1 Wed Dec 11 11:11:44 2002 Owen Taylor <otaylor@redhat.com>
3 * glib/grand.c: Include string.h, supress a warning.
5 2002-12-10 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
7 * glib/grand.c (g_rand_int_range): Improve generation of
8 pseudo-random integers. (#99720, Morten Welinder <terra@diku.dk>)
10 * README.in, docs/reference/glib/running.sgml,
11 docs/reference/glib/tmpl/random_numbers.sgml,
12 docs/reference/glib/changes.sgml: Added notes about the new
15 2002-12-09 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
17 * configure.in: use 0 instead of $defattr, as it isn't defined
18 anymore on platforms using -pthread and linux, where this is used,
19 only uses 0 anyway and not "pthread_attr_default".
21 Sun Dec 8 23:36:12 2002 Owen Taylor <otaylor@redhat.com>
23 * glib/gfileutils.c (g_file_test): Guard against
24 the POSIX allowed behavior where access (file, X_OK)
25 succeeds for uid==0 when no executable bits are set.
27 * glib/gfileutils.c (g_file_test): Add doc notes
28 about the possibility of race conditions, and the
29 fact that EXISTS and IS_EXECUTABLE give results
30 for the real uid not the effective user ID.
31 (#81854, Morten Welinder.)
33 Fri Dec 6 14:34:42 2002 Owen Taylor <otaylor@redhat.com>
35 Avoid literal UTF-8. (Allow it to build with
36 IRIX cc, #72757, Tomas Ogren)
38 * tests/patterntest.c (main): Convert literal
39 iso-8859-1 high bit characters in tests to
42 * glib/gunichartable.h glib/gen-unicode-tables.pl:
43 Use hex \xMN escapes rather than literal UTF-8
46 * configure.in: AC_PREREQ(2.53). needed for recent
47 GLIB_SIZEOF() changes.
49 Fri Dec 6 12:40:00 2002 Owen Taylor <otaylor@redhat.com>
51 * glib/libcharset/Makefile.am (EXTRA_DIST): Don't
52 distribute ref-add/del.sed., fixing problem with
53 srcdir != destdir. (#99813, Phuc LeHong)
55 Fri Dec 6 12:31:30 2002 Owen Taylor <otaylor@redhat.com>
57 * configure.in: chmod 0755 scripts in AC_OUTPUT(), since
58 install -c won't necessarily do that. (#74376, Martin
59 Gansser, Joshua Weage)
61 Thu Dec 5 16:05:55 2002 Owen Taylor <otaylor@redhat.com>
63 * configure.in: Use AC_LINK_IFELSE(), not AC_TRY_RUN()
64 for va_copy(), __va_copy() checks.
66 * configure.in: When there is a reasonable default,
67 provide that in the 3rd argument of AC_TRY_RUN(),
68 instead of just leaving the 3rd argument empty.
70 2002-12-05 Matthias Clasen <maclas@gmx.de>
72 * glib/gstrfuncs.c (g_strncasecmp): Break long deprecation
73 info into two paragraphs.
75 Thu Dec 5 15:43:46 2002 Owen Taylor <otaylor@redhat.com>
77 * configure.in: Fix multiple problems with the programs
78 in the argument of AC_TRY_RUN() having preprocessor
79 defines not in the first column.
81 Thu Dec 5 15:24:14 2002 Owen Taylor <otaylor@redhat.com>
83 Start of fixes for cross-compilation. Based on
84 patches from Dimi Shahbaz, Dan Kegel, Johannes Stenzenbach,
87 * configure.in: Don't run the Digital-Unix 4 -std1 check
90 * configure.in: Use AC_TRY_COMPILE() rather than
91 AC_TRY_RUN() for inline checks. (Daniel Egger)
93 * configure.in: use AC_CHECK_SIZEOF rather than
94 GLIB_SIZEOF for size_t. Remove unused checks for
95 size of ptrdiff_t/intmax_t.
97 * acglib.m4: Resync GLIB_SIZEOF of to the current
98 AC_CHECK_SIZEOF, which handles cross-compilation.
100 * acglib.m4: Remove no longer needed GLIB_SYSDEFS(),
101 add GLIB_CHECK_VALUE() as a wrapper around the
102 amazing _AC_COMPUTE_INT() autoconf 2.5x internal.
104 * configure.in: Redo the POLL* value checks in a
105 cross-compilation friendly way.
107 Thu Dec 5 15:28:37 2002 Owen Taylor <otaylor@redhat.com>
109 * glib/gtimer.c: Include glibconfig.h early so
110 that we include <windows.h> when needed.
112 * glib/gstrfuncs.c: Fix typo.
114 Tue Dec 3 21:10:28 2002 Owen Taylor <otaylor@redhat.com>
116 * configure.in: Add 3rd argument to remaining AC_DEFINES(),
117 so we can get rid of acconfig.h. (#76334, Daniel Egger)
119 * acconfig.h: Removed.
121 Tue Dec 3 20:22:27 2002 Owen Taylor <otaylor@redhat.com>
123 * glib/*.c: Patch from Sven Neumann to make the
124 include order consistent. (#71704)
126 2002-12-03 Matthias Clasen <maclas@gmx.de>
128 * configure.in: Check that strlcpy behaves as per the OpenBSD
129 man page before wrapping it. (#53933)
131 2002-12-03 Frederic Crozat <fcrozat@mandrakesoft.com>
133 * glib/gunicollate.c: (g_utf8_collate_key):
134 Fix typo from previous commit
136 Tue Dec 3 01:05:00 2002 James M. Cape <jcape@ignore-your.tv>
138 * glib/gunicollate.c (g_utf8_collate, g_utf8_collate_key):
139 Check for NULL before doing anything.
141 Mon Dec 2 16:34:13 2002 Owen Taylor <otaylor@redhat.com>
143 * === Released 2.1.3 ===
147 [ Unmerged change from stable from May 16 ]
149 * autogen.sh: Remove check for GNU gettext, since it
150 was causing problems for Solaris CVS builds.
151 (#81885, Hidetoshi Tajima.)
