0800 14069 14070 Convert Xkb API man pages to ANSI prototypes 14071 14072 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14073 14074commit fa2eecca85baab9616f7143cc1a853a594b4a90c 14075Author: James Cloos <cloos@jhcloos.com> 14076Date: Mon Aug 24 06:35:17 2009 -0400 14077 14078 Add some (Serbian) Cyrillic NFD sequences. 14079 14080 A number of characters in use in the various countries which use the 14081 Cyrillic script do not appear as pre-composed characters in The UCS 14082 or Unicode; they are only available as combining-character sequences. 14083 14084 This commit adds support for using (prefix) dead keys and Multi_key- 14085 initiated sequences to enter a number of these combining-character 14086 sequences. This ensures that users can enter these scripts even 14087 when using the current Cyrillic keymaps, which lack support for 14088 the combining characters. 14089 14090 Please see the discussions on the xkb mailing list. 14091 14092 Signed-off-by: James Cloos <cloos@jhcloos.com> 14093 14094commit d1bdc909f9246119696c8b0d9afb7bd8afb71b60 14095Author: Julien Cristau <jcristau@debian.org> 14096Date: Wed Aug 5 18:14:23 2009 +0200 14097 14098 man/xkb: delete spurious newline in .TH headers 14099 14100commit ee723b83b24682db833a2f0abd96cd319b8a62af 14101Author: Julien Cristau <jcristau@debian.org> 14102Date: Wed Aug 5 16:45:19 2009 +0200 14103 14104 man: use __libmansuffix__ instead of 3X11 for references to other pages 14105 14106commit 595e204feb82c798a92eea41fea03be6476ac181 14107Author: Julien Cristau <jcristau@debian.org> 14108Date: Wed Aug 5 16:43:36 2009 +0200 14109 14110 man/xkb: use __libmansuffix__ instead of hardcoding 3Xkb for manpage sections 14111 14112commit 9da7e230d5320e1556ad2084fcd06ee7994385ea 14113Author: Peter Hutterer <peter.hutterer@who-t.net> 14114Date: Wed Aug 5 14:15:02 2009 +1000 14115 14116 Bump to 1.2.99.901 (1.3 RC1) 14117 14118 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14119 14120commit 8f78c7b4e3570cd46c5a220982963c17fe2157b8 14121Author: Filippo Giunchedi <filippo@debian.org> 14122Date: Sat Jun 6 16:56:54 2009 +0200 14123 14124 nls: add {left,right}wards arrow to compose table 14125 14126 Debian bug#532117 <http://bugs.debian.org/532117> 14127 14128 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14129 14130commit 7949bfa00390241d994f32463e50d4bd78920568 14131Author: Julien Cristau <jcristau@debian.org> 14132Date: Fri Jul 31 13:33:52 2009 +0200 14133 14134 Update library version for new symbols 14135 14136 Commit 554f755e5545f63d3c8f299297927238da155773 added generic event 14137 cookie handling. Bump libX11 version number accordingly. 14138 14139 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14140 14141commit 640fec5f4feacd01a00eea3dcd4edb220907d3dc 14142Author: Julien Cristau <jcristau@debian.org> 14143Date: Sun Aug 2 17:18:31 2009 +0200 14144 14145 Add _XFUNCPROTOBEGIN/END to Xlib-xcb.h 14146 14147 X.Org bug#22252 <https://bugs.freedesktop.org/show_bug.cgi?id=22252> 14148 14149 Reported-by: Riku Salminen <rsalmin2@cc.hut.fi> 14150 Signed-off-by: Julien Cristau <jcristau@debian.org> 14151 14152commit bc06d49e9dac1836d6824769ddb2ac5ba9f14df7 14153Author: Peter Hutterer <peter.hutterer@who-t.net> 14154Date: Wed Jul 29 08:44:09 2009 +1000 14155 14156 Fix compiler warning 'unused variable qelt' 14157 14158 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14159 14160commit 03f4907e14f5755e72309f08742977b871e81e33 14161Author: Peter Hutterer <peter.hutterer@who-t.net> 14162Date: Wed Jul 29 08:34:57 2009 +1000 14163 14164 Add utlist.h to the Makefile.am 14165 14166 utlist.h contains the linked list macros, it was added with the recent 14167 addition of event cookies but utlist.h wasn't added to the Makefile.am. As a 14168 result, make dist failed. 14169 14170 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14171 14172commit 554f755e5545f63d3c8f299297927238da155773 14173Author: Peter Hutterer <peter.hutterer@who-t.net> 14174Date: Fri Jun 26 11:27:43 2009 +1000 14175 14176 Add generic event cookie handling to libX11. 14177 14178 Generic events require more bytes than Xlib provides in the standard XEvent. 14179 Memory allocated by the extension and stored as pointers inside the event is 14180 prone to leak by simple 'while (1) { XNextEvent(...); }' loops. 14181 14182 This patch adds cookie handling for generic events. Extensions may register 14183 a cookie handler in addition to the normal event vectors. If an extension 14184 has registered a cookie handler, _all_ generic events for this extensions 14185 must be handled through cookies. Otherwise, the default event handler is 14186 used. 14187 14188 The cookie handler must return an XGenericEventCookie with a pointer to the 14189 data.The rest of the event (type, serialNumber, etc.) are to be filled as 14190 normal. When a client retrieves such a cookie event, the data is stored in 14191 an internal queue (the 'cookiejar'). This data is freed on the next call to 14192 XNextEvent(). 14193 14194 New extension interfaces: 14195 XESetWireToEventCookie(display, extension_number, cookie_handler) 14196 14197 Where cookie_handler must set cookie->data. The data pointer is of arbitray 14198 size and type but must be a single memory block. This memory block 14199 represents the actual extension's event. 14200 14201 New client interfaces: 14202 XGetEventData(display, *cookie); 14203 XFreeEventData(display, *cookie); 14204 14205 If the client needs the actual event data, it must call XGetEventData() with 14206 the cookie. This returns the data pointer (and removes it from the cookie 14207 jar) and the client is then responsible for freeing the event with 14208 XFreeEventData(). It is safe to call either function with a non-cookie 14209 event. Events unclaimed or not handled by the XGetEventData() are cleaned up 14210 automatically. 14211 14212 Example client code: 14213 XEvent event; 14214 XGenericEventCookie *cookie = &ev; 14215 14216 XNextEvent(display, &event); 14217 if (XGetEventData(display, cookie)) { 14218 XIEvent *xievent = cookie->data; 14219 ... 14220 } else if (cookie->type == GenericEvent) { 14221 /* handle generic event */ 14222 } else { 14223 /* handle extension/core event */ 14224 } 14225 XFreeEventData(display, cookie); 14226 14227 Cookies are not multi-threading safe. Clients that use XGetEventData() must 14228 lock between XNextEvent and XGetEventData to avoid other threads freeing 14229 cookies. 14230 14231 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14232 14233commit d7675cb8fa7155e7aff1459636a117a97aa1bf28 14234Author: Peter Hutterer <peter.hutterer@who-t.net> 14235Date: Mon Jul 6 13:17:35 2009 +1000 14236 14237 Bump to 1.2.99.1 14238 14239 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14240 14241commit 75fe48e7a42a685d7098e8d7108b9b956c471563 14242Author: Peter Hutterer <peter.hutterer@who-t.net> 14243Date: Fri Jul 10 14:07:34 2009 +1000 14244 14245 Bump to 1.2.2 14246 14247 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14248 14249commit 5d0fe0e0e92759af5667c5dca2eacb1b6f2d66ea 14250Author: Peter Hutterer <peter.hutterer@who-t.net> 14251Date: Thu Jul 2 09:10:25 2009 +1000 14252 14253 XMaskEvent/XCheckMaskedEvents must not check for GenericEvents. 14254 14255 GenericEvent cannot be selected for in the core event masks and they must 14256 thus be treated like extension events. 14257 14258 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14259 14260commit c1bf65b89f4e361f6178a73dd5334c8f2bd95732 14261Author: Peter Hutterer <peter.hutterer@who-t.net> 14262Date: Thu Jul 2 09:06:05 2009 +1000 14263 14264 XWindowEvent/XCheckWindowEvent must not return GenericEvents. 14265 14266 GenericEvents have no fixed position for the window, so they must be treated 14267 like extension events. 14268 14269 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 14270 14271commit 38f9054554d63525d2dd51aafb5eb57821158ab9 14272Author: Alan Coopersmith <alan.coopersmith@sun.com> 14273Date: Mon Jun 15 19:00:43 2009 -0700 14274 14275 Drop ancient USG SysV #ifdefs 14276 14277 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14278 14279commit 7bfe1323f16a1a69cc474659f7ac0c2570b1cf42 14280Author: Adam Jackson <ajax@redhat.com> 14281Date: Fri Jun 12 12:44:01 2009 -0400 14282 14283 Remove X_NOT_STDC_ENV usage. (#6527) 14284 14285commit 6dd74d7fb414ca1e99bae5c13e333961f396eb36 14286Author: Julien Cristau <jcristau@debian.org> 14287Date: Fri May 29 11:18:11 2009 +0200 14288 14289 nls: remove more duplicated aliases 14290 14291 Signed-off-by: Julien Cristau <jcristau@debian.org> 14292 14293commit e4b0899f516da224010e68bd2d953d5293d94993 14294Author: parag <parag@rawhideTM.pnq.redhat.com> 14295Date: Thu May 28 11:29:35 2009 +0530 14296 14297 nls: Add pa_PK locale information and make pa_IN as default for pa. 14298 14299 X.Org bug#21954 <http://bugs.freedesktop.org/show_bug.cgi?id=21954> 14300 14301 [jcristau: removed the pa_PK.UTF-8 alias to itself] 14302 14303 Signed-off-by: parag <pnemade@redhat.com> 14304 Signed-off-by: Julien Cristau <jcristau@debian.org> 14305 14306commit f0ea1f6d51145592f8617854f9320ec5dbff3299 14307Author: Julien Cristau <jcristau@debian.org> 14308Date: Fri May 29 10:58:20 2009 +0200 14309 14310 nls: remove broken sd_IN.UTF-8 alias 14311 14312 Signed-off-by: Julien Cristau <jcristau@debian.org> 14313 14314commit e29e010dabdb17d6498f2ef1786f69b8830c18ca 14315Author: Julien Cristau <jcristau@debian.org> 14316Date: Fri May 29 10:57:43 2009 +0200 14317 14318 nls: remove duplicated en_US* aliases 14319 14320 Signed-off-by: Julien Cristau <jcristau@debian.org> 14321 14322commit a89a300d87852c84389ad97db66dcb8930cb45dd 14323Author: Caolan McNamara <caolanm@redhat.com> 14324Date: Thu May 21 18:41:05 2009 +0200 14325 14326 man: missing space in XAllocColor man page 14327 14328 X.Org bug#21854 <http://bugs.freedesktop.org/show_bug.cgi?id=21854> 14329 14330commit c1c001e36504fd304f76f69bf6af3643225c49ea 14331Author: James Cloos <cloos@jhcloos.com> 14332Date: Wed May 13 13:03:54 2009 -0400 14333 14334 [nls] Replace remaining UCS Combining Characters in Compose sequences. 14335 14336 The replaces the instances of keysyms which match <U03[0-6][0-9A-Fa-f]>, 14337 where the keysym is used as a dead_key, with an actual dead_key symbol. 14338 14339 The only remaining instances of UCS combining characters in the 14340 compose sequences are of U0338 COMBINING LONG SOLIDUS OVERLAY 14341 used as a suffix in Multi_key-initiated sequences to create 14342 mathematics characters such as ∉ U+2209 NOT AN ELEMENT OF 14343 from ∈ U+2208 ELEMENT OF. 14344 14345commit e2b0bad3d3b9e9ca781fc264eb7584afbe2a1a4f 14346Author: James Cloos <cloos@jhcloos.com> 14347Date: Wed May 13 09:58:59 2009 -0400 14348 14349 [nls] Remove extraneous instances of UCS Combining Characters in Compose sequences. 14350 14351 This removes those instances of keysyms which match <U03[0-6][0-9A-Fa-f]>, 14352 where the matching keysym is used as a dead_key, and for which alternative 14353 compose sequences exist. 14354 14355commit 79f47e6dff2f0a0b673bbfecc47528edca814baa 14356Author: James Cloos <cloos@jhcloos.com> 14357Date: Fri May 8 20:11:54 2009 -0400 14358 14359 [nls] Remove combining_ keysyms from the Compose files 14360 14361 Some of the UTF-8 Compose tables included combining_ keysyms in 14362 the compose sequences as though they were dead symbols. This 14363 is contrary to how combining characters are used in the UCS. 14364 Therefore, those lines have been removed from the Compose tables. 14365 14366 There were also some combining_ keysyms as targets. As those 14367 are not included in x11proto’s keysymdef.h, and as those do 14368 exist there as Uxxxx keysyms, they are replaced with the Uxxxx 14369 keysym names. 14370 14371 This addresses http://bugzilla.freedesktop.org/show_bug.cgi?id=5107 14372 and is based on attachment 25644 by samuel.thibault@ens-lyon.org. 14373 14374 Signed-off-by: James Cloos <cloos@jhcloos.com> 14375 14376commit 4a08a3dfbda497b2be46e3e5fe6b777815ea27f9 14377Author: parag <pnemade@redhat.com> 14378Date: Tue May 5 16:50:47 2009 +0530 14379 14380 libX11: Add new Indic language information to nls directory files. #21560 14381 14382 Signed-off-by: parag <pnemade@redhat.com> 14383 14384commit aaf81096eb44b4c2812108721ba02738391884da 14385Author: Alan Coopersmith <alan.coopersmith@sun.com> 14386Date: Sat May 2 01:38:14 2009 -0700 14387 14388 Correct return type in XkbGetKeyboard man page 14389 14390 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14391 14392commit 2bef065b70f70af520b5de8fb23529254d15f003 14393Author: Christoph Pfister <christophpfister@gmail.com> 14394Date: Mon Apr 27 22:32:57 2009 -0700 14395 14396 X.Org Bug #21117: crash in get_rotate_fontname (omGeneric.c) 14397 14398 http://bugs.freedesktop.org/show_bug.cgi?id=21117 14399 14400 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14401 14402commit 5c1cde105db10df24d3c532f032cbc59050c7313 14403Author: Alan Coopersmith <alan.coopersmith@sun.com> 14404Date: Fri Apr 17 22:14:47 2009 -0700 14405 14406 Use AC_USE_SYSTEM_EXTENSIONS instead of hand-rolled check for _GNU_SOURCE 14407 14408 Raises minimum autoconf version required to 2.60 14409 14410 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14411 14412commit 84b7a91ef84f345384e4b0e13907385ca3ca3255 14413Author: Julien Cristau <jcristau@debian.org> 14414Date: Tue Apr 14 15:59:57 2009 +0100 14415 14416 Fix fi_FI.UTF-8, again 14417 14418 Commit 97fc6babd4ccaf300e25708868aa2a738893dc30 "NLS: Add UTF-8 compose 14419 file for Finnish" made fi_FI.UTF-8 use a broken empty XLC_LOCALE file. 14420 This reverts it back to using the en_US.UTF-8 one. 14421 14422commit 128daff4422f973ea40dd1e31b2db230e643549e 14423Author: Theppitak Karoonboonyanan <thep@linux.thai.net> 14424Date: Thu Apr 9 12:01:07 2009 +0700 14425 14426 Thai XIM not retrieve MB surrounding on UTF-8 LC 14427 14428 On th_TH.UTF-8 locale, Thai XIM rejects all combining characters for GTK+ apps 14429 that use X Input Method. 14430 14431 This is because GTK+ imxim immodule passes surrounding text in locale encoding, 14432 which is UTF-8 for UTF-8 locales. But current Thai XIM in Xlib assumes the 14433 multi-byte StringConversionText response for the StringConversionCallback to 14434 always be TIS-620, by retrieving a single byte and using it as-is. 14435 14436 If the Thai XIM tries to convert the multi-byte text based on locale codeset 14437 before using it, it will work again. 14438 14439 X.Org But 12759 <http://bugs.freedesktop.org/show_bug.cgi?id=12759> 14440 14441 Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net> 14442 Signed-off-by: Julien Cristau <jcristau@debian.org> 14443 14444commit e09f0d227fbf95b6252759af9d426efd57686f9f 14445Author: Theppitak Karoonboonyanan <thep@linux.thai.net> 14446Date: Thu Apr 9 11:47:55 2009 +0700 14447 14448 Thai XIM not filters when NumLock or CapsLock is on 14449 14450 The Thai XIM component in libx11 activated on 'th*' locales normally filters 14451 input sequence according to orthographic rules. However, when NumLock/CapsLock 14452 is on, this stops working. All sequences are passed through. 14453 14454 This is caused by missing masks in _XimThaiFilter(), which normally screens out 14455 certain special keys from entering orthographic rules. Unfortunately, this 14456 included events with NumLock/CapsLock on. Negating the masks from the check 14457 allows the events to be tested by the rules. 14458 14459 X.Org Bug 12517 <http://bugs.freedesktop.org/show_bug.cgi?id=12517> 14460 14461 Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net> 14462 Signed-off-by: Julien Cristau <jcristau@debian.org> 14463 14464commit d108d3c706af3502820b5202564488ea19908b77 14465Author: Theppitak Karoonboonyanan <thep@linux.thai.net> 14466Date: Thu Apr 9 11:25:25 2009 +0700 14467 14468 CharSet-to-CompoundText Conversion Failed for Thai Locales 14469 14470 SCIM fails to commit Thai input characters on Thai locales, because it commits 14471 string in compound text form, which was converted via 14472 XwcTextListToTextProperty(). But the XLC_LOCALE for th_TH and th_TH.UTF-8 14473 declares cs1's ct_encoding as TIS620-0:GR, which was commented out in 14474 src/xlibi18n/lcCT.c default_ct_data, in favor of ISO8859-11 ESC sequence. 14475 So, declaring cs1 as ISO8859-11:GR instead makes it work. 14476 14477 Besides, for th_TH.UTF-8, adding cs2 class with ISO10646-1 encoding also adds 14478 support for UTF-8 input. 14479 14480 And, along discussion in the bug, a similar problem was found for fontset, too, 14481 by causing delays on X apps startups on systems without tis620-0 fonts. This 14482 is normally the case, as mkfontdir and mkfontscale generate iso8859-11 entries 14483 by default for Thai X fonts. So, Thai fontset charset is also patched. 14484 14485 X.Org Bug 16475 <http://bugs.freedesktop.org/show_bug.cgi?id=16475> 14486 14487 Signed-off-by: Theppitak Karoonboonyanan <thep@linux.thai.net> 14488 Signed-off-by: Julien Cristau <jcristau@debian.org> 14489 14490commit fd62d3318c846cd43d66a505946e94704d7d83dc 14491Author: Alan Coopersmith <alan.coopersmith@sun.com> 14492Date: Wed Apr 8 19:42:25 2009 -0700 14493 14494 Revert "Change masculine to ordmasculine in Compose file comments" 14495 14496 This reverts commit 892b401d5acc055803a20e349ede0d64490f2230. 14497 14498 As Julien Cristau correctly points out, I misread the Compose file grammar 14499 in modules/im/ximcp/imLcPrs.c, and those are keysyms, not comments, and 14500 the keysym is named XK_masculine in keysymdef.h. This change is thus a 14501 bug in the Solaris compose tables to be fixed, not an improvement to bring 14502 upstream. 14503 14504commit eac57c77afdf44f50692225b8b0345a7c927bc84 14505Author: Alan Coopersmith <alan.coopersmith@sun.com> 14506Date: Tue Apr 7 15:38:45 2009 -0700 14507 14508 Version bump: 1.2.1 14509 14510 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14511 14512commit bfbec08baec33c5024510b0bcbbee6e4a8473e79 14513Author: Yaakov Selkowitz (Cygwin Ports maintainer) <yselkowitz@users.sourceforge.net> 14514Date: Tue Apr 7 13:46:57 2009 -0700 14515 14516 Bug 20773: Xcursor dynamic loading on Cygwin 14517 14518 X.Org Bug #20773 <http://bugs.freedesktop.org/show_bug.cgi?id=20773> 14519 Patch #24096 <http://bugs.freedesktop.org/attachment.cgi?id=24096> 14520 14521 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14522 14523commit c8c41614911be4fa222fa22478677d263b41c751 14524Author: Alan Coopersmith <alan.coopersmith@sun.com> 14525Date: Mon Apr 6 16:52:46 2009 -0700 14526 14527 Fix a several sparse warnings: Using plain integer as NULL pointer 14528 14529 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14530 14531commit b336c3d0cc2aefc8926500cff5f76b5a3e803886 14532Author: Alan Coopersmith <alan.coopersmith@sun.com> 14533Date: Mon Apr 6 16:32:05 2009 -0700 14534 14535 Further ansify prototypes & reduce #ifdefs in locking.c 14536 14537 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14538 14539commit 892b401d5acc055803a20e349ede0d64490f2230 14540Author: Alan Coopersmith <alan.coopersmith@sun.com> 14541Date: Mon Apr 6 10:50:09 2009 -0700 14542 14543 Change masculine to ordmasculine in Compose file comments 14544 14545 Matches the ordfeminine name used for the matching character, 14546 and the ordmasculine name used in many font descriptions of the glyph. 14547 14548 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14549 14550commit e3198b55dc16ec57346cc28aa8d34165ce8cde96 14551Author: Julien Cristau <jcristau@debian.org> 14552Date: Fri Mar 13 13:52:33 2009 +0100 14553 14554 Add a ru_RU.UTF-8 locale 14555 14556 Based on patch by Eugene Konev <ejka@imfi.kspu.ru> for X.Org 6.9.0. 14557 14558 Debian bug#330144 <http://bugs.debian.org/330144> 14559 X.Org bug#15887 <http://bugs.freedesktop.org/show_bug.cgi?id=15887> 14560 14561commit d239de9452691d6f875e6e5ace3d499ec3bf14d9 14562Author: Alan Coopersmith <alan.coopersmith@sun.com> 14563Date: Wed Mar 25 17:59:09 2009 -0700 14564 14565 Delete some unused "#ifdef notdef" static functions 14566 14567 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14568 14569commit 4281892e31058ef3aecc96a5767824b34d88d415 14570Author: Alan Coopersmith <alan.coopersmith@sun.com> 14571Date: Wed Mar 25 17:52:48 2009 -0700 14572 14573 Remove _XP_PRINT_SERVER_ #ifdefs from Xrm.c 14574 14575 This copy of Xrm.c is never compiled into the Xprint server any more, so 14576 this old code-sharing #ifdef from the monolith tree isn't needed. 14577 14578 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14579 14580commit 27a2e16e8ea11c1604e28645fae4d6ba4371d513 14581Author: Alan Coopersmith <alan.coopersmith@sun.com> 14582Date: Tue Mar 17 18:38:58 2009 -0700 14583 14584 makekeys: combine malloc(strlen)+strcpy into strdup 14585 14586 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14587 14588commit 78a894231ae8ec4959007b6d9b8d2a15d6333a1e 14589Author: Alan Coopersmith <alan.coopersmith@sun.com> 14590Date: Tue Mar 17 15:42:19 2009 -0700 14591 14592 When makekeys fails to find a good hash, print error instead of divide-by-zero 14593 14594 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14595 14596commit c9f84af591d15fbc3fa890bcd955d94f1ff82a0b 14597Author: Alan Coopersmith <alan.coopersmith@sun.com> 14598Date: Tue Mar 17 14:59:16 2009 -0700 14599 14600 Remove ifdef checks for macII left over from ancient A/UX 3.0 support 14601 14602 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14603 14604commit 4b90da0ffb32b791d915ecae11996cc2f2cac9a7 14605Author: Alan Coopersmith <alan.coopersmith@sun.com> 14606Date: Mon Mar 16 18:37:49 2009 -0700 14607 14608 XErrorDB additions for DRI2 requests 14609 14610 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14611 14612commit 273f45af1441be9d2135e4cac8c46ceb33470236 14613Author: Alan Coopersmith <alan.coopersmith@sun.com> 14614Date: Mon Mar 16 18:35:44 2009 -0700 14615 14616 XErrorDB additions for XInput errors and new XInput 1.5 requests 14617 14618 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14619 14620commit 90ef36f6336068183bf9d13ca972202db35b1202 14621Author: Alan Coopersmith <alan.coopersmith@sun.com> 14622Date: Mon Mar 16 18:23:05 2009 -0700 14623 14624 XErrorDB additions for RANDR 1.3 14625 14626 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14627 14628commit 124a4efaccf329f1a44f6b824e70278bdd1cad83 14629Author: Lubos Lunak <l.lunak@suse.cz> 14630Date: Mon Mar 16 18:03:36 2009 -0700 14631 14632 XErrorDB updates for XTEST, RANDR, DAMAGE extensions 14633 14634 From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html 14635 14636 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14637 14638commit 80811846e37b805fddb37c71589fd5f6f6037b3f 14639Author: Lubos Lunak <l.lunak@suse.cz> 14640Date: Mon Mar 16 17:57:52 2009 -0700 14641 14642 XGetErrorText() fails for extension error codes equal to the error base 14643 14644 From http://lists.freedesktop.org/archives/xorg/2008-January/031937.html 14645 14646 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14647 14648commit da95ecbbdcacc483cd0b5fd7db1fb2e2543341bd 14649Author: Milos Komarcevic <miloskomarcevic@netscape.net> 14650Date: Mon Mar 16 17:43:26 2009 -0700 14651 14652 Bug 11456: Serbian locale updates (sr_RS and sr_ME) 14653 14654 X.Org Bug #11456 <http://bugs.freedesktop.org/show_bug.cgi?id=11456> 14655 Patch #23937 <http://bugs.freedesktop.org/attachment.cgi?id=23937> 14656 14657 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14658 14659commit 934a6c0519a9e32505beee55b004f36c2a710217 14660Author: Alan Coopersmith <alan.coopersmith@sun.com> 14661Date: Mon Mar 16 14:55:22 2009 -0700 14662 14663 Bug 10082: Compose entries for some standard mathematical operators 14664 14665 X.Org Bug #10082 <http://bugs.freedesktop.org/show_bug.cgi?id=10082> 14666 14667 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14668 14669commit abf4da1ed0f735ca7ce471dc13a0ec3677391486 14670Author: Alan Coopersmith <alan.coopersmith@sun.com> 14671Date: Mon Mar 16 14:27:46 2009 -0700 14672 14673 Bug 14651: We need to add new locale specification for Belarusian Latin locale 14674 14675 X.Org Bug #14651 <http://bugs.freedesktop.org/show_bug.cgi?id=14651> 14676 14677 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14678 14679commit 837703c8651e1321a50147a8311c56e4758ce08a 14680Author: Caolan McNamara <caolanm@redhat.com> 14681Date: Mon Mar 16 14:15:50 2009 -0700 14682 14683 Bug 20575: man page for XCreatePixmapFromBitmapData doesn't match signature 14684 14685 X.Org Bug #20575 <http://bugs.freedesktop.org/show_bug.cgi?id=20575> 14686 Patch #23717 <http://bugs.freedesktop.org/attachment.cgi?id=23717> 14687 14688 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14689 14690commit 22199018879055d8653e59d8236bef57164fac66 14691Author: Alan Coopersmith <alan.coopersmith@sun.com> 14692Date: Mon Mar 16 13:28:18 2009 -0700 14693 14694 Correct locale alias for sh_BA.ISO8859-2@bosnia (should be sr, not nr) 14695 14696 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14697 14698commit 4c63c27eab2b88f5556dbf72c36321f50f6de35e 14699Author: Alan Coopersmith <alan.coopersmith@sun.com> 14700Date: Thu Mar 12 18:57:20 2009 -0700 14701 14702 Bug 9953: Please provide locale alias hu_HU.utf8 14703 14704 X.Org Bug #9953 <http://bugs.freedesktop.org/show_bug.cgi?id=9953> 14705 Debian Bug #407573 <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=407573> 14706 14707 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14708 14709commit 501f4e0ada1690783ada05ad412e4b191ad55336 14710Author: Alan Coopersmith <alan.coopersmith@sun.com> 14711Date: Thu Mar 12 17:38:21 2009 -0700 14712 14713 Bug 6820: Xlib shouldn't handle EAGAIN as a fatal IO error 14714 14715 X.Org Bug #6820 <http://bugs.freedesktop.org/show_bug.cgi?id=6820> 14716 Patch #17637 <http://bugs.freedesktop.org/attachment.cgi?id=17637> 14717 14718 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14719 14720commit 7b9c543e7210c7da204871c31a160e79d3a949b6 14721Author: Paul Bender <pebender@gmail.com> 14722Date: Thu Mar 12 17:11:42 2009 -0700 14723 14724 Bug 15664: xau & xdmcp not needed in x11.pc dependencies when built with xcb 14725 14726 X.Org bug #15664 <https://bugs.freedesktop.org/show_bug.cgi?id=15664> 14727 Patch #16128 <https://bugs.freedesktop.org/attachment.cgi?id=16128> 14728 14729 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14730 14731commit fd2cf1ef66c2aff3dc758956c9e9e567b9892c06 14732Author: Xue Wei <Wei.Xue@Sun.COM> 14733Date: Wed Mar 4 19:32:29 2009 -0800 14734 14735 Add UTF-8 locale entries for es_US, kk_KZ, mt_MT, and sh_BA 14736 14737 Sun bug 6809309 Add new utf8 locales supported by Xlib 14738 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6809309> 14739 14740 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14741 14742commit cb70c9bc43267577859a3674ca9de9be396ba69e 14743Author: Alan Coopersmith <alan.coopersmith@sun.com> 14744Date: Mon Feb 23 19:29:15 2009 -0800 14745 14746 Add --with-locale-lib-dir configure option to set locale lib install dir 14747 14748 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14749 14750commit ccd3584f0330db8dac90b9313c33ab8b5b2ec6af 14751Author: Alan Coopersmith <alan.coopersmith@sun.com> 14752Date: Mon Feb 23 18:33:51 2009 -0800 14753 14754 Incorporate more locale names/aliases from Solaris libX11 14755 14756 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14757 14758commit 83ce4daefdf544f801c7d666c89162690a36ce41 14759Author: Alan Coopersmith <alan.coopersmith@sun.com> 14760Date: Mon Feb 23 18:32:34 2009 -0800 14761 14762 Incorporate char range comments from Solaris version of ksc5601.h 14763 14764 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14765 14766commit ee279c84e34f1ebb8a6ff17d54ee146d11e29764 14767Author: Chris Ball <cjb@laptop.org> 14768Date: Sat Feb 21 14:48:42 2009 -0500 14769 14770 Fix fi_FI locale install directory. 