1 2001-02-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3 * gthread.c, gthread.h: Added g_static_mutex_init to allow
4 initialization of a GStaticMutex, that can not be initialized with
5 G_STATIC_MUTEX_INIT, for example in allocated structures.
7 Wed Jan 31 13:46:58 2001 Owen Taylor <otaylor@redhat.com>
9 * acinclude.m4 glib-gettext.m4: Fix problem with --disable-nls.
11 2001-01-30 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
13 * gthread.c, gthread.h: Added g_static_mutex_free to allow using
14 GStaticMutexes with limited lifetime without leaking.
16 * configure.in: GStaticMutex doesn't have to provide extra space
17 for debugging information for G_ERRORCHECK_MUTEXES, as then the
18 non-default implementation (runtime_mutex) is used anyway.
20 * gthread.h (g_cond_wait): Added debug information for
21 g_mutex_free calls, if G_ERRORCHECK_MUTEXES is defined.
23 2001-01-29 Tor Lillqvist <tml@iki.fi>
25 * glibconfig.h.win32.in: Use the same GMutex structure as the
26 configure-generated glibconfig.h does.
28 * gstrfuncs.c (g_strsignal): Declare strsignal() on Cygwin, too,
29 says jbdoll@kepri.re.kr.
31 2001-01-29 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
33 * gthread.c: Broadcast the condition, if there are waiting
34 readers, as all might read at the same time. Only signal the
35 writer thread, if there are no more readers.
37 2001-01-25 Tor Lillqvist <tml@iki.fi>
40 * */makefile.mingw.in: Protect the rule to rebuild makefile.mingw
41 if makefile.mingw.in has changed with a check if said .in file
42 exists. (This rule is mainly a convenience for yours truly.)
44 * giowin32.c: Socket support rewritten. It was utterly broken, and
45 untested in fact. We still do use a thread for each socket being
46 watched, but instead of blocking in recv() (which of course was
47 plain stupid for sockets being listen()ed on in a server-type
48 application), we block in select(). The read method for sockets
49 calls recv(). It is now possible for the application to call
50 accept(), recv() or send() in the callback, just like on
51 Unix. Tested with code kindly provided by Andrew Lanoix.
53 Rename g_io_channel_win32_new_stream_socket() to
54 g_io_channel_win32_new_socket() as it isn't restricted to stream
57 * gmain.c (g_poll): Related changes in the Win32 version of
58 g_poll(). When polling for messages, always do a PeekMessage()
59 first. We used to miss messages if several were posted between
62 * giochannel.h: Improve Win32-related comments.
64 * gutf8.c: (Win32) Include <stdio.h> for sprintf.
66 * tests/gio-test.c: (Win32) Add tests for polling for Windows
69 * tests/makefile.mingw.in: Remove superfluous compilation command
72 2001-01-23 Alex Larsson <alexl@redhat.com>
74 * gmain.c (g_source_callback_unref): Free the callback
75 (g_source_set_callback): Initialize the callback refcount
77 2001-01-20 Tor Lillqvist <tml@iki.fi>
79 * gutils.c (g_get_codeset): (Win32) Even if g_get_codeset() is
80 currently commented out from gutils.h, fix it to return the same
81 CP%d value as g_get_charset().
83 2001-01-19 Kjartan Maraas <kmaraas@gnome.org>
85 * gconvert.c: Fix typo.
86 * gfileutils.c: Same here.
87 * configure.in: Added no to ALL_LINGUAS
89 2001-01-17 Tor Lillqvist <tml@iki.fi>
91 * gwin32.c (g_win32_getlocale, g_win32_error_message): Add doc
93 (g_win32_get_package_installation_directory): Add one parameter,
94 the name of a DLL in the package. Add possibility to use that to
95 deduce the installation directory if not entered into the
96 Registry. Make the return value dynamically allocated.
97 (g_win32_get_package_installation_subdirectory): New convenience
100 * gutils.c (GLIB_LOCALE_DIR)
101 * gwin32.h: Adapt accordingly.