153 Tue Nov 26 09:51:43 2002 Owen Taylor <otaylor@redhat.com>
155 * glib/gstrfuncs.c (g_strchomp): Avoid non-ANSI pointer
156 comparison. (#54344, Morten Welinder)
158 * tests/strfunc-test.c (main): Add tests for strchomp().
160 2002-11-30 Ole Laursen <olau@hardworking.dk>
162 * configure.in: Added "da" to ALL_LINGUAS.
164 2002-11-29 Matthias Clasen <maclas@gmx.de>
166 * glib/gstrfuncs.c (g_ascii_strtoull): It is a 2.2 addition.
168 * glib/gtimer.c (g_time_val_add): Adjust to changed parameter names.
170 * glib/gutils.c (g_get_application_name): Mark as 2.2. addition.
172 * glib/gstrfuncs.c (g_ascii_strtoull): Mark as 2.0.7 addition.
174 2002-11-28 Matthias Clasen <maclas@gmx.de>
180 * glib/giochannel.c: Move some docs inline, and add deprecation
181 information. To see the list of affected functions, grep for
184 * configure.in: Fix the definitions around printf: either we use
185 system printf in which case HAVE_VASPRINTF, HAVE_C99_VSNPRINTF and
186 HAVE_UNIX98_PRINTF have already been determined by earlier tests,
187 or we use the included printf, in which case we know that these
188 three can be defined as 1. (#99826)
190 * glib/gutf8.c (g_utf8_strreverse):
191 * glib/gstrfuncs.c (g_str_has_suffix):
193 * glib/gprintf.c (g_printf):
199 * glib/gmarkup.c (g_markup_parse_context_get_element):
200 Mark as 2.2 API additions.
202 * configure.in: Fix the gtk-doc version check.
204 2002-11-27 Matthias Clasen <maclas@gmx.de>
206 * glib/gmessages.h: Use G_LIKELY without surrounding parentheses
207 in g_assert() and g_return_[val]_if_fail() so that we always trigger
208 the gcc warning about "assignment used as truth value".
210 * glib/gmacros.h: Always put parentheses in G_LIKELY and G_UNLIKELY.
212 2002-11-26 Matthias Clasen <maclas@gmx.de>
214 * glib/gmessages.h: Only use G_LIKELY in g_assert() and
215 g_return_[val]_if_fail() if it is actually doing something.
217 2002-11-26 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
219 * glib/grand.c, gthread/gthread-impl.c, tests/rand-test.c:
220 Changed the seeding algorithm. Old behaviour can be achived by
221 setting envvar G_RANDOM_VERSION to "2.0". (#99262)
223 * docs/reference/glib/glib-docs.sgml,
224 docs/reference/glib/Makefile.am: Renamed
225 docs/reference/glib/changes-2.0.sgml to
226 docs/reference/glib/changes.sgml and added section for changes
227 from 2.0 to 2.2 (Also corrected 1.0 to 1.2).
229 * README.in, docs/reference/glib/running.sgml,
230 docs/reference/glib/tmpl/random_numbers.sgml,
231 docs/reference/glib/changes.sgml: Added notes about the new
234 * configure.in: Make CPPFLAGS, not CFLAGS, include
235 G_THREAD_CFLAGS. CFLAGS is used while linking too and thus GLib
236 programs would link to the threads library on some platforms. Also
237 fixed a bug manifesting through this change. (#77981)
239 2002-11-26 Matthias Clasen <maclas@gmx.de>
241 * glib/gmacros.h: Fix the non-gcc-3.x definitions of G_LIKELY
242 and G_UNLIKELY. (Reported by Dan Mills)
244 2002-11-25 Matthias Clasen <maclas@gmx.de>
246 * glib/gfileutils.c (g_build_path):
247 (g_build_filename): Document that the varargs must be
248 NULL-terminated. (#99510)
250 * glib/gmessages.h (g_assert):
252 (g_return_val_if_fail): Remove the (no longer effective) empty
255 * glib/gmacros.h: Change the definition of G_LIKELY, so that
256 g_return_if_fail() and friends still trigger a gcc warning if
257 the expr is an assignment.
259 2002-11-23 Matthias Clasen <maclas@gmx.de>
261 * configure.in: Generate docs/reference/*/version.xml.
263 * glib/gdir.h: Add Copyright notice.
265 2002-11-22 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
267 * configure.in: Fixed typo: PTHREAD_PRIO_MIN ->
268 PTHREAD_PRIO_MAX. (Laurent Vivier, #99293)
270 Fri Nov 22 09:39:09 2002 Owen Taylor <otaylor@redhat.com>
272 * glib/gmacros.h: Remove broken G_HIDDEN_SYMBOL
273 definition which wasn't supposed to be committed
276 Thu Nov 21 16:19:21 2002 Owen Taylor <otaylor@redhat.com>
278 * glib/ghash.c: Patch from Morten Welinder to
279 make ghash.c properly obey DISABLE_MEM_POOLS. (#96600)
281 Thu Nov 21 14:09:44 2002 Owen Taylor <otaylor@redhat.com>
283 * glib/gmacros.h: Add G_GNUC_DEPRECATED. (Tom Tromey,
286 2002-11-21 Tor Lillqvist <tml@iki.fi>
288 * config.h.win32.in: Update to match what is currently produced by
291 * configure.in: Remove superfluous spaces on two shell variable
292 assignment lines. Don't define HAVE_GOOD_PRINTF as 0 if we don't
293 have a good printf, it is tested with #ifdef.
295 * glib/glib.def: Add new functions.
297 * glib/Makefile.am: If !HAVE_GOOD_PRINTF, add libtrio.la to LIBADD
301 * gmodule/gmodule.rc.in
302 * gobject/gobject.rc.in
303 * gthread/gthread.rc.in
304 * {glib,gmodule,gobject,gthread}/makefile.{mingw,msc}.in:
305 Hardcode 2.0 in the names, as that is what Makefile.am does.
307 2002-11-21 Matthias Clasen <maclas@gmx.de>
309 Include a printf implementation supporting C99 snprintf and SUS
310 positional parameters: (#79488)
313 * glib/gspawn-win32.c:
316 * glib/gbacktrace.c: Use _g_printf wrappers.
318 * glib/gutils.c (g_vsnprintf): Simplify, since we can assume C99
319 snprintf semantics now.