14771 14772 fi_FI was setting "x11thislocaledir" to en_US, with the result that its 14773 locale data was written in that locale dir. 14774 14775 Signed-off-by: Chris Ball <cjb@laptop.org> 14776 14777commit da6bbca07c796c69172a649405474f03bee66754 14778Author: Emilio Jesús Gallego Arias <egallego@babel.ls.fi.upm.es> 14779Date: Sat Feb 21 20:17:23 2009 +0100 14780 14781 xcb_io: Avoid datatype overflow on AMD64 and friends. 14782 14783commit 4ef6491afa69e8441caee7bbebc583e6e796275e 14784Author: Chris Ball <cjb@laptop.org> 14785Date: Sat Feb 21 12:51:03 2009 -0500 14786 14787 Build fix for fi_FI. 14788 14789 Commit 642c4e928e770e0.. instructs make to enter nls/fi_FI, but no 14790 Makefile is written there by configure. 14791 14792 Signed-off-by: Chris Ball <cjb@laptop.org> 14793 14794commit 642c4e928e770e012379539a6ce09e11c02f09a6 14795Author: Julien Cristau <jcristau@debian.org> 14796Date: Sat Feb 21 03:12:05 2009 +0100 14797 14798 nls: actually use the fi_FI.UTF-8 files 14799 14800 The subdir wasn't added to nls/Makefile.am 14801 14802commit 9bad8309ef289bb943651abf6967b24fa2252aac 14803Author: Alan Coopersmith <alan.coopersmith@sun.com> 14804Date: Fri Feb 20 14:45:54 2009 -0800 14805 14806 flags member of Display structure needs to be marked volatile 14807 14808 Since the Xlib multithreaded code checks the flags variable in _XFlushInt 14809 to see if the other threads are done yet, it has to be marked volatile so 14810 the compiler doesn't optimize out re-loading it on each trip through the 14811 while loop and end up in an impossible-to-exit infinite loop of CPU chewing. 14812 14813 Part of fix for Sun bug 6409332: infinite loop in XFlushInt() on x86/32-bit 14814 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6409332> 14815 14816 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14817 14818commit ef5c446395aa30d7b1096a112e241a81c5b358e7 14819Author: Xue Wei <Wei.Xue@Sun.COM> 14820Date: Fri Feb 20 15:12:35 2009 -0800 14821 14822 Add nn_NO.UTF-8 to compose.dir.pre for Norwegian Nynorsk 14823 14824 Sun bug 6691236: Swing applications dump core when locale is nn_NO.UTF-8 14825 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691236> 14826 14827 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14828 14829commit 3822f2654e9630167f0c6cae317b472c09771672 14830Author: Xue Wei <Wei.Xue@Sun.COM> 14831Date: Fri Feb 20 15:03:51 2009 -0800 14832 14833 Add locale aliases for no_NO & sh_BA locale variants 14834 14835 Fixes Sun bug id 6691219: xterm refuses to start in some locales 14836 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6691219> 14837 14838 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14839 14840commit 63b6b5b5f522c0a606b32163c643edb64ca91d54 14841Author: Xue Wei <Wei.Xue@Sun.COM> 14842Date: Fri Feb 20 14:23:11 2009 -0800 14843 14844 Add kk_KZ.UTF-8 to locale.dir.pre for Kazakhstan 14845 14846 Fixes Sun bug id 6737254 ("kk_KZ.UTF-8 locale: In Java applications 14847 changing keyboard layout with gimlet does not work") 14848 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6737254> 14849 14850 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14851 14852commit d497301707962f3b94542e999a36abffcfbd303d 14853Author: Julien Cristau <jcristau@debian.org> 14854Date: Tue Feb 17 16:09:41 2009 +0100 14855 14856 Bump to 1.2 14857 14858commit 990e71361d1d7b79bf07b1dc93e4e905d4f1bdaf 14859Author: Julien Cristau <jcristau@debian.org> 14860Date: Tue Feb 17 15:23:40 2009 +0100 14861 14862 Check Xmalloc return value in _XConnectXCB 14863 14864 X.Org bug#19137 <http://bugs.freedesktop.org/show_bug.cgi?id=19137> 14865 14866 Signed-off-by: Julien Cristau <jcristau@debian.org> 14867 14868commit b4b5893f69419ff577bbaa4d18f78e4ffd729a0c 14869Author: James Cloos <cloos@jhcloos.com> 14870Date: Sat Feb 14 12:35:56 2009 -0500 14871 14872 dolt: allow older versions of bash to compile the library 14873 14874 Cf xserver commit 7be6520d and bugzilla #19031. 14875 14876commit 20982d6866e24453642b0b592fa0f13a88aa747c 14877Author: Will Thompson <will@willthompson.co.uk> 14878Date: Thu Feb 5 02:53:06 2009 +1100 14879 14880 NLS: Compose: Non-aliasing CCCP 14881 14882 Oops, cccp aliased cc for question mark. Upper-case it to avoid fail. 14883 14884 Signed-off-by: Will Thompson <will@willthompson.co.uk> 14885 Signed-off-by: Daniel Stone <daniel@fooishbar.org> (sorry) 14886 14887commit f052665394f3f0319e93a98f1d5d4ea287e1dd07 14888Author: Will Thompson <will@willthompson.co.uk> 14889Date: Wed Feb 4 14:51:11 2009 +0000 14890 14891 Add two essential compose sequences 14892 14893 Signed-off-by: Will Thompson <will@willthompson.co.uk> 14894 Signed-off-by: Daniel Stone <daniel@fooishbar.org> 14895 14896commit d7bea6fa909bf34c43efe0ca8239ab0f9f3a415f 14897Author: Alan Coopersmith <alan.coopersmith@sun.com> 14898Date: Mon Feb 2 20:34:31 2009 -0800 14899 14900 Add README with pointers to mailing list, bugzilla & git repos 14901 14902 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 14903 14904commit f682c27e93512773122887d2cbabb1657af45d2e 14905Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14906Date: Mon Feb 2 16:36:39 2009 -0200 14907 14908 Check if a function argument is NULL. 14909 14910 This was an addition to patch (also by me) 14911 https://bugs.freedesktop.org/attachment.cgi?id=14660 14912 that was not added when rediscovering/correcting the problem. 14913 14914commit 427e9d45d424b84efd9fc499aebf8d72392844c5 14915Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14916Date: Thu Feb 28 15:58:12 2008 -0300 14917 14918 Allow multiple inclusions of cursorfont.h, cosmetic patch. 14919 14920commit b91524a53e691f6a5d278fd8972b48a14ebeedeb 14921Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14922Date: Thu Feb 28 15:54:43 2008 -0300 14923 14924 Don't add prototypes for functions that don't exist. 14925 14926 Note that a full review was not done, only for functions that receive 14927 char/short arguments, or one of it's parameters is a function pointer 14928 that requires char/short arguments. 14929 14930commit 537eb52fe266ac439c4b383bb04a70017b709911 14931Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14932Date: Thu Feb 28 15:50:27 2008 -0300 14933 14934 WORD64 compile fix. This bug catched on a overview of the code. 14935 14936 The code is wrong since the first git revision, so it seens that it has 14937 not been compiled with WORD64 for quite some time, there is also another 14938 interesting code in xkb/XKBRdBuf.c: 14939 <hash>ifdef WORD64 14940 _XkbWriteCopyData32 Not Implemented Yet for sizeof(int)==8 14941 <hash>endif 14942 and possibly there are other similar problems. 14943 14944commit ffd0300fb74c6183208ae599133f2ded09e08d97 14945Author: Brian Rogers <brian@xyzw.org> 14946Date: Sat Jan 31 10:37:51 2009 -0800 14947 14948 Initialize event_notify after allocating the memory for it. 14949 14950 An uninitialized or otherwise invalid condition variable can apparently 14951 cause a hang in pthread_cond_broadcast. Ekiga, openoffice, and xine 14952 at least are freezing as a result of event_notify never being initialized. 14953 14954 Signed-off-by: Brian Rogers <brian@xyzw.org> 14955 Signed-off-by: Bart Massey <bart@cs.pdx.edu> 14956 14957commit 97fc6babd4ccaf300e25708868aa2a738893dc30 14958Author: James Cloos <cloos@jhcloos.com> 14959Date: Thu Jan 29 20:10:41 2009 -0500 14960 14961 NLS: Add UTF-8 compose file for Finnish 14962 14963 From bug report: 14964 14965 https://bugs.freedesktop.org/show_bug.cgi?id=18747 14966 14967commit 1bd2966ed88f83479a066c6ca7da23a515979550 14968Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14969Date: Thu Jan 29 20:25:15 2009 -0200 14970 14971 patches to avoid gcc warnings for libX11 (#4) 14972 14973 Author is Peter Breitenlohner <peb@mppmu.mpg.de> 14974 Bug #17946, attachment #19443 14975 14976 This patch avoids the gcc warning 14977 ../../../../libX11-1.1.5/modules/im/ximcp/imDefLkup.c:223: warning: passing arg 1 of `_XimProcSyncReply' from incompatible pointer type 14978 (same as already done at other places) 14979 14980 BTW: what is the difference between XIM (the type of ic->core.im) 14981 and Xim ? 14982 14983commit f16dd6af3eb17a25b8ee03d6617a7acc6e919fb0 14984Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 14985Date: Thu Jan 29 20:22:21 2009 -0200 14986 14987 patches to avoid gcc warnings for libX11 (#3) 14988 14989 Author is Peter Breitenlohner <peb@mppmu.mpg.de> 14990 Bug #17946, attachment #19441 14991 14992 This patch avoids the two gcc warnings 14993 ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:413: warning: assignment discards qualifiers from pointer target type 14994 ../../../../libX11-1.1.5/modules/im/ximcp/imRm.c:450: warning: assignment discards qualifiers from pointer target type 14995 14996 Note, that this as a rather crude fix of the problem (and it is really a 14997 shame to cast name_table to non-const). 14998 14999 The right solution would be to declare XIMValuesList.supported_values 15000 (in include/X11/Xlib.h) as 'const char **' (or '_Xconst char **'). 15001 This will, however, require extensive modifications in various places. 15002 15003commit cce75c5dce73fe1f8626ed9e6798138ada09a860 15004Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 15005Date: Thu Jan 29 20:20:18 2009 -0200 15006 15007 patches to avoid gcc warnings for libX11 (#2) 15008 15009 Author is Peter Breitenlohner <peb@mppmu.mpg.de> 15010 Bug #17946, attachment #19440 15011 15012 Avoid a preprocessor message 15013 <stdin>:194: warning: no newline at end of file 15014 15015 Two more such warnings (in XkbSAGroup.man and XkbSASetGroup.man) 15016 seem to be caused by a truncated (or otherwise incomplete) 15017 manpage. 15018 15019commit 692baebcc50f1e952800bfe4e2e6bc42f54e62fe 15020Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 15021Date: Thu Jan 29 20:12:24 2009 -0200 15022 15023 patches to avoid gcc warnings for libX11 (#1) 15024 15025 Author is Peter Breitenlohner <peb@mppmu.mpg.de> 15026 Bug #17946, attachment #19439 15027 15028 Define as 1 (one) as done by autoconf and the command line 15029 option, e.g. -DX11_t, not as empty. 15030 15031 This avoids the gcc (3.4.6) warnings: 15032 ../../libX11-1.1.5/src/x11_trans.c:27:1: warning: "X11_t" redefined 15033 <command line>:7:1: warning: this is the location of the previous definition 15034 ../../libX11-1.1.5/src/x11_trans.c:28:1: warning: "TRANS_CLIENT" redefined 15035 <command line>:8:1: warning: this is the location of the previous definition 15036 15037 Similarly, follow the autoconf convention to define XTHREADS 15038 and XUSE_MTSAFE_API as one. 15039 15040 This avoids analogous warnings when compiling libXcomposite, 15041 libXcursor, and libXdamage. 15042 15043 No reason to AC_SUBST XTHREADS and XUSE_MTSAFE_API (unused). 15044 15045commit a1977883c9f5ef0e515569d6e2ebccb07411f98c 15046Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 15047Date: Thu Jan 29 15:01:06 2009 -0200 15048 15049 Janitor: Correct some gcc/sparse warnings. 15050 15051 Most remaining warnings are about XIM/Xim to/from conversion 15052 and discarding const from pointers. 15053 15054commit 8ba0ca32a63c532f128bdca7f1bf982cab8e12be 15055Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 15056Date: Wed Jan 28 20:31:42 2009 -0200 15057 15058 Janitor: ansification, make distcheck, compiler warnings. 15059 15060 Only convert to use "ansi prototypes" the functions warned from 15061 compilation with "./autogen.sh --prefix=/usr", on a Linux computer. 15062 15063 Also, only address "trivial" compiler warning fixes in this commit. 15064 15065 The new .gitignore is the output of a command like: 15066 % find . -name .gitignore -exec cat {} \; | sort | uniq 15067 and only the toplevel .gitignore file was kept. 15068 15069commit 091c1624fd2f9d933329d6152e4ecd865aa7903a 15070Author: Peter Hutterer <peter.hutterer@who-t.net> 15071Date: Tue Jan 13 12:05:54 2009 +1000 15072 15073 Fix wrong implies symbol. 15074 15075 Quote Simos Xenitellis: 15076 > I checked the gtk+ commit logs that go back to 2000, and I see that still it 15077 > was 15078 > 15079 > { 0x08ce, 0x21d2 }, /* implies ⇒ RIGHTWARDS 15080 > DOUBLE ARROW */ 15081 > 15082 > In XFree86, it appears there was an error when they converted the 15083 > original table to 15084 > http://cvsweb.xfree86.org/cvsweb/xc/lib/X11/imKStoUCS.c?rev=1.1&content-type=text/vnd.viewcvs-markup 15085 > and the problem still shows up there (November 2000). 15086 15087 http://lists.freedesktop.org/archives/xorg/2008-October/039743.html 15088 15089 Reported by Erik Streb del Toro. 15090 15091commit e32521f19e0b07649b7e3a03d56a2bd556b138fb 15092Author: James Cloos <cloos@jhcloos.com> 15093Date: Sun Dec 7 04:13:34 2008 -0500 15094 15095 [i18n] s/U00DC/Udiaeresis/g 15096 15097 The xkeyboard-config keyboards generate the symbol Udiaeresis, not 15098 U00DC. Make sure the relevant Compose sequences expect the symbol 15099 which the keyboards actually send. 15100 15101commit b7502abfe22f8dc009b21cda1172af221d8f9f32 15102Author: James Cloos <cloos@jhcloos.com> 15103Date: Sun Dec 7 04:09:46 2008 -0500 15104 15105 Revert "For nls/*.pre, allow people to comment lines by starting them with '##'." 15106 15107 As of commit c9d20e3 the initial double-hashes are replaced with proper C comments. 15108 15109 This reverts commit a225a0be48770beb689d5ac5da97073634f7deab. 15110 15111commit c9d20e3f697c9cfae5511412023362c1db7449b1 15112Author: James Cloos <cloos@jhcloos.com> 15113Date: Sun Dec 7 04:08:23 2008 -0500 15114 15115 Use C comments rather than initial doubled hashes to exclude lines from .pre files 15116 15117commit c34ce54d9eac2d8052dc5f205a2ab09866ef5d25 15118Author: vehemens <vehemens@verizon.net> 15119Date: Sun Dec 7 01:18:26 2008 -0500 15120 15121 [i18n] Distribute new headers which were added for gb18030 support. 15122 15123 big5hkscs.h and gbk.h, added in 67e34d7a, need to be in SOURCES to 15124 make it into the tar. 15125 15126 Completes 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51. 15127 15128 Signed-off-by: James Cloos <cloos@jhcloos.com> 15129 15130commit 418819558d2c60e58b4e3022ce0fadf2143488ac 15131Author: Stefan Dirsch <sndirsch@suse.de> 15132Date: Sat Nov 22 22:01:07 2008 +0100 15133 15134 Fixed strange font mixups, when fontsets are still used (#2676, Novell #74299). 15135 15136commit 67e34d7a82ccd31f1208c0c43a6d58c3c05bf51a 15137Author: Stefan Dirsch <sndirsch@suse.de> 15138Date: Sat Nov 22 19:40:54 2008 +0100 15139 15140 Added remaining xlib patch required for gb18030 support (#1573). 15141 15142commit 55782a0a1fe1560f1a9c0ed78bc7f2575c15abcf 15143Author: Stefan Dirsch <sndirsch@suse.de> 15144Date: Sat Nov 22 17:53:06 2008 +0100 15145 15146 Added remaining hunk of Egbert's patch to prevent XIM deadlocks (#1182). 15147 15148commit c859446c500c883a67f7a86cab1a44844e24dade 15149Author: Ken Thomases <ken@codeweavers.com> 15150Date: Fri Nov 21 13:58:10 2008 -0500 15151 15152 [i18n] Provide translation from XK_partialderivative (8ef) to Unicode (U2202) 15153 15154 Signed-off-by: James Cloos <cloos@jhcloos.com> 15155 15156commit 5e68e94d852c730ef9264fc0d8ca61a2ffe98b53 15157Author: Jeremy Huddleston <jeremyhu@freedesktop.org> 15158Date: Mon Nov 17 20:47:26 2008 -0800 15159 15160 Force local transport when using the launchd socket. 15161 15162 Fixes a regression due to 15163 bf53987eaf0fbd7f7627783fc39e7ee99e8361ba 15164 15165commit 7aca689ce14d314b5c8c72c8df76f53f76ab467c 15166Author: Peter Hutterer <peter.hutterer@redhat.com> 15167Date: Fri Nov 14 10:32:50 2008 +1000 15168 15169 Add XF86Suspend, XF86Hibernate to KeysymDB. 15170 15171commit d16b11f25f8265e651def8d80bcd430c0448e664 15172Author: James Cloos <cloos@jhcloos.com> 15173Date: Tue Nov 11 19:43:39 2008 -0500 15174 15175 [nls] Annotate the Bépo compose sequences 15176 15177 Add comments with the UCS names. 15178 Add utf-8 strings for each result. 15179 Format for easy reading. 15180 15181commit 730298464240be6f65b32416b3f9b20062c61825 15182Author: James Cloos <cloos@jhcloos.com> 15183Date: Tue Nov 11 16:41:34 2008 -0500 15184 15185 [nls] Add some UTF-8 Compose sequences 15186 15187 As requested for the Bépo keyboard layout (http://clavier-dvorak.org/wiki/). 15188 15189 Cf. bugs: 15190 15191 https://bugs.freedesktop.org/show_bug.cgi?id=17821 15192 https://bugs.freedesktop.org/show_bug.cgi?id=17822 15193 15194commit ad6008a0c92733826983eb93f063d3d6276007d8 15195Author: Colin Harrison <colin.harrison@virgin.net> 15196Date: Thu Nov 6 17:48:21 2008 -0500 15197 15198 Fix copy/paste typo in imLcPrs 15199 15200 Signed-off-by: James Cloos <cloos@jhcloos.com> 15201 15202commit 0f0168ad18f8a280fc5a689eb02cfaa62d022ea6 15203Author: Adam Jackson <ajax@redhat.com> 15204Date: Thu Nov 6 14:54:13 2008 -0500 15205 15206 Fix leak in _XimXGetReadData 15207 15208 Spotted by Denis Dzyubenko 15209 15210commit ae23c25b9349ab1d7ff81f3075c000cf35fc442b 15211Author: Julien Cristau <jcristau@debian.org> 15212Date: Wed Nov 5 22:09:37 2008 +0100 15213 15214 Bump to 1.1.99.2 15215 15216commit d31e644c65c52828ea3e7abd94a8cf9aee12265c 15217Author: Julien Cristau <jcristau@debian.org> 15218Date: Wed Nov 5 21:33:13 2008 +0100 15219 15220 Fix distcheck 15221 15222commit e6a7b70cdb2ae8b713012839a0a0bbb93817b8ef 15223Author: Jamey Sharp <jamey@minilop.net> 15224Date: Wed Oct 29 14:00:33 2008 -0700 15225 15226 Support multiple independent internal sync handlers 15227 15228 Xlib has several independent tasks that need to be performed with the 15229 display unlocked. It does this by replacing the existing sync handler with 15230 one of a variety of internal sync handlers. However, if multiple internal 15231 sync handlers need to run, then the last one registering wins and 15232 previously registered internal sync handlers are never invoked. This 15233 manifested as a bug with DRI applications on Xlib/XCB as that requires 15234 both an XID handler after every XID allocation, and the periodic sequence 15235 number handler. The XID handler would win, and the sequence number handler 15236 would never be invoked. 15237 15238 Fix this by unifying the internal sync handler mechanism into a single 15239 function that calls all of the known internal sync handlers. They all need 15240 to deal with being called when not strictly necessary now. 15241 15242 Signed-off-by: Keith Packard <keithp@keithp.com> 15243 Signed-off-by: Jamey Sharp <jamey@minilop.net> 15244 Signed-off-by: Josh Triplett <josh@freedesktop.org> 15245 15246commit 2dbaaab9c4e3894b33dcae850551dee5473431d5 15247Author: Keith Packard <keithp@keithp.com> 15248Date: Sat Oct 11 21:44:21 2008 -0700 15249 15250 Ensure that _XReadEvents always leaves an event in the queue on return 15251 15252 XNextEvent assumes that the event queue will be non-empty on return from 15253 _XReadEvents, but with multiple event readers running, the previous change 15254 could leave the queue empty on return from process_responses. Re-invoke 15255 process_responses until the queue is non-empty. 15256 15257 Signed-off-by: Keith Packard <keithp@keithp.com> 15258 15259commit bedfe68259037c5564fe52758c92b9c97729640a 15260Author: Keith Packard <keithp@keithp.com> 15261Date: Sat Oct 11 21:10:23 2008 -0700 15262 15263 Permit only one Xlib thread to block waiting for events 15264 15265 As Xlib queues events internally, we must prevent multiple Xlib threads from 15266 entering XCB to wait for an event in case the queued event is to be 15267 delivered to the thread which didn't manage to read it. In other words, let 15268 only one Xlib thread into xcb_wait_for_event at a time. 15269 15270 Jamey Sharp looked over my shoulder while making this fix and, while hating 15271 my whitespace conventions, appears happy enough with the actual code. 15272 15273 Signed-off-by: Keith Packard <keithp@keithp.com> 15274 15275commit cc19618d2eb3ed92a0b574aee26a7da8b4aed5d2 15276Author: Jamey Sharp <jamey@minilop.net> 15277Date: Sun Mar 23 16:33:50 2008 -0700 15278 15279 Fix XAllocID race: hold the user display lock until we have a new XID. 15280 15281 Xlib built --without-xcb is also vulnerable to this race, and a similar 15282 fix might work there too. 15283 15284 Also, use an XID that's truly invalid while waiting for the next XID to be 15285 requested. 15286 15287commit 54e5c0941b0ded1628d559a9f0a3451ea96c299b 15288Author: Josh Triplett <josh@freedesktop.org> 15289Date: Sat Mar 15 17:22:23 2008 -0700 15290 15291 Use XCB's new socket handoff mechanism rather than the old XCB Xlib lock. 15292 15293 Previously, Xlib/XCB used XCB's Xlib lock to prevent XCB from sending 15294 requests between calls to Xlib's LockDisplay and UnlockDisplay macros. 15295 Xlib/XCB then sent all of its requests using XCB's xcb_send_request, and 15296 had to flush its requests when unlocking the display. 15297 15298 XCB 1.2 adds a new socket handoff mechanism, xcb_take_socket. Replace 15299 much of the existing Xlib/XCB implementation with the use of 15300 xcb_take_socket to take ownership of the write side of the X connection 15301 socket, and a return_socket callback which writes any outstanding requests 15302 with xcb_writev. This approach allows Xlib/XCB to use the same buffering 15303 as traditional Xlib did. In particular, programs which use Xlib/XCB and 15304 never make XCB calls will never need to hand the socket back to XCB, and 15305 vice versa. 15306 15307 This allows us to discard large quantities of synchronization code from 15308 Xlib/XCB, together with the synchronization bugs present in that code. 15309 Several test cases which previously failed now work perfectly, including 15310 multi-threaded ico. In addition, the infamous locking correctness 15311 assertions, triggered when double-locking or when unlocking without a 15312 previous lock, no longer exist, because Xlib/XCB no longer has any reason 15313 to care more about application locking than traditional Xlib does. 15314 15315 Furthermore, the handoff approach provides great improvements to 15316 performance. Results from x11perf's XNoOp test, which represented the 15317 worst case for the lock-based Xlib/XCB: 15318 15319 Traditional Xlib: average 19100000/sec 15320 Lock-based Xlib/XCB: average 3350000/sec 15321 Handoff-based Xlib/XCB: average 17400000/sec 15322 15323 Thus, for no-ops, the handoff mechanism provides more than a 4x speedup to 15324 Xlib/XCB, bringing Xlib/XCB within 9% of traditional Xlib no-op 15325 performance. Of course, real-world workloads do not use no-op, so your 15326 mileage may vary. In particular, since no-ops represent the worst case, 15327 we expect real workloads to more closely match the performance of 15328 traditional Xlib. 15329 15330 While removing synchronization code, we changed _XReply to not drop any 15331 locks when calling xcb_wait_for_reply; previously, we had to carefully 15332 avoid a deadlock between the Display lock and the XCB Xlib lock. Holding 15333 the locks reduces implementation complexity and should not impact 15334 applications. 15335 15336 Commit by Jamey Sharp and Josh Triplett. 15337 XCB's handoff mechanism inspired by Keith Packard. 