103 * testglib.c (main): Adapt tests accordingly.
105 Tue Jan 16 23:20:38 2001 Owen Taylor <otaylor@redhat.com>
107 * gutils.c glibintl.h: Optimize for size rather than speed by
108 making _() always call _glib_gettext() instead of conditionally
109 calling gettext() or _glib_gettext_init. glib only uses translated
110 strings in slow error handling code anyways.
112 * glibintl.h: g'ify types.
114 * Makefile.am: include glibintl.h in _SOURCES
116 2001-01-17 Tor Lillqvist <tml@iki.fi>
118 * config.h.win32.in: Define ENABLE_NLS and GETTEXT_PACKAGE.
121 * makefile.msc.in: Use the GNU intl library.
123 * gwin32.c (g_win32_get_package_installation_directory): New
124 function. To be used by various GLib-using packages to get their
125 installation directory, which should be stored in the Registry by
128 * gwin32.h: Declare it.
130 * testglib.c (main): Test it.
132 * gutils.c: On Win32, define GLIB_LOCALE_DIR using
133 g_win32_get_package_installation_directory().
137 Mon Jan 15 21:39:06 2001 Owen Taylor <otaylor@redhat.com>
139 * glib-2.0.pc.in (Libs) glib-config-2.0.in: Add @INTLLIBS@.
141 Mon Jan 15 21:12:49 2001 Owen Taylor <otaylor@redhat.com>
143 * configure.in acconfig.h glibintl.h gutils.c
144 po/{Makefile.in.in,POTFILES.in,po2tbl.in}: Add gettext
147 * glib-gettext.m4 acinclude.m4: Clean up the GTK+ gettext macros
148 some more and put them in this file, though they also need
149 to be included in acinclude.m4 due to the brokeness of
152 * gspawn.c gspawn-win32.c gutf8.c gconvert.c gfileutils.c
153 gshell.c: Remove dummy _() #defines, include glibintl.m4.
155 2001-01-09 Tor Lillqvist <tml@iki.fi>
157 * giowin32.c: Rework the changes needed to pass mainloop-test. Now
158 we don't need to call TerminateThread() after all, which is a
159 relief, as the docs have a BIG RED WARNING SIGN about using that
160 API. Instead, when closing a fd channel that has a reader thread
161 running, just mark it as non-running and additionally mark the fd
162 as ripe for closing. When the reader thread hopefully eventually
163 gets something (and EOF or some actual data), it will note that it
164 shouldn't be running, break out of the loop, and close the fd.
166 The socket channel closing code should probably be changed
167 similarily, but that will have to wait until I have a test case.
169 (g_pipe_readable_msg, g_io_channel_win32_new_pipe,
170 g_io_channel_win32_new_pipe_with_wakeups,
171 g_io_channel_win32_pipe_request_wakeups,
172 g_io_channel_win32_pipe_readable): Remove these, have been
173 obsolete for some time.
175 * gutils.c (g_basename, g_dirname): Don't warn about deprecation
176 on Win32. Code written for GLib 1.2 doesn't have much choice but
177 to use GLib >= 1.3 on Win32.
181 2001-01-09 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
183 * gmem.c: Made g_profile_mutex a GMutex* instead of
184 G_LOCK_DEFINE_STATIC to avoid deadlock for thread implementations
185 without native static mutexes. Contruct g_profile_mutex in
188 2001-01-06 Tor Lillqvist <tml@iki.fi>
190 * gconvert.c (g_locale_to_utf8, g_locale_from_utf8): Get len using
191 strlen() if arg is negative in the Win32 code, too.