321 * glib/gmessages.c (printf_string_upper_bound): No longer needed,
322 since we can assume C99 snprintf semantics now.
325 * acinclude.m4 (AC_FUNC_PRINTF_UNIX98): New macro to check wether
326 printf supports SUS positional parameters.
328 * configure.in: New option --enable-included-printf to force
329 compilation of trio; otherwise trio is compiled if the system
330 printf misses either C99 snprintf semantics of SUS positional
333 * glib/Makefile.am (SUBDIRS): Conditionally compile trio.
334 (libglib_2_0_la_SOURCES): Add gprintf.c and gprintfint.h.
335 (glibsubinclude_HEADERS): Add gprintf.h.
337 * glib/gprintfint.h: New private wrapping either system printf
338 or trio printf variants in _g_printf wrappers for use inside glib.
340 * glib/gprintf.h: New public header declaring g_printf variants.
341 * glib/gprintf.c: Corresponding implementations.
343 * glib/trio/*: New directory, containing the trio-1.9 sources.
345 2002-11-20 Matthias Clasen <maclas@gmx.de>
347 * glib/gmessages.h (g_return_if_fail):
348 (g_return_val_if_fail): Use G_LIKELY. (#69022)
350 * glib/gmacros.h (G_LIKELY):
351 (G_UNLIKELY): New macros for hinting the compiler about the
352 expected result of expressions. For gcc 3.x, define these
353 using __builtin_expect. (#69022)
355 Tue Nov 19 14:38:18 2002 Owen Taylor <otaylor@redhat.com>
357 * glib/gstrfuncs.c (g_ascii_strtoull): Fix
358 strtull/strtoull type in docs (#99012, Morten
359 Welinder.) Add copyright information for code
362 2002-11-18 Tor Lillqvist <tml@iki.fi>
364 * glib/gspawn-win32.c (do_spawn_with_pipes): Do handle
365 G_SPAWN_DO_NOT_REAP_CHILD after all, similarily as on Unix. If the
366 flag is not set, don't call DuplicateHandle() on the handle
367 returned by the helper process, and set the "child pid" returned
368 to the called to zero. Close the handle to the helper process in
371 * glib/gspawn.c (g_spawn_async_with_pipes): Document Windows
372 behaviour of G_SPAWN_DO_NOT_REAP_CHILD.
374 2002-11-18 Tor Lillqvist <tml@iki.fi>
376 [Win32] Fix the asynchronous g_spawn* to return the process handle
377 of the started program properly. (Note: not the process id. The
378 spawn*() functions in the C runtime return the created process's
379 handle. There doesn't seem to be any way to get the process id of
380 a child process if you have the handle. But then, the process
381 handle usually is more useful anyway.)
383 * glib/gspawn-win32-helper.c (WinMain): If the spawning of the
384 child process succeeded, and if asynchronous spawn (P_NOWAIT),
385 write the result handle up to the parent process, waiting to read
386 it in do_spawn_with_pipes().
388 * glib/gspawn-win32.c (do_spawn): Use return value from spawning
389 the helper. If it is -1 the helper wasn't found or couldn't be run
390 for some reason. Otherwise it is the helper's process handle.
392 (g_spawn_async_with_pipes): Pass the child_pid parameter on to
393 do_spawn_with_pipes().
395 (do_spawn_with_pipes): Take also a child_pid parameter. If
396 do_spawn() returned -1, fail immediately. Otherwise make the
397 handle passed to us by the helper process into a handle valid in
398 this process by calling DuplicateHandle().
400 2002-11-17 Tor Lillqvist <tml@iki.fi>
402 * glib/gspawn-win32.c (g_spawn_async_with_pipes): Ignore the
403 G_SPAWN_DO_NOT_REAP_CHILD flag, can't be meaninfully implemented
404 on Windows, at least not now. Always pass dont_wait as TRUE to
405 do_spawn_with_pipes(). The semantics of the dont_wait parameter is
406 very different from the semantics of the intermediate_child
407 parameter to fork_exec_with_pipes() in the Unix version. This
408 fixes a serious bug, g_spawn_async() in fact behaved
411 (do_spawn_with_pipes, do_spawn): Rename from
412 fork_exec_with_pipes() and do_exec(), those names were from the
413 Unix bersion, and misleading.
415 (close_and_invalidate): Don't try to close invalid fds.
417 * glib/gspawn.c (g_spawn_async_with_pipes): Add warning about
418 Windows behaviour. There is no fork(), so the child_setup()
419 function is in fact called in the parent.
421 * glib/gspawn-win32-helper.c (WinMain): Insert spaces in argv
424 * tests/spawn-test-win32-gui.c: New file. Test program to be
425 linked as a GUI application. Behaves differently depending on how
426 invoked (by spawn-test).
428 * tests/spawn-test.c (run_tests): On Win32, run the
429 spawn-test-win32-gui program, too, in several ways, synchronously
432 * tests/Makefile.am: Corresponding change.
434 Fri Nov 8 19:44:20 2002 Soeren Sandmann <sandmann@daimi.au.dk>
436 * docs/reference/glib/tmpl/arrays.sgml:
437 * docs/reference/glib/tmpl/arrays_byte.sgml:
438 * docs/reference/glib/tmpl/arrays_pointer.sgml:
439 * docs/reference/glib/tmpl/date.sgml:
440 * docs/reference/glib/tmpl/linked_lists_double.sgml:
441 * docs/reference/glib/tmpl/linked_lists_single.sgml:
442 * docs/reference/glib/tmpl/main.sgml:
443 * docs/reference/glib/tmpl/queue.sgml:
444 * docs/reference/glib/tmpl/random_numbers.sgml:
445 * docs/reference/glib/tmpl/relations.sgml:
446 * docs/reference/glib/tmpl/scanner.sgml:
447 * docs/reference/gobject/tmpl/gtype.sgml:
448 * docs/reference/gobject/tmpl/value_arrays.sgml glib/garray.h:
449 * glib/gdate.h glib/giochannel.h glib/glist.h glib/gmain.c:
450 * glib/gmain.h glib/gqueue.c glib/gqueue.h glib/grand.c glib/grand.h:
451 * glib/grel.h glib/gslist.h glib/gtimer.h gobject/gvaluearray.h:
453 Trivial s/foo/foo_/ fixes to make <glib.h> includable with
454 -Wshadow without warnings (#91680)
456 Thu Nov 7 19:32:26 2002 Owen Taylor <otaylor@redhat.com>
458 * glib/gutils.[ch] (g_set/get_application_name):
459 Patch from Havoc Pennington to add functions for
460 setting and getting a human readable application
463 * configure.in: Up to version 2.1.3, since we'll
464 need to depend on last addition for GTK+.