15338 15339commit 5a19ac473f7a8046b0421fbd5d53da160c22ed75 15340Author: Chris Ball <cjb@laptop.org> 15341Date: Mon Nov 3 22:57:29 2008 -0500 15342 15343 Remove configure check for xcb-xlib. 15344 15345 xcb-xlib has been intentionally removed from libxcb; stop checking for 15346 it at configure-time. 15347 15348commit 34b35dda0bb7f3cf0ad9ab95ad7953d35d24f71b 15349Author: Josh Triplett <josh@freedesktop.org> 15350Date: Wed Oct 29 14:37:44 2008 -0700 15351 15352 .gitignore: Add dolt files 15353 15354commit 1290cccf2d90083eba852f5f413f7e3dff48ccd2 15355Author: Peter Hutterer <peter.hutterer@redhat.com> 15356Date: Tue Oct 28 11:56:55 2008 +1030 15357 15358 man: fix formatting error in XkbGetIndicatorState man page. 15359 15360commit b1022fa6d7e97640049e93ffa108083fc8d71b05 15361Author: James Cloos <cloos@jhcloos.com> 15362Date: Sat Oct 25 09:13:08 2008 -0400 15363 15364 Increase size of working arrays in the makekeys utility program. 15365 15366 Makekeys is used to create an optimal hash of the keysyms defined 15367 in x11proto’s keysymdef.h. 15368 15369 The recent addition of new keysyms there has triggered a bug in 15370 makekeys where it tries to use a zero on the rhs of the % (mod) 15371 operator (resulting in a divide by zero error) whenever it fails 15372 to find a solution within its constraints. 15373 15374 Increasing the size of the arrays allows it to find a solution for 15375 the current set of keysyms. 15376 15377 Makekeys is only run durring the build process, so this has no impact 15378 on users of libX11, only on the amount of VM needed to build it. 15379 15380 It still needs a more complete fix, but this allows compiles to 15381 progress until that is completed. 15382 15383commit 3e9afd501e40d76040635bd9a3045bcaf5a03b60 15384Author: James Cloos <cloos@jhcloos.com> 15385Date: Sat Oct 11 01:03:14 2008 -0400 15386 15387 Dolt-ify 15388 15389 Add dolt to acinclude.m4 and call it it configure.ac to speed compiles. 15390 15391commit 39c0b266cac8cbc15bf501d7869186862f01d823 15392Author: Peter Hutterer <peter.hutterer@redhat.com> 15393Date: Wed Oct 15 14:30:20 2008 +1030 15394 15395 Add more keysyms for PS3 BD remotes, Ericsson Phones #16519 15396 15397 X.Org Bug 16519 <https://bugs.freedesktop.org/show_bug.cgi?id=16519> 15398 15399commit d23aad31338e7d869d878d5aa1b6b91d20287005 15400Author: Peter Hutterer <peter.hutterer@redhat.com> 15401Date: Mon Oct 13 09:41:59 2008 +1030 15402 15403 Add XF86Battery, XF86Bluetooth, XF86WLAN, XF86UWB to keysymdb. 15404 15405commit 214ea6f5fd6aeaa7303ea4a69f9aedabf219ec4c 15406Author: Peter Hutterer <peter.hutterer@who-t.net> 15407Date: Thu Jul 24 15:44:26 2008 +0930 15408 15409 xkb: fix out-by-1 error in _XkbWriteKeyExplicit. 15410 15411 Thanks to Michael Meeks, Novell Bug 369263. 15412 https://bugzilla.novell.com/show_bug.cgi?id=369263 15413 15414commit e7ece39afc8e0adc3b6b1e70b337b98376754462 15415Author: Alan Coopersmith <alan.coopersmith@sun.com> 15416Date: Tue Oct 7 15:41:38 2008 -0700 15417 15418 Sun bug #6739431: double free in _X11TransConnectDisplay() 15419 15420 Double free() introduced in bf53987eaf0fbd7f7627783fc39e7ee99e8361ba 15421 After copying original_hostname to phostname, set original_hostname 15422 to NULL, so we don't free the same pointer twice when we free both 15423 original_hostname and phostname. 15424 15425 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6739431> 15426 15427commit 0877bc916afbd1ff8f1833edc930b765ea783576 15428Author: Daniel Stone <daniel@fooishbar.org> 15429Date: Tue Sep 23 19:02:02 2008 +0300 15430 15431 configure.ac: Fix CC_FOR_BUILD logic error 15432 15433 Turns out we were accidentally smashing it so that you couldn't set it 15434 externally at all. Oops. 15435 15436commit 58bf3aa746908f01c37be7045699e43a4e270944 15437Author: John Tapsell <johnflux@gmail.com> 15438Date: Tue Sep 23 17:30:13 2008 +0300 15439 15440 Build: Use native compiler for makekeys 15441 15442 makekeys needs to be run during the build process, as opposed to on the 15443 target, so build it with either of gcc or cc to fix cross-compiling. 15444 This can be overridden by setting $CC_FOR_BUILD. 15445 15446commit 340422a5c7a413faef18666cada27cee14615250 15447Author: Adam Jackson <ajax@redhat.com> 15448Date: Wed Sep 17 12:54:34 2008 -0400 15449 15450 Fix the previous patch for the BadFont case. 15451 15452commit 2335eafe4b53c27f6f9ee1bab3e1f5842f896428 15453Author: Matthias Clasen <mclasen@redhat.com> 15454Date: Wed Sep 17 10:43:52 2008 -0400 15455 15456 Bug #17616: Fix an XCB leak when the client has a non-fatal error handler. 15457 15458commit db0b85db29699be6bf7e78dede655d59ba926dfc 15459Author: Rafael Ávila de Espíndola <rafael.espindola@gmail.com> 15460Date: Sun Sep 14 19:15:26 2008 -0400 15461 15462 Fix problem with <dead_acute> <c> in pt_BR.UTF-8 15463 15464 The <dead_acute> <C> and <dead_acute> <c> lines in the pt_BR UTF-8 15465 Compose file show "Ç" and "ç" (c with cedilla accent) (akin to the 15466 ISO 8859 pt_BR Compose file) as the string but specify the keysym 15467 and comment for Ć and ć (c with acute accent). 15468 15469 This commit normalizes those two lines to match the specified string. 15470 15471 Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=4671 15472 15473 Signed-off-by: James Cloos <cloos@jhcloos.com> 15474 15475commit b065c011baa69b69e3ea77c30d5e153c0d103e2d 15476Author: Michael Verret <michael.verret@gmail.com> 15477Date: Mon Sep 8 16:33:35 2008 -0400 15478 15479 Fix documentation typo 15480 15481 Signed-off-by: James Cloos <cloos@jhcloos.com> 15482 15483commit 4213ea95185377bdd1b51e82933f331fc0f52e5b 15484Author: James Cloos <cloos@jhcloos.com> 15485Date: Sat Sep 6 04:19:19 2008 -0400 15486 15487 Remove extraneous <angle brackets> from the Ethiopic Compose file. 15488 15489 The am_ET.UTF-8 Compose file submitted in: 15490 15491 https://bugs.freedesktop.org/show_bug.cgi?id=11307 15492 15493 for the OLCP project used incorrect syntax. (It has angle brackets around the 15494 Uxxxx symbols on the right hand side rather than only on the left hand side). 15495 15496 This bug is noted in OLPC’s ticket: 15497 15498 http://dev.laptop.org/ticket/7474 15499 http://dev.laptop.org/attachment/ticket/7474/olpc_7474_dead_vowels_libX11.patch 15500 15501commit 9df84b513dd2b6e65e6d528cfac6d4cc3ea46918 15502Author: James Cloos <cloos@jhcloos.com> 15503Date: Mon Sep 1 17:49:33 2008 -0400 15504 15505 Complete the set of vulgar fractions 15506 15507 Unicode 1.1 added thirds, fifths, sixths and eights; 15508 we might as well catch up. 15509 15510 (Unicode and ISO 10646 have 1/7 (U2150), 1/9 (U2151), 1/10 (U2152) 15511 and 0/3 (U2189) in their pipelines, but those four can be added 15512 here after they are published.) 15513 15514commit a788792e9de95f8db0639557859722a35087481d 15515Author: James Cloos <cloos@jhcloos.com> 15516Date: Wed Aug 20 15:28:07 2008 -0400 15517 15518 nls (en_US) Re-remove long compositions that override shorter 15519 15520 As reported in <https://bugs.freedesktop.org/show_bug.cgi?id=17228>: 15521 15522 Commit a6f4bbf7 15523 nls (en_US): remove long compositions that override shorter [...] 15524 removed some longer compose sequences because there are shorter 15525 ones which take preference over the longer. For example the 15526 sequences: 15527 15528 <Multi_key> <apostrophe> <comma> <c> : U1E09 # ḉ 15529 <Multi_key> <apostrophe> <comma> <C> : U1E08 # Ḉ 15530 15531 were removed becase there already was: 15532 15533 <Multi_key> <apostrophe> <comma> : U201A # ‚ 15534 15535 Then commit 4ba09125 15536 Work on making the en_US and pt_BR UTF-8 Compose as similar as 15537 possible added exactly the same key sequences again. Obviusly 15538 they won't work. 15539 15540commit 55248e5c84c3fd8c349a3bb4cb15a1ec86989d74 15541Author: James Cloos <cloos@jhcloos.com> 15542Date: Thu Jul 17 21:01:42 2008 -0400 15543 15544 Add more <Multi_key> <cedilla> Compose tuples 15545 15546 The last commit missed the el_GR UTF-8 Compose.pre as well as 15547 the various ISO 8859 locales which have compose sequences 15548 generating ‘WITH CEDILLA’ characters. 15549 15550 (Interestingly, some of the 8859 locales already supported 15551 <Multi_key> <cedilla> for some CEDILLA characters, but not 15552 for Ç or ç.) 15553 15554 This is further work on bug 10397. 15555 15556commit 4ba091255bb953d53078ba5619d6751052c739f7 15557Author: James Cloos <cloos@jhcloos.com> 15558Date: Thu Jul 17 17:16:50 2008 -0400 15559 15560 Work on making the en_US and pt_BR UTF-8 Compose as similar as possible. 15561 15562 The eventual goal here is to have a single primary UTF-8 Compose 15563 file which the locale-specific UTF-8 Compose.pre files can #include. 15564 15565commit 254522d3c24e0590732fc03cdd61ff4564819d94 15566Author: James Cloos <cloos@jhcloos.com> 15567Date: Thu Jul 17 17:13:36 2008 -0400 15568 15569 Add <Multi_key> <cedilla> Compose tuples 15570 15571 The en_US and pt_BR UTF-8 Compose tables had support for using <comma> 15572 with <Multi_key> to enter CEDILLA characters. Bug 10397 requests 15573 support for using <cedilla> instead of <comma> in said sequences. 15574 15575 This commit makes both styles work. 15576 15577commit 7dc907f6032e1d5cbe4da0e414bdf2c569c04b44 15578Author: James Cloos <cloos@jhcloos.com> 15579Date: Sat Jun 28 15:25:23 2008 -0400 15580 15581 Fix commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01 15582 15583 The new block was added twice to the en_US.UTF-8 Compose.pre; 15584 delete the duplicate. 15585 15586commit 596e081b7457dcd1c4ad555ac140e6999239bc0d 15587Author: Peter Hutterer <peter@cs.unisa.edu.au> 15588Date: Sat Jun 28 20:14:05 2008 +0930 15589 15590 Fix unbalanced parenthesis in XKBlib.h # 16551 15591 15592 X.Org Bug 16551 <http://bugs.freedesktop.org/show_bug.cgi?id=16551> 15593 15594commit f6af6dd2f76c12b56ec166bb771457b9f08fe246 15595Author: Adam Jackson <ajax@redhat.com> 15596Date: Tue Jun 24 13:16:53 2008 -0400 15597 15598 Bug #14898: Don't abuse the sprintf() implementation. 15599 15600 The thing you're printing into should not itself appear in the list of 15601 things to print from, that's bad juju. Just use strcat(). 15602 15603commit 21e464ec682ab23ba20ddf6bd72c6db214cfbe01 15604Author: Khaled Hosny <khaledhosny@eglug.org> 15605Date: Thu Jun 19 18:26:11 2008 -0400 15606 15607 NLS: Add Arabic Lam-Alef ligature compose sequences (bug #16426) 15608 15609 Add some Arabic digraphs to utf-8 locales with a Compose.pre 15610 15611 Signed-off-by: James Cloos <cloos@jhcloos.com> 15612 15613commit bf53987eaf0fbd7f7627783fc39e7ee99e8361ba 15614Author: Alan Coopersmith <alan.coopersmith@sun.com> 15615Date: Wed Jun 18 20:00:25 2008 -0700 15616 15617 Rework code to choose local connection types and fallback to others 15618 15619 Adds --with-local-transport-order configure flag if you don't like the 15620 default ordering (which is platform dependent) 15621 15622 Includes fixes for these Sun/Solaris bug ids: 15623 6678250 X Commands returning incorrect display value unix:0.0 not <system>:0.0 15624 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6678250> 15625 6716481 libX11 should prefer Unix domain sockets over named pipes on Solaris 15626 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6716481> 15627 15628commit cf49e537014c3cb5aaee07e57400933e0bb72b6b 15629Author: Alan Coopersmith <alan.coopersmith@sun.com> 15630Date: Tue Jun 17 14:41:17 2008 -0700 15631 15632 Strip whitespace from end of lines in source files 15633 15634commit f76fd81dfbbd5cfae75c87ce0511e88e08529cf3 15635Author: Jeff Smith <whydoubt@yahoo.com> 15636Date: Sun Jun 15 23:52:20 2008 -0500 15637 15638 Fix memory leak in XOpenDisplay 15639 15640 Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au> 15641 15642commit fca0b0ba3f72b7284601d4690bba99fc80a92614 15643Author: Jens Herden <jens@khmeros.info> 15644Date: Tue Jun 10 20:07:30 2008 +0300 15645 15646 NLS: Add Khmer compose sequences (bug #5706) 15647 15648 Add some Khmer digraphs to all locales with a Compose.pre. 15649 15650commit e54cffb649b1622c17457e470cfab8cc56d38c97 15651Merge: 19802ccd 721b574d 15652Author: Daniel Stone <daniel@fooishbar.org> 15653Date: Tue Jun 10 20:04:30 2008 +0300 15654 15655 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11 15656 15657commit 721b574d36f1884c3f1bf7bd933646e2ed6680b5 15658Author: Peter Hutterer <peter@cs.unisa.edu.au> 15659Date: Thu May 29 10:57:21 2008 +0930 15660 15661 Bump to 1.1.99.1 15662 15663commit 631d32d13247d1cf52c0833d438c5b38b01b17a4 15664Author: Peter Hutterer <peter@cs.unisa.edu.au> 15665Date: Wed May 28 17:31:59 2008 +0930 15666 15667 Require xproto 7.0.13 and libxcb 1.1.90 (for GenericEvents) 15668 15669commit e9195db7257e418f83707233baeeb84b225caf4f 15670Merge: c34f76f4 a7f85567 15671Author: Peter Hutterer <peter@cs.unisa.edu.au> 15672Date: Thu May 22 12:14:28 2008 +0930 15673 15674 Merge branch 'master' into xge 15675 15676commit 19802ccd3909145e3ba2f6c073271cb5f3701685 15677Author: Daniel Stone <daniel@fooishbar.org> 15678Date: Mon May 19 19:22:31 2008 +0300 15679 15680 gitignore: Update with loads more bits from server 15681 15682commit a7f85567a3e850fba0c44571453d2852ab1a09be 15683Author: Adam Jackson <ajax@redhat.com> 15684Date: Tue May 13 10:28:39 2008 -0400 15685 15686 Bug #15884: Remove useless sleep()'s from the connection code. 15687 15688 For network transports, there's enough delay in the network layer 15689 already without adding more. For local transports, just hurry up 15690 and fail if the server isn't there. 15691 15692commit c34f76f475bc632490122e67b5a82575d69d5569 15693Author: Peter Hutterer <peter@cs.unisa.edu.au> 15694Date: Mon May 12 21:46:24 2008 +0930 15695 15696 Pull down extra bytes when reading a GenericEvent (non-xcb). 15697 15698 I refuse to take any responsibily for this code. It works, I guess. 15699 But - all the flushing is done somewhere before that, so we might need to 15700 flush here. Under some circumstances anyway. Don't ask me, I'm an optical 15701 illusion. 15702 15703 Build with xcb as transport layer highly recommended. 15704 15705commit c9b2ff1e6a607463993afa4a8d085857d97cc2f3 15706Merge: 17d7dcbf 9129057b 15707Author: Peter Hutterer <peter@cs.unisa.edu.au> 15708Date: Mon May 12 17:58:37 2008 +0930 15709 15710 Merge branch 'master' into xge 15711 15712commit 9129057bdbff0ec9cd8bb780cf7f85f134a291eb 15713Author: Teemu Likonen <tlikonen@iki.fi> 15714Date: Wed May 7 21:44:22 2008 +0300 15715 15716 Change <dead_belowdot> to <dead_belowring> for U+1E00 and U+U1E01 15717 15718 Commit 6b6caeea830a977bdb54688cfb648d879821e752 added <dead_belowdot> 15719 <A> and <dead_belowdot> <a> compose sequences for letters U+1E00 and 15720 U+U1E01 (LATIN CAPITAL/SMALL LETTER A WITH RING BELOW). This caused 15721 duplicate compose sequences since these have already been defined. Also, 15722 using <dead_belowring> is more logical since the diacritic is indeed 15723 a "RING BELOW". 15724 15725commit 01a9cb58888d290cc3d319feec4ee4a0297a844c 15726Author: Daniel Stone <daniel@fooishbar.org> 15727Date: Wed May 7 20:04:44 2008 +0300 15728 15729 NLS: Make UTF-8 the default for Russian 15730 15731 No-one uses 8859-5 anymore, so make the default for Russian UTF-8; the 15732 only other possible answer would be KOI8-R. 15733 15734 Signed-off-by: Sergey V. Udaltsov <sergey.udaltsov@gmail.com> 15735 15736commit 407b81bfbbabf6feb565d6da22f9ef9a69016ab8 15737Author: Ross Burton <ross@burtonini.com> 15738Date: Tue Apr 29 13:38:10 2008 +0300 15739 15740 NLS: Add interrobang to UTF-8 compose tables (bug #15653) 15741 15742 It is what it says on the box. 15743 15744commit 0b6682303e9c61fefc3818acfda616b1e3691abf 15745Author: Theppitak Karoonboonyanan <thep@linux.thai.net> 15746Date: Mon Apr 28 11:51:25 2008 +0300 15747 15748 IM: Respect XMODIFIERS for Thai locale (bug #15719) 15749 15750 When looking at Thai input methods, make sure XMODIFIERS is checked 15751 before jumping straight into built-in Thai processing, so external XIM 15752 servers such as SCIM can be used with Thai. 15753 15754commit c13aded1b2f830ba5004abb0ec5518f9ea16087e 15755Author: Colin Harrison <colin.harrison-at-virgin.net> 15756Date: Sat Apr 26 18:56:05 2008 +0100 15757 15758 Fix missing error condition 15759 15760commit f5c5ffc175cb383c92ea0fa8c08cfb087c5f3083 15761Author: Colin Harrison <colin.harrison-at-virgin.net> 15762Date: Mon Apr 21 17:24:33 2008 +0100 15763 15764 Xlib warning fixes 15765 15766commit 6b6caeea830a977bdb54688cfb648d879821e752 15767Author: James Cloos <cloos@jhcloos.com> 15768Date: Fri Apr 18 02:50:55 2008 -0400 15769 15770 Add some dead_key sequences to en_US.UTF-8 Compose table 15771 15772 Make use of the new dead key symbols added to x11proto’s 15773 commit 44e24a27bca023cf7b799f191fe6d52e12efbe5f (which 15774 was in responce to bug #15446). 15775 15776commit 8f9b039580deaf658e464b7d6254064fcf183df6 15777Author: Alan Coopersmith <alan.coopersmith@sun.com> 15778Date: Mon Apr 14 19:09:42 2008 -0700 15779 15780 Update ac_define_dir macro in acinclude.m4 to 2008-04-12 version 15781 15782commit 9f5e96eb91ab55dd441c3e94b75caf48c588778f 15783Author: Alan Coopersmith <alan.coopersmith@sun.com> 15784Date: Mon Apr 14 18:21:14 2008 -0700 15785 15786 Fix mismatched brace indenting 15787 15788commit a19f9c65ee9e5e5d783feaa84998c36439b0288b 15789Author: Bart Massey <bart@cs.pdx.edu> 15790Date: Fri Apr 4 18:58:45 2008 -0700 15791 15792 added error check in Xcms color file parser; closes bug #15305 15793 15794commit 12e8d0d01dd72ce98e7683ddb1bde181b7ed246f 15795Author: Christian Weisgerber <naddy@mips.inka.de> 15796Date: Tue Mar 18 07:30:05 2008 +0100 15797 15798 ConnDis: properly cast 'addr' before accessing it as a byte array. 15799 15800 If you use XDM-AUTHORIZATION-1 authorization keys for remote X11 15801 clients over IPv6, the clients are liable to segfaults. 15802 15803commit 64325f38bab082a8e0e9ce779a8e582de5c8588e 15804Author: Josh Triplett <josh@freedesktop.org> 15805Date: Sat Mar 15 12:29:33 2008 -0700 15806 15807 Fix fd.o bug 15023: make Xlib sync correctly given many void requests 15808 15809 If given many requests without replies, Xlib may not sync until it flushes 15810 the output buffer. Thus, if Xlib can fit enough requests in the buffer to 15811 pass by the number of requests it would normally sync after (65536 - 15812 BUFSIZE/sizeof(xReq)), it will sync too late. The test case in bug 15023 15813 demonstrated this by issuing a request with a reply (ListExtensions) at 15814 just the right time to get confused with the GetInputFocus reply issued in 15815 response to the sync 65,536 requests later; the test case used an async 15816 handler to watch the replies, since otherwise it could not issue a request 15817 without waiting for the response. When the test case failed, Xlib's sync 15818 handler would eat the ListExtensions reply, and the test case's async 15819 handler would see the GetInputFocus reply. 15820 15821 Fix this by replacing SEQLIMIT with a function sync_hazard() that uses the 15822 buffer size to figure out when the sequence numbers could potentially wrap 15823 before the next flush. 15824 15825 With this commit, the test case consistently passed, and the async reply 15826 handler always saw the ListExtensions reply. 15827 15828 Commit by Jamey Sharp and Josh Triplett. 15829 15830commit a5395563bbee15fabe1e8fd7aa86f9f314d8d30e 15831Author: Colin Harrison <colin.harrison@virgin.net> 15832Date: Sat Mar 15 13:39:13 2008 -0400 15833 15834 Fix typo 15835 15836 Signed-off-by: James Cloos <cloos@jhcloos.com> 15837 15838commit f07585ca27a8487bc66dfe41486c823f0fdcea7d 15839Author: Daniel Stone <daniel@fooishbar.org> 15840Date: Sat Mar 15 17:32:57 2008 +0200 15841 15842 configure.ac: Don't search for legacy X11 headers 15843 15844 This can actually break cross-compiles, so don't do it anymore. 15845 15846commit bf69541238c7df6606340c0f389e5c47149b29c7 15847Author: Matthieu Herrb <matthieu.herrb@laas.fr> 15848Date: Sun Mar 9 09:08:07 2008 +0100 15849 15850 nuke RCS Ids 15851 15852commit 5e98aed13e529638df744e45893c471d5f2014fb 15853Author: Adam Jackson <ajax@redhat.com> 15854Date: Thu Mar 6 16:10:33 2008 -0500 15855 15856 libX11 1.1.4 15857 15858commit 8e085971dc661da9f80ff6b67747459c0fb15c08 15859Author: Alan Coopersmith <alan.coopersmith@sun.com> 15860Date: Thu Feb 28 20:17:41 2008 -0800 15861 15862 Man page typo fixes 15863 15864commit 1a1a42a3ca1dfaf42f1094936b71c140fc030fcb 15865Author: Søren Sandmann Pedersen <sandmann@redhat.com> 15866Date: Sun Feb 24 20:03:35 2008 -0500 15867 15868 XIM: Fix a hand when switching input context. 15869 15870 Red Hat bug #201284. 15871 15872commit e02e4ccafcaf3eb8993152dfcbfbee0240ea2db2 15873Author: Adam Jackson <ajax@redhat.com> 15874Date: Sun Feb 24 20:00:43 2008 -0500 15875 15876 Bug #14029: Don't LockDisplay() recursively. 15877 15878 See also Red Hat bugzilla #326461. 15879 15880commit e5892467ae3308c8651be76e06db322dcbc08522 15881Author: Alan Coopersmith <alan.coopersmith@sun.com> 15882Date: Fri Feb 15 17:27:53 2008 -0800 15883 15884 Add support for building lint library with --enable-lint-library 15885 15886commit e3eb83ec6a9bffa63cdffd94f077c12f85ad7240 15887Author: Alan Coopersmith <alan.coopersmith@sun.com> 15888Date: Mon Feb 11 20:11:43 2008 -0800 15889 15890 Spell out number in XkbGetKeyVirtualModMap man page to avoid cpp errors from # 15891 15892commit d5ceed7a73a6b61758ddb6ff4e194955fbd5c185 15893Author: Alan Coopersmith <alan.coopersmith@sun.com> 15894Date: Fri Feb 8 16:46:46 2008 -0800 15895 15896 Add WM_LOCALE_NAME to list of properties set in XSetWMProperties comment 15897 15898commit 416a812200f24d19149dcc497e5c51a0608120f6 15899Author: Alan Coopersmith <alan.coopersmith@sun.com> 15900Date: Fri Feb 8 15:31:31 2008 -0800 15901 15902 XErrorDB updates for Render 0.9 & XFixes 4.0 15903 15904commit 16a76091cd632e5a3708e235ff864b58f3e4613e 15905Author: Kim Woelders <kim@woelders.dk> 15906Date: Sat Dec 22 21:45:23 2007 +0100 15907 15908 Fix bs_BA entries in locale.dir.pre 15909 15910 X.Org bug#13786 <http://bugs.freedesktop.org/show_bug.cgi?id=13786> 15911 15912commit 32115c563b87d2f37e3f9de70fbd0f4d9e424aea 15913Merge: 8f0bd3f4 e8d4cefa 15914Author: James Cloos <cloos@jhcloos.com> 15915Date: Fri Dec 14 22:43:47 2007 -0500 15916 15917 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11 15918 15919commit 8f0bd3f445cbdcc67650f6e8e3baf2ed89cb3695 15920Author: James Cloos <cloos@jhcloos.com> 15921Date: Fri Dec 14 22:42:59 2007 -0500 15922 15923 add a comment to en_US.UTF-8/Compose 15924 15925commit 4d6c45e60ed13d3b0fea10413873d6a74f9d6a3b 15926Author: James Cloos <cloos@jhcloos.com> 15927Date: Fri Dec 14 22:40:13 2007 -0500 15928 15929 Add <dead_stroke> compose sequences 15930 15931 The added sequences match the existing <Multi_key> <slash> sequences. 15932 15933 This is related to bug #12765¹. 15934 15935 1] https://bugs.freedesktop.org/show_bug.cgi?id=12765 15936 15937commit e8d4cefa0837afa149a10e981528b368485a9e38 15938Author: Jeremy Huddleston <jeremy@yuffie.local> 15939Date: Mon Dec 10 23:00:44 2007 -0800 15940 15941 Added launchd support. 15942 15943commit 17d7dcbfced4a9417b33507bd3fd9b7dd8268242 15944Merge: 5dfefd38 13ac8046 15945Author: Peter Hutterer <peter@cs.unisa.edu.au> 15946Date: Thu Dec 6 14:08:29 2007 +1030 15947 15948 Merge branch 'master' into xge 15949 15950commit 5dfefd3829d3ba7e41d5db0ad28e9dfee92fadd5 15951Merge: a68a1cd7 eff33ae5 15952Author: Peter Hutterer <peter@cs.unisa.edu.au> 15953Date: Thu Dec 6 13:57:09 2007 +1030 15954 15955 Merge branch 'master' into xge 15956 15957 Conflicts: 15958 15959 src/xcb_io.c 15960 15961commit 13ac80469f6958cabac596834e203bd9cb6d4c94 15962Author: James Cloos <cloos@jhcloos.com> 15963Date: Wed Dec 5 20:14:03 2007 -0500 15964 15965 Update the currency symbols block of en_US.UTF-8/Compose.pre 15966 Add XCOMM lines a la the existing NEW SHEQEL SIGN entry for 15967 the KIP, TUGRIK, DRACHMA. GERMAN PENNY, PESO, GUARANI, 15968 AUSTRAL, HRYVNIA and CEDI SIGNs. 15969 15970commit b0a8f2ec4ba698841683f8ce389f9d72e6bce53e 15971Author: Anton Zinoviev <anton@lml.bas.bg> 15972Date: Wed Dec 5 19:56:03 2007 -0500 15973 15974 Additions to the Compose file for UTF-8 15975 From bug #5371¹ 15976 15977 Commit 5cf5bc76642bfece7cb5b76faf414bf445f14489 left out this change 15978 from those in attachment #4122². The post³ on xorg resulted in only 15979 a positive reply⁴ from Daniel, so this block is now also commited. 15980 15981 1] https://bugs.freedesktop.org/show_bug.cgi?id=5371 15982 2] https://bugs.freedesktop.org/attachment.cgi?id=4122 15983 3] http://article.gmane.org/gmane.comp.freedesktop.xorg/20628 15984 4] http://article.gmane.org/gmane.comp.freedesktop.xorg/23966 15985 15986 Signed-off-by: James Cloos <cloos@jhcloos.com> 15987 Acked-by: Daniel Stone <daniel@fooishbar.org> 15988 15989commit 438d02ebc08ee171cf1d3936f4c81050d428ab92 15990Author: James Cloos <cloos@jhcloos.com> 15991Date: Tue Dec 4 17:25:39 2007 -0500 15992 15993 Fix the <U\x+> keysyms in the en_US.UTF-8 Compose file 15994 15995 Based on src/KeysymStr.c and src/StrKeysym.c and comments in 15996 bugs #11930¹ and #5129² it is clear that <U100XXXXX> is invalid; 15997 those should be in the form U plus the hex of the UCS Code Point. 15998 15999 The 0x01000000 is ORed in by the code. 16000 16001 This update fixes all of those. 