193 * giowin32.c: Changes necessary to be able to run
194 mainloop-test. We can't close the fd that our (internal) reader
195 thread is sitting doing a blocking read() from. We must terminate
196 the thread first. Keep track of thread handle, and close it when
197 thread is dying. Start reader thread with the lower-level
198 CreateThread() instead of _beginthreadex() from the C runtime, in
199 order to be able to use TerminateThread(). Hopefuly this isn't
204 * tests/makefile.{mingw,msc}.in (TESTS): Add mainloop-test and
207 * tests/mainloop-test.c: Portability: <unistd.h>, need <fcntl.h>
210 * tests/unicode-encoding.c (process): Add missing "line" argument
211 to fail(). On Win32, convert UTF-16LE, as libiconv'c ivonf always
212 converts to UTF-16BE if we ask for unspecific UTF-16.
213 (main) Handle also '\r'.
215 Fri Jan 5 11:25:42 2001 Owen Taylor <otaylor@redhat.com>
217 * configure.in (PACKAGE): move $enable_debug down below
218 checks for GCC to avoid setting CFLAGS prematurely,
219 change checks to avoid adding -g twice.
221 * gutf8.c (g_ucs4_to_utf8): Support len < 0 to mean
224 * gutf8.c (g_utf8_to_ucs4): Terminate result with 0.
226 * tests/mainloop-test.c (main): Fix uses of
227 g_main_loop_destroy().
229 * tests/unicode-encoding.c tests/Makefile.am tests/utf8.txt:
230 Tests for unicode-conversion code.
232 * gconvert.c (g_convert, g_convert_with_fallback): work around
233 a couple of GNU libc bugs.
235 * gconvert.[ch] (g_{locale,filename}_{to,from}_utf8): Standardize
236 arguments to match g_convert(). Document.
239 - Implement conversion functions to and from UTF-16
240 - Standardize unicode conversion functions on prototype like
242 - Add a lot of error checking to unicode conversion functions.
244 * gunicode.[ch] (g_utf8_to_ucs4_fast): Add fast, non-checking
245 variant of g_utf8_to_ucs4.
247 * gutf8.c (g_utf8_validate):
248 - add g_return_if_fail (str != NULL).
249 - add checks for overlong strings, non-valid Unicode characters (>= 110000)
250 and single surrogates.
252 2001-01-05 Tor Lillqvist <tml@iki.fi>
254 * testglib.c (main): Add test for g_path_skip_root().
256 * gfileutils.c (g_file_open_tmp): (Win32:) Look also for (illegal)
257 forward slashes in the template.
259 * gutils.c (g_path_skip_root): On Win32, skip the \\server\share
260 part of UNC paths. On all platforms, skip several initial
261 slashes. Add a few comments.
262 (g_get_any_init): On Win32, in case HOME is Unix-style with
263 (forward) slashes (some other applications apparently set it up
264 this way, convert to backslashed form.
266 * configure.in (glib_os): Remove stray 'v'. Add case for mingw,
267 although using configure for mingw surely doesn't work yet.
271 2001-01-02 Havoc Pennington <hp@redhat.com>
273 * configure.in: remove glib-config-2.0
275 * Makefile.am: remove glib-config-2.0
277 2001-01-03 Havoc Pennington <hp@pobox.com>
279 * tests/Makefile.am (INCLUDES): -I$(top_srcdir)/gmodule,
280 fix from Michael Meeks
282 * Makefile.am (INCLUDES): DISABLE was spelled wrong
284 Wed Jan 3 14:10:49 2001 Owen Taylor <otaylor@redhat.com>
286 * gmain.[ch]: Switch GMainLoop to be ref/unref, use to
287 make dropping reference to running loop safe.
289 Wed Dec 13 20:41:49 2000 Owen Taylor <otaylor@redhat.com>
291 * gmain.c (g_source_unref_internal): Unref callback->cb_data
292 if it was still set when the source is freed. (Usually, this
293 will be done by g_source_destroy.)
295 2001-01-02 Dan Winship <danw@helixcode.com>
297 * garray.h (g_array_append_val, g_array_prepend_val,
298 g_array_insert_val): Use parentheses around an argument to make
299 these cause an error if you pass a non-lvalue for the value,
300 rather than silently doing the wrong thing.