466 2002-11-06 Tor Lillqvist <tml@iki.fi>
468 * glib/glib.def: Add g_main_thread_init.
470 2002-11-06 Matthias Clasen <maclas@gmx.de>
472 * glib/gstrfuncs.c (g_str_has_suffix):
473 (g_str_has_prefix): Minor doc markup fix.
475 Mon Nov 4 10:45:48 2002 Owen Taylor <otaylor@redhat.com>
477 * configure.in: Add -DG_DISABLE_CAST_CHECKS for
478 everything but --enable-debug.
480 * configure.in: Require pkg-config 0.14. (#97553)
482 Mon Nov 4 14:41:48 2002 Owen Taylor <otaylor@redhat.com>
484 * glib/gbsearcharray.c: Include config.h
485 so DISABLE_MEMPOOLS actually has an effect.
486 (#96437, Morten Welinder)
488 * tests/uri-test.c: Include <config.h>
490 2002-11-03 Dmitry G. Mastrukov <dmitry@taurussoft.org>
492 * configure.in: Added Belarusian to ALL_LINGUAS
494 2002-11-02 Daniel Elstner <daniel.elstner@gmx.net>
496 * glib/giochannel.c (g_io_channel_write_chars): Fix left_len
497 calculation in the from UTF-8 to UTF-8 case: left_len should
498 be the number of bytes left in the input buffer rather than
499 channel->write_buf. (#96373)
501 2002-10-27 Tor Lillqvist <tml@iki.fi>
503 * configure.in (G_MODULE_LDFLAGS): Don't set on Win32, only causes
506 * glib/gmain.c (g_poll): Fix for bug reported by Herman Bloggs
507 (http://mail.gnome.org/archives/gtk-devel-list/2002-October/msg00101.html)
508 and others. We waited for events only for GPollFDs whose events
509 field had G_IO_IN set. We need to wait also for events for
510 GPollFDs that have just G_IO_OUT set. Non-blocking sockets in the
511 process of being connect()ed are one such case. Also silence a
512 couple of gcc warnings.
514 Fri Oct 18 13:41:30 2002 Manish Singh <yosh@gimp.org>
516 * glib/giochannel.c (g_io_channel_read_line_backend): avoid
517 creating negative values out of unsigned values using MAX,
518 check to see if the result would be positive before doing
521 Tue Oct 15 15:28:47 2002 Manish Singh <yosh@gimp.org>
523 * tests/iochannel-test.c: use gsize instead of int where appropriate
524 (64-bit cleanliness fix). Removed leftover line_term cruft.
526 Tue Oct 15 15:07:45 2002 Manish Singh <yosh@gimp.org>
528 * gmodule/Makefile.am gobject/Makefile.am gthread/Makefile.am:
529 add -DG_DISABLED_DEPRECATED
531 * tests/gio-test.c tests/mainloop-test.c tests/string-test.c
532 tests/testglib.c test/tree-test.c tests/unicode-collate.c
533 tests/unicode-normalize.c: Deprecation cleanup
535 Mon Oct 14 15:51:05 2002 Owen Taylor <otaylor@redhat.com>
537 * glib/gdate.c (g_date_fill_parse_tokens): Fix a memory
538 leak. (#94550, Sebastian Rittau)
540 Mon Oct 14 15:36:11 2002 Owen Taylor <otaylor@redhat.com>
542 * glib/gcompletion.[ch] (g_completion_complete): Make
543 prefix argument const. (#91662, Gustavo Carneiro)
545 Mon Oct 14 15:32:14 2002 Owen Taylor <otaylor@redhat.com>
547 * tests/mainloop-test.c (adder_response): Fix a minor memory
550 Sat Oct 12 21:30:41 2002 Tim Janik <timj@gtk.org>
552 * merged up from glib-2-0:
554 * glib/gstrfuncs.c (g_ascii_strtod): fix comment.
555 (g_ascii_strtoull): new function, acting like strtoull(3) in the C
558 * glib/gscanner.[hc]: fix 32bit issues with integer parsing and
559 support storing 64bit values in GTokenValue by
560 using g_ascii_strtoull().
562 Sat Oct 12 12:34:22 2002 Soeren Sandmann <sandmann@daimi.au.dk>
564 * glib/gtree.c (g_tree_search), glib/gspawn.c
565 (g_spawn_async_with_pipes):
567 Documentation fixes: #71778, Owen Taylor; #85095, Bill Janssen,
570 Thu Oct 10 23:27:02 2002 Tim Janik <timj@gtk.org>
572 * glib/gscanner.c (g_scanner_msg_handler): if input_name is NULL,
573 print out "<memory>" instead of completely skipping input specification
574 and thusly loosing error line information.