16002 16003 1] https://bugs.freedesktop.org/show_bug.cgi?id=11930 16004 2] https://bugs.freedesktop.org/show_bug.cgi?id=5129 16005 16006commit 02e04059c89e175f51647e3b031344f743286b34 16007Merge: 1254c57d b57129ef 16008Author: James Cloos <cloos@jhcloos.com> 16009Date: Tue Dec 4 06:55:04 2007 -0500 16010 16011 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11 16012 16013commit 1254c57dd3a8d6ea87041b2f63024f99094f290f 16014Author: James Cloos <cloos@jhcloos.com> 16015Date: Tue Dec 4 06:53:55 2007 -0500 16016 16017 Use the new dead_psili and dead_dasia keysyms added to proto/x11proto 7.0.11 16018 Inspired by bug 11930¹: 16019 16020 Commit 40ed4eef92e31fcf7ea0a436e1a00cdf49484c1b to x11proto added dead_psili 16021 and dead_dasia keysyms. Make use of them in the en_US.UTF-8 and el_GR.UTF-8 16022 Compose files. 16023 16024 This was done with a pair of perl scripts based on the one quoted in the 16025 log for commit c76d30253f1483ac8200ad5c032a818907e65030. 16026 16027 1] https://bugs.freedesktop.org/show_bug.cgi?id=11930 16028 16029commit b57129ef324c73ee91c2a796b800c4b45f4d4855 16030Author: Jeremy Huddleston <jeremy@yuffie.local> 16031Date: Mon Dec 3 20:04:19 2007 -0800 16032 16033 Use __APPLE__ instead of __DARWIN__ 16034 16035commit b9c032e1d5ed08510863dfb91b8bda588e6c8c9e 16036Author: Jeremy Huddleston <jeremy@yuffie.local> 16037Date: Wed Nov 28 16:43:49 2007 -0800 16038 16039 Define __DARWIN__ when host_os == darwin* as needed in SetLocale.c 16040 16041commit 4b91ed099554626f1ec17d5bdf7bd77ce1a70037 16042Author: Alan Coopersmith <alan.coopersmith@sun.com> 16043Date: Wed Nov 21 16:50:04 2007 -0800 16044 16045 X.Org Bug #4312: incorrect comment asterisk in XAnyEvent.3x man page 16046 16047 X.Org Bugzilla #4312 <http://bugs.freedesktop.org/show_bug.cgi?id=4312> 16048 16049 Protect /* sequences from cpp pre-processor removal without transforming 16050 to Unicode mathematical asterisk character 16051 16052commit 2af660c2fcd15c86c66459bfc074c190ea1462e6 16053Author: Jamey Sharp <jamey@minilop.net> 16054Date: Mon Oct 29 10:46:20 2007 -0700 16055 16056 Two threads can request sequence sync and XID fetch simultaneously. 16057 16058 So don't assert that they can't. 16059 16060 This makes the Xlib/XCB implementation of _XAllocID more closely 16061 resemble the traditional Xlib version. 16062 16063commit 6e5485e0a5e3ab738becad12193e760c5fee83a4 16064Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 16065Date: Sun Oct 28 04:44:00 2007 -0800 16066 16067 X.Org Bug #12983: Typos in ./man/XChangeKeyboardControl.man 16068 16069 <http://bugs.freedesktop.org/show_bug.cgi?id=12983> 16070 16071commit 24527c92fd1f433ea135e85ec876a94a529fe500 16072Author: Yann Droneaud <ydroneaud@mandriva.com> 16073Date: Wed Oct 24 19:26:07 2007 +0300 16074 16075 XIM: Properly initialise client event 16076 16077 Make sure all ClientMessage fields are initialised to 0 before we send it. 16078 16079commit e41477f37b0d9b74a056d22dbf0073a94eecb9d6 16080Author: Daniel Stone <daniel@fooishbar.org> 16081Date: Wed Oct 24 19:22:22 2007 +0300 16082 16083 Colours: Fix --disable-xcms 16084 16085 Disable large tracts of colour management code when passing 16086 --disable-xcms. 16087 16088commit 11ea09745efa8de7dc82fe30ebd2393f08390957 16089Author: Eric Anholt <eric@anholt.net> 16090Date: Fri Aug 31 17:30:33 2007 -0700 16091 16092 Bug #2081: Note the range limitation of XSetScreenSaver arguments. 16093 16094commit 31540f1438ec63faf37044f2fd654b335ddf80f0 16095Author: Kristian Høgsberg <krh@redhat.com> 16096Date: Wed Aug 29 19:50:57 2007 -0400 16097 16098 Add GLX 1.4 requests and errors. 16099 16100commit 7c996f78914c77fe17e9f4feede980d895d9df51 16101Author: Eric Anholt <eric@anholt.net> 16102Date: Tue Aug 28 15:15:11 2007 -0700 16103 16104 Add XF86 keyboard/monitor brightness keysyms to the keysymbdb. 16105 16106commit fa4effe82759f864a22a2dc6c920fa72ddb175a8 16107Author: Eric S. Raymond <esr@thyrsus.com> 16108Date: Sun Jan 14 10:48:00 2007 -0800 16109 16110 Bug #9658: Bad markup on XIfEvent.3x 16111 16112 X.Org Bugzilla #9658 <https://bugs.freedesktop.org/show_bug.cgi?id=9658> 16113 16114commit 74cba78daa738ef4d92096107d95c1a585933666 16115Author: Eric S. Raymond <esr@thyrsus.com> 16116Date: Sun Jan 14 10:43:00 2007 -0800 16117 16118 Bug #9655: Bad markup in XrmUniqueQuark.3x 16119 16120 X.Org Bugzilla #9655 <https://bugs.freedesktop.org/show_bug.cgi?id=9655> 16121 16122commit 4341d1a34b2a2e460b58131b6fd81935f3355bbc 16123Author: Eric S. Raymond <esr@thyrsus.com> 16124Date: Sun Jan 14 10:43:00 2007 -0800 16125 16126 Bug #9654: Bad markup in XrmGetFileDatabase.3x 16127 16128 X.Org Bugzilla #9654 <https://bugs.freedesktop.org/show_bug.cgi?id=9654> 16129 16130commit 2db713252090cae08f0200fecad4fc25fb64c8b1 16131Author: Eric S. Raymond <esr@thyrsus.com> 16132Date: Sun Jan 14 10:40:00 2007 -0800 16133 16134 Bug #9653: Bad markup in XQueryColor.3x 16135 16136 X.Org Bugzilla #9653 <https://bugs.freedesktop.org/show_bug.cgi?id=9653> 16137 16138commit 9d3ceea4b902e0471824c1e07ad64342b9a1114e 16139Author: Eric S. Raymond <esr@thyrsus.com> 16140Date: Sun Jan 14 10:40:00 2007 -0800 16141 16142 Bug #9652: Bad markup in XDrawArc.3x 16143 16144 X.Org Bugzilla #9652 <https://bugs.freedesktop.org/show_bug.cgi?id=9652> 16145 16146commit 2e7e0748d353d7f53bbd65ec6bf0df8758528ddc 16147Author: Eric S. Raymond <esr@thyrsus.com> 16148Date: Sun Jan 14 10:39:00 2007 -0800 16149 16150 Bug #9651: Bad markup in XcmsColor.3x 16151 16152 X.Org Bugzilla #9651 <https://bugs.freedesktop.org/show_bug.cgi?id=9651> 16153 16154commit c316aaf0aab06951db9dc5c9c1148bfea835d885 16155Author: Eric S. Raymond <esr@thyrsus.com> 16156Date: Sun Jan 14 10:38:00 2007 -0800 16157 16158 Bug #9650: Bad markup in XLoadFont.3x manual page 16159 16160 X.Org Bugzilla #9650 <https://bugs.freedesktop.org/show_bug.cgi?id=9650> 16161 16162commit ca5d9a625ea0965853fa9e74a448b8c29c78ec95 16163Author: Eric S. Raymond <esr@thyrsus.com> 16164Date: Tue Jan 2 10:18:00 2007 -0800 16165 16166 Bug 9523: Markup problems in XQueryExtension.3x 16167 16168 X.Org Bugzilla #9523 <https://bugs.freedesktop.org/show_bug.cgi?id=9523> 16169 16170commit fe713c616e29ba19c179b43c18eca1035079ce18 16171Author: Ian Romanick <idr@us.ibm.com> 16172Date: Tue Aug 21 14:56:33 2007 -0700 16173 16174 Make sure nls/am_ET.UTF-8/Makefile is created by configure. 16175 16176commit e3430616f26b68e1439143cbe10732f3fc329d20 16177Author: Eric S. Raymond <esr@thyrsus.com> 16178Date: Tue Jan 2 08:40:00 2007 -0800 16179 16180 Bug #9516: Markup error in XAllocWMHints.3x 16181 16182 X.Org Bugzilla #9516 <https://bugs.freedesktop.org/show_bug.cgi?id=9516> 16183 16184commit d8fe979fc929833e8c754aed32641786d5a0622b 16185Merge: 21ca9533 4ec1723f 16186Author: James Cloos <cloos@jhcloos.com> 16187Date: Mon Aug 20 15:34:50 2007 -0400 16188 16189 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/lib/libX11 16190 16191commit 21ca953337fb221b85345bf35ce1a98a0dcb2bf2 16192Author: James Cloos <cloos@jhcloos.com> 16193Date: Mon Aug 20 15:34:03 2007 -0400 16194 16195 Fix typo 16196 The code <U1000000D> was used where <U10000DC> was obviously intended. 16197 16198 It is possible that <Udiaeresis> should be used instead, if that will 16199 not break anyone’s setup. 16200 16201commit 4ec1723fff729440cd3349c1f95d87d2a6ba89cf 16202Author: James Cloos <cloos@jhcloos.com> 16203Date: Mon Aug 20 15:25:48 2007 -0400 16204 16205 Add compose file for Ethiopic to match new keyboard in xkeyboard-config 16206 From bug report: 16207 16208 https://bugs.freedesktop.org/show_bug.cgi?id=11307 16209 16210commit eff33ae525337ce2026be135a26464c7b1237113 16211Author: James Cloos <cloos@jhcloos.com> 16212Date: Sat Aug 18 17:58:23 2007 -0400 16213 16214 Patch for Catalan locales 16215 From bugzilla bug 10943¹: 16216 16217 There are several Catalan locale codes which presently can 16218 be used in X11 systems; especially after they were accepted 16219 in belocs-locale-data². 16220 16221 In the following patches, I³ add ca_AD, ca_FR and ca_IT Catalan 16222 locale codes. For instance, without this, using ca_AD (actually 16223 a quite used locale⁴) some applications (eg. Emacs or Skype) 16224 cannot display Catalan diacritic marks as you type them. 16225 16226 1] https://bugs.freedesktop.org/show_bug.cgi?id=10943 16227 2] http://lists.debian.org/debian-devel-changes/2005/07/msg01429.html 16228 3] Toni Hermoso Pulido <toniher@softcatala.org> 16229 4] https://launchpad.net/~ubuntu.cat/+members 16230 16231commit 1f980cb7d022f53d0aee9e793b08203fb888e86e 16232Author: James Cloos <cloos@jhcloos.com> 16233Date: Sat Aug 18 17:47:04 2007 -0400 16234 16235 Add additional Euro signs to compose 16236 Inspired by bug 7419¹ make all of: 16237 16238 C=, =C, c=, =c, E=, =E, e=, E= 16239 16240 after <Multi_key> generate € U+20AC EURO SIGN. 16241 16242 1] https://bugs.freedesktop.org/show_bug.cgi?id=7419 16243 16244commit 4b0a14521449dfce8b4347bd17243efd1d3eae2d 16245Author: James Cloos <cloos@jhcloos.com> 16246Date: Sat Aug 18 17:29:08 2007 -0400 16247 16248 Compose fix for Latin-1 (from Debian) 16249 The description from bugzilla bug 7417¹ is: 16250 16251 We've been shipping this patch for some time in Debian now. The 16252 problem description from the patch header is reproduced below. You 16253 may want to note the licensing issue mentioned below, but we've been 16254 shipping it because the method by which this particular patch was 16255 generated and updated was also given below. 16256 16257 This patch by Denis Barbier. 16258 16259 The X11 protocol states that Unicode keysyms are in the range 16260 0x01000100 - 0x0110FFFF. If the result of composing characters is a 16261 Unicode codepoint, X returns the corresponding Unicode keysym, which 16262 is its Unicode codepoint augmented by 0x01000000. Latin-1 16263 characters must not appear with their Unicode codepoints in compose 16264 files, otherwise the returned composed character lies in the range 16265 0x01000000 - 0x010000FF which is not valid. 16266 16267 There are two solutions: either fix composing routines to return 16268 0xZZ instead of 0x010000ZZ (where Z is an hexadecimal digit), or 16269 replace U00ZZ by their corresponding keysyms in compose files. The 16270 latter is more logical and less error prone, so compose files will 16271 be patched. Many applications accept these invalid Unicode keysyms, 16272 but few of them don't, most notably xemacs. Only UTF-8 locales are 16273 affected. 16274 16275 This has been fixed very recently in XFree86 CVS (but not xorg), but 16276 for licensing reasons, this patch is not grabbed. Instead automatic 16277 conversion is performed by: 16278 16279 sed -e '/XK_LATIN1/,/XK_LATIN1/!d' /usr/include/X11/keysymdef.h \ 16280 | grep -v deprecated | grep 0x0 \ 16281 | sed -e 's/0x0/U0/' -e 's/XK_//' \ 16282 | awk '{ printf "s/\\b%s\\b/%s/ig\n", $3, $2; }' > sedfile 16283 for f in nls/*.UTF-8/Compose.pre 16284 do 16285 sed -f sedfile $f > $f.tmp && mv $f.tmp $f 16286 done 16287 16288 [I edited the quoted script to update it for the current location of 16289 the installed keysymdef.h and the current layout of the libX11 16290 repo. -JimC] 16291 16292 I applied the script, not the patch attached to the bugreport. 16293 16294 1] https://bugs.freedesktop.org/show_bug.cgi?id=7417 16295 16296commit 5cf5bc76642bfece7cb5b76faf414bf445f14489 16297Author: James Cloos <cloos@jhcloos.com> 16298Date: Sat Aug 18 17:13:41 2007 -0400 16299 16300 Add some compose sequences 16301 16302 Add some compose sequences from the patch in bug 5371 (attachment 4122). 16303 16304 Cf: 16305 16306 https://bugs.freedesktop.org/show_bug.cgi?id=5371 16307 https://bugs.freedesktop.org/attachment.cgi?id=4122 16308 16309commit d4002e389dd69780dfc7c2f7bd3cb0c57f05d4f8 16310Author: James Cloos <cloos@jhcloos.com> 16311Date: Sat Aug 18 13:57:31 2007 -0400 16312 16313 Fix SMP Compose targets 16314 The compose targets from the SMP (plane 1) were incorrect. 16315 16316 At some point the 0x10000 bit had been lost. 16317 16318commit f1ed3da9a30a1f0264fdc7d1c6466f27fe2a3d7d 16319Author: Jeremy C. Reed <reed@glacier.reedmedia.net> 16320Date: Thu Aug 16 17:37:22 2007 -0500 16321 16322 Fix a mutex reference-counting bug. 16323 16324 I was told that some systems have a much more permissive libpthread. 16325 I was asked to commit this. This is from NetBSD's X source. 16326 (I didn't receive any feedback on xorg list for over two weeks 16327 about this.) 16328 16329commit ac00a44b4875de70382da5a40dd87f976e5b9327 16330Author: Tilman Sauerbeck <tilman@code-monkey.de> 16331Date: Wed Aug 1 20:36:03 2007 +0200 16332 16333 Bumped version to 1.1.3. 16334 16335commit 76fae9cba1e9bdf7f0eb2ff2b90153d622136cf0 16336Author: Joerg Sonnenberger <joerg@netbsd.org> 16337Date: Fri Jul 27 11:15:47 2007 -0700 16338 16339 Fail properly on errors in recursive make. 16340 16341commit 1a18319b3bde08dd9ef69c7cd735a76000cf3177 16342Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 16343Date: Thu Jul 26 23:31:15 2007 +0200 16344 16345 Add missing override parameter in XrmCombineDatabase prototype in the manpage 16346 16347 Reported by Arnaud Giersch <arnaud.giersch@iut-bm.univ-fcomte.fr> 16348 in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=393434 16349 and https://bugs.freedesktop.org/show_bug.cgi?id=9948 16350 16351 Also add the type of the second argument in XrmMergeDatabases. 16352 16353commit 6f0764d4b56f64786b4980839ca262f10a51af6f 16354Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 16355Date: Thu Jul 26 22:53:52 2007 +0200 16356 16357 Clarify return value of XGetCommand in case of error in the manpage 16358 16359 Reported by Sean Perry <shalehperry@attbi.com> 16360 in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=133348 16361 and https://bugs.freedesktop.org/show_bug.cgi?id=9828 16362 16363commit b8bef57342632cc2d25580bb7daa3839bae04d89 16364Author: Alan Coopersmith <alan.coopersmith@sun.com> 16365Date: Wed Jul 25 17:44:06 2007 -0700 16366 16367 Include comment/copyright/license for AC_DEFINE_DIR in acinclude.m4 16368 16369commit 590cde811a79375231c59ed8583e02b111ed567c 16370Author: Kean Johnston <kean@armory.com> 16371Date: Thu Jul 19 16:44:20 2007 -0700 16372 16373 Reset hostname when falling back from :0 to hostname/localhost:0 as well 16374 16375commit d334665e619e9db657a2ea2764a8b852401d4a3f 16376Author: Kean Johnston <kean@armory.com> 16377Date: Tue Jul 17 16:34:30 2007 -0700 16378 16379 LOCALCONN fallback changes DisplayString() output, breaks KDE 16380 16381 See <http://lists.freedesktop.org/archives/xorg/2007-July/026443.html> 16382 16383commit 8e76bcf3cafda85058ce5f35f1f81929f4772f72 16384Author: Olivier Blin <blino@mandriva.org> 16385Date: Mon Jun 11 18:32:09 2007 +1000 16386 16387 fix XGetMotionEvents arguments order - Fixes bug 11222 16388 16389commit c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8 16390Author: Jamey Sharp <jamey@minilop.net> 16391Date: Sun Jun 10 16:19:59 2007 -0700 16392 16393 Fix locking in _XimGetWindowEventmask. 16394 16395 Now that XFilterEvent drops the Display lock before invoking callback 16396 functions, _XimGetWindowEventmask is called without the lock held. So 16397 when it called _XGetWindowAttributes, a variant of XGetWindowAttributes 16398 that does not lock the Display, Xlib/XCB would assert: 16399 16400 xcb_xlib.c:50: xcb_xlib_unlock: Assertion `c->xlib.lock' failed. 16401 16402 Should fix Gentoo #156367, Ubuntu #87947, Debian #427296. And others? 16403 16404commit 65f0ab5d46d80d55fc04d4eb14fa05f130eb8b9c 16405Author: Jamey Sharp <jamey@minilop.net> 16406Date: Sun Jun 3 21:41:47 2007 -0700 16407 16408 Move security fixes to the top of NEWS, and fix spacing. 16409 16410 Commit by Josh Triplett and Jamey Sharp. 16411 16412commit a549a258b8fcb1ba9d0c1b01b72967e385f67cab 16413Author: Jamey Sharp <jamey@minilop.net> 16414Date: Sun Jun 3 21:29:40 2007 -0700 16415 16416 Add NEWS item for bugfix in commit e2c1d788d1fe7bd2d34756493951552441e59b8c. 16417 16418 Commit by Josh Triplett and Jamey Sharp 16419 16420commit 5123b77a3d32d3ad479462f319762c328278aed9 16421Author: Jamey Sharp <jamey@minilop.net> 16422Date: Sun Jun 3 21:24:54 2007 -0700 16423 16424 Xlib/XCB: Inline and simplify handle_event, since only process_responses calls it now. 16425 16426 Commit by Josh Triplett and Jamey Sharp. 16427 16428commit 582ca690ea4f3ffd2b94826c4db97229bd3c7238 16429Author: Jamey Sharp <jamey@minilop.net> 16430Date: Sun Jun 3 20:59:12 2007 -0700 16431 16432 Xlib/XCB: Fix _XReadEvents to always enqueue a new event, even if an error occurs 16433 16434 Commit c337f9de7cfd89f983f83956b7457a274dd412f3 broke the invariant that 16435 _XReadEvents always enqueues at least one event even if an error occurred, 16436 because the one call to xcb_wait_for_event would then return an error, not an 16437 event, and nothing else ensured that process_responses would obtain an event. 16438 Fix this by reverting most of c337f9de7cfd89f983f83956b7457a274dd412f3 and 16439 f417570735aac865eb6b576d1ea76b5bfcd8573b and implementing the correct fix. In 16440 process_responses, wait_for_first_event now serves as a flag, cleared when 16441 actually handling an event. 16442 16443 Commit by Josh Triplett and Jamey Sharp. 16444 16445commit e2c1d788d1fe7bd2d34756493951552441e59b8c 16446Author: Jamey Sharp <jamey@minilop.net> 16447Date: Sun Jun 3 17:33:23 2007 -0700 16448 16449 Xlib/XCB: Only remove pending_requests when there are provably no more responses. 16450 16451commit 7a6dbd4b07ca0a49c30ca7a1d2437eafb2e15eab 16452Author: Josh Triplett <josh@freedesktop.org> 16453Date: Sun Jun 3 15:39:39 2007 -0700 16454 16455 Bump version number to 1.1.2, and add NEWS entry for 1.1.2 16456 16457 Signed-off-by: Josh Triplett <josh@freedesktop.org> 16458 16459commit 416f38f2e67ee1979b3d2feac6f06b3670238804 16460Author: Josh Triplett <josh@freedesktop.org> 16461Date: Sun Jun 3 12:13:44 2007 -0700 16462 16463 Revert "Revert "include: don't distribute XlibConf.h"" 16464 16465 This reverts commit 79fa3d8070d95b960ba486f2439225872471dadd. 16466 16467 Re-revert the XlibConf.h change, which prevented distribution, not 16468 installation. 16469 16470commit 79fa3d8070d95b960ba486f2439225872471dadd 16471Author: Josh Triplett <josh@freedesktop.org> 16472Date: Sat Jun 2 22:05:16 2007 -0700 16473 16474 Revert "include: don't distribute XlibConf.h" 16475 16476 This reverts commit c9e28e05ae01ce8a29bea09df759b6271865b44c. 16477 16478 The installed XlibInt.h includes XlibConf.h , so libX11 should ship 16479 XlibConf.h. (Commit c9e28e05ae01ce8a29bea09df759b6271865b44c didn't actually 16480 prevent automake from shipping XlibConf.h, because it used 16481 nodist_x11include_HEADERS rather than nodist_HEADERS.) 16482 16483commit f417570735aac865eb6b576d1ea76b5bfcd8573b 16484Author: Jamey Sharp <jamey@minilop.net> 16485Date: Sat Jun 2 17:59:15 2007 -0700 16486 16487 Xlib/XCB: inline wait_or_poll_for_event, which now had only one caller. 16488 16489 Commit by Josh Triplett and Jamey Sharp. 16490 16491commit c337f9de7cfd89f983f83956b7457a274dd412f3 16492Author: Jamey Sharp <jamey@minilop.net> 16493Date: Sat Jun 2 17:46:41 2007 -0700 16494 16495 Xlib/XCB: Ensure _XReadEvents reads at least one new event and blocks for exactly one event. 16496 16497 Commit by Jamey Sharp and Josh Triplett. 16498 16499commit 7f66c897f04806b75e574b55b48921b48045e3f9 16500Author: Jamey Sharp <jamey@minilop.net> 16501Date: Sat Jun 2 16:43:39 2007 -0700 16502 16503 Update _XReply's copy of _XCBUnlockDisplay's guts. 16504 16505 We introduced this bug in 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5. 16506 16507 Also add a comment in _XCBUnlockDisplay to discourage this problem from 16508 respawning. 16509 16510 Commit by Josh Triplett and Jamey Sharp. 16511 16512commit 740ead23512f8d2eaafaa69e514f1ebafad475b9 16513Author: Jamey Sharp <jamey@minilop.net> 16514Date: Sat Jun 2 16:01:01 2007 -0700 16515 16516 Xlib/XCB: Avoid re-crashing after _XIOError. 16517 16518 Commit by Josh Triplett and Jamey Sharp. 16519 16520commit 6b81cbbedfb521ce046b77ee3cc54e884a1dc0c5 16521Author: Jamey Sharp <jamey@minilop.net> 16522Date: Sat Jun 2 12:30:30 2007 -0700 16523 16524 Hold XCB's Xlib lock even when only the user lock (XLockDisplay) is held. 16525 16526 An Xlib client can query Display state, such as with NextRequest, while 16527 it holds only the Xlib user lock (between XLockDisplay and 16528 XUnlockDisplay), so XCB requests in other threads should be blocked when 16529 the Xlib user lock is held. 16530 16531 We acquire the lock even when XInitThreads was not called, so that pure 16532 XCB code can use multiple threads even in an otherwise single-threaded 16533 Xlib application. 16534 16535 Commit by Josh Triplett and Jamey Sharp. 16536 16537commit 95523387d619af5b400748898d722e080b5ce1a6 16538Author: Jamey Sharp <jamey@minilop.net> 16539Date: Sat Jun 2 11:57:39 2007 -0700 16540 16541 Allow re-entrant Xlib calls from _XIOError. 16542 16543 Some libraries try to clean up X resources from atexit handlers, _fini, 16544 or C++ destructors. To make these work, the Display lock should be 16545 downgraded to a user lock (as in XLockDisplay) before calling exit(3). 16546 This blocks Xlib calls from threads other than the one calling exit(3) 16547 while still allowing the exit handlers to call Xlib. 16548 16549 This assumes that the thread calling exit will call any atexit handlers. 16550 If this does not hold, then an alternate solution would involve 16551 registering an atexit handler to take over the lock, which would only 16552 assume that the same thread calls all the atexit handlers. 16553 16554 Commit by Josh Triplett and Jamey Sharp. 16555 16556commit 91b02b8064f4e0bcc56019f0722914850008a597 16557Author: Tilman Sauerbeck <tilman@code-monkey.de> 16558Date: Thu May 17 19:07:34 2007 +0200 16559 16560 More constification. 16561 16562commit a4f3841940158351f9424c3f59b305cce877177d 16563Author: Tilman Sauerbeck <tilman@code-monkey.de> 16564Date: Thu May 17 00:29:43 2007 +0200 16565 16566 Constified composite text charset table. 16567 16568commit 0581c0aa6039e6b2abb9f7b0a4f9904d8e01f00e 16569Author: Tilman Sauerbeck <tilman@code-monkey.de> 16570Date: Wed May 16 23:19:22 2007 +0200 16571 16572 Have the compiler fill in hexTable so we don't have to do it at runtime. 16573 16574commit 0e8d9ca47dab0d069e305d5784d05f2ade04f0a8 16575Author: Tilman Sauerbeck <tilman@code-monkey.de> 16576Date: Wed May 16 19:36:23 2007 +0200 16577 16578 More constification. 16579 16580commit 6d2bed8f04942b4de086a519ac693e729c9fdeea 16581Author: Tilman Sauerbeck <tilman@code-monkey.de> 16582Date: Wed May 16 19:18:20 2007 +0200 16583 16584 Constify and clean up token table. 16585 16586commit 6c508eab5df5d517f7e4cbe6087308cd53a564b2 16587Author: Tilman Sauerbeck <tilman@code-monkey.de> 16588Date: Wed May 16 18:24:42 2007 +0200 16589 16590 Constified more tables. 16591 16592commit e699c4231c205ef00d687b6412308d031b99806b 16593Author: Tilman Sauerbeck <tilman@code-monkey.de> 16594Date: Wed May 16 18:01:23 2007 +0200 16595 16596 Constified error list. 16597 16598commit a68a1cd7cb990ba276fbc36a7591044d78b3d3c1 16599Author: Peter Hutterer <peter@cs.unisa.edu.au> 16600Date: Tue May 15 16:54:01 2007 +0930 16601 16602 Add XGenericEvent definition and handling for long events. 16603 16604commit c76d30253f1483ac8200ad5c032a818907e65030 16605Author: Jan Willem Stumpel <jstumpel@planet.nl> 16606Date: Fri May 4 12:00:49 2007 -0700 16607 16608 Update el_GR.UTF-8/Compose.pre to match changes in xkeyboard-config cvs 16609 16610 Cf: 16611 16612 https://bugs.freedesktop.org/show_bug.cgi?id=10851 16613 https://bugs.freedesktop.org/show_bug.cgi?id=10824 16614 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=386385 16615 16616 The greek keyboard definition was changed to replace dead_horn and 16617 dead_ogonek with U0313 COMBINING COMMA ABOVE (aka Psili) and U0314 16618 COMBINING REVERSED COMMA ABOVE (aka Dasia). 16619 16620 This patch modifies the Greek Compose.pre to match. 16621 16622 It is generated by the script Jan Willem Stumpel <jstumpel@planet.nl> 16623 posted to 386385@bugs.debian.org: 16624 16625 #!/usr/bin/perl 16626 while (<>) { 16627 print $_; 16628 if (/dead_horn/) { 16629 s/dead_horn/U0313/; 16630 print $_; 16631 } 16632 elsif (/dead_ogonek/) { 16633 s/dead_ogonek/U0314/; 16634 print $_; 16635 } 16636 } 16637 16638commit a48386ce6b5f8fd2d9dc11a966c9bf5da59f3831 16639Author: Magnus Kessler <Magnus.Kessler@gmx.net> 16640Date: Tue May 1 15:20:08 2007 +0200 16641 16642 Switched function definitions from K&R to ANSI style. 16643 16644commit 605d357074d556a05a3fba2e85cbea36a3204248 16645Author: Tilman Sauerbeck <tilman@code-monkey.de> 16646Date: Tue May 1 14:47:03 2007 +0200 16647 16648 Tweaked configure output about the man pages suffix. 16649 16650commit 9824b40d2af4ca2376512c1be7743da0d5065900 16651Author: Alan Coopersmith <alan.coopersmith@sun.com> 16652Date: Sat Apr 28 00:42:18 2007 -0700 16653 16654 Fix typo in nroff macro in XkbAddGeomOverlayKey.man 16655 16656commit f93849dcc68bd5042ea0884e5190dc7c35b31d68 16657Author: Alan Coopersmith <alan.coopersmith@sun.com> 16658Date: Sat Apr 28 00:30:55 2007 -0700 16659 16660 Protect C comments and #defines in XKB man pages from being mangled by cpp 16661 16662commit f2f27d4763c7665e422fab10b96b4cf5ad6c0a6f 16663Author: Alan Coopersmith <alan.coopersmith@sun.com> 16664Date: Sat Apr 28 00:14:50 2007 -0700 16665 16666 Add Makefile to process/install XKB man pages 16667 16668commit d9954c6f6f3a8c406b946acd0d034ff83c656156 16669Author: Dennis Arellano <Dennis.Arellano@Sun.COM> 16670Date: Thu Aug 19 00:00:00 1999 -0700 16671 16672 Add man pages for XKB API's 16673 16674 Man pages originally written for X11R6.4 integration to Solaris 7 11/99 16675 Sun bug id 4258344: Add new XKB API manpages for 6.4 upgrade 16676 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4258344> 16677 16678 (Volunteer needed to convert prototypes in man pages to ANSI C style...) 