302 2000-12-29 Tor Lillqvist <tml@iki.fi>
304 * glibconfig.h.win32.in: Add GLIB_SIZEOF_VOID_P and GLIB_SIZEOF_LONG.
308 * {.,*}/makefile.{mingw,msc}.in: Add -DG_ENABLE_DEBUG.
310 Fri Dec 29 14:53:18 2000 Tim Janik <timj@gtk.org>
312 * configure.in: we can't grow _cv_ variables by using a backticked
313 expr that refers back to the variable (glib_cv_sizeof_system_thread for
314 G_THREAD_USE_PID_SURROGATE), that'd keep the variable growing every time
315 it's evaluated. quantum states, anyone?)
317 Thu Dec 28 10:21:46 2000 Tim Janik <timj@gtk.org>
319 * gmem.[hc]: got rid of outdated dmalloc support. provide g_try_malloc()
320 and g_try_realloc() which _may_ fail and return NULL.
321 nuked g_mem_check(), provided GMemVTable for memory function
322 virtualization, alterable at program startup with g_mem_set_vtable().
323 provided glib_mem_profiler_table and g_mem_profile() to support limited
324 profiling information out of the box (uses mprotect() for free()ed areas
326 provide globally visible G_MEM_ALIGN.
329 * docs/macros.txt: file to get a clue about the various configuration
332 * docs/debugging.txt: explain debugging traps.
334 * configure.in: got rid of --enable-mem-check and --enable-mem-profile,
335 define GLIB_SIZEOF_VOID_P and GLIB_SIZEOF_LONG. check malloc prototypes
336 and define SANE_MALLOC_PROTOS is we can use them.
337 <boy, is this file a mess>
339 * gutils.c, gscanner.c: fix up compatibility warnings, use g_message().
341 2000-12-27 Tor Lillqvist <tml@iki.fi>
343 * README.win32: Update.
345 2000-12-25 Tor Lillqvist <tml@iki.fi>
347 * gmessages.c: (Win32) Use a MessageBox for fatal
348 messages. Collect eror message into a buffer, and display that.
352 * glibconfig.h.win32.in: Update. Remove unused wchar and wctype
353 macros, add G_MODULE_SUFFIX.
355 2000-12-24 Ali Abdin <aliabdin@aucegypt.edu>
357 * Makefile.am, gcompat.h, glib.h: New gcompat.h header file
358 as recommended by Havoc.
360 * gdate.c, gdate.h, testgdate.c,
361 docs/reference/glib/glib-sections.txt,
362 docs/reference/glib/tmpl/date.sgml, tests/date-test.c: Rename some
363 of the gdate functions to use the '_get' in their name. Patch
366 2000-12-22 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
368 * configure.in: Determine the suffix of the shared librarries for
369 this system. This is done analogous to
370 ltconfig.sh. G_MODULE_SUFFIX in glibconfig.h is set to either
371 "sl", "dll", or (most often) "so".
373 * tests/Makefile.am, tests/module-test.c,
374 tests/libmoduletestplugin_a.c, tests/libmoduletestplugin_b.c:
375 Added new testcase for gmodule. This is mostly copied from
376 gmodule/testgmodule.c, but unlike that is is quiet. (Why BTW are
377 some tests that verbose, not to say loquacious...)
379 2000-12-19 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
381 * grand.c: Updated G_RAND_DOUBLE_TRANSFORM to be more
382 accurate. Redid g_rand_double() such that it returns 52 bits after
383 the point instead of 32 as before. That OTOH requires calling
384 g_rand_int() twice. Overhauled g_rand_int_range(), which is easier
385 now thanks to the new precision of g_rand_double(). Thanks to
386 Sverre Johansen <sj@ifi.uio.no> for the hint.
388 * grand.h: Added g_rand_boolean() and g_random_boolean()
389 macros. While they could be omitted due to extreme simplicity,
390 they make intention clearer in code and are therefore good to have.