576 Sun Sep 29 12:15:44 2002 Manish Singh <yosh@gimp.org>
578 * tests/mainloop-test.c: use gsize instead of int where appropriate
579 (64-bit cleanliness fix)
581 2002-09-29 Tor Lillqvist <tml@iki.fi>
583 * configure.in: Instead of forcing -fnative-struct into CFLAGS
584 when using gcc for Win32, check for gcc version 3.x which uses
585 -mms-bitfields instead. Also check if either of these switches is
586 actually available at all, and warn if not. Thanks to Soren
587 Andersen for the inspiration.
589 * HACKING: Say we require autoconf 2.52 as that is what
592 2002-09-23 Arvind Samptur <arvind.samptur@wipro.com>
593 * glib/gspawn.c (fork_exec_with_pipes) : when the child fails
594 we need to reap it to avoid a zombie. This would
595 happen in case of g_spawn_sync. Fixes #92658
597 2002-09-20 Matthias Clasen <maclas@gmx.de>
599 * glib/gscanner.c (g_scanner_msg_handler): Don't print
600 scanner->input_name when it is NULL. (#93752)
602 2002-09-06 Havoc Pennington <hp@redhat.com>
604 * autogen.sh: automake 1.4
606 2002-09-01 Soeren Sandmann <sandmann@daimi.au.dk>
608 * docs/reference/glib/tmpl/datalist.sgml, glib/gmain.c:
610 Documentation fixes: (#75255, Martin Schulze; #76104, Daryll Strauss)
612 2002-08-26 Tor Lillqvist <tml@iki.fi>
614 * configure.in (G_LIBS_EXTRA): Don't link with -lwsock32 on
615 Cygwin (#91696, Masahiro Sakai).
617 Tue Aug 20 16:01:03 2002 HideToshi Tajima <hidetoshi.tajima@sun.com>
619 * glib/gconvert.c (strdup_len): validate 'len' argument properly
620 for the case that input string is not null-terminated. (#91222)
622 2002-08-10 Gustavo Noronha Silva <kov@debian.org>
624 * configure.in: added pt_BR to ALL_LINGUAS
626 2002-08-07 Matthias Clasen <maclas@gmx.de>
628 * glib/gmarkup.c (g_markup_parse_context_get_element):
629 Use g_return_val_if_fail, no g_return_if_fail.
631 2002-08-06 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
633 * glib/gthread.c: Set the normal PID surrogate priority according
634 to getpid() to avoid errors for niced processes. (#86116)
636 * gthread/gthread-impl.c, gthread/gthread-posix.c,
637 gthread/gthread-solaris.c: Do the same for the posix/dce
638 implementation. Solaris still needs to set priority of the main
639 thread, because all unbound threads will schedule according to
640 that value and it defaults to 0 (the minimal value).
642 * glib/gmain.c: Factor out g_main_context_init_pipe from
643 g_main_context_new to initialize the wakeup pipe of a
644 context. Call that function from g_main_context_new if threads are
645 initialized and for all contexts when threads are initialized
646 in g_main_thread_init. (#86872)
648 * gthread/gthread-impl.c: Call g_main_thread_init. (#86872)
650 Thu Jul 25 14:23:15 2002 Owen Taylor <otaylor@redhat.com>
652 * glib/gfileutils.c: Clarify the behavior of g_build_path()
653 for empty elements and for leading and trailing copies
654 of the separator in the docs.
656 * glib/gfileutils.c: Fix problems with leading elements
657 consisting only of "/" characters. (#85928, Guillaume Chazarain)
659 * tests/strfunc-test.c (main): Add more test cases
660 for g_build_filename().
662 2002-07-26 Matthias Clasen <maclas@gmx.de>
665 * glib/gutf8.c (g_utf8_strreverse): New function to revert
667 * glib/gpattern.c (g_utf8_reverse): Gone, replaced by
668 g_utf8_strreverse. (#87725)
670 Thu Jul 25 20:57:20 2002 Owen Taylor <otaylor@redhat.com>
672 * glib/giochannel.c (g_io_channel_read_line): Fix
673 docs for @str_return to conform to the main docs
674 and the implementation as to whether the terminator
675 is included. (#87964, Jacob Berkman)
677 Thu Jul 25 19:52:41 2002 Owen Taylor <otaylor@redhat.com>
679 * glib/guniprop.c (real_tolower): Handle the
680 end of the string properly when max_len is set.
681 (#88545, Morten Welinder.) Rename the next_t
682 variable to next_type to make cc-mode happy.
684 * glib/gconvert.c (g_iconv_open): Fix missing
685 label from previous commit.
687 Thu Jul 25 19:41:03 2002 Owen Taylor <otaylor@redhat.com>
689 * glib/giochannel.c (g_io_channel_get_flags): Fix
690 incorrect return value in g_return_val_if_fail
691 (#85545, David L. Cooper II)
693 Thu Jul 25 19:12:43 2002 Owen Taylor <otaylor@redhat.com>
695 * glib/giochannel.c (g_io_channel_set_encoding): Clarify
696 that only one of the conditions listed in the docs
697 need to be true, not all of them. (#87176,
700 * glib/giochannel.c (g_io_channel_read_line): Fix
701 description of the return value. (#87754, Manuel Clos)
703 Thu Jul 25 19:02:53 2002 Owen Taylor <otaylor@redhat.com>
705 * glib/gmain.c (g_timeout_add): Remove reference
706 in the docs to @notify parameter. (#87768,
709 Thu Jul 25 17:57:07 2002 Owen Taylor <otaylor@redhat.com>
711 * glib/gconvert.c (g_iconv_open): Document the
712 (GIConv)-1 return value on failure. (#87559,
715 * glib/gconvert.c (g_iconv_open): Fix potential
716 problems with the assumption that (GIConv)(iconv_t)-1
717 is the same as (GIConv)-1.
719 2002-07-15 Matthias Clasen <maclas@gmx.de>
722 * glib/gmarkup.c (g_markup_parse_context_get_element): New
723 function to get the currently open element. (#70448)
725 2002-07-04 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
727 * tests/*.c: Added #undef G_DISABLE_ASSERT and #undef G_LOG_DOMAIN
728 throughout the files, which didn't already have them. (#87312)
730 2002-07-03 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
732 * configure.in: Made the thread flags determination tests first
733 check for the canonical form -pthread[s], then for other
734 possibilities. Also recognize some more platforms. Modeled after
735 patch from Miroslaw Dobrzanski-Neumann <mne@mosaic-ag.com>
738 * configure.in: Fixed bugs in posix thread priority
739 resolution. Also added AIX priorities. Fix from
740 Laurent Vivier <Laurent.Vivier@bull.net>. (#82599)
742 * gthread/gthread-solaris.c: Use thr_min_stack() instead of
743 sysconf (_SC_THREAD_STACK_MIN). if stack_size is 0 on entry to
744 g_thread_create_solaris_impl, also pass 0 to thr_create. Otherwise
745 the stack might be to small for any nontrivial thread. Discovered
746 on intel-solaris by Rajkumar Sivasamy <rajkumar.siva@wipro.com>.