16679 16680commit f640a49b5e2ebf29f9d655df544c63bf826f619a 16681Author: Tilman Sauerbeck <tilman@code-monkey.de> 16682Date: Fri Apr 20 18:39:59 2007 +0200 16683 16684 Markup tweak for XOpenIM. 16685 16686commit e972b0bb255af4f3258217852542faf5afa60b28 16687Author: Julien Cristau <jcristau@debian.org> 16688Date: Fri Apr 20 18:35:09 2007 +0200 16689 16690 Bug #9695: Fixed a few argument types in the XOpenIM manpage. 16691 16692commit b4e2276f329fa42397cb8609cfcd34ebafd3d96b 16693Author: Julien Cristau <jcristau@debian.org> 16694Date: Fri Apr 20 18:28:52 2007 +0200 16695 16696 Bug #9697: Fixed documentation of XVisualInfo struct. 16697 16698 The "depth" member was said to be unsigned int, but it's signed. 16699 16700commit 4068f3dae01c630f825002673b1d3a047ad61863 16701Author: Julien Cristau <jcristau@debian.org> 16702Date: Fri Apr 20 16:41:21 2007 +0200 16703 16704 Bug #9696: refer to XDefineCursor() instead of XDefineCusor(). 16705 16706commit 603c2f88d4e57ce1a3c16e8b6246866e6edd8fa8 16707Author: Tilman Sauerbeck <tilman@code-monkey.de> 16708Date: Sat Apr 7 14:42:55 2007 +0200 16709 16710 Use unistd.h to get getresuid() and friends. 16711 16712 This works since we now have _GNU_SOURCE defined. 16713 16714commit 0300f295bbd3a0c7c46baac8e0a27aeaf53c9d9b 16715Author: Tilman Sauerbeck <tilman@code-monkey.de> 16716Date: Sat Apr 7 14:27:12 2007 +0200 16717 16718 Bug #10562: Define _GNU_SOURCE on glibc systems. 16719 16720commit a225a0be48770beb689d5ac5da97073634f7deab 16721Author: Tilman Sauerbeck <tilman@code-monkey.de> 16722Date: Fri Apr 6 12:46:25 2007 +0200 16723 16724 For nls/*.pre, allow people to comment lines by starting them with '##'. 16725 16726 This fixes a bunch of cpp warnings from nls/en_US.UTF-8/Compose.pre. 16727 Fixing that file would have been a larger diff, and using ## may be 16728 nicer to use anyway. 16729 16730commit 1c75a9479011e5f1ad01c950628d0ef5a302d8b6 16731Author: Tilman Sauerbeck <tilman@code-monkey.de> 16732Date: Fri Apr 6 11:59:31 2007 +0200 16733 16734 Bug #10475: Fixed lots of char*/const char* mixups. 16735 16736 I didn't fix all of them, as that would require touching 16737 public headers. 16738 16739commit 680dd50193b5b3fcabdd3f1fcbd6a889d5a95c54 16740Author: Matthias Hopf <mhopf@suse.de> 16741Date: Thu Apr 5 11:53:18 2007 +0200 16742 16743 Fix 64bit issues with reallocation. 16744 16745commit 4d38aeaca42d0bdfe34a833a142ee4d895de03bf 16746Author: Tilman Sauerbeck <tilman@code-monkey.de> 16747Date: Wed Apr 4 18:41:18 2007 +0200 16748 16749 Fixed a few warnings. 16750 16751commit 7dc7ef398b6ad90ccd1680ed9cd1cfdd47312f5a 16752Author: Matthieu Herrb <matthieu@roadrock.(none)> 16753Date: Tue Apr 3 15:39:52 2007 +0200 16754 16755 Multiple integer overflows in the XGetPixel() and XInitImage functions 16756 16757 CVE-2007-1667 16758 16759commit 0284b144340a455a4b5b5011d81ac5a610372291 16760Author: David Baron <dbaron@dbaron.org> 16761Date: Fri Mar 30 17:07:46 2007 +0200 16762 16763 Bug #7703: Fixed XSetSizeHints() et al wrt use of uninitialized data. 16764 16765 Now only those fields of the respective hint struct are set that 16766 are actually valid in the input data. 16767 The changed functions are: 16768 XSetSizeHints(), XSetWMHints() and XSetWMSizeHints(). 16769 16770commit 0994faa0c76c45b106442db461b8a30a3e1c9395 16771Author: Tilman Sauerbeck <tilman@code-monkey.de> 16772Date: Thu Mar 29 17:31:25 2007 +0200 16773 16774 Fixed the change from the previous SendEvent commit. 16775 16776 Testing a different patch than the one you commit is bad, right? 16777 16778commit 398d75528a84f4b8414eb0e363cf53b1b16f6fdf 16779Author: Tilman Sauerbeck <tilman@code-monkey.de> 16780Date: Wed Mar 28 22:23:44 2007 +0200 16781 16782 Bug #10292: Fixed a memory leak related to XOpenDisplay() in the XCB code. 16783 16784commit ab0bcd07957cecc8e7c0e75d5160a625e91264fe 16785Author: David Baron <dbaron@dbaron.org> 16786Date: Wed Mar 28 22:21:40 2007 +0200 16787 16788 Bug #7713: Initialize all of the event's fields before sending it. 16789 16790commit bc80f9fe3ccce40ee41246b97470c4f0519756ad 16791Author: Julien Cristau <jcristau@debian.org> 16792Date: Sun Mar 18 13:14:48 2007 +0100 16793 16794 Bug #9279: Fixed a file descriptor leak. 16795 16796commit c9e28e05ae01ce8a29bea09df759b6271865b44c 16797Author: Daniel Stone <daniel@fooishbar.org> 16798Date: Sat Dec 16 00:45:19 2006 +0200 16799 16800 include: don't distribute XlibConf.h 16801 16802 Since XlibConf.h is built by configure, don't distribute it. 16803 16804commit dd1705ced2cac6b4b6b21e79272fcf9bed4bf376 16805Merge: 129bbb9f 769b9854 16806Author: Jeremy C. Reed <reed@glacier.reedmedia.net> 16807Date: Thu Dec 14 14:23:20 2006 -0600 16808 16809 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/lib/libX11 16810 16811commit 129bbb9f9114a571556fa3a24f15ba58a5cdb2de 16812Author: Jeremy C. Reed <reed@glacier.reedmedia.net> 16813Date: Thu Dec 14 14:21:19 2006 -0600 16814 16815 For NetBSD, define the XTHREADLIB and XTHREAD_CFLAGS. 16816 16817commit 769b9854f7eb1d6d20dd0b4a1c1215ad8e1b77b6 16818Author: Daniel Stone <daniel@fooishbar.org> 16819Date: Wed Dec 6 18:53:00 2006 +0200 16820 16821 Makefile.am: make ChangeLog hook safer 16822 16823 Make ChangeLog hook as safe as possible. 16824 (cherry picked from f5d6a3d24095c7ffed86705995d0874c885e7676 commit) 16825 16826commit 8a8185a649e93b90ffa820387ffdca831227f5a9 16827Author: Josh Triplett <josh@freedesktop.org> 16828Date: Thu Nov 30 18:33:13 2006 -0800 16829 16830 Add autogen.sh to EXTRA_DIST. 16831 16832commit efe817f95ef8d05e863c83147e903140bc860de4 16833Author: Jamey Sharp <jamey@minilop.net> 16834Date: Thu Nov 30 17:58:35 2006 -0800 16835 16836 Release libX11 1.1.1. 16837 16838commit f637a5b03164263a3af2e644cf655e52b015f1bb 16839Author: Ross Combs <rocombs@cs.nmsu.edu> 16840Date: Sat Nov 25 14:45:17 2006 -0800 16841 16842 Debian bug #354315: Clarify return value in XGetWindowAttributes man page 16843 16844 This man page does not discuss the actual return values of the 16845 function, but says they are of type "Status". One might assume 16846 that this means you could compare it with the "Success" macro. 16847 One would be wrong. 16848 16849 The X functions seem to have two three types representing status. 16850 If it is an "int" there are a number of error codes or "Success" 16851 which can be compared against. If it is a bool, the result can be 16852 compared with "True" or "False". If the return type is "Status" it 16853 appears that the return type is either 0 or 1. Unfortunately the 16854 value for Success is zero, so it is important to distinguish 16855 between the first two types of return values and the third; 16856 otherwise the conditional will be inverted. 16857 16858 XGetWindowAttributes() is one of the functions which returns zero 16859 for failure. The man page should make this clear. 16860 16861commit c6a0b0f18ed1242eeb908f5cf767ab8381edd456 16862Author: Jamey Sharp <jamey@minilop.net> 16863Date: Sat Nov 25 14:23:45 2006 -0800 16864 16865 Bug #9154: Always process an event for _XReadEvents, even if an error occurs 16866 16867 Previously, process_responses (in the wait_for_first_event case called 16868 from _XReadEvents) considered any return from xcb_wait_for_event 16869 sufficient to think it had processed an event. If xcb_wait_for_event 16870 returned an error, and no more events occurred before process_responses 16871 called xcb_poll_for_event, process_responses would try to return with 16872 dpy->head NULL, and would fail an assertion for the _XReadEvents 16873 postcondition. Now, process_responses continues using xcb_wait_for_event 16874 until it gets an event. 16875 16876commit d56e78acce9b2aa1dd1bf172afedaa3bccd5e1c8 16877Author: Tilman Sauerbeck <tilman@code-monkey.de> 16878Date: Sat Nov 25 05:29:31 2006 -0800 16879 16880 Bug #9153: Fix access to freed memory. 16881 16882 The fix for bug #8622 introduced a smaller bug where _XReply would 16883 read memory shortly after freeing it. This patch caches the needed 16884 value in a stack-allocated variable before the heap-allocated memory 16885 is freed. 16886 16887commit 934ca763bbc0dd7ae460469bfc000ba101602bcc 16888Author: Josh Triplett <josh@freedesktop.org> 16889Date: Fri Nov 24 19:57:58 2006 -0800 16890 16891 libx11 doesn't use inputproto in public headers; don't require it in x11.pc 16892 16893 Based on a Debian patch. 16894 16895commit 4255997ef2d92740d51f6e63e9eabcfa089683f0 16896Author: Josh Triplett <josh@freedesktop.org> 16897Date: Thu Nov 23 07:19:32 2006 -0800 16898 16899 Release libX11 1.1 16900 16901commit a1168e11ec9377307c51a7271faec3bf88a63a66 16902Author: Jamey Sharp <jamey@minilop.net> 16903Date: Tue Nov 21 17:52:34 2006 -0800 16904 16905 Add note in man-page that XListFontsWithInfo is not thread-safe. 16906 16907 _XReply drops the Display lock, so the value of dpy->request may change 16908 before _XReply is called again. 16909 16910 I discovered this by inspection a year or two ago. I'm pretty confident 16911 in the claim, and nobody has come up with an argument for why it's safe 16912 despite appearances. 16913 16914commit 67abe024268c6b1fdee516e5d3a046ccffd7e80a 16915Author: Jamey Sharp <jamey@minilop.net> 16916Date: Sat Nov 18 15:39:26 2006 -0800 16917 16918 Bug #8622: Fix response processing order for threaded apps 16919 16920 Previously, process_responses (the common code for _XReply, 16921 _XReadEvents, and _XEventsQueued) took the current request sequence 16922 number as an argument, and did some highly complicated processing to 16923 attempt to process responses in order across threads. This didn't 16924 actually work. 16925 16926 Now, process_responses handles responses in order, by adding condition 16927 variables to the list of outstanding requests in 16928 dpy->xcb->pending_requests, and blocking on them when those requests 16929 should get processed to allow _XReply to process them; if actually 16930 called from _XReply, it returns when _XReply's request should get 16931 processed. _XReply broadcasts on the condition variable after it has 16932 read its reply and re-acquired the display lock. 16933 16934 Another great commit brought to you by Jamey Sharp, Josh Triplett, the 16935 Gobby collaborative text editor, conspy, and ridiculous amounts of SSH 16936 forwarding. 16937 16938 Signed-off-by: Josh Triplett <josh@freedesktop.org> 16939 Signed-off-by: Jamey Sharp <jamey@minilop.net> 16940 16941commit 941f02ede63baa46f93ed8abccebe76fb29c0789 16942Author: Lars Knoll <lars@trolltech.com> 16943Date: Wed Nov 8 12:17:41 2006 -0800 16944 16945 Don't hold the display lock around callbacks to the application. 16946 16947 This fixes an XCB locking assertion failure, particularly with emacs. 16948 16949commit e494ecaac1ec8a22bd9a85f800fca74d02e9d358 16950Author: Diego 'Flameeyes' Pettenò <flameeyes@gentoo.org> 16951Date: Tue Nov 7 09:32:00 2006 -0800 16952 16953 Add xcb-xlib dependency to x11.pc when built against XCB. 16954 16955commit 2302008a3793eb4df8ede777d54fe06505c47eaf 16956Author: Eric Anholt <eric@anholt.net> 16957Date: Mon Nov 6 17:11:42 2006 -0800 16958 16959 XCB: Allocate the right amount of memory for dpy->lock_fns. 16960 16961 Fixes a crash I was experiencing on startup of anything using gdk. 16962 16963commit 5f860655be88108b03ccd97470a0814819254bf0 16964Author: Jamey Sharp <jamey@minilop.net> 16965Date: Thu Nov 2 17:55:31 2006 -0800 16966 16967 Release libX11 1.1 RC2 (1.0.99.2). 16968 16969commit a6f4bbf7b1d725b0f04bd660f57b861a76b19831 16970Author: Benno Schulenberg <bensberg@justemail.net> 16971Date: Sun Oct 29 03:10:30 2006 +0300 16972 16973 nls (en_US): remove long compositions that override shorter (bug #2286) 16974 Remove long compositions that override (or get overriden by) later shorter 16975 compositions, e.g. a four-key compose sequence that gets overriden by a 16976 later three-key compose sequence. 16977 16978commit 0280bf11ef88673a9b5bba3a91a599260f1f0949 16979Author: Benno Schulenberg <bensberg@justemail.net> 16980Date: Sun Oct 29 03:08:36 2006 +0300 16981 16982 nls: remove shadowed compose entries (bug #2286) 16983 Remove compose entries shadowed by others later on. 16984 16985commit d118f2b1ef10997194b281524177dea7396da7dd 16986Author: Benno Schulenberg <bensberg@justemail.net> 16987Date: Sun Oct 29 03:07:15 2006 +0300 16988 16989 nls: remove duplicate compose entries (bug #2286) 16990 Remove a bunch of duplicate entries from various Compose files. 16991 16992commit 5e1cc2fe20e5904ca1e05a4cb7be13d450a593bb 16993Author: Caolan McNamara <caolanm@redhat.com> 16994Date: Sun Oct 29 02:46:15 2006 +0300 16995 16996 XKB geometry: fix leaks in GetKeyboardByName (bug #8099) 16997 Don't leak the name and value of every property we parse, as well as the 16998 name of every colour. 16999 17000commit 686bb8b35acf6cecae80fe89b2b5853f5816ce19 17001Author: Matthias Hopf <mhopf@suse.de> 17002Date: Wed Oct 18 14:25:04 2006 +0200 17003 17004 Fix double open of compose file. 17005 17006 Issue found by Kees Cook <kees@canonical.com>. 17007 17008commit d3e65cb8cddf08913d83c9df2bb9b1517f2ad3a8 17009Author: Jamey Sharp <jamey@minilop.net> 17010Date: Sat Oct 14 21:25:10 2006 -0700 17011 17012 XCB: check for and handle I/O errors in _XGetXCBBuffer. 17013 17014commit 256eba6b40c5f811a03b04abf5f85f728ee3ab5d 17015Author: Jamey Sharp <jamey@minilop.net> 17016Date: Wed Oct 11 00:06:50 2006 -0700 17017 17018 XKB bugfix: SyncHandle must be called after UnlockDisplay, not before. 17019 17020commit 1eedf1bd033e496843cfde42ae4ae5a119298605 17021Author: Jamey Sharp <jamey@minilop.net> 17022Date: Tue Oct 10 23:03:28 2006 -0700 17023 17024 Add correct Display locking to XKB functions. 17025 17026 Some XKB functions didn't correctly call LockDisplay or UnlockDisplay. 17027 This patch fixes at least some instances of that problem. 17028 17029 Thanks to Magnus Kessler <Magnus.Kessler@gmx.net> for finding these bugs 17030 and proposing a fix, which this patch is based on. 17031 17032commit e17c2cbe9fbaa1600d4b9463ec800a874b0d87cd 17033Author: David Nusinow <dnusinow@debian.org> 17034Date: Tue Oct 10 22:11:05 2006 -0400 17035 17036 Dynamically generate internal manpage section using __libmanpagesuffix__ so that it actually matches the section if you don't use 3X11 17037 17038commit e53557da969b706dbc843f6fde3db31ffe382e0f 17039Author: Jamey Sharp <jamey@minilop.net> 17040Date: Sat Oct 7 21:00:36 2006 -0700 17041 17042 Release libX11 1.1 RC1 (1.0.99.1). 17043 17044commit bf237409c5fce32c557d298f62f44d456c2b5bc8 17045Merge: ba477191 2d426d1f 17046Author: Jamey Sharp <jamey@minilop.net> 17047Date: Sat Oct 7 21:07:16 2006 -0700 17048 17049 As XCB support is about to be released in libX11, stable is now subsumed by master. 17050 17051commit ba477191c67ce93e61423cc1abe35275704cce50 17052Author: Jamey Sharp <jamey@minilop.net> 17053Date: Sat Oct 7 03:48:13 2006 -0700 17054 17055 XCB: Don't rely on having the definition of struct xcb_setup_t available. 17056 17057commit 117b55cbd0b0ce51362df88363ed83d44a493ac7 17058Author: Jamey Sharp <jamey@minilop.net> 17059Date: Fri Oct 6 16:27:31 2006 -0700 17060 17061 xcb_poll_for_event no longer takes an 'int *error' out-parameter. 17062 17063commit caaa8e8a55e837b3585c1dee7bef194fc4c79d16 17064Author: Josh Triplett <josh@freedesktop.org> 17065Date: Fri Oct 6 16:26:11 2006 -0700 17066 17067 Actually ship Xxcbint.h 17068 17069commit 7b027e53b5e393082f4f515c8ba18077eb97163f 17070Author: Josh Triplett <josh@freedesktop.org> 17071Date: Fri Oct 6 16:25:50 2006 -0700 17072 17073 Clean ChangeLog only in "make maintainer-clean", not "make clean" 17074 17075 ChangeLog requires a git repo to generate; make clean and make distclean 17076 shouldn't get rid of it. 17077 17078commit cab22e02e78b3e5b8a73d1cd55cf6686426b47e0 17079Author: Josh Triplett <josh@freedesktop.org> 17080Date: Fri Oct 6 16:24:58 2006 -0700 17081 17082 Add ChangeLog and "make dist"-generated files to .gitignore 17083 17084commit e4c7cfdee4a40e466c0c6b370cabd432e9e855a0 17085Author: Josh Triplett <josh@freedesktop.org> 17086Date: Fri Oct 6 15:53:27 2006 -0700 17087 17088 Add manual pages for XGetXCBConnection and XSetEventQueueOwner 17089 17090commit 688224cea95e453f94c5a602dc6fce84bc93dfc0 17091Author: Josh Triplett <josh@freedesktop.org> 17092Date: Fri Oct 6 13:41:16 2006 -0700 17093 17094 Remove unnecessary prototype for _XFreeDisplayStructure in xcb_disp.c 17095 17096commit ab728ca372288d0db1b486c265e34c1376f29104 17097Author: Josh Triplett <josh@freedesktop.org> 17098Date: Fri Oct 6 12:36:39 2006 -0700 17099 17100 XCL is dead; long live Xlib/XCB 17101 17102 Rename all instances of "XCL" to Xlib/XCB-derived names. 17103 17104 The only user-visible change: rename the include file <X11/xcl.h> to 17105 <X11/Xlib-xcb.h>; programs will need to change their #include lines to match. 17106 17107 Remove the XCL cast inlines from Xlib-xcb.h. 17108 17109commit 5b73093203039d307eb7ab3845c3ced207e9e26c 17110Author: Josh Triplett <josh@freedesktop.org> 17111Date: Fri Oct 6 11:33:49 2006 -0700 17112 17113 Add XCB developers to AUTHORS 17114 17115commit 1cb71ff139276a0a58c60ea5f261f64b94706b9b 17116Author: Josh Triplett <josh@freedesktop.org> 17117Date: Fri Oct 6 11:21:28 2006 -0700 17118 17119 Fix email addresses in README 17120 17121commit 12f038669278019594ca0ed53dadcf4e84092422 17122Author: Jamey Sharp <jamey@minilop.net> 17123Date: Fri Oct 6 02:13:05 2006 -0700 17124 17125 XCB: Handle all responses in order of monotonically increasing sequence number. 17126 17127commit f392680273278b43079302206897f794e60f3c70 17128Author: Josh Triplett <josh@freedesktop.org> 17129Date: Fri Oct 6 01:11:08 2006 -0700 17130 17131 Actually install x11-xcb.pc, and ship x11-xcb.pc.in 17132 17133commit f1fcad2e3fd17aaf1294f1d8e9f406fd5b32a863 17134Author: Jamey Sharp <jamey@minilop.net> 17135Date: Thu Oct 5 18:32:29 2006 -0700 17136 17137 XCB: correctly handle failure to connect to X server. 17138 17139commit e754b3b078d556c7861da56aad47d244e9199e06 17140Author: Josh Triplett <josh@freedesktop.org> 17141Date: Thu Oct 5 17:44:22 2006 -0700 17142 17143 Split public Xlib/XCB functions into libX11-xcb 17144 17145 We can never change the libX11 soname, and we don't want to commit to never 17146 changing the public Xlib/XCB functions, so split them into a separate library 17147 libX11-xcb. This also means that a program linked solely against libX11 17148 should work with either Xlib or Xlib/XCB, which will make life easier for 17149 package maintainers. 17150 17151 Signed-off-by: Josh Triplett <josh@freedesktop.org> 17152 Acked-by: Jamey Sharp <jamey@minilop.net> 17153 17154commit ffd367f708b295abaedf3a23a1bfd4710d171d6f 17155Author: Jamey Sharp <jamey@minilop.net> 17156Date: Wed Oct 4 17:16:46 2006 -0700 17157 17158 No longer #include Xmd from xcl.h: we do not need it. 17159 17160commit 3aff149d42ba8ed620091971b3766bdf96c62aaf 17161Author: Jamey Sharp <jamey@minilop.net> 17162Date: Wed Oct 4 16:58:32 2006 -0700 17163 17164 XCB: Revert locking to simple wrapper around libX11's normal locks. 17165 17166 No more recursive mutexes, no more banging XCB's I/O lock in-place, and 17167 reduces the differences between the previous stable release and an 17168 XCB-enabled one. Sadly, Xlib's pluggable thread functions work again 17169 too, now. I apologize to the world. 17170 17171commit 8ff122fb529bdb1c2b9a86b12d06b6da1b35d708 17172Author: Jamey Sharp <jamey@minilop.net> 17173Date: Mon Sep 25 04:54:52 2006 -0700 17174 17175 Link explicitly against XCB's Xlib compatibility functions. 17176 17177commit bde3cd123d65a2f36ee0c417f5f231b7e01d0671 17178Author: Jamey Sharp <jamey@minilop.net> 17179Date: Mon Sep 25 04:13:20 2006 -0700 17180 17181 libxcb now installs header files in <xcb>, not <X11/XCB>. 17182 17183commit 87d00207f5a1f25a45a153618739cd6481814f89 17184Author: Ian Osgood <iano@quirkster.com> 17185Date: Sun Sep 24 23:39:01 2006 -0700 17186 17187 Track XCB's "Great Renaming". 17188 17189commit 85a5e98dff9b7752fae157fad9b8c9825cb0efab 17190Author: Jamey Sharp <jamey@minilop.net> 17191Date: Tue Sep 12 23:02:42 2006 -0700 17192 17193 Quit using XCBGetQueuedRequestRead. 17194 17195commit a61936fc4e9bd93b108764bbacd5b8f786e51915 17196Author: Tollef Fog Heen <tfheen@err.no> 17197Date: Wed Aug 30 00:05:54 2006 +0300 17198 17199 nls: use _XlcUtf8Loader for en_US (bug #7982) 17200 Use _XlcUtf8Loader instead of _XlcUnicodeLoade,r bringing it into line with 17201 every other locale. 17202 17203commit abcc7e1865cdfbd591f6520cfe4257f0b0b1c03e 17204Author: Alan Coopersmith <alan.coopersmith@sun.com> 17205Date: Wed Aug 23 18:49:30 2006 -0700 17206 17207 When opening display, if LOCALCONN fails, fall back to UNIXCONN, then TCPCONN 17208 17209 Port to X11R7 of Sun bug fix 4061225 by Alex Chen for X11R6 - when failing to 17210 connect on a named pipe, try a Unix socket first, to better support people who 17211 replace their X servers with ones that don't support named pipe transport. 17212 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4061225> 17213 17214commit 1ddc44c1cada7e926bd4787406444ce7c36b61e1 17215Author: Mark Brown <mark.brown@sun.com> 17216Date: Thu Jul 27 19:17:10 2006 -0700 17217 17218 Sun bug 1149809: Document event delivery when grab is terminated. 17219 17220commit 171107b03ac89d94f9006c7cda242aeefb9ecd16 17221Author: Dennis Arellano <Dennis.Arellano@Sun.COM> 17222Date: Thu Jul 27 18:47:06 2006 -0700 17223 17224 Sun bug 4091271: XGetWindowProperty is missing a crucial prop_return description 17225 17226 Document that 32-bit format properties are always returned in arrays of type 17227 long, even on systems where long is 64-bits. 17228 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4091271> 17229 17230commit 8309efe6550877cd0bf22979904b3f6bd3e6cffa 17231Author: Alan Coopersmith <alan.coopersmith@sun.com> 17232Date: Mon Jul 24 15:52:00 2006 -0700 17233 17234 Add support for "make lint" to check code with lint/sparse/etc. 17235 17236commit 931e02fbd1acd09aae2b0954c34342c86c72dff6 17237Author: Alan Coopersmith <alan.coopersmith@sun.com> 17238Date: Mon Jul 24 15:50:52 2006 -0700 17239 17240 ANSIfy some static function definitions 17241 17242commit 20b7abcaac324d90454de63f32f4a2b398e69e63 17243Author: Alan Coopersmith <alan.coopersmith@sun.com> 17244Date: Mon Jul 24 15:01:40 2006 -0700 17245 17246 Fix sparse warning: Using plain integer as NULL pointer 17247 17248commit d158ab29930513c4097f5b67e7bea08ed2bfd62c 17249Author: Alan Coopersmith <alan.coopersmith@sun.com> 17250Date: Mon Jul 24 14:00:24 2006 -0700 17251 17252 Remove unused variable 17253 17254commit 30377000375bdb958042dcb1f38503c94ef21eaf 17255Author: Eric Anholt <anholt@FreeBSD.org> 17256Date: Fri Jul 21 18:55:36 2006 -0400 17257 17258 Bug #7188: Fix the documentation of XUrgencyHint (not UrgencyHint). 17259 17260commit 4eba45879aea4e415ab550ee56b900d060099110 17261Author: Matthieu Herrb <matthieu.herrb@laas.fr> 17262Date: Sun Jul 16 10:55:39 2006 +0200 17263 17264 set GIT_DIR=${srcdir}/.git for git-log 17265 17266commit 2d426d1f2608fedb77bd7d010dabece76b8d4a60 17267Author: Aaron Plattner <aplattner@nvidia.com> 17268Date: Tue Jul 11 13:27:49 2006 -0700 17269 17270 Add a .PHONY to ensure the ChangeLog isn't stale. 17271 17272 Setting the ChangeLog rule as phony forces it to be re-run even when the 17273 ChangeLog file already exists. Research indicates .PHONY is portable to BSD and 17274 Solaris make. 17275 (cherry picked from b8a98809ed81e1226775e6447ef219ffc01334b5 commit) 17276 17277commit b8a98809ed81e1226775e6447ef219ffc01334b5 17278Author: Aaron Plattner <aplattner@nvidia.com> 17279Date: Tue Jul 11 13:27:49 2006 -0700 17280 17281 Add a .PHONY to ensure the ChangeLog isn't stale. 17282 17283 Setting the ChangeLog rule as phony forces it to be re-run even when the 17284 ChangeLog file already exists. Research indicates .PHONY is portable to BSD and 17285 Solaris make. 17286 17287commit 8f2be66089b88e4ed6acb0211ea107f4bb910bd3 17288Author: Mayank Jain <majain@redhat.com> 17289Date: Tue Jul 11 19:42:47 2006 +0100 17290 17291 add Indian language locales 17292 Add as, kn, ml, or, ur, and te locales. 17293 17294commit a92eb6785699bbc9c6c3813f6af3bb9431a3f6b0 17295Author: Matthias Hopf <mhopf@suse.de> 17296Date: Tue Jul 4 12:16:30 2006 +0200 17297 17298 Fix for autoconf 2.60 issue. 17299 17300 Updated AC_DEFINE_DIR. 17301 Reverted datarootdir change. 17302 17303commit 644f4828b15bce42b597eb123ba0bbc372c46c03 17304Author: Keith Packard <keithp@neko.keithp.com> 17305Date: Sat Jul 1 21:31:23 2006 -0700 17306 17307 Work around recent autoconf (2.59?) changes in directory expansion. 17308 17309 Recent autoconf versions have changed how directory names are managed in the 17310 configure.ac script; automatic 'eval' invocations now occur as a part of the 17311 AC_DEFINE_DIR macro which make it imperative that AC_DEFINE_DIR be executed 17312 before the variables are used in further macro definitions. Also, ${datadir} 17313 is apparantly an old name for ${datarootdir} as ${datadir} doesn't get 17314 expanded correctly by AC_DEFINE_DIR. This looks like an autoconf bug, but it 17315 is easy to work around by just using ${datarootdir} instead of ${datadir}. 