392 * grand.c, grand.h: Renamed all 'min' and 'max' parameters to'
393 begin' and 'end' resp. to avoid making people think, that 'max' is
394 included in the interval. 'end' now isn't, whereas 'begin'
395 is. That's similar to the use in the STL.
397 * gslist.c, glist.c: Ok, I'm a moron. When I originally
398 implemented ENABLE_GC_FRIENDLY, I forgot to include config.h into
399 the affected files. Now that Alex did that for those two,
400 inevitable typos surfaced, which are now fixed.
402 * garray.c, ghash.c, gqueue.c, gtree.c: Include config.h as well,
403 as ENABLE_GC_FRIENDLY should be known.
405 2000-12-19 Alexander Larsson <alexl@redhat.com>
408 Added --disable-mem-pools option.
414 Disable free list and memory chunks if DISABLE_MEM_POOLS is defined.
416 2000-12-17 Tor Lillqvist <tml@iki.fi>
418 * gutf8.c (g_utf8_get_charset_internal): (Win32) Use GetACP to get
419 the current ANSI codepage.
421 * gunicode.h: Add comment that the static string g_get_charset
422 sets the parameter to point to should be copied in case the
423 charset might be changed later in the program.
425 2000-12-14 Tor Lillqvist <tml@iki.fi>
427 * makefile.{mingw,msc}.in: No need to -DGSPAWN_HELPER when
428 compiling gspawn-win32-helper any longer.
430 * giowin32.c (g_io_win32_dispatch): Warn if no callback. Call
432 (g_io_win32_create_watch): Fix typo.
433 (g_io_win32_fd_create_watch): Ditto.
434 (g_io_channel_unix_new): If it is a file descriptor (i.e., a Unix
435 fd lookalike provided by the C library), call
436 g_io_channel_win32_new_fd(). If it is a socket (from WinSock),
437 call g_io_cahnnel_win32_new_stream_socket(). Hopefully sockets and
438 fds don't overlap. TODO: Implement also datagram sockets.
439 (g_io_channel_win32_poll): Call g_main_context_get_poll_func().
441 * gcompletion.h: Include <unistd.h> only on Unix. Is this
442 inclusion really needed here? OTOH, do include <stddef.h>, for
445 * gmessages.c: (Win32) Don't define a function called "write" that
446 might clash with the prototype from <io.h>, use a #define.
450 * gmain.c (g_source_add_poll): Don't return a value from void
452 (g_main_context_get_poll_func): Compile also for non-Win32, as
453 presumably was intended. The result var is a GPollFunc, not a
454 GPollFunc*. Return the result!
456 2000-12-13 Havoc Pennington <hp@redhat.com>
458 * gconvert.c (open_converter): make static
460 * gutf8.c (g_utf8_validate): Simplify logic a bit, maybe
461 speeding it up - now we just return FALSE if we had to bail out
462 for any reason before getting to the end of the string, as defined
463 by a nul byte if len was -1, defined by the len otherwise. This
464 also fixes a bug where nul bytes were not treated as invalid
465 when the length was specified.
467 2000-12-12 Havoc Pennington <hp@pobox.com>
469 * gmain.c (g_main_context_destroy): don't try to use thread stuff
470 unless G_THREADS_ENABLED
471 (g_main_context_query): ditto
472 (g_main_context_check): ditto
473 (g_main_loop_quit): ditto
475 Tue Dec 12 18:58:22 2000 Tim Janik <timj@gtk.org>
477 * ghash.c (g_hash_table_remove): return whether a value
480 Tue Dec 12 15:18:10 2000 Owen Taylor <otaylor@redhat.com>
482 * gmain.[ch]: Revert unauthorized changes.