748 2002-07-03 Anders Carlsson <andersca@gnu.org>
750 * configure.in: Add a forgotten trailing ` in the
751 GLIB_BINARY_AGE expression. Also fixup LT_CURRENT.
753 2002-07-02 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
755 * configure.in: Set binary age to 0 to make it compile.
757 * configure.in: Improve checks for dce/posix threads to also do
758 the right thing, if the thread functions are not declared in
759 pthread.h. Idea from Miroslaw Dobrzanski-Neumann
760 <mne@mosaic-ag.com> in #77981.
762 2002-07-01 Anders Carlsson <andersca@gnu.org>
764 * configure.in: Up version to 2.1.0.
766 2002-07-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
768 * glib/garray.h: Add parenthesis to protect macro
769 argument. Spotted by Sam Couter <sam@topic.com.au>. (#86826)
771 2002-06-28 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
773 * glib/gmessages.h (g_assert, g_assert_not_reached,
774 g_return_if_fail, g_return_val_if_fail): In case of
775 G_DISABLE_ASSERT replace "G_STMT_START{ }G_STMT_END" with
776 "G_STMT_START{ (void)0; }G_STMT_END", because the former chokes
777 gcc prior to 3.0. (#86664)
779 2002-06-26 Yanko Kaneti <yaneti@declera.com>
781 * configure.in: (ALL_LINGUAS) Added Bulgarian (bg).
783 Thu Jun 13 16:12:04 2002 Owen Taylor <otaylor@redhat.com>
785 * glib/gmain.c (struct _GTimeoutSource): Make
786 interval unsigned to match g_timeout_source_new().
789 * glib/gmain.c (g_timeout_prepare): Add some
790 extra gymnastics to avoid signed/unisgned integer
793 Thu Jun 13 15:57:46 2002 Owen Taylor <otaylor@redhat.com>
795 * INSTALL.in: Add a note about the deficiency
796 of Tru64 iconv. (#81699, Manuel Op de Coul)
798 Tue Jun 11 17:03:39 2002 Owen Taylor <otaylor@redhat.com>
800 * tests/patterntest.c (test_compilation)
801 * glib/gmem.c (profiler_try_realloc): A couple
802 of 64-bit printf format fix from George Lebl.
805 * glib/gconvert.c (open_converter): Fix gsize/gint
806 mixup for g_iconv() arguments. (#83270,
809 2002-06-03 Matthias Clasen <maclas@gmx.de>
811 * glib/gwin32.c (g_win32_get_package_installation_directory):
812 Replace homegrown "hash" entity by standard ISO entity "num".
814 2002-05-26 Matthias Clasen <maclas@gmx.de>
816 * glib/gmain.c: Remove references to nonexisting functions
817 g_source_set_callback_closure(), g_source_poll(), g_source_add()
820 * glib/gdir.c (g_dir_open): Typo fix in docs.
822 * glib/gasyncqueue.c (g_async_queue_lock):
823 (g_async_queue_unref_and_unlock): Fix markup to avoid erroneous
826 * glib/gwin32.c: Escape #'s leading to erroneous <link>s in docs.
828 * glib/gtree.c: Replace some occurances of Gtree by GTree in docs.
830 * glib/gstring.c (g_string_insert_unichar): Typo fix in docs.
832 2002-05-23 Havoc Pennington <hp@redhat.com>
834 * glib/gspawn.c (fork_exec_with_pipes): on success, close the
835 pipes from the child. Fix from Tim.
837 2002-05-22 jacob berkman <jacob@ximian.com>
839 * m4macros/glib-gettext.m4 (AM_GLIB_WITH_NLS): fix tyop
841 Wed May 22 15:40:47 2002 Owen Taylor <otaylor@redhat.com>
843 * README.in: Remove notes about now-fixed-bugs.
845 * m4macros/glib-gettext.m4 acinclude.m4: Get rid
846 of AC_MSG_NOTICE() usage, since some broken systems
847 (Hi Debian!) might not be using autoconf-2.5x for
848 downstream packages even though we require it for
851 * INSTALL.in: Add a note about installing extra
852 converters for Solaris.
854 * glib/gutils.h (g_bit_nth_msf): Fix termination
855 condition. (#82582, Paolo Molaro)
857 Tue May 21 15:51:17 2002 Owen Taylor <otaylor@redhat.com>
859 * configure.in: $with_libiconv, not $with_iconv.
861 Mon May 20 18:02:46 2002 Owen Taylor <otaylor@redhat.com>
863 * configure.in: Move iconv tests before gettext
866 * m4macros/glib-gettext.m4 acinclude.m4: If we can't
867 link to gettext, try adding in -liconv. (#80076,
870 * m4macros/glib-gettext.m4 acinclude.m4: Suppress
871 warnings about xgettext not being GNU gettext when
872 libintl wasn't found at all.
873 (#79016, Andrew P. Lentvorski, Jr.)
875 2002-05-21 Matthias Clasen <maclas@gmx.de>
877 * glib/gmarkup.c (g_markup_parse_context_parse): Added
878 proper support for CDATA sections.
880 2002-05-20 jacob berkman <jacob@ximian.com>
882 * glib/gdir.c: include sys/types.h before dirent.h to build on
883 darwin (fixes #72859)
885 Mon May 20 15:35:59 2002 Owen Taylor <otaylor@redhat.com>
887 * glib/gfileutils.c (get_contents_regfile): Fix double
888 close of file descriptor on error. (#82139, Wayne Schuller)
890 * glib/gspawn.c (close_and_invalidate): Don't close
891 fd's that have already been closed. (Fix from Michael
894 * glib/giochannel.c (g_io_channel_close): If not flushing,
895 dump the contents of the write buffers, so we won't try
896 to write them to an invalid fd later. (Patch from Ron Steinke,
897 fixing #78290, Andreas Persenius.)
899 2002-05-18 Matthias Clasen <maclas@gmx.de>
901 * tests/markups/fail-36.gmarkup:
902 * tests/markups/fail-35.gmarkup:
903 * tests/markups/valid-8.gmarkup:
904 * tests/markups/valid-7.gmarkup:
905 * tests/markups/valid-6.gmarkup:
906 * tests/markups/valid-5.gmarkup: New testcases.