17316 17317commit be70a31229aa106aff0a09d78c00812682cd3475 17318Author: Keith Packard <keithp@neko.keithp.com> 17319Date: Sat Jul 1 01:56:05 2006 -0700 17320 17321 Xlib/XCB: handle 32-bit sequence wrap. 17322 17323 Replace broken sequence compares with XCB_SEQUENCE_COMPARE (copied from 17324 XCB). 17325 Account for XCB sequence 0 handling. 17326 17327commit 99c711707ad08e1396e123b1c7df687c560a489a 17328Author: Donnie Berkholz <spyderous@gentoo.org> 17329Date: Thu Jun 29 19:43:20 2006 -0700 17330 17331 Bump version to 1.0.3. 17332 17333commit cde3c0dd72af2b490e80cffca962e3487dd31be4 17334Author: Donnie Berkholz <spyderous@gentoo.org> 17335Date: Thu Jun 29 19:39:36 2006 -0700 17336 17337 Bug #7349: Missed one of the setuid fixes. 17338 (cherry picked from e9614c963b532f46a7932c2305a4b177a996a222 commit) 17339 17340commit df3fef8983d96c59d481c4cdaf1f271d54a116d2 17341Author: Matthias Hopf <mhopf@suse.de> 17342Date: Thu Jun 29 18:59:57 2006 +0200 17343 17344 Update to final Compose cache directory location. 17345 (cherry picked from abda4d223e9cce9ac6e7b5d82a5680d9a502e52a commit) 17346 17347commit 912ef198292d3053daa810f842510e5d62ded0f0 17348Author: Matthias Hopf <mhopf@suse.de> 17349Date: Thu Jun 29 17:41:41 2006 +0200 17350 17351 Fix alignment of trees and wide chars in the cache. 17352 (cherry picked from 40a64c61f8bc33d497e1224e02c41dea2d424d97 commit) 17353 17354commit 2ece832118b3ee5d8ed19f1ee9b1c822b70ec6e9 17355Author: Matthias Hopf <mhopf@suse.de> 17356Date: Wed Jun 28 19:17:03 2006 +0200 17357 17358 First (dummy) entry of compose caches was not initialized and thus contained varying garbage. 17359 (cherry picked from f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3 commit) 17360 17361commit e9614c963b532f46a7932c2305a4b177a996a222 17362Author: Donnie Berkholz <spyderous@gentoo.org> 17363Date: Thu Jun 29 19:39:36 2006 -0700 17364 17365 Bug #7349: Missed one of the setuid fixes. 17366 17367commit abda4d223e9cce9ac6e7b5d82a5680d9a502e52a 17368Author: Matthias Hopf <mhopf@suse.de> 17369Date: Thu Jun 29 18:59:57 2006 +0200 17370 17371 Update to final Compose cache directory location. 17372 17373commit 40a64c61f8bc33d497e1224e02c41dea2d424d97 17374Author: Matthias Hopf <mhopf@suse.de> 17375Date: Thu Jun 29 17:41:41 2006 +0200 17376 17377 Fix alignment of trees and wide chars in the cache. 17378 17379commit f442dcaa56f8ecb7443e8e51c88ed97e10dbdba3 17380Author: Matthias Hopf <mhopf@suse.de> 17381Date: Wed Jun 28 19:17:03 2006 +0200 17382 17383 First (dummy) entry of compose caches was not initialized and thus contained varying garbage. 17384 17385commit bdbe464d774e01d317f67c63ebbda2fd0edbbdd1 17386Author: Donnie Berkholz <spyderous@gentoo.org> 17387Date: Thu Jun 22 23:59:03 2006 -0700 17388 17389 Delete ChangeLog, and add a rule to autogenerate it for distribution. Also 17390 add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a 17391 ChangeLog. 17392 (cherry picked from b0edfb8df16ab8c9959b83a4c966d55a59c4e295 commit) 17393 17394commit b0edfb8df16ab8c9959b83a4c966d55a59c4e295 17395Author: Donnie Berkholz <donnie@comet.(none)> 17396Date: Thu Jun 22 23:47:38 2006 -0700 17397 17398 Delete ChangeLog, and add a rule to autogenerate it for distribution. Also 17399 add 'foreign' to AM_INIT_AUTOMAKE to make autotools happy with not having a 17400 ChangeLog. 17401 17402commit dd54981aa76e9dfdc4c3302d6105b4b229447c84 17403Author: Donnie Berkholz <donnie@comet.(none)> 17404Date: Thu Jun 22 15:47:38 2006 -0700 17405 17406 Bump version to 1.0.2. 17407 17408commit c93539d974a67f596a5eb5b65042d26602546c72 17409Author: Matthieu Herrb <matthieu@deville.herrb.com> 17410Date: Tue Jun 20 21:04:03 2006 +0200 17411 17412 Check setuid() return value. 17413 (cherry picked from 5169d0e08ff6acb350a6ea768623f5ff0b85b05f commit) 17414 17415commit 605533f814ab7892991578706a6458f61a89ca4d 17416Author: Derek Wang <derek.wang@sun.com> 17417Date: Mon Jun 19 11:05:37 2006 -0700 17418 17419 Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded 17420 (cherry picked from eff50c94a07194297e705da53d9fbb3a40fb9ad4 commit) 17421 17422commit 5bbd0822c5a926de0ed293437fb9f2b75cf3c4f4 17423Author: Scott Revelt <scott.revelt@sun.com> 17424Date: Fri Jun 16 19:11:13 2006 -0700 17425 17426 Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators 17427 based on locale that doesn't match those used in the Xcms.txt 17428 (cherry picked from 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa commit) 17429 17430commit 0b05cd4da6134df527fb010384a9fd569bd5d6a3 17431Author: Alan Coopersmith <alanc@alf.(none)> 17432Date: Fri Jun 16 18:53:33 2006 -0700 17433 17434 Add *~ to ignore emacs droppings 17435 (cherry picked from c33d7b8282ac196b36229be26442296768c16f3e commit) 17436 17437commit 214658b76b56768f69c3959a11525aae7813f448 17438Author: Matthias Hopf <mhopf@suse.de> 17439Date: Fri Jun 16 15:36:40 2006 +0200 17440 17441 Bug #3104: Compose table cache for faster X11 application starts. 17442 (cherry picked from 1f4c9893ade08bad30c9bd12a36bee57d30b001e commit) 17443 17444commit 13968a23aaea838ba4b69e42e8900f803499e091 17445Author: Matthias Hopf <mhopf@suse.de> 17446Date: Tue Jun 13 20:23:46 2006 +0200 17447 17448 Bug #3104: Compose table cache for faster X11 application starts. 17449 (cherry picked from 4fe22647e6010a2886c2f3a7093adeaeb6678faa commit) 17450 17451commit e7f8bca08f5476d0ca262097639ac7d424bb4d10 17452Author: Lubos Lunak <llunak@suse.de> 17453Date: Mon Jun 12 18:48:08 2006 +0200 17454 17455 Bug #3104: Compose table cache for faster X11 application starts. 17456 (cherry picked from 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be commit) 17457 17458commit f506aaf8ac9aae1ee8daaef6cde34ee85aecd641 17459Author: Matthias Hopf <mhopf@suse.de> 17460Date: Fri Jun 9 18:24:02 2006 +0200 17461 17462 Bug #3104: Compose table cache for faster X11 application starts. Part 1: Pointerless compose data structure, using indices instead of pointers, needed for mmap()ing data structure. 17463 (cherry picked from 9354351fcb8baeaab85250d14409cfb4fa50f3e9 commit) 17464 17465commit 0e6d5e979aacb0c295ce79369ecc5f22ffa7922c 17466Author: Jamey Sharp <jamey@minilop.net> 17467Date: Wed Jun 7 20:29:05 2006 -0700 17468 17469 Fix bug #7035: unnecessary memmove in XOpenDisplay. 17470 Using memmove on the connection setup data causes a problem for XCB, but making 17471 Xlib stop doing that should be harmless for non-XCB as well. 17472 (cherry picked from b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0 commit) 17473 17474commit cd7328c46ae72903ed02832828891b2dab4d5ee0 17475Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17476Date: Sat Jun 3 13:51:51 2006 +0300 17477 17478 Fix threading support on GNU/kFreeBSD systems. (Robert Millan) 17479 (cherry picked from e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb commit) 17480 17481commit c336eb6b80a6f91da1d0b3d28634a2cfde324670 17482Merge: 33556ca8 ad9ebbd2 17483Author: Donnie Berkholz <donnie@comet.(none)> 17484Date: Thu Jun 22 14:25:35 2006 -0700 17485 17486 Merge branch 'stable' of http://people.freedesktop.org/~jamey/libX11 into stable 17487 17488commit 213dacad21740466e7ab31b01d3fc513fe4b3e74 17489Author: Daniel Stone <daniel@fooishbar.org> 17490Date: Thu Jun 22 17:20:59 2006 +0300 17491 17492 Bump to 1.0.99.0 to avoid confusion. 17493 17494commit efedfd68e31bcee2d21ac340be8dc9e1825ec890 17495Merge: e3acee88 4b8eb5d4 17496Author: Daniel Stone <daniel@fooishbar.org> 17497Date: Thu Jun 22 16:53:45 2006 +0300 17498 17499 Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11 17500 17501commit 4b8eb5d4a1da73a94b5a6ab12e34784aae4c79c5 17502Merge: 5169d0e0 eff50c94 17503Author: Matthieu Herrb <matthieu@deville.herrb.com> 17504Date: Tue Jun 20 21:05:15 2006 +0200 17505 17506 Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11 17507 17508commit 5169d0e08ff6acb350a6ea768623f5ff0b85b05f 17509Author: Matthieu Herrb <matthieu@deville.herrb.com> 17510Date: Tue Jun 20 21:04:03 2006 +0200 17511 17512 Check setuid() return value. 17513 17514commit eff50c94a07194297e705da53d9fbb3a40fb9ad4 17515Author: Derek Wang <derek.wang@sun.com> 17516Date: Mon Jun 19 11:05:37 2006 -0700 17517 17518 Sun bug 6209243: XExtentsOfFontSet causes segfault when font set not loaded 17519 17520commit 94f3213fc4bd6ec49bfb68e8b4a4fddea2bf3baa 17521Author: Scott Revelt <scott.revelt@sun.com> 17522Date: Fri Jun 16 19:11:13 2006 -0700 17523 17524 Sun bug 4022903: Xcms routines may fail if sscanf() is looking for separators 17525 based on locale that doesn't match those used in the Xcms.txt 17526 17527commit c33d7b8282ac196b36229be26442296768c16f3e 17528Author: Alan Coopersmith <alanc@alf.(none)> 17529Date: Fri Jun 16 18:53:33 2006 -0700 17530 17531 Add *~ to ignore emacs droppings 17532 17533commit 1f4c9893ade08bad30c9bd12a36bee57d30b001e 17534Author: Matthias Hopf <mhopf@suse.de> 17535Date: Fri Jun 16 15:36:40 2006 +0200 17536 17537 Bug #3104: Compose table cache for faster X11 application starts. 17538 17539commit 4fe22647e6010a2886c2f3a7093adeaeb6678faa 17540Author: Matthias Hopf <mhopf@suse.de> 17541Date: Tue Jun 13 20:23:46 2006 +0200 17542 17543 Bug #3104: Compose table cache for faster X11 application starts. 17544 17545commit 1d28a655629a11ea7fd1e5df4c7b77dd4b63e3be 17546Author: Lubos Lunak <llunak@suse.de> 17547Date: Mon Jun 12 18:48:08 2006 +0200 17548 17549 Bug #3104: Compose table cache for faster X11 application starts. 17550 17551commit 9354351fcb8baeaab85250d14409cfb4fa50f3e9 17552Author: Matthias Hopf <mhopf@suse.de> 17553Date: Fri Jun 9 18:24:02 2006 +0200 17554 17555 Bug #3104: Compose table cache for faster X11 application starts. Part 1: Pointerless compose data structure, using indices instead of pointers, needed for mmap()ing data structure. 17556 17557commit b18713ec3f36a10b3cdb1e16f9550e1d2e05dff0 17558Author: Jamey Sharp <jamey@minilop.net> 17559Date: Wed Jun 7 20:29:05 2006 -0700 17560 17561 Fix bug #7035: unnecessary memmove in XOpenDisplay. 17562 Using memmove on the connection setup data causes a problem for XCB, but making 17563 Xlib stop doing that should be harmless for non-XCB as well. 17564 17565commit ad9ebbd2424bc2699944ffdf4e19e13f9dd8ab84 17566Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17567Date: Sat Jun 3 12:57:55 2006 +0300 17568 17569 Bug #2186: Add cs_CZ.iso8859-2 alias. 17570 17571commit 9e7765e0b1cbaae6643072d91066ba1201b36227 17572Author: Daniel Stone <daniel@fooishbar.org> 17573Date: Fri Jun 2 02:46:29 2006 +0300 17574 17575 nls: Serbian (sr_CS) update (#5575) 17576 17577 Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a 17578 corresponding locale change. Update compose.dir.pre, locale.alias.pre, 17579 and locale.dir.pre. (Milos Komarcevic) 17580 17581commit a4ac2242b588da23044a20aa999ae84d4de7b2d8 17582Author: Daniel Stone <daniel@fooishbar.org> 17583Date: Fri Jun 2 02:24:25 2006 +0300 17584 17585 im: add Braille input method (#6296) 17586 17587 Bug #6296: Add a Braille input method. (Samuel Thibault) 17588 17589commit 90de1e2e141ec591048a76cb695579ef809a28d3 17590Author: Daniel Stone <daniel@fooishbar.org> 17591Date: Fri Jun 2 02:22:17 2006 +0300 17592 17593 xkb support: small typo 17594 17595commit 4c3e34bece7402f08139d34d1ef5834e3cf533c7 17596Author: Daniel Stone <daniel@fooishbar.org> 17597Date: Fri Jun 2 01:50:24 2006 +0300 17598 17599 en_US.UTF-8 Compose.pre: updates from Simos (#5129) 17600 17601 Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms, 17602 adding Unicode character names, removing duplicate entries, et al. 17603 17604commit 0c6473dd329c7334ae511884bcb6e73e632c784f 17605Author: Daniel Stone <daniel@fooishbar.org> 17606Date: Fri Jun 2 01:44:53 2006 +0300 17607 17608 nls: fix use of non-keysym dead_space (#5107) 17609 17610 Bug #5107: Change users of dead_space to space. 17611 17612commit 6f99f6349de5120f1cb7e02fbc97849341bc48e8 17613Author: Daniel Stone <daniel@fooishbar.org> 17614Date: Fri Jun 2 01:41:18 2006 +0300 17615 17616 optional XKB support fix 17617 17618 Fix compilation with --disable-xkb. 17619 17620commit 217d43ed44ced901122093af3ef1294e1736bb77 17621Author: Daniel Stone <daniel@fooishbar.org> 17622Date: Fri Jun 2 01:39:12 2006 +0300 17623 17624 i18n: separate data and lib directories 17625 17626 Break out locale data into separate data and library directories, under 17627 $(datadir) and $(libdir), respectively, by default. 17628 17629commit 92fa7fcde8df22830fca7c0275ab201033f7909c 17630Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17631Date: Thu May 11 14:04:48 2006 -0400 17632 17633 libXcursor.so.1, not libXcursor.so 17634 17635commit 135b4df13ed7c35dbae8975f302fc1fb8412d7c0 17636Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17637Date: Wed May 10 18:06:03 2006 +0300 17638 17639 XKBMisc.c: use Xfree, not xfree 17640 17641 Use Xfree() instead of xfree() when freeing interps. 17642 17643commit 3518d772b08e3433bc28b4d8d293fa53ca25f0ee 17644Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17645Date: Wed May 10 14:51:37 2006 +0300 17646 17647 locale.alias.pre: bg_BG typo fix 17648 17649 Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre. 17650 17651commit cc533db60cb64dc163c66451933a9bf77c519062 17652Author: Daniel Stone <daniels@endtroducing.localdomain> 17653Date: Sun Apr 9 22:22:03 2006 +0300 17654 17655 Coverity #203, #204: Fix potential NULL dereferences. 17656 17657commit b83adf7dfd6157694fe4f232012fef36cd9666da 17658Author: Daniel Stone <daniels@endtroducing.localdomain> 17659Date: Sun Apr 9 22:20:25 2006 +0300 17660 17661 Coverity #205: Fix potential NULL dereference. 17662 17663commit 2d0cd10ad907864d0136739eaac459779c9a5332 17664Author: Daniel Stone <daniels@endtroducing.localdomain> 17665Date: Sun Apr 9 22:18:20 2006 +0300 17666 17667 Coverity #209: Fix potential NULL dereference. (Alan Coopersmith) 17668 17669commit dc2f3966068d66a564aa452cab9f0c26657fa1df 17670Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17671Date: Fri Apr 7 18:11:52 2006 +0300 17672 17673 Coverity #826: Fix potential memory leak. 17674 17675commit 23df609ec451a01c77e8f31ecc85c5af7c62efed 17676Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17677Date: Fri Apr 7 17:49:41 2006 +0300 17678 17679 Bug #1625: Include keysym.h from Xutil.h. 17680 17681commit 5262a1945c543a3419ed626e1deb09ef5b4584c1 17682Author: Eric Anholt <anholt@leguin.anholt.net> 17683Date: Wed Apr 5 17:12:15 2006 -0700 17684 17685 Check if visualList == NULL, not nVisualsMatched == 0. NULL happens in more 17686 17687 cases (allocation failure) than nVisualsMatched == 0. Noticed from inspection 17688 of Coverity #599, #600. 17689 17690commit 8b42635f577468bb143ca593cdd9fb3450ad712c 17691Author: Eric Anholt <anholt@leguin.anholt.net> 17692Date: Wed Apr 5 16:42:26 2006 -0700 17693 17694 Coverity #558: Free newly-allocated Database in error path. 17695 17696commit 6d06e41d1f431b3f1a1fcf69161e0af411325e9f 17697Author: Eric Anholt <anholt@leguin.anholt.net> 17698Date: Wed Apr 5 16:38:52 2006 -0700 17699 17700 Coverity #582: Free newly-allocated region in error path. 17701 17702commit 5fd8f79ad3e38df74d9a6cb573617542c101df1a 17703Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17704Date: Fri Mar 17 15:58:39 2006 +0200 17705 17706 Properly clip bounds when only one point is defining an outline. 17707 17708commit 1e1572eb7f8394ce152e99d96f711ccf9083baf9 17709Author: Matthieu Herrb <matthieu@bluenote.herrb.com> 17710Date: Sat Mar 11 15:39:15 2006 +0100 17711 17712 Fix prototype of XConfigureWindow(). Bugzilla #6023. 17713 17714commit 1da8bd904f1fc79d63d368473531b438d08bfe37 17715Author: Jeremy C. Reed <reed@reedmedia.net> 17716Date: Tue Feb 21 14:10:22 2006 -0800 17717 17718 Set XTHREADLIB correctly for dragonfly platforms. 17719 17720commit c9768133e3f65ac4bb64e3941e2d6ae36897ec9c 17721Author: Jamey Sharp <jamey@minilop.net> 17722Date: Sun Feb 19 12:29:27 2006 -0800 17723 17724 Update .gitignores for *.o and nls/locale.dir*. 17725 17726commit efcbde6ba0b770bb0d4c7818e307712161011f10 17727Author: Jamey Sharp <jamey@minilop.net> 17728Date: Sun Feb 19 12:28:41 2006 -0800 17729 17730 Move .cvsignore to .gitignore. 17731 17732commit e3acee88cfcc4ef0fa8a7db39763a5ebe2e985cb 17733Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17734Date: Sat Jun 3 13:51:51 2006 +0300 17735 17736 Fix threading support on GNU/kFreeBSD systems. (Robert Millan) 17737 17738commit f1bd3152359ddfadd0d043006036c239f3e2907b 17739Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17740Date: Sat Jun 3 12:57:55 2006 +0300 17741 17742 Bug #2186: Add cs_CZ.iso8859-2 alias. 17743 17744commit 2b1b79d90db1d7f0472718b8c010c36275994195 17745Author: Daniel Stone <daniel@fooishbar.org> 17746Date: Fri Jun 2 02:46:29 2006 +0300 17747 17748 nls: Serbian (sr_CS) update (#5575) 17749 Bug #5575: 'Yugoslavia' has changed to Serbia & Montenegro, along with a 17750 corresponding locale change. Update compose.dir.pre, locale.alias.pre, 17751 and locale.dir.pre. (Milos Komarcevic) 17752 17753commit d6fba1f44d404362d3be1b421f57d7ccc3c8cdac 17754Author: Daniel Stone <daniel@fooishbar.org> 17755Date: Fri Jun 2 02:24:25 2006 +0300 17756 17757 im: add Braille input method (#6296) 17758 Bug #6296: Add a Braille input method. (Samuel Thibault) 17759 17760commit 0fed7d3185addd610e917dcdaa0676f0256c0ec5 17761Author: Daniel Stone <daniel@fooishbar.org> 17762Date: Fri Jun 2 02:22:17 2006 +0300 17763 17764 xkb support: small typo 17765 17766commit cf7d9f9e46f3ce01ac04a95978918d5c0f3f3cf9 17767Author: Daniel Stone <daniel@fooishbar.org> 17768Date: Fri Jun 2 01:50:24 2006 +0300 17769 17770 en_US.UTF-8 Compose.pre: updates from Simos (#5129) 17771 Bug #5129: Numerous updates from Simos Xenitellis, fixing Unicode keysyms, 17772 adding Unicode character names, removing duplicate entries, et al. 17773 17774commit 332d45fce9fdbf59168d90a133af1f580a589e54 17775Author: Daniel Stone <daniel@fooishbar.org> 17776Date: Fri Jun 2 01:44:53 2006 +0300 17777 17778 nls: fix use of non-keysym dead_space (#5107) 17779 Bug #5107: Change users of dead_space to space. 17780 17781commit 34f59ce3d1e2eb2971b732d11871c6fff8a1c75b 17782Author: Daniel Stone <daniel@fooishbar.org> 17783Date: Fri Jun 2 01:41:18 2006 +0300 17784 17785 optional XKB support fix 17786 Fix compilation with --disable-xkb. 17787 17788commit c5940a0b85edec4003f91a59fc3c44f538accfe1 17789Author: Daniel Stone <daniel@fooishbar.org> 17790Date: Fri Jun 2 01:39:12 2006 +0300 17791 17792 i18n: separate data and lib directories 17793 Break out locale data into separate data and library directories, under 17794 $(datadir) and $(libdir), respectively, by default. 17795 17796commit 33556ca81db2419b9d2a37664c4cea2069414b37 17797Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17798Date: Fri May 12 14:49:17 2006 -0400 17799 17800 Bump to 1.0.1 17801 17802commit 80d88557626fae9debc404de33d7fb5d69a6049d 17803Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17804Date: Thu May 11 14:06:28 2006 -0400 17805 17806 libXcursor.so.1, not libXcursor.so 17807 17808commit 5384f27dfe3c94c462c137ab3540eaa5646ca4c6 17809Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17810Date: Thu May 11 14:04:48 2006 -0400 17811 17812 libXcursor.so.1, not libXcursor.so 17813 17814commit 01f4d433eed6b70c6e9636157acac022054fdeb6 17815Author: Jamey Sharp <jamey@minilop.net> 17816Date: Wed May 10 17:02:52 2006 -0700 17817 17818 Count any partial request towards the current Xlib sequence number. 17819 17820commit 770cfbd1fcc80a83a9be0c4f68727b8af2c8f4a4 17821Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17822Date: Wed May 10 18:06:03 2006 +0300 17823 17824 XKBMisc.c: use Xfree, not xfree 17825 Use Xfree() instead of xfree() when freeing interps. 17826 17827commit 22a5255b80b80772612279bc840a953edd0e3442 17828Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17829Date: Wed May 10 14:51:37 2006 +0300 17830 17831 locale.alias.pre: bg_BG typo fix 17832 Fix typo (be_BG.UTF-8 rather than bg_BG.UTF-8) in locale.alias.pre. 17833 17834commit 9cac8c9824874ca7d835f001a4efa910b7fdd822 17835Merge: 19b8840a e514bc87 17836Author: Daniel Stone <daniels@preemptive.fooishbar.org> 17837Date: Wed May 10 14:50:37 2006 +0300 17838 17839 Merge branch 'master' of git+ssh://git.freedesktop.org/srv/git.freedesktop.org/git/xorg/lib/libX11 17840 17841commit e514bc875f27f4bf197b06b8315eeca526195915 17842Author: Jamey Sharp <jamey@minilop.net> 17843Date: Tue May 9 12:41:59 2006 -0700 17844 17845 Assert that dpy->request does not go backwards. Catches #5839 earlier. 17846 17847commit fc1159137365a599bf611ee001f439416952c4e0 17848Author: Jamey Sharp <jamey@minilop.net> 17849Date: Sun May 7 17:40:01 2006 -0700 17850 17851 In _XPutXCBBuffer, set aside any trailing partial request until the last byte is available. 17852 17853commit c394480a4247213239822808e3f6e7c6cd6decd9 17854Author: Jamey Sharp <jamey@minilop.net> 17855Date: Sun May 7 16:58:13 2006 -0700 17856 17857 Update for XCB ConnSetupSuccessRep name change. 17858 17859commit 7672bf93bc1200905461aeb0a2dc2c2696410b93 17860Author: Daniel Stone <daniels@endtroducing.localdomain> 17861Date: Sun Apr 9 22:22:03 2006 +0300 17862 17863 Coverity #203, #204: Fix potential NULL dereferences. 17864 17865commit cfcafbe48d22d9a0cd50eb9454ce0ff88f0129d3 17866Author: Daniel Stone <daniels@endtroducing.localdomain> 17867Date: Sun Apr 9 22:20:25 2006 +0300 17868 17869 Coverity #205: Fix potential NULL dereference. 17870 17871commit b6771501feea98d037032f82117b6d4f15779f07 17872Author: Daniel Stone <daniels@endtroducing.localdomain> 17873Date: Sun Apr 9 22:18:20 2006 +0300 17874 17875 Coverity #209: Fix potential NULL dereference. (Alan Coopersmith) 17876 17877commit 19b8840af241087bb17b1edabcaa9b28fdd0a1dc 17878Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17879Date: Fri Apr 7 18:11:52 2006 +0300 17880 17881 Coverity #826: Fix potential memory leak. 17882 17883commit 3a16f262abe48b44ed641525e894bc22e13bf72a 17884Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17885Date: Fri Apr 7 17:49:41 2006 +0300 17886 17887 Bug #1625: Include keysym.h from Xutil.h. 17888 17889commit 2481b767ae96e2f2503c0390545932c8397b090f 17890Author: Eric Anholt <anholt@leguin.anholt.net> 17891Date: Wed Apr 5 17:12:15 2006 -0700 17892 17893 Check if visualList == NULL, not nVisualsMatched == 0. NULL happens in more 17894 cases (allocation failure) than nVisualsMatched == 0. Noticed from inspection 17895 of Coverity #599, #600. 17896 17897commit 152b17e47d878c2d928eb74581aa69d925a29123 17898Author: Eric Anholt <anholt@leguin.anholt.net> 17899Date: Wed Apr 5 16:42:26 2006 -0700 17900 17901 Coverity #558: Free newly-allocated Database in error path. 17902 17903commit bc62b99ef36edb34035911c42104be7f6f9d2333 17904Author: Eric Anholt <anholt@leguin.anholt.net> 17905Date: Wed Apr 5 16:38:52 2006 -0700 17906 17907 Coverity #582: Free newly-allocated region in error path. 17908 17909commit d47f0b3cec1388f7ce60ab2af91df0dea0f221c5 17910Author: Jamey Sharp <jamey@minilop.net> 17911Date: Fri Mar 31 22:53:07 2006 -0800 17912 17913 Fix buggy interaction with XCB when running out of XIDs. 17914 17915commit e3f452571824d6a875bbf582946de185de9e01e9 17916Author: Jamey Sharp <jamey@minilop.net> 17917Date: Fri Mar 31 22:52:14 2006 -0800 17918 17919 Add explicit include of Xmd.h to work around bug including both xcb.h and Xmd.h simultaneously. 17920 17921commit 2363b74ca795c1b3a73c9e572532ba5191adec5b 17922Author: Daniel Stone <daniels@preemptive.research.nokia.com> 17923Date: Fri Mar 17 15:58:39 2006 +0200 17924 17925 Properly clip bounds when only one point is defining an outline. 17926 17927commit e876efb8aa410f2f5c87aaaa7042f847c4ff96f7 17928Merge: be266b20 f71ea0bc 17929Author: Matthieu Herrb <matthieu@bluenote.herrb.com> 17930Date: Sat Mar 11 15:40:35 2006 +0100 17931 17932 Merge branch 'master' of git+ssh://herrb@git.freedesktop.org/git/xorg/lib/libX11 17933 17934commit be266b201dc13530a302a7572283ccd3f32aad87 17935Author: Matthieu Herrb <matthieu@bluenote.herrb.com> 17936Date: Sat Mar 11 15:39:15 2006 +0100 17937 17938 Fix prototype of XConfigureWindow(). Bugzilla #6023. 17939 17940commit f71ea0bc737c5a42e9e022b86e7ec3b4f846d31c 17941Author: Jamey Sharp <jamey@minilop.net> 17942Date: Fri Mar 3 11:08:41 2006 -0800 17943 17944 Update for XCBSendRequest sequence number API changes. 17945 17946commit a11d1b0ae674320cf9897f6a83ec08c65eca8d9b 17947Author: Jamey Sharp <jamey@minilop.net> 17948Date: Fri Mar 3 01:42:49 2006 -0800 17949 17950 Use the full_sequence from XCBGenericError/Event for setting last_request_read, and quit replacing _XSetLastRequestRead with an XCB-specific version. 17951 17952commit d8ba4ae7045b227f8b675628b9094dded02f1c08 17953Author: Jamey Sharp <jamey@minilop.net> 17954Date: Thu Mar 2 23:43:26 2006 -0800 17955 17956 Bugfix: Rely on XCBSendRequest to leave iov in a well-defined state, and place the spare iovecs at the beginning of the array. 17957 17958commit f9afb5a54435c30961306080e9358d4240ecb844 17959Author: Jamey Sharp <jamey@minilop.net> 17960Date: Thu Mar 2 23:34:19 2006 -0800 17961 17962 assert() that we will not infinite loop or read uninitialized memory. 17963 17964commit d3512ef3aae5b036a8ce6579318108f1ec20ee22 17965Author: Jamey Sharp <jamey@minilop.net> 17966Date: Thu Mar 2 15:58:52 2006 -0800 17967 17968 Quit relying on XCBSendRequest to pad to 4-byte boundaries and do it ourselves. 17969 17970commit fb590c15a740264ee867d15a2547072e43b21eed 17971Author: Jamey Sharp <jamey@minilop.net> 17972Date: Thu Mar 2 12:06:04 2006 -0800 17973 17974 Handle errors correctly when Xlib owns the event queue and XCB has the checked error feature. 