484 2000-12-12 Elliot Lee <sopwith@redhat.com>
486 * gmain.c, gmain.h (g_main_context_new, g_main_context_destroy):
487 GMainContext useful in implementing some additional styles of
488 main loop usage. To do this, however, Joe Hacker needs to be able
489 to create/destroy GMainContext's at will. This is just an export
490 of existing functionality, rather than any new functionality.
492 They are listed in the "Low level functions for implementing custom
493 main loops" section of the header file, to avoid confusing people.
495 Sun Dec 10 10:47:11 2000 Owen Taylor <otaylor@redhat.com>
497 * gmain.c (g_source_destroy_internal): Remove pollfds
498 from the context here, not when actually freeing the
501 * gmain.c (g_source_unref_internal): Free source list
502 and source, call source->source_funcs->destroy().
504 * giochannel.c: Unreference io_channel properly.
506 Thu Dec 7 15:22:30 2000 Owen Taylor <otaylor@redhat.com>
508 * tests/mainloop-test.c (recurser_start): Add a bunch
511 * gmain.c (g_source_attach): Reference the source
512 when adding (pointed out by Elliot)
514 2000-12-08 Raja R Harinath <harinath@cs.umn.edu>
516 * acglib.m4 (GLIB_AC_DIVERT_BEFORE_HELP):
517 Rename from GLIB_DIVERT_BEFORE_HELP.
518 Update to track autoconf 2.49b.
519 * configure.in: Reflect above change.
520 (AC_EGREP_HEADER): Rename from really obselete AC_HEADER_EGREP.
521 (debug_default): Replace "if test `expr ...`" with "case".
523 * tests/Makefile.am (TESTS): Rearrange into other variables, and
524 include run-markup-tests.sh.
525 (TESTS_ENVIRONMENT): New. Pass $srcdir to tests.
526 (noinst_PROGRAMS): Rename to ...
527 (check_PROGRAMS): ... this. 'automake' ensures that these are
528 built before running the tests.
529 * tests/run-markup-tests.sh: Support $srcdir != $builddir.
531 2000-12-08 Havoc Pennington <hp@pobox.com>
533 * tests/Makefile.am (TESTS): move markup-test to noinst_PROGRAMS;
534 it isn't a proper test, the proper test would be
535 run-markup-tests.sh, but that can't go in tests, so we need
536 a manual make check rule. Didn't do that yet.
538 2000-12-07 Raja R Harinath <harinath@cs.umn.edu>
540 * gmain.h: Don't put anything after an #endif.
543 2000-12-06 Havoc Pennington <hp@pobox.com>
545 * tests/strfunc-test.c (main): add g_strdupv test
547 * gstrfuncs.c (g_strdupv): Add a function to copy
550 Tue Dec 5 12:23:04 2000 Owen Taylor <otaylor@redhat.com>
552 * gmain.[hc]: Major change in API for creating sources
553 to handle multiple main loops (GMainContext *).
555 GSources are now exposed as GSource * and implemented
556 with structure derivation.
558 * giochannel.[ch]: Changed vtable for GIOChannel to correspond
559 to the new mainloop API, add g_io_channel_create_watch().
561 * gtypes.h: Move GTimeVal here.
563 * gthread.h: Remove gmain.h include to avoid circularity.
565 * giounix.c: Update for new GMain API.
567 * giowin32.c: Update for new GMain API. (No check for
568 proper compilation or working.)
570 * timeloop.c timeloop-basic.c: A benchmarking program for
571 the main loop comparing the main loop against a
572 hand-written (timeloop-basic.c) variant.
574 * tests/mainloop-test.c: New torture test of mainloop.
576 * docs/Changes-2.0.txt: Started. Added text about
579 * gmain.c (g_main_add_poll_unlocked): Initial fd->revents
580 to zero. (#8482, Benjamin Kahn)
582 2000-12-01 Tor Lillqvist <tml@iki.fi>
584 * {.,*}/makefile.msc.in: Include make.msc from GLib's build subdir.