908 * tests/markup-test.c (passthrough_handler):
909 (text_handler): Don't ignore the text_len parameter.
911 * glib/gmarkup.c (find_current_text_end): Don't hang on embedded
913 (g_markup_parse_context_parse): Fix passthrough handling to
914 correctly skip processing instructions, comments, doctype
915 declarations and CDATA marked sections. (#81977)
917 Sat May 18 00:21:51 2002 Tim Janik <timj@gtk.org>
921 * glib/gscanner.c (g_scanner_unexp_token): fix missing cases for
924 2002-05-17 Tor Lillqvist <tml@iki.fi>
926 * config.h.win32.in: Match new additions in a configure-produced
929 * glibconfig.h.win32.in: gssize and gsize as in a
930 configure-produced glibconfig.h.
932 * glib/glib.def: Add new functions.
934 Thu May 16 12:24:00 2002 Owen Taylor <otaylor@redhat.com>
936 * acinclude.m4 m4macros/glib-gettext.m4: Incorporate
937 AM_LC_MESSAGES, AM_PATH_PROG_WITH_TEST as AM_GLIB_* to avoid
938 dependencies on gettext m4 files. (#81885, Hidetoshi Tajima.)
940 2002-05-14 Alex Larsson <alexl@redhat.com>
944 New functions g_str_has_suffix and g_str_has_prefix.
946 * tests/string-test.c: (main):
947 Test the new functions.
949 Mon May 13 23:20:00 2002 Owen Taylor <otaylor@redhat.com>
951 * autogen.sh (have_gettext): Add a check for GNU gettext.
952 (Pointed out by Dan Winship in #59386.)
954 Mon May 13 11:55:33 2002 Owen Taylor <otaylor@redhat.com>
956 * configure.in acglib.m4: If sizeof(int) == sizeof(long)
957 run compilation tests to determine which way gsize should
958 be defined. (#74413, reported by Miroslaw Dobrzanski-Neumann)
960 Mon May 13 11:42:23 2002 Owen Taylor <otaylor@redhat.com>
962 * glib/libcharset/*: Update from libcharset CVS.
963 Includes additional encodings for Solaris (#80396,
966 * configure.in: Add getc_unlocked to CHECK_FUNCS().
968 * glib/gmessages.c (strdup_convert): If
969 g_convert_with_fallback() fails, print the error
970 message to stderr the first time, then return
971 the original string. (#78197)
973 2002-05-10 Naba Kumar <kh_naba@users.sourceforge.net>
975 * configure.in: Added "hi" to ALL_LINGUAS.
977 2002-05-08 Michael Natterer <mitch@gimp.org>
979 * configure.in: fixed yesterdays fix for cross compiling: simply
980 check the variable "cross_compiling" which gets set by autoconf.
982 2002-05-07 Matthias Clasen <maclas@gmx.de>
984 * glib/gmarkup.c (unescape_text): Report unfinished entity
985 references as errors rather than running into an assert. (#80441)
987 2002-05-07 Michael Natterer <mitch@gimp.org>
989 * configure.in: added a new conditional CROSS_COMPILING which
990 indicates ($build != $host). If it is set, look for
991 glib-genmarshal in PATH. Error out if it was not found.
993 Tue May 7 11:24:22 2002 Owen Taylor <otaylor@redhat.com>
995 Fixes for #79347, Ron Arts.
997 * glib/gqsort.c (g_qsort_with_data): Handle 0 elements,
998 don't g_return_if_fail().
1000 * tests/qsort-test.c (main): Add a 0 element test.
1002 * glib/garray.c (g_[ptr_]array_sort_with[_data]):
1003 Remove invalid assertions that array->pdata != NULL ..
1004 it's NULL for 0 elements which is a valid case.
1006 Mon May 6 16:00:41 2002 Owen Taylor <otaylor@redhat.com>
1008 * glib/gbacktrace.h: Exclude OSF from alpha definition
1009 of G_BREAKPOINT(), since the assembler apparently
1010 doesn't support bpt. (#77852, Gareth Pierce) [from stable]
1012 Mon May 6 11:48:08 2002 Owen Taylor <otaylor@redhat.com>
1014 [ merged from stable ]
1016 * Makefile.am (EXTRA_DIST): Add ChangeLog.pre-2-0 (#78641)
1018 * m4macros/glib-2.0.m4: save CFLAGS/LIBS properly when
1019 compiling "what went wrong" test case. (#79330)
1021 * m4macros/glib-2.0.m4: Remove reference to editing
1022 pkg-config script. (From Jim Gettys)
1024 * configure.in: (Look for dyld interfaces before dlopen()
1025 since OS X can have both. Patch from Jacob Berkman,
1028 2002-05-01 jacob berkman <jacob@ximian.com>
1030 * configure.in: fix typo on 64-bit printf formatting string
1031 configure message (fixes #80389)
1033 2002-04-29 Pablo Saratxaga <pablo@mandrakesoft.com>
1035 * configure.in: Added Vietnamese (vi) to ALL_LINGUAS
1037 2002-04-19 Tor Lillqvist <tml@iki.fi>
1039 * glib/gspawn.c (g_spawn_command_line_sync): Add Windows-specific
1040 note to the gtk-doc comment.
1042 * glib/gspawn-win32.c: Remove the copy-pasted gtk-doc comment
1043 blocks. It's enough to have them in gspawn.c.
1045 2002-04-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1047 * gthread/gthread-impl.c (g_thread_init): Fixed typo. (#78985)
1049 2002-04-13 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1051 * glib/gthreadpool.c (g_thread_pool_thread_proxy): Unlock the pool
1052 for all threads leaving it. (#78348)
1054 2002-04-11 Matthias Clasen <maclas@gmx.de>
1056 * tests/patterntest.c: Remove manual UTF-8 -> Latin1 conversion for
1057 stuff fed to g_print.
1059 2002-04-08 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1061 * tests/gio-test.c, tests/markup-test.c: Use gsize instead of
1062 gint. From Miroslaw Dobrzanski-Neumann
1063 <mne@mosaic-ag.com>. (#77982)
1065 2002-04-08 Stanislav Brabec <utx@penguin.cz>
1067 * cs.po: Added Czech (cs) to ALL_LINGUAS.