17975 17976commit 8356ba37d307a9eda895a6bf41ef727bbfc9a695 17977Author: Jamey Sharp <jamey@minilop.net> 17978Date: Mon Feb 27 11:51:47 2006 -0800 17979 17980 Use the new XCBSendRequest flag, XCB_REQUEST_RAW, to hand a bag-o-bytes down uninterpreted. 17981 17982commit 07bdf1fbbf2418f866df1a2140d514dd3f035139 17983Author: Jamey Sharp <jamey@minilop.net> 17984Date: Sun Feb 26 15:46:01 2006 -0800 17985 17986 Update for new XCBSendRequest API. 17987 17988commit 409a08cff8347d39e0e6c53c9f380d21f221f5ac 17989Author: Jamey Sharp <jamey@minilop.net> 17990Date: Thu Feb 23 18:12:31 2006 -0800 17991 17992 Performance fix: Replace calls to XCBGetRequestRead with the new XCBGetQueuedRequestRead. Cuts a lot of syscalls. 17993 17994commit ec30a27341b97620b07dd886f98d1d7664a67685 17995Author: Jamey Sharp <jamey@minilop.net> 17996Date: Thu Feb 23 18:01:46 2006 -0800 17997 17998 Minor performance fix: Access dpy->xcl->connection directly instead of calling XCBConnectionOfDisplay. It happens a lot. 17999 18000commit 53c471c6a835d5cedeca99f2c97058d196a3fd7e 18001Author: Jamey Sharp <jamey@minilop.net> 18002Date: Thu Feb 23 11:46:09 2006 -0800 18003 18004 XCBFlush used to return non-positive on failure, and this test did not catch 0. Now it returns boolean: 0 or 1. Testing <= 0 covers both cases. I probably want to switch to a boolean test eventually. 18005 18006commit 41c0121a8718b530feaf7fe315b673d9b8defce2 18007Author: Jamey Sharp <jamey@minilop.net> 18008Date: Tue Feb 21 21:25:41 2006 -0800 18009 18010 Refactor the code that passes requests down to XCB into a separate issue_complete_request function. 18011 18012commit 67d06e0fe468dca22847aa14d3f917128f89f9cf 18013Author: Jamey Sharp <jamey@minilop.net> 18014Date: Tue Feb 21 15:33:05 2006 -0800 18015 18016 If we have not actually put the buffer back, deferred invariants may not hold. This is OK. 18017 18018commit 434bf80b4053ff1ba82adf65de1f76b4d3731bf1 18019Author: Jeremy C. Reed <reed@reedmedia.net> 18020Date: Tue Feb 21 14:10:22 2006 -0800 18021 18022 Set XTHREADLIB correctly for dragonfly platforms. 18023 18024commit b24834762e975bd319f9ab5c7cf790b2a02a9474 18025Author: Jamey Sharp <jamey@minilop.net> 18026Date: Tue Feb 21 14:03:26 2006 -0800 18027 18028 Sometimes functions other than _XUnlockDisplay call _XPutXCBBuffer. Some invariants appropriate for Unlock are not appropriate otherwise: move them to _XUnlockDisplay. 18029 18030commit 99b8defd0d5e6993071e21638128c9de2574b37d 18031Author: Jamey Sharp <jamey@minilop.net> 18032Date: Tue Feb 21 13:04:21 2006 -0800 18033 18034 Execute BeforeFlush hooks on complete buffers, not request-at-a-time. 18035 Traditional Xlib worked this way; I dunno why I changed it. 18036 18037commit 9b01e7849775749182052fe324df9d8e6ceeee99 18038Author: Jamey Sharp <jamey@minilop.net> 18039Date: Tue Feb 21 12:51:44 2006 -0800 18040 18041 Factor the XCBSendRequest call out of the conditional in _XPutXCBBuffer. 18042 18043commit 7ce7ac882de128955751a5307889db9d712d8a72 18044Author: Jamey Sharp <jamey@minilop.net> 18045Date: Tue Feb 21 12:44:30 2006 -0800 18046 18047 Quit using a triple-pointer. Almost as if I were a sensible person. 18048 18049commit 35a858be218cdbfa4593d44a67663d5c25297016 18050Author: Jamey Sharp <jamey@minilop.net> 18051Date: Tue Feb 21 12:21:01 2006 -0800 18052 18053 Remove the XCL_PAD macro. 18054 18055commit e741b70ed2542c5463c57dac44bc37328616733b 18056Author: Jamey Sharp <jamey@minilop.net> 18057Date: Sun Feb 19 12:29:27 2006 -0800 18058 18059 Update .gitignores for *.o and nls/locale.dir*. 18060 18061commit f25b4b00e1683b0d97dba46dac46d65a9c2270a6 18062Author: Jamey Sharp <jamey@minilop.net> 18063Date: Sun Feb 19 12:28:41 2006 -0800 18064 18065 Move .cvsignore to .gitignore. 18066 18067commit c7cda56eebaf6ab11403363be14d4948d7d8be38 18068Author: Jamey Sharp <jamey@minilop.net> 18069Date: Sun Feb 19 11:49:15 2006 -0800 18070 18071 Land XCB support on X.org HEAD. 18072 18073commit 881467b3032261791ef5ec61b3879bb68d0a3d8c 18074Author: Jamey Sharp <jamey@minilop.net> 18075Date: Sun Feb 19 02:14:11 2006 +0000 18076 18077 Merge the X11 and BIGREQS package sets: they were used in exactly the same places. 18078 18079commit 6b0158dfad714db5b89c04dbea3aedeafa0fb146 18080Author: Jamey Sharp <jamey@minilop.net> 18081Date: Tue Feb 14 19:37:36 2006 +0000 18082 18083 Refactor _XFlush and _XSend code that sets dpy->synchandler to _XSeqSyncFunction into a new function, _XSetSeqSyncFunction. It makes the patch for XCB cleaner, but is arguably a good idea anyway. 18084 18085commit 010c3acbb3a6993d39274f42d88c00849acb0fb0 18086Author: Alan Coopersmith <alan.coopersmith@sun.com> 18087Date: Sun Feb 12 18:19:17 2006 +0000 18088 18089 Bug #5628 <https://bugs.freedesktop.org/show_bug.cgi?id=5628> Shadow pages not created correctly when MANDIR & MANSUFFIX don't match. 18090 18091commit fe8c01c80263457f01b70dc1511b2bd9466b7c96 18092Author: Alan Coopersmith <alan.coopersmith@sun.com> 18093Date: Fri Feb 3 23:34:43 2006 +0000 18094 18095 Fix typo in .TH line 18096 18097commit b091c217f3c3f60dde78b09a95c150df6c83d7ba 18098Author: Kevin E Martin <kem@kem.org> 18099Date: Thu Dec 15 00:24:28 2005 +0000 18100 18101 Update package version number for final X11R7 release candidate. 18102 18103commit e7c04e0e65a2a0c70c6ad29ec2d6f4350fd81c2a 18104Author: Kevin E Martin <kem@kem.org> 18105Date: Tue Dec 6 22:48:41 2005 +0000 18106 18107 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 18108 18109commit b5c495854d5270e64e6d588388ffa906bfcaac22 18110Author: Kevin E Martin <kem@kem.org> 18111Date: Sat Dec 3 05:49:42 2005 +0000 18112 18113 Update package version number for X11R7 RC3 release. 18114 18115commit 9e96dbc343c7f27ff47607acd75378ab23903e2a 18116Author: Kevin E Martin <kem@kem.org> 18117Date: Sat Dec 3 04:41:47 2005 +0000 18118 18119 Add check and cflags for malloc(0) returning NULL. 18120 18121commit 649c37b47909620ccafde3e983de8321cddd74ce 18122Author: Alan Coopersmith <alan.coopersmith@sun.com> 18123Date: Mon Nov 28 22:03:04 2005 +0000 18124 18125 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 18126 18127commit 8d1500df66d796ebff2b0d8c02205e5fa6796d4a 18128Author: Alan Coopersmith <alan.coopersmith@sun.com> 18129Date: Wed Nov 23 22:33:06 2005 +0000 18130 18131 Bug #5003 <https://bugs.freedesktop.org/show_bug.cgi?id=5003> Patch #3763 <https://bugs.freedesktop.org/attachment.cgi?id=3763> Xorg code misuses S_IF* macros 18132 18133commit 93cf3747f9ae8d30bd485b41c5ff10397f68f078 18134Author: Eric Anholt <anholt@freebsd.org> 18135Date: Sun Nov 20 23:17:39 2005 +0000 18136 18137 Add/improve libs .cvsignores. 18138 18139commit fa1f4a08112bfa14d3758f4702733dd3892966c3 18140Author: Kevin E Martin <kem@kem.org> 18141Date: Sat Nov 19 07:15:39 2005 +0000 18142 18143 Update pkgconfig files to separate library build-time dependencies from application build-time dependencies, and update package deps to work with separate build roots. 18144 18145commit 7012f9b56c594cf40855ba0dbf93af1263417a8c 18146Author: Kevin E Martin <kem@kem.org> 18147Date: Mon Nov 14 21:51:07 2005 +0000 18148 18149 Fix xf86bigfont pkgconfig dep typo. 18150 18151commit 60a72f47951f46b4e0505d9903a94af3b6bed8ca 18152Author: Kevin E Martin <kem@kem.org> 18153Date: Wed Nov 9 21:19:12 2005 +0000 18154 18155 Update package version number for X11R7 RC2 release. 18156 18157commit 6bb0c3796b6e1beddc376a896e865704886e1462 18158Author: Kean Johnson <kean@armory.com> 18159Date: Tue Nov 8 06:33:25 2005 +0000 18160 18161 See ChangeLog entry 2005-11-07 for details. 18162 18163commit 14be0098ad90c3e68bd2d21b00ffabb76f1fd780 18164Author: Kevin E Martin <kem@kem.org> 18165Date: Tue Nov 1 15:11:50 2005 +0000 18166 18167 Update pkgcheck dependencies to work with separate build roots. 18168 18169commit 1bf71462a972e0fc56de63f5c7fd613b37fc70f1 18170Author: Donnie Berkholz <spyderous@gentoo.org> 18171Date: Fri Oct 28 10:44:03 2005 +0000 18172 18173 Revert that, it's redundant. But it is worth noting that --disable-xkb is broken, if anyone cares. 18174 18175commit be627a39fe373e2e81fdc263780e70a271d9e0c5 18176Author: Donnie Berkholz <spyderous@gentoo.org> 18177Date: Fri Oct 28 08:28:08 2005 +0000 18178 18179 Add dependency on inputproto for XI.h. 18180 18181commit 5fae4cb456cb03fb70cd065dbc2ca94c8ed99082 18182Author: Adam Jackson <ajax@nwnk.net> 18183Date: Fri Oct 21 18:44:24 2005 +0000 18184 18185 Bug #4736: Error decoding for Damage extension. (Jonathan Lennox) 18186 18187commit 1171fa9dc77413f0e90933a565ec07068052afb4 18188Author: Kevin E Martin <kem@kem.org> 18189Date: Wed Oct 19 02:48:08 2005 +0000 18190 18191 Update package version number for RC1 release. 18192 18193commit 2a2d905706308b9d5a1c16af1067fb390f43850c 18194Author: Alan Coopersmith <alan.coopersmith@sun.com> 18195Date: Tue Oct 18 00:00:08 2005 +0000 18196 18197 Use @LIB_MAN_SUFFIX@ instead of $(LIB_MAN_SUFFIX) in macro substitutions to work better with BSD make 18198 18199commit 66d35b6971fb26762392a2a8e2c47db46c11116a 18200Author: Alan Coopersmith <alan.coopersmith@sun.com> 18201Date: Mon Oct 17 21:13:15 2005 +0000 18202 18203 Rename .shadows.DONE to shadows.DONE to avoid some make's thinking it's a suffix rule (reported by Matthieu Herrb) 18204 18205commit a316995a17c084e98ef1b7f25d287c2c08b6d749 18206Author: Donnie Berkholz <spyderous@gentoo.org> 18207Date: Sun Oct 16 03:03:35 2005 +0000 18208 18209 Change '==' to portable '='. 18210 18211commit b76a072530e884bcbea6ed8fed5aef39361dcfc0 18212Author: Alan Coopersmith <alan.coopersmith@sun.com> 18213Date: Wed Oct 12 00:04:50 2005 +0000 18214 18215 configure.ac Use XORG_MAN_SECTIONS instead of custom man section configuration. Add shadow man pages for man pages that document multiple functions. 18216 18217commit 8ee5c1429af98206e05a0536f87c0f21a529cdf0 18218Author: Eric Anholt <anholt@freebsd.org> 18219Date: Tue Oct 11 02:18:36 2005 +0000 18220 18221 Add appropriate pthread libs/flags for FreeBSD, fixing the build of ico and probably others. 18222 18223commit 41ff3b9d1f194a7b56437b650d5f589225c078c6 18224Author: Alan Coopersmith <alan.coopersmith@sun.com> 18225Date: Sun Oct 9 22:28:39 2005 +0000 18226 18227 Bug #3021 <https://bugs.freedesktop.org/show_bug.cgi?id=3021> Requests and Errors for XFixes are not in XErrorDB (Jonathan Lennox) 18228 18229commit 91ed79852e790049ab54e68f288afb3c953194c9 18230Author: Kevin E Martin <kem@kem.org> 18231Date: Fri Oct 7 15:00:00 2005 +0000 18232 18233 Clean up generated files 18234 Add missing dist tarball files 18235 18236commit 440399b470c97b159a530602fff11c315aca8d97 18237Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18238Date: Wed Oct 5 18:27:10 2005 +0000 18239 18240 Add el_GR.UTF-8 compose file 18241 18242commit 010f0647e25ac617d0f92c8d2b8dda684da545db 18243Author: Daniel Stone <daniel@fooishbar.org> 18244Date: Fri Sep 30 07:52:46 2005 +0000 18245 18246 Bug #2609: Add Kyrgyz locale (Ilyas Bakirov). 18247 18248commit 3ef2fb67bd8905b208ad7eb790c3843e14cea7ed 18249Author: Daniel Stone <daniel@fooishbar.org> 18250Date: Fri Sep 30 07:47:55 2005 +0000 18251 18252 Bug #1640: Kinyarwanda locale support. (Steve Murphy) 18253 18254commit 4ae0decabe0960870df0ec165f495166c10a053c 18255Author: Daniel Stone <daniel@fooishbar.org> 18256Date: Fri Sep 30 07:40:03 2005 +0000 18257 18258 Bug #2268: Add South African locales (Dwayne Bailey). Some whitespace cleanups, as the parser is a little touchy. 18259 18260commit 3f79eb4c99844f618f1889741d1631c2ffe5385f 18261Author: Daniel Stone <daniel@fooishbar.org> 18262Date: Fri Sep 30 07:11:19 2005 +0000 18263 18264 Generate locale.alias and compose.dir exactly like we do in the monolith, which fixes non-UTF-8 locales in particular (so we get foo: bar, as well as foo bar). Switch to generating locale.dir the same way. 18265 18266commit d1237d1483ff972c76a0ac344ec97d5280db0007 18267Author: Alan Coopersmith <alan.coopersmith@sun.com> 18268Date: Thu Sep 29 21:27:12 2005 +0000 18269 18270 Add Xcms.txt to lib/X11 18271 18272commit 4a86f299693f7376cbe98175f0b0c44d691802b3 18273Author: Alan Coopersmith <alan.coopersmith@sun.com> 18274Date: Sat Sep 24 20:11:06 2005 +0000 18275 18276 Add XQueryExtension.man 18277 18278commit a057a66e2041d45198a13a4ece7c07068f76f21b 18279Author: Alan Coopersmith <alan.coopersmith@sun.com> 18280Date: Sat Sep 24 00:16:32 2005 +0000 18281 18282 Add XTHREAD_CFLAGS for platforms that need special defines like 18283 -D_REENTRANT or -D_POSIX_whatever to get re-entrant function definitions. Set XDMCP_LIBS correctly for later libXdmcp tests. 18284 18285commit 3e920a65a7c376ad63eae2240fd06904d25d18bf 18286Author: Alan Coopersmith <alan.coopersmith@sun.com> 18287Date: Fri Sep 2 23:00:30 2005 +0000 18288 18289 Issue an error if XTRANS macros were not found when generating configure from configure.ac to flag errors early, instead of when people wonder why libX11 can't talk to an Xserver because it has no transports defined 18290 18291commit 3eb9f2d693af89d04e2fd92492c8205dce332c9c 18292Author: Kristian Høgsberg <krh@redhat.com> 18293Date: Thu Sep 1 19:24:13 2005 +0000 18294 18295 Use $(X11_LOCALEDATADIR) instead of @X11_LOCALEDATADIR@ so this install destination can be overridden at make install time. 18296 18297commit 7afa64325183b78d2d6a4862821f8b3e9866105c 18298Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18299Date: Sun Aug 28 19:45:48 2005 +0000 18300 18301 OpenBSD needs -lpthread for threaded applications too. 18302 18303commit 2b2f3d3877cb7927f196d01a5df6a27bf8d0518a 18304Author: Daniel Stone <daniel@fooishbar.org> 18305Date: Fri Aug 26 05:16:46 2005 +0000 18306 18307 Define ERRORDB/KEYSYMDB to XERRORDB/XKEYSYMDB if the former is undefined but the latter is. (Gerte Hoogewerf) 18308 18309commit 07066da0902df91c71f2adb81d1a17ec29165553 18310Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18311Date: Sun Aug 21 15:45:04 2005 +0000 18312 18313 update 18314 18315commit de44d8b111f57bd2f015e085fd8298c5f2a15ef3 18316Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18317Date: Sun Aug 21 15:38:39 2005 +0000 18318 18319 Threads support for BSD systems: 18320 - need to check for gewtpwuid_r to define mtsafeapi 18321 - build UIThrstubs if needed. 18322 18323commit 60217fdb918bafb2082519efe5cba3b13ad3082a 18324Author: Adam Jackson <ajax@nwnk.net> 18325Date: Wed Aug 17 19:46:08 2005 +0000 18326 18327 Add xthreadlib variable to x11.pc. Bump to 0.99.1. 18328 18329commit e1f4c6f5e36c1511f66fa1fac76520fd97eecbad 18330Author: Alan Coopersmith <alan.coopersmith@sun.com> 18331Date: Wed Aug 17 01:27:08 2005 +0000 18332 18333 Fix more broken multi-line .ds macros. Remove extraneous ;'s . 18334 18335commit 1909786f4a7d686369edcfc05a938df115fab37c 18336Author: Alan Coopersmith <alan.coopersmith@sun.com> 18337Date: Tue Aug 16 19:23:15 2005 +0000 18338 18339 Bugzilla #4112 <https://bugs.freedesktop.org/show_bug.cgi?id=4112> Patch #2687 <https://bugs.freedesktop.org/attachment.cgi?id=2897> Fix multi-line macros in XPutImage man page. (Debian bug #323210, fix by David Mart?nez Moreno) 18340 18341commit 83406d69c62070d2eeef23eb47f1ca887f711ee5 18342Author: Alan Coopersmith <alan.coopersmith@sun.com> 18343Date: Mon Aug 15 19:53:37 2005 +0000 18344 18345 Move RAWCPP macro to xorg-macros.m4 so other modules can use it Add check for whether or not RAWCPP needs -traditional instead of hardcoding it, so non-gcc cpp's can be used 18346 18347commit 5bb43de17de8e71d967488a713bf2b3448533444 18348Author: Alan Coopersmith <alan.coopersmith@sun.com> 18349Date: Sat Aug 6 18:59:49 2005 +0000 18350 18351 Typo fix in output message 18352 18353commit afe34b95862bb3c06cdbe724cb5ec3001a4a5215 18354Author: Alan Coopersmith <alan.coopersmith@sun.com> 18355Date: Thu Aug 4 02:55:49 2005 +0000 18356 18357 //bugs.freedesktop.org/show_bug.cgi?id=1887> Patch #3005 <https://bugs.freedesktop.org/attachment.cgi?id=3005> libX11 locale defs severely adrift from glibc: adding new aliases (From Debian via Nathanael Nerode) 18358 18359commit 3979a0b88edf6475ce5cfaa386e18ef980bda13c 18360Author: Alan Coopersmith <alan.coopersmith@sun.com> 18361Date: Thu Aug 4 02:51:30 2005 +0000 18362 18363 //bugs.freedesktop.org/show_bug.cgi?id=1887> Patch #3002 <https://bugs.freedesktop.org/attachment.cgi?id=3002> libX11 locale defs severely adrift from glibc: locales bugfix for bs_BA (From Debian via Nathanael Nerode) 18364 18365commit 34b454df192f4563499c453ccdb8c079f4a20cbe 18366Author: Alan Coopersmith <alan.coopersmith@sun.com> 18367Date: Sat Jul 30 20:30:46 2005 +0000 18368 18369 Include config.h so Xtrans knows which transport types to build code for 18370 18371commit cd9c9936b49c125eda779b99887d7e6ae4cf56cd 18372Author: Alan Coopersmith <alan.coopersmith@sun.com> 18373Date: Sat Jul 30 19:15:16 2005 +0000 18374 18375 Add -D flags to clear various warnings (Stefan Dirsch) 18376 18377commit e7fef67b4531faddd805d8f2157903006d3117ed 18378Author: Kevin E Martin <kem@kem.org> 18379Date: Fri Jul 29 21:22:50 2005 +0000 18380 18381 Various changes preparing packages for RC0: 18382 - Verify and update package version numbers as needed 18383 - Implement versioning scheme 18384 - Change bug address to point to bugzilla bug entry form 18385 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to reenable it) 18386 - Fix makedepend to use pkgconfig and pass distcheck 18387 - Update build script to build macros first 18388 - Update modular Xorg version 18389 18390commit 2ebb00244928237088e68325b1032b3550455ce9 18391Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18392Date: Sat Jul 23 20:19:31 2005 +0000 18393 18394 remove orphan TAB at begin of line 18395 18396commit dd7a9cdecda73e024ca84c5b9a22b18688038d94 18397Author: Kevin E Martin <kem@kem.org> 18398Date: Sat Jul 23 18:09:39 2005 +0000 18399 18400 Modify modular libs to use Xregion.h instead of region.h 18401 18402commit 6d635a88d91647b1b63611c3591f74916f88cd1c 18403Author: Kevin E Martin <kem@kem.org> 18404Date: Sat Jul 23 18:06:16 2005 +0000 18405 18406 lib/Xrender/Picture.c Change region.h to Xregion.h and modify internal references to include <X11/Xregion.h>. 18407 18408commit 36283f50fd9748733ae84cb7fb52ca8d9e661c15 18409Author: Daniel Stone <daniel@fooishbar.org> 18410Date: Sat Jul 16 06:25:35 2005 +0000 18411 18412 Set soversion to 6.2.0. 18413 18414commit 75fd5ae6e4683b9b9dcc13bc2f0faf223610a74f 18415Author: Keith Packard <keithp@keithp.com> 18416Date: Fri Jul 15 04:27:32 2005 +0000 18417 18418 Add missing Makefile.am to lib/X11/modules 18419 18420commit 6d84a8b1329af1fcfe86b198f1a6e7dd6ff616c3 18421Author: Keith Packard <keithp@keithp.com> 18422Date: Fri Jul 15 04:08:51 2005 +0000 18423 18424 Move i18n modules to top-level so they can be built in the right order (before xlib for non-loadable, after xlib for loadable). 18425 Link i18n modules against xlib to resolve Xlib symbols used by them. 18426 18427commit 45f40126a73295345bb5eb187b1167874842ab6e 18428Author: Alexander Gottwald <ago@freedesktop.org> 18429Date: Thu Jul 14 19:50:00 2005 +0000 18430 18431 Add $(top_srcdir)/src to include list 18432 18433commit 0aed7d91f5928d09d541617aad03709b5090658d 18434Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18435Date: Thu Jul 14 17:04:49 2005 +0000 18436 18437 Build fix for non-GNU make. 18438 18439commit b79422ccb02ab44548d1038956ab0cd4e2638645 18440Author: Adam Jackson <ajax@nwnk.net> 18441Date: Thu Jul 14 15:12:44 2005 +0000 18442 18443 typo fixes (Matthieu Herrb) 18444 18445commit 419304cde2fda19457c667870edefc0b227651b3 18446Author: Alan Coopersmith <alan.coopersmith@sun.com> 18447Date: Wed Jul 13 02:41:36 2005 +0000 18448 18449 Add missing backslashes to xlocale_la_SOURCES 18450 18451commit d14cc5c8964c4539b57c6cb51ef653292e410b79 18452Author: Keith Packard <keithp@keithp.com> 18453Date: Mon Jul 11 20:32:55 2005 +0000 18454 18455 Ammend AM_CFLAGS with all needed includes; cannot figure out an easy way to automate this. 18456 18457commit 424c2d8905eb2ad0a2df15b1da2f96140bfbcffb 18458Author: Lars Knoll <lars@trolltech.com> 18459Date: Mon Jul 11 15:24:32 2005 +0000 18460 18461 compile 18462 18463commit de9784eb1bde46efc316da279e3da27c6cc288a7 18464Author: Keith Packard <keithp@keithp.com> 18465Date: Mon Jul 11 09:26:40 2005 +0000 18466 18467 Ignore built man page files 18468 18469commit 0c258c36d1523113790c599b16d2947d7aa6469a 18470Author: Keith Packard <keithp@keithp.com> 18471Date: Mon Jul 11 09:18:31 2005 +0000 18472 18473 Minor changes to help modular Xlib build i18n modules 18474 18475commit 6e752ea1203b786423e40f43340bce15ca3de0f0 18476Author: Keith Packard <keithp@keithp.com> 18477Date: Mon Jul 11 08:29:18 2005 +0000 18478 18479 Enable loadable i18n modules, making them configurable on the configure command line. 18480 Clean up conditionals for XKB, XCMS, XLOCALEDIR 18481 Create new lib directory for locale modules in ${X11_LIBDIR}/locale/lib. Add this to the default XLOCALEDIR search path. 18482 Create separate X11_LOCALEDATADIR variable pointing at ${datadir}/X11/locale for installing locale data. 18483 Split out xcms, xkb, xlibi18n sources from main xlib bits so they can be conditionally included more easily. Lots of source files have been moved with this step; the result seems like it might be easier to maintain. 18484 Display message at end of configure script with selected options. 18485 Fix manual building with cpp to add -traditional in cpprules.in. This isn't conditionalized at all, so it will break on systems not using GNU cpp. 18486 18487commit b46cf0d879f1dbf92dcf5a0305d18986c766ed84 18488Author: Daniel Stone <daniel@fooishbar.org> 18489Date: Sun Jul 10 22:37:33 2005 +0000 18490 18491 Fix segfault when _XimProtoCreateIC() fails to create a context; Debian #239991. (Chung-chieh Shan) 18492 18493commit 6f2132b18e61ca9755e4b45550f3f5097dd1fbb7 18494Author: Alan Coopersmith <alan.coopersmith@sun.com> 18495Date: Sat Jul 9 20:06:04 2005 +0000 18496 18497 Set __libmansuffix__ & __xorgversion__ correctly when cpp processing man pages 18498 18499commit 3939ac4410446b46071c9d714f4270c12bf904fe 18500Author: Alan Coopersmith <alan.coopersmith@sun.com> 18501Date: Sat Jul 9 18:44:14 2005 +0000 18502 18503 - Since all but one line of all the nls/*/Makefile.am files are identical, move common bits to nls/localerules.in for easier updating and use automake includes to include in all the nls/*/Makefile.am files 18504 - Don't assume $(CPP) can take gcc-only -traditional flag 18505 - CPP process man pages as is done in the monolithic tree 18506 18507commit 19ba9d0df86c688319377467254b9ea9c4b0eccc 18508Author: Keith Packard <keithp@keithp.com> 18509Date: Sat Jul 9 06:01:49 2005 +0000 18510 18511 Clean up .cvsignore files 18512 18513commit f4bba6fa9b9ce637be7662754750054567e9250a 18514Author: Adam Jackson <ajax@nwnk.net> 18515Date: Sat Jul 9 02:53:29 2005 +0000 18516 18517 typo fix. i suck. 18518 18519commit 3305da61a61695e24c1ea11d3f59dddb52873d47 18520Author: Adam Jackson <ajax@nwnk.net> 18521Date: Sat Jul 9 02:41:18 2005 +0000 18522 18523 Check for bigfont proto headers via pkgconfig (Arwed von Merkatz) 18524 18525commit afdae2e8ccb93de2987b5a5e850695af15ece7e0 18526Author: Keith Packard <keithp@keithp.com> 18527Date: Fri Jul 8 07:11:56 2005 +0000 18528 18529 Create and use XlibConf.h to match modular build which needs XTHREADS defined when building extensions 18530 18531commit 86fa88dc594ca2093030acf5c84973ee2b33eb4c 18532Author: Keith Packard <keithp@keithp.com> 18533Date: Fri Jul 8 06:57:06 2005 +0000 18534 18535 Create and install XlibConf.h to provide external users of Xlibint.h the defines necessary to correctly use the library. Xlibint.h should #include this new file. 18536 Add a bunch of .cvsignore files 18537 18538commit 30f6ffedeaf899e699f0c6b14c9471ce9bcc430c 18539Author: Chris Less <clee@c133.org> 18540Date: Mon Jul 4 23:01:48 2005 +0000 18541 18542 Fixing bug #380 - add a man page for XQueryExtension, XListExtensions, and XFreeExtensionList. 18543 18544commit 55328bb6caabd37885cd19d7a5821ed425daf321 18545Author: Eric Anholt <anholt@freebsd.org> 18546Date: Sun Jul 3 10:17:38 2005 +0000 18547 18548 Avoid a gmake-specific construct, and instead just write the name of the current directory into each Makefile.am. 18549 18550commit 726538ba21b631bfb0b8094a5546acdaf57379ff 18551Author: Daniel Stone <daniel@fooishbar.org> 18552Date: Sun Jul 3 07:37:33 2005 +0000 18553 18554 Fix more include paths; add dix-config.h to XKB code. 18555 18556commit 8c7677138e70e14eca0255f2168583f0ecc21994 18557Author: Daniel Stone <daniel@fooishbar.org> 18558Date: Sun Jul 3 07:00:55 2005 +0000 18559 18560 Add Xtrans definitions (FONT_t, TRANS_CLIENT) to clean up warnings. 18561 Add XSERV_t, TRANS_SERVER, TRANS_REOPEN to quash warnings. 18562 Add #include <dix-config.h> or <xorg-config.h>, as appropriate, to all source files in the xserver/xorg tree, predicated on defines of HAVE_{DIX,XORG}_CONFIG_H. Change all Xfont includes to <X11/fonts/foo.h>. 18563 18564commit 197697c92a63091a4cc3cc04dcb7fa29d2655758 18565Author: Daniel Stone <daniel@fooishbar.org> 18566Date: Fri Jul 1 22:13:35 2005 +0000 18567 18568 Fix objdir != srcdir, as well as make distcheck. 18569 Don't attempt to create Compose.pre files; formatting fixes. 18570 Added if not already present. 18571 18572commit c162d60ad8f124563f94a2a266de59373936266c 18573Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18574Date: Wed Jun 22 22:46:31 2005 +0000 18575 18576 Apply these patches from Theo van Klaveren: 18577 lib-dmx.patch lib-FS.patch lib-X11.patch lib-XRes.patch lib-XScrnSaver.patch lib-xtrans.patch 18578 to make the libraries distcheck. 