586 * makefile.{mingw,msc}.in (glib_OBJECTS): Add gunibreak.
588 * glib.def: Update correspondingly.
590 2000-11-21 Havoc Pennington <hp@pobox.com>
592 * gmacros.h: Provide G_CONST_RETURN which is 'const' by default,
593 and nothing when G_DISABLE_CONST_RETURNS is defined.
595 2000-11-29 Havoc Pennington <hp@redhat.com>
597 * gunidecomp.c (COMBINING_CLASS):
598 s/UNICODE_LAST_CHAR/G_UNICODE_LAST_CHAR/
600 * gunichartables.h: Update for data in Unicode 3.0.1
602 * gunidecomp.h: Ditto
604 * gunicode.h (GUnicodeBreakType): Enum for line break properties
605 (g_unichar_break_type): Get the break property for a char
607 * gunibreak.h: Autogenerated line break property tables
609 * gunibreak.c (g_unichar_break_type): added
611 * Makefile.am (EXTRA_DIST): dist gen-unicode-tables.pl
613 * Makefile.am (libglib_1_3_la_SOURCES): Add gunibreak.h, gunibreak.c
615 * gen-unicode-tables.pl: Include the script to update the unicode
618 2000-11-28 Elliot Lee <sopwith@redhat.com>
620 * gmarkup.c: Fix warnings.
622 * guniprop.c, gunidecomp.c: Make warnings go away by using
623 GPOINTER_TO_INT() instead of (int).
625 * gcompletion.[ch]: Add g_completion_set_compare(),
626 to allow (for example) using case-insensitive completion.
628 2000-11-28 Tor Lillqvist <tml@iki.fi>
630 Patches by Hans Breuer:
632 * gspawn-win32.c: Move the code for gspawn-win32-helper to its own
635 * makefile.{mingw,msc}.in: Change accordingly.
637 * gspawn-win32-helper.c: New file.
639 * Makefile.am (EXTRA_DIST): Add it.
641 * gmarkup.c (find_current_text_end): Fix assertion not to check an
642 uninitialised variable.
644 2000-11-28 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
646 * gthread.c: Set the thread data before locking the mutex, because
647 the locking call might use g_thread_self ().
649 * gthread.h: Do only show the location of the locking/unlocking
650 for -DG_ERRORCHECK_MUTEXES and not the name of the mutex. Add the
651 errorcheck capability for g_cond_wait and g_cond_timed_wait as
654 2000-11-27 Havoc Pennington <hp@redhat.com>
656 * gthread.h: Add void in empty function arg list
658 2000-11-22 Tor Lillqvist <tml@iki.fi>
660 * makefile.{mingw,msc}.in: Add gqsort.
664 2000-11-21 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
666 * configure.in: Add a surrogate for thread priorities using PID
667 niceness for systems with no thread priorities and different PIDs
668 for threads of the same process (most notably: Linux). Define
669 G_THREAD_USE_PID_SURROGATE in that case, as used by
670 gthread-posix.c. Also make the system thread bigger by
671 sizeof (long) to contain the thread's PID.
673 * gfileutils.c: Include stdlib.h for mkstemp prototype.
675 * gthread.c: Add priority range checks to the affected functions.
677 * gthreadpool.c: Remove unused variable.
679 Mon Nov 20 18:55:17 2000 Jonathan Blandford <jrb@redhat.com>
681 * gtree.[hc]: Patch from David Benson <daveb@idealab.com> to add
682 user_data support to gtree functions.
684 Mon Nov 13 18:35:52 2000 Jonathan Blandford <jrb@redhat.com>
686 * gtypes.h (GCompareFuncData): new func type to let you use user
687 data when comparing nodes.
689 * gslist.c (g_list_sort_with_data): new function to sort with
692 * glist.c (g_list_sort_with_data): new function to sort with
695 * garray.[ch]: Added convenience functions to sort arrays.