1069 2002-03-31 Hasbullah Bin Pit <sebol@ikhlas.com>
1071 * configure.in: Added Malay (ms)to ALL_LINGUAS.
1073 2002-03-29 Tor Lillqvist <tml@iki.fi>
1075 * glib/glib.def: Add g_convert_init.
1077 Thu Mar 28 18:25:14 2002 Owen Taylor <otaylor@redhat.com>
1081 * configure.in: Version 2.0.1, interface, binary age 1.
1083 Thu Mar 28 18:22:53 2002 Owen Taylor <otaylor@redhat.com>
1085 * README: Remove warning about g_print, etc, encoding.
1089 * glib/gmessages.c (g_log_default_handler): Use %lu
1090 (plus a cast) when printing out pid_t arguments.
1091 (#76770, Morten Welinder)
1093 * glib/gstrfuncs.c (g_strdup_vprintf): Check the
1094 result of vasprintf(), return NULL on failure.
1095 (#76802, Akira Tagoh)
1097 * tests/testglib.c (TEST): Supress a warning with
1100 Thu Mar 28 20:31:51 2002 Tim Janik <timj@gtk.org>
1104 (g_printerr): convert prtinf() strings to local charset
1105 when writing them to stdout or stderr.
1107 Wed Mar 27 18:42:22 2002 Tim Janik <timj@gtk.org>
1110 major cleanups. introduced _g_log_fallback_handler() to handle
1111 recursive messages which really doesn't call any GLib functions.
1112 this allowes the default handler to use normal GLib functions and
1113 also fixes user supplied log level handlers.
1114 fixed locking issues, based on a patch from Sebastian Willhelmi,
1115 attached to #74356. translate log messages from UTF-8, based
1116 on the same patch. save fatal and recursion flags across flag
1117 loop. use new integer format code from above patch.
1118 move GLib functions out of locked mutex state to avoid deadlocks.
1119 move the level prefix and filedescriptor logic into mklevel_prefix().
1120 move _g_debug_init() into a place where we can figure and handle
1123 Mon Mar 25 18:13:06 2002 Owen Taylor <otaylor@redhat.com>
1125 * glib/gtypes.h (GUINT*_SWAP_LE_BE_X86): Remove __const__
1126 qualifier from __asm__ statements... GCC since at least
1127 2.96 has assumed no side effects automaticaly, and gcc-3.1
1128 will warn about this usage. (#73308, Cody Russell)
1130 Fri Mar 22 17:59:27 2002 Owen Taylor <otaylor@redhat.com>
1132 * glib/gmem.c (g_mem_chunk_reset): Fix problem where
1133 if g_mem_chunk_reset() is called on an alloc-only
1134 memchunk, then a useles GTree was created.
1136 Wed Mar 20 18:20:21 2002 Owen Taylor <otaylor@redhat.com>
1138 * glib/gconvert.c (open_converter): Try to work around segfaults
1139 on Solaris if NULL is passed for outbuf... supposedly gchar
1140 *outbuf = NULL... &outbuf works. (#74336, Lauri Alanko)
1142 Wed Mar 20 11:17:32 2002 Owen Taylor <otaylor@redhat.com>
1144 * glib/gfileutils.c (get_contents_regfile): Close the
1145 file descriptor. (#75507, Matthias Clasen)
1147 Wed Mar 20 11:00:59 2002 Owen Taylor <otaylor@redhat.com>
1149 * configure.in: Use $PKG_CONFIG, not pkg-config.
1150 (LEE Sau Dan, #75572)
1152 2002-03-20 Sven Neumann <sven@gimp.org>
1154 * glib/ghash.c (g_hash_table_resize): avoid repeated call of
1155 g_spaced_primes_closest() by moving it out of the CLAMP macro
1156 (spotted by Salmaso Raffaele).
1158 2002-03-17 Tor Lillqvist <tml@iki.fi>
1160 * README.win32: Add MSVC-specific text by Hans Breuer.
1162 2002-03-16 Tor Lillqvist <tml@iki.fi>
1164 * glib/giowin32.c: Some debugging output formatting changes. The
1165 following changes fix a problem with buffered GIOChannels, noticed
1166 with the help of Owen's test program, thanks! (#59969)
1167 (g_io_win32_prepare): Return value that takes the buffer condition
1168 into account, like g_io_unix_prepare() does.
1169 (g_io_win32_check): Ditto, like g_io_unix_check().
1170 (g_io_win32_dispatch): Ditto, like g_io_unix_dispatch().
1172 Thu Mar 14 17:37:45 2002 Owen Taylor <otaylor@redhat.com>
1174 * glib/giochannel.c (g_io_channel_write_chars): Fix some
1175 gsize/gint mismatches. (#74422, Miroslaw Dobrzanski-Neumann)
1177 2002-03-14 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1179 * glib/gmessages.c: Fixed threading issues brought up by
1180 #74577. Make g_log_find_domain, g_log_domain_new,
1181 g_log_domain_check_free and g_log_domain_get_handler require being
1182 called with g_messages_lock held and remove all internal locking
1183 in them. Then added proper locking to g_log_set_handler,
1184 g_log_remove_handler and g_logv. Problem spotted by Miroslaw
1185 Dobrzanski-Neumann <mne@mosaic-ag.com>. (#74577)
1187 2002-03-13 Erwann Chenede <erwann.chenede@sun.com>
1189 glib/gen-unicode-tables.pl
1190 glib/gunidecomp.h : fixed cast/type problems to
1191 avoid warnings (with forte compiler) (#73898)
1193 2002-03-12 Alexander Larsson <alla@lysator.liu.se>
1196 Cache getenv("G_BROKEN_FILENAMES") in have_broken_filenames() and
1197 use instead. Add g_convert_init() that calls have_broken_filenames()
1199 * gthread/gthread-impl.c:
1200 Have g_thread_init call g_convert_init.
1202 2002-03-11 Matthias Clasen <maclas@gmx.de>
1204 * glib/gstrfuncs.c (g_strtod):
1207 (g_ascii_formatd): Doc formatting fixes.