18579 18580commit 9ee8abdab03ea605a6327118ab7dacab6adf8876 18581Author: Alan Coopersmith <alan.coopersmith@sun.com> 18582Date: Sat Jun 18 07:48:43 2005 +0000 18583 18584 Move Secure RPC flags from X11/configure.ac to xtrans/xtrans.m4 since multiple modules will need them 18585 18586commit 12afc57b7d455781eee305e9ed6a899ceec8729e 18587Author: Daniel Stone <daniel@fooishbar.org> 18588Date: Wed Jun 15 16:50:47 2005 +0000 18589 18590 Typo fix to locale/error/keysym location declarations. 18591 18592commit 9a895777e30762f61b98d25be3b5d8b1169baa17 18593Author: Daniel Stone <daniel@fooishbar.org> 18594Date: Wed Jun 15 13:37:43 2005 +0000 18595 18596 Typo in ImUtil.h commit -- I AM CAPTAIN SKILL. 18597 18598commit add49285663684875ab7a5c58ec7a2cf8b775f67 18599Author: Daniel Stone <daniel@fooishbar.org> 18600Date: Wed Jun 15 13:32:35 2005 +0000 18601 18602 Define locations for XErrorDB, XKeysymDB, and locale data in configure.ac. Add AC_DEFINE_DIR macro from autoconf-archive.cryp.to towards this end. 18603 Move ImUtil.h from src/ to include/X11/. 18604 18605commit 845dfc6b42b950890866ee4df27761e086f50dca 18606Author: Daniel Stone <daniel@fooishbar.org> 18607Date: Wed Jun 15 13:27:48 2005 +0000 18608 18609 Move ImUtil.h from src/ to include/X11/. Additionally, copy Cmap.h as a distribution file. 18610 18611commit bba117f0d98f62cfb060d0fab97b407a3a0bfda9 18612Author: Daniel Stone <daniel@fooishbar.org> 18613Date: Fri Jun 10 14:11:36 2005 +0000 18614 18615 Remove pointless include of Xlib.h. 18616 Fix #include path to bigreqstr.h. 18617 18618commit 1a0de49da1274882bab05b0f7240936b37955e5c 18619Author: Alexander Gottwald <ago@freedesktop.org> 18620Date: Thu Jun 9 21:30:15 2005 +0000 18621 18622 Use $(srcdir) for Compose.pre and XLC_LOCALE.pre 18623 18624commit af4f0f302644ebfbb0ca9f4016a4aee85c973d37 18625Author: Alexander Gottwald <ago@freedesktop.org> 18626Date: Thu Jun 9 15:55:33 2005 +0000 18627 18628 Replace <X11/transport.c> with <X11/Xtrans/transport.c> 18629 18630commit fd5f58e0baf692e34b9b622286f18762cc2500d3 18631Author: Alexander Gottwald <ago@freedesktop.org> 18632Date: Thu Jun 9 15:52:02 2005 +0000 18633 18634 Replace <X11/Xtrans.h> with <X11/Xtrans/Xtrans.h> 18635 Copy Xtrans.h to exports/include/X11/Xtrans only 18636 18637commit 44538f9940f969d46c0e5e4b201c684cde2ba611 18638Author: Alan Coopersmith <alan.coopersmith@sun.com> 18639Date: Sun Jun 5 03:29:33 2005 +0000 18640 18641 Port Imake flags to autoconf tests & --enable-* flags: HASSETUGID, HASGETRESUID, NO_XLOCALEDIR, HAS_SHM and XF86BIGFONT 18642 18643commit 588e30e9ec65fa6205a34be650b79d5e2243edec 18644Author: Alan Coopersmith <alan.coopersmith@sun.com> 18645Date: Sat Jun 4 22:53:21 2005 +0000 18646 18647 Add --enable-secure-rpc flag and checks for needed functions for Secure RPC ("SUN-DES-1") authentication method 18648 18649commit a547afee2ef49cc41bbb67f9cff5a52a283c0854 18650Author: Alan Coopersmith <alan.coopersmith@sun.com> 18651Date: Sat Jun 4 21:20:20 2005 +0000 18652 18653 Bug #3436 <https://bugs.freedesktop.org/show_bug.cgi?id=3436> Conndis.c uses incorrect type for args to XdmcpWrap() (Mike Harris - mharris@www.linux.org.uk) 18654 18655commit ca93c761559ae464189c7ea7cf11c6a6679f2431 18656Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18657Date: Sat May 28 01:02:32 2005 +0000 18658 18659 Don't use $< in explicit rules. This only works with GNU make. <https://bugs.freedesktop.org/show_bug.cgi?id=3383> 18660 18661commit 09ebb349359e3dd9131fa2fa8b07559faa173654 18662Author: Alan Coopersmith <alan.coopersmith@sun.com> 18663Date: Sun May 22 19:05:11 2005 +0000 18664 18665 Convert man pages to long file names in lib/X11, lib/Xt, & lib/Xext 18666 18667commit 761219b1ef9befc350c8a35b6f96d047e5f008cc 18668Author: Alan Coopersmith <alan.coopersmith@sun.com> 18669Date: Sat May 21 23:07:48 2005 +0000 18670 18671 xtrans: 18672 Create autoconf macro XTRANS_CONNECTION_FLAGS to provide standard set of --enable flags for which transports to support and common place to update for required libraries for platforms that need certain libs for certain transports 18673 ICE: 18674 Add ICE_t #define required by Xtrans headers. Replace static defines of LOCALCONN & UNIXCONN with new XTRANS_CONNECTION_FLAGS macro. 18675 X11: 18676 Moved transport type checks to new macro XTRANS_CONNECTION_FLAGS in xtrans.m4 in xtrans module so they can be shared by all modules using xtrans. 18677 18678commit 4b1ba6eb975b547b808f4d5c8825a3261de5e02a 18679Author: Alan Coopersmith <alan.coopersmith@sun.com> 18680Date: Sat May 21 04:26:12 2005 +0000 18681 18682 Quote $ac_cv_search_* variables to prevent errors from test when they are set to "none required" (as happens on Solaris since dlopen is in libc) 18683 Comment out "override CC = gcc" line as it breaks builds with non-GNU makes, and its incorrect to force a specific compiler. Change LINK line from gcc to $(CC). 18684 18685commit 1d425d5e2092dd18a7dd599b37ed9af61cf59819 18686Author: Adam Jackson <ajax@nwnk.net> 18687Date: Thu May 19 00:22:32 2005 +0000 18688 18689 revert last change, didn't do right thing at all, sorry for the noise 18690 18691commit 1b0c46c1ae61d751dd3ca96de8e2c3fe21c5f4f2 18692Author: Adam Jackson <ajax@nwnk.net> 18693Date: Thu May 19 00:10:07 2005 +0000 18694 18695 Require automake 1.7 in AM_INIT_AUTOMAKE 18696 18697commit cd4657c175dbab6aaca36f18a0ca92c95b5567dd 18698Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18699Date: Tue May 17 22:32:09 2005 +0000 18700 18701 - Check for xproto as its CFLAGS are needed in the .pc file 18702 18703commit 46e8d8a65430dd87c10b066b5cff99a689c22241 18704Author: Egbert Eich <eich-at-freedesktop-dot-org> 18705Date: Tue May 17 08:10:10 2005 +0000 18706 18707 gcc4 allows to check if sentinels are correct (a sentinel is the terminating element in a varargs list). A sentinel needs to be NULL, not 0 - which doesn't make a difference on 32bit but matters on 64bit. Furthermore it can be told that functions have a printf-like format string and argument list so that they can verify that both match. To use these features certain attributes need to be set - which are compiler specific. To do this we define macros which are expanded depending on the compiler version. For now we put those in include/Xfuncproto.h (the XFree86 DDX layer contains a file compiler.h which however is not visible outside the DDX) (Bugzilla #3268). 18708 18709commit 23198d2bfbf0049b2630235cd4d4a4ffba7ec6c1 18710Author: <ssp@aware-of-vacuity.boston.redhat.com> 18711Date: Mon May 16 22:35:27 2005 +0000 18712 18713 Make Xdmcp unconditionally required, require various protocol modules. 18714 Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com> 18715 Check for kbproto if using XKB. 18716 Require xextproto rather than xextensions 18717 Remove the entries from the xlibs tree, as they are not relevant here. 18718 18719commit 8bd3aea84ce54b8b76a898f3ae00e2b499c14a5e 18720Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18721Date: Mon May 16 21:48:36 2005 +0000 18722 18723 Mon May 16 17:48:03 2005 Søren Sandmann <sandmann@redhat.com> 18724 Check for kbproto if using XKB. 18725 18726commit 9b1fa9ca3ed852ed40860f137511683f4bdae6fa 18727Author: <ssp@aware-of-vacuity.boston.redhat.com> 18728Date: Mon May 16 21:27:35 2005 +0000 18729 18730 Require xextproto rather than xextensions 18731 Remove the entries from the xlibs tree, as they are not relevant here. 18732 18733commit 7eee605e3aeed549d1053325a03027c5e8cbf71b 18734Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18735Date: Fri May 13 22:53:36 2005 +0000 18736 18737 - For now put xtrans in X11/Xtrans/X11, since libX11 is looking for it in <X11/...> 18738 - For Xcomposite and Xdamage, don't link the build system out of the xc tree 18739 - Link the public X11 headers into their own directory 18740 - Add links to XKeysymDB and XErrorDB 18741 - Add links to all the Xlib man pages 18742 - Add links to the lcUniConv subdirectory 18743 - Conditionally include config.h in Xlib source 18744 18745commit 6769ccda88caf27d1441d335ef2b318a047a612b 18746Author: Alan Coopersmith <alan.coopersmith@sun.com> 18747Date: Mon Mar 21 04:58:21 2005 +0000 18748 18749 xc/lib/X11/ErrDes.c 18750 //bugs.freedesktop.org/show_bug.cgi?id=132) Patch #2168 (https://bugs.freedesktop.org/attachment.cgi?id=2168) Replace a couple of BUFSIZE uses with better values to check against. Fixes by Stuart Anderson <anderson@netsweng.com> 18751 18752commit 3b9e8ece93b916c55a82df53e85f097418edf471 18753Author: Roland Mainz <roland.mainz@nrubsig.org> 18754Date: Sat Mar 19 22:04:55 2005 +0000 18755 18756 xc/nls/Compose/iso8859-2 18757 xc/nls/Compose/iso8859-3 18758 xc/nls/Compose/iso8859-9 18759 //bugs.freedesktop.org/show_bug.cgi?id=2592) attachment #2156 (https://bugs.freedesktop.org/attachment.cgi?id=2156) Fix a couple of typos in ISO8859-* Compose files ("asciicircum" instead "of asciicirum"). Patch by Matthias Hopf <mhopf@suse.de>. 18760 18761commit 0ce5950a08b9ab23ca8a32effdd40c421e92df84 18762Author: Alan Coopersmith <alan.coopersmith@sun.com> 18763Date: Tue Mar 8 02:53:36 2005 +0000 18764 18765 Bugzilla Bug 2006 (https://bugs.freedesktop.org/show_bug.cgi?id=2006) Patch #2031 (https://bugs.freedesktop.org/attachment.cgi?id=2031) XEmbed client doesn't receive key events from XIM: Use | to set a bit, not &. Patch by Hidetoshi Tajima <hidetoshi.tajima@sun.com>. 18766 18767commit df341cd2f2f263f13323e8c0936ea8aa0d7fbba1 18768Author: Roland Mainz <roland.mainz@nrubsig.org> 18769Date: Tue Feb 1 03:12:28 2005 +0000 18770 18771 xc/nls/Compose/pt_BR.UTF-8 18772 //bugs.freedesktop.org/show_bug.cgi?id=2400) attachment #1762 (https://bugs.freedesktop.org/attachment.cgi?id=1762): Fix build bustage caused by broken patch for brazillian locale support (see Bugzilla #1896). Patch by Kevin E. Martin <kem@freedesktop.org>. 18773 18774commit 483d3973f56dda51f9251d3ea808aaf7f3b48c10 18775Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> 18776Date: Fri Jan 28 18:31:31 2005 +0000 18777 18778 cleaned up boundary-case handling for Uxxxx Unicode keysym notation 18779 18780commit f234188a4c1e6b655aef0e3957ccad20d4c5847f 18781Author: Roland Mainz <roland.mainz@nrubsig.org> 18782Date: Wed Jan 19 01:53:55 2005 +0000 18783 18784 xc/nls/compose.dir 18785 xc/nls/locale.dir 18786 xc/nls/Compose/Imakefile 18787 xc/nls/Compose/pt_BR.UTF-8 18788 xc/nls/XI18N_OBJS/Imakefile 18789 xc/nls/XI18N_OBJS/pt_BR.UTF-8 18790 xc/nls/XLC_LOCALE/Imakefile 18791 xc/nls/XLC_LOCALE/pt_BR.UTF-8 18792 //bugs.freedesktop.org/show_bug.cgi?id=1896) attachment #1675 (https://bugs.freedesktop.org/attachment.cgi?id=1675): Add support for pt_BR.UTF-8 locale. Patch by Gustavo Noronha Silva, Branden Robinson, Julien Lafon. 18793 18794commit 7448ea7ef425d35cfc31eb41d46f4d879774f376 18795Author: Egbert Eich <eich-at-freedesktop-dot-org> 18796Date: Fri Jan 14 18:03:09 2005 +0000 18797 18798 Made some security enhancements: 18799 - no writing past end of buffer caused by bogus locale. 18800 - explicitely add a \0 character at end of string. (Bugzilla #2262) 18801 18802commit 5557d47fcf22a6f3adf327691158f2270e3d5094 18803Author: Roland Mainz <roland.mainz@nrubsig.org> 18804Date: Fri Jan 14 04:59:05 2005 +0000 18805 18806 xc/nls/compose.dir 18807 xc/nls/locale.alias 18808 xc/nls/locale.dir 18809 //bugs.freedesktop.org/show_bug.cgi?id=1830) attachment #1674 (https://bugs.freedesktop.org/attachment.cgi?id=1674): Adding support for the bs_BA (bs, bs_BA, bs_BA.iso88592, bs_BA.ISO-8859-2, bs_BA.ISO_8859-2, bs_BA.UTF-8) locale. Patch by Vedran Ljubovic <vljubovic@smartnet.ba>. 18810 18811commit 1fa3737f042f798fa11a9ff5b03f3b3ba3529824 18812Author: Egbert Eich <eich-at-freedesktop-dot-org> 18813Date: Tue Jan 11 17:37:57 2005 +0000 18814 18815 'Normalize' locale names (ie. remove any '-' and '_' and convert to lower case after the <language>_<territory> part) before matching against locale.alias (Bugzilla #2262). This needs adequate testing that we don't accidentally introduce undesirable side effects. 18816 18817commit 642cd269f94a234aa470a1d43385bd42625d89a1 18818Author: Egbert Eich <eich-at-freedesktop-dot-org> 18819Date: Tue Dec 14 08:59:20 2004 +0000 18820 18821 Removed #ifdef'ed out code together with the comment explaining why it was #ifdef'ed out. 18822 Fixed typo. 18823 Added comment to a changed that's been committed with one of the previous commits. 18824 18825commit a07ccae36e629741d2e48de7730114d30a975a41 18826Author: Matthieu Herrb <matthieu.herrb@laas.fr> 18827Date: Sun Dec 12 08:42:50 2004 +0000 18828 18829 Fix missing XChangeProperty() prototype missing in synopsis section. 18830 18831commit c73adb1c8ad806c2f9f0ba32f4b449e7a871501c 18832Author: Alexander Gottwald <ago at freedesktop dot org> 18833Date: Wed Dec 8 13:42:01 2004 +0000 18834 18835 Bugzilla #1980 (https://bugs.freedesktop.org/show_bug.cgi?id=1980) Handle XERRORDB only on WIN32 platform 18836 18837commit d5e7ab194103e22fd6c5094aaa107d90210bb600 18838Author: Jim Gettys <jg@freedesktop.org> 18839Date: Thu Dec 2 16:18:16 2004 +0000 18840 18841 fix comment to indicate additional possible mode. i bug 1756 reported by Owen Taylor. 18842 18843commit 03940d7330cb1bbf93d49c650aefb19de457da7c 18844Author: Alexander Gottwald <ago at freedesktop dot org> 18845Date: Wed Dec 1 13:06:55 2004 +0000 18846 18847 Bugzilla #1980, https://bugs.freedesktop.org/show_bug.cgi?id=1980 Make location of XErrorDB configurable 18848 18849commit 11a03ab908bfcfab8a3492684e8e9320e492d552 18850Author: Alexander Gottwald <ago at freedesktop dot org> 18851Date: Wed Dec 1 12:42:17 2004 +0000 18852 18853 Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory 18854 18855commit f3d83ee153f42e8899b844377e6b842d93411e62 18856Author: Alexander Gottwald <ago@freedesktop.org> 18857Date: Mon Nov 15 15:06:54 2004 +0000 18858 18859 Bufzilla #1802, http://freedesktop.org/bugzilla/show_bug.cgi?id=1802 Added mingw (Win32) port 18860 18861commit b798ea11911ac58a8e6e7d15a2a643b023859749 18862Author: Alexander Gottwald <ago at freedesktop dot org> 18863Date: Mon Nov 15 13:29:56 2004 +0000 18864 18865 Bugzilla #1864, http://freedesktop.org/bugzilla/show_bug.cgi?id=1864 Initialize pointer to NULL to avoid freeing random memory 18866 18867commit 436108cd6c84053698e5ca629096f59b34f50c2a 18868Author: Kristian Høgsberg <krh@redhat.com> 18869Date: Thu Nov 11 15:37:01 2004 +0000 18870 18871 Fix #1818 18872 18873commit bf2e6ef66ba55f90efa4a4ba8c8b6d3ec0d1531c 18874Author: Roland Mainz <roland.mainz@nrubsig.org> 18875Date: Tue Nov 9 00:56:56 2004 +0000 18876 18877 xc/nls/compose.dir 18878 xc/nls/locale.alias 18879 xc/nls/locale.dir 18880 //freedesktop.org/bugzilla/show_bug.cgi?id=1544): Adding support for the si_LK (si, sinhala, si_LK, si_LK.UTF-8) locale. Patch by Anuradha Ratnaweera <gnu.slash.linux@gmail.com>. 18881 18882commit 443890ceefbd6dafe68e30d103ec4f9d316ed655 18883Author: Roland Mainz <roland.mainz@nrubsig.org> 18884Date: Fri Nov 5 00:58:49 2004 +0000 18885 18886 xc/nls/compose.dir 18887 xc/nls/locale.alias 18888 xc/nls/locale.dir 18889 Adding some of the major indic locales (bn_IN.UTF-8, bn_IN.utf8, gu_IN.UTF-8, gu_IN.utf8, pa_IN.UTF-8, pa_IN.utf8) to X. Patch by Leon Ho <llch@redhat.com>. 18890 18891commit af7467ec734321f127b957921cce7792902b6794 18892Author: Egbert Eich <eich-at-freedesktop-dot-org> 18893Date: Mon Oct 18 17:29:03 2004 +0000 18894 18895 Correcting font encodings for GB18030, GBK and BIG5-HKSCS. Adding nls support for those encodings (Bugzilla 1573, James Su). 18896 18897commit 1b900b59cb24fe8be3db0d434b04d462c0eeb206 18898Author: Egbert Eich <eich-at-freedesktop-dot-xorg> 18899Date: Mon Oct 18 14:21:45 2004 +0000 18900 18901 Made handling of DevelDrivers for x86-64 more conformant to other platforms. 18902 Compress all font encodings (Stefan Dirsch). 18903 Fixed warnings. 18904 Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and Root mode (bugzilla #1580, James Su). 18905 Added another compose key combination for the Euro symbol (Stefan Dirsch). 18906 Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian). 18907 Changed default encoding for ru from KOI8-R to ISO8859-5 (Mike Fabian). This is the encoding that is also used by glibc. We may break other libcs - lets see who complains. 18908 Added explanation for DESTDIR to install to a different directory than /. 18909 Added some early bailouts to atiprobe if PCI structure pointer is NULL to prevent sig11. 18910 XV support doesn't depend on 2D acceleration any more. This patch removes this limitation from the NSC driver. This is a patch that I have committed to XFree86 a while ago but never ported over to X.Org. Matthieu Herrb contributed some build fixes. 18911 Fixing SetDPMSTimers() so that DPMS*Time == 0 disables the appropriate timer. This takes advantage of the fact that TimerSet() with a timeout argument 0 behaves like TimerCanel(). 18912 Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink). 18913 Create missing links for backward compatibility to XFree86 (Stefan Dirsch). 18914 Changed comment to mention xorg. 18915 Changed cursor for the 'kill' action from XC_draped_box to XC_Pirate. If you don't like it we can change it back (original author unkown). 18916 Added 'pic' to the man page rendering command pipeline (Werner Fink). 18917 Added missing return value (Stefan Dirsch, Roland Mainz) 18918 18919commit 1ba103c3cad2329db3a31f88d7798b913affc570 18920Author: Eric Anholt <anholt@freebsd.org> 18921Date: Wed Oct 13 07:38:00 2004 +0000 18922 18923 Use attributes instead of pragmas for weak symbols on gcc 2.95 as well. Using pragmas may result in the symbols being undefined with big -O. (FreeBSD ports/69708, Masakazu HIGAKI <higamasa@dream.com>) 18924 18925commit d80237db627bf71ad5615ff4ba05e6ab436aa740 18926Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18927Date: Fri Oct 8 22:57:56 2004 +0000 18928 18929 Fri Oct 8 18:53:11 2004 Soeren Sandmann <sandmann@redhat.com> 18930 Move iso10646 last so the "fallback" fonts will actually be used if they are better matches. 18931 18932commit d1f76d17ecf418396627b1a58319f2b9b255548a 18933Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> 18934Date: Sun Sep 26 22:54:57 2004 +0000 18935 18936 bug fix for previous patch 18937 18938commit 48932d9b71d10e15812f47d5b842ab6aa8dac625 18939Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> 18940Date: Sun Sep 26 20:46:17 2004 +0000 18941 18942 The big keysym cleanup, to bring implementation in line with the recent revision of Appendix A of the protocol spec. (Markus Kuhn) 18943 18944commit 2e02a95dcc43dd3ec7bbaf4675ffb94f5074f543 18945Author: Egbert Eich <eich@freedesktop.org> 18946Date: Wed Sep 15 09:05:22 2004 +0000 18947 18948 Unregistering events in XSelectInput() when unregistering IM filter callbacks may be a bad idea as others may be interested in this event. Removed the call to XSelectInput() altogether as we are in root window anyway (Lubos Lunak). 18949 Fix size of a variable that gets assigned the value of SmartScheduleTime (long) to long. This should help to prevent smart scheduler lockup on 64 bit systems due to overruns (Andreas Schwab). 18950 18951commit 2d3afb68a104a80a21ee622b9abb9c95e83505d3 18952Author: Egbert Eich <eich@freedesktop.org> 18953Date: Tue Aug 31 11:37:03 2004 +0000 18954 18955 Fixed some lockups in XIM code when the application is running with multi thread support. These lockups occur deep down in XFilterEvents() which itself locks when another Xlib function gets called that also locks. This fixes two instances by separating those Xlib functions into an internal (non-locking) call and a locking wrapper that is used as an external function. There may be several other such instances therefore another more general patch is eventually required (Bugzilla #1182). 18956 18957commit e689746c8d0e21e9011e8b91a3071d235d3a2a74 18958Author: Kevin E Martin <kem@kem.org> 18959Date: Thu Aug 19 06:48:06 2004 +0000 18960 18961 Fix header file to #ifdef the XKB keysyms when they are used. This fixes the X test suite build failure. 18962 18963commit d558a53a6f57eecfcaadce5141fe3a08860defcb 18964Author: Keith Packard <keithp@keithp.com> 18965Date: Sat Aug 14 07:12:36 2004 +0000 18966 18967 Use XLIB_SKIP_ARGB_VISUALS environment variable to disable all depth 32 visuals. Necessary to keep Flash from crashing. 18968 Must call ValidateGC/ValidatePicture on "real" GC/Picture to ensure pCompositeClip is set correctly. 18969 Need to take the composite clip from the "real" GC/Picture and turn it into the clientClip for the backing version. 18970 Adjust pixmap screen origin to account for drawable->x/y Change debugging output a bit (disabled by default) 18971 18972commit 85c2d81f299ed3444658011b7d6fb0a7ab8a6f55 18973Author: Alexander Gottwald <ago@freedesktop.org> 18974Date: Fri Aug 13 16:28:19 2004 +0000 18975 18976 Set most significant bit to be a one. (Bug #1024, Kensuke Matsuzaki) 18977 Fix conversion from sjis and euc. (Bug #1024, Toshio Takabe) 18978 18979commit c4d56e4e288d4e48b84b021a61638f46e9a45e27 18980Author: Adam Jackson <ajax@nwnk.net> 18981Date: Wed Aug 11 05:25:13 2004 +0000 18982 18983 Bug #372: Prevent a crash in XPolygonRegion when called with a bogus point count. Reported by Andreas Luik. 18984 18985commit 92487437173f600f208d825f65756d3ad14a4f7e 18986Author: Kevin E Martin <kem@kem.org> 18987Date: Mon Aug 9 22:37:22 2004 +0000 18988 18989 Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, Harold L. Hunt II, Alexander Gottwald). 18990 Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, David Dawes). 18991 Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson). 18992 HP-PA build fix (Bug #828, Guy Martin, Paul Anderson). 18993 Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine). 18994 Fix attempt to read video ROM before enabling it (Bug #843, Ivan Kokshaysky, Mike A. Harris). 18995 Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. Harris). 18996 Clarify xset man page description of how to use the keyboard repeat rate settings (Bug #846, Mike A. Harris). 18997 Fix problem where print-screen key would get remapped to sys-req in certain keymaps, which broke GNOME printscreen functionality (Bug #847, Owen Taylor). 18998 Fix several render problems: 18999 - MMIO mode support 19000 - Hang on IGP chips 19001 - VT switching hang 19002 - 3D render corruption (Bug #922, Hui Yu). 19003 19004commit 55c2ee568e7d3903258286a13bdf96ce5348ffda 19005Author: Matthieu Herrb <matthieu.herrb@laas.fr> 19006Date: Tue Jul 27 06:06:05 2004 +0000 19007 19008 - remove remaining AMOEBA references. 19009 - remove unused file. 19010 19011commit 6e884b12911eedfb003e90a3829ce66f7fc9cf2d 19012Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 19013Date: Tue Jul 20 17:48:09 2004 +0000 19014 19015 Tue Jul 20 19:38:06 2004 Soeren Sandmann <sandmann@daimi.au.dk> 19016 Set font_data->xlfd_data to NULL after XFree(). (#837, patch from Bastien Nocera). 19017 Tue Jul 20 18:23:32 2004 Soeren Sandmann <sandmann@daimi.au.dk> 19018 Use /dev/urandom on Linux.(#761). 19019 19020commit 6f0bc97aa4e6de5a4b001f40ac10795cfdf09fc1 19021Author: Eric Anholt <anholt@freebsd.org> 19022Date: Fri May 28 23:26:44 2004 +0000 19023 19024 Forced commit to note repocopy from xc/lib/XThrStub, will be connected to the build after this. 19025 19026commit aa7010c43ae9f39fb84b5ff155f76117c9e527a0 19027Author: Egbert Eich <eich@pdx.freedesktop.org> 19028Date: Mon May 24 19:02:11 2004 +0000 19029 19030 Improve 'uniqueness' of authorization cookie sent by client for XDM-AUTHORIZATION-1. Old 'uniquness' consisted of the PID of the client, a time stamp (in seconds) and a number obtained by starting to count down from 0xffff. When a client did an XOpenDisplay() then execv'ed a child and did XOpenDisplay() again within the same second, the cookie was identical to the previous one (as the PID did not change but the static 'count down' variable was reinitialized) and thus refused by the server. 19031 19032commit 720702da29769d80ad1254d92edbad5b30f8a8da 19033Author: Alan Coopersmith <alan.coopersmith@sun.com> 19034Date: Sat May 22 03:47:42 2004 +0000 19035 19036 Bugzilla #658: XStringToKeysym fails for Greek_IOTAdiaeresis (Fixes VSW5 testcase XStringToKeysym-7 failure) 19037 19038commit 3aed873292424b497d9a7dcee2975b95bf5ac966 19039Author: Alan Coopersmith <alan.coopersmith@sun.com> 19040Date: Sat Apr 24 23:39:25 2004 +0000 19041 19042 XOpenDisplay should try tcp connection if local connections fail (aka Sun bug id #4624183). 19043 Also includes fix from NetBSD Problem Report #25098 (Michael van Elst) (Xlib segfaults with IPv6 if compiled with HASXDMAUTH). 19044 xc/config/cf/sunLib.tmpl 19045 xc/lib/FS/Imakefile Add missing shared library dependencies for Solaris 19046 19047commit c6349f43193b74a3c09945f3093a871b0157ba47 19048Author: Egbert Eich <eich@freedesktop.org> 19049Date: Fri Apr 23 18:42:09 2004 +0000 19050 19051 Merging XORG-CURRENT into trunk 19052 19053commit c3c4ddc682950a01b80825021f3e2503ab01ea7f 19054Author: Kaleb Keithley <kaleb@freedesktop.org> 19055Date: Tue Nov 25 19:28:07 2003 +0000 19056 19057 Initial revision 19058 19059commit dc4268a7dadc8da0d561757a68461246728613d3 19060Author: Kaleb Keithley <kaleb@freedesktop.org> 19061Date: Fri Nov 14 16:48:47 2003 +0000 19062 19063 Initial revision 19064 19065commit deae12c6b683898f5213992d561a59d4ea889cca 19066Author: Kaleb Keithley <kaleb@freedesktop.org> 19067Date: Fri Nov 14 15:54:30 2003 +0000 19068 19069 R6.6 is the Xorg base-line 19070