697 2000-11-16 Havoc Pennington <hp@redhat.com>
699 * guniprop.c (g_unichar_isspace): Use a switch here, maybe helps
700 the compiler optimize things. Also, ' ' is a SPACE_SEPARATOR,
701 so don't special case it.
703 2000-11-17 Tor Lillqvist <tml@iki.fi>
705 * glib.def: Add g_trash_stack entry points.
707 Fri Nov 17 15:43:00 2000 Owen Taylor <otaylor@redhat.com>
713 * tests/Makefile.am (dist-hook): Add code to distribute
716 2000-11-15 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
718 * configure.in: Check for the sched.h header and include it on
719 gthread/gthread-posix.c if available.
721 * configure.in: Add -D_POSIX4_DRAFT_SOURCE to
722 GTHREAD_COMPILE_IMPL_DEFINES. Also add -D_POSIX4A_DRAFT10_SOURCE
723 to G_THREAD_CFLAGS. Really deploy GTHREAD_COMPILE_IMPL_DEFINES,
724 when searching for thread libs. Look for sched_* functions in
725 -lrte as well. All of that is necessary on DG/UX.
727 * configure.in: Use AC_TRY_COMPILE instead of AC_EGREP_HEADERS in
728 various places to make it work more reliable, to make it accept
729 macros instead of functions etc.
731 * configure.in: Replace some NULL's for checks with 0 to make it
732 work without stdio.h everywhere.
734 * configure.in, gutils.c: changed the test for getpwuid_r to first
735 test for a posix version and then for a non-posix version. No code
736 change in gutils.c. Again this change deals better with getpwuid_r
737 being a macro and not a function. Most of the above with kind help
738 from Tethys <tet@isengard.europe.dg.com>. This fixes Bug #13403.
740 2000-11-14 Tor Lillqvist <tml@iki.fi>
742 * gwin32.h: Make #endif comment match #ifdef.
744 Mon Nov 13 14:00:20 2000 Owen Taylor <otaylor@redhat.com>
746 * configure.in: Up version to 1.3.2
748 * gconvert.h (enum GConvertError): Remove trailing ,
750 * gfileutils.c (g_file_open_tmp): Fix comment to
751 properly describe return value.
753 2000-11-13 Tor Lillqvist <tml@iki.fi>
755 * config.h.win32.in: Add USE_LIBICONV.
757 * gconvert.c: Check G_OS_WIN32 only after including glib.h.
761 2000-11-13 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
763 * gthread.c (g_static_rec_mutex_*): Made recursive mutexes also
764 work when the thread system is not (yet) initialized.
766 Sun Nov 12 18:34:32 2000 Owen Taylor <otaylor@redhat.com>
768 * gconvert.[ch]: Create wrapper functions for iconv()
769 so that we can transparently use the native iconv,
770 libiconv, or (in the future) a mini-iconv included
773 * glib-config-2.0.in glib-2.0.pc.in: Include @ICONV_LIBS@
775 * INSTALL: Added note about libiconv.
777 * configure.in: Add checks for libiconv from pango. If
778 EILSEQ is not defined in errno.h add define for it into
779 glibconfig.h so g_iconv can use it. (Note, recompiling
780 from a system without EILSEQ to a system with EILSEQ
781 will break binary compatibility)
783 2000-11-12 Robert Brady <robert@suse.co.uk>
785 * gstrfuncs.c, gstrfuncs.h: Remove g_filename_{to,from}_utf8
787 * gconvert.c, gconvert.h: Add g_filename_{to,from}_utf8 and
788 g_locale_{to.from}_utf8. The locale_ variant honours
789 nl_langinfo(CODESET), the filename_ variant uses UTF-8 unless
792 (g_convert): Add G_CONVERT_ERROR_PARTIAL_INPUT error, if bytesread
793 != length and no bytesread pointer passed.
795 Sun Nov 12 15:29:53 2000 Owen Taylor <otaylor@redhat.com>
797 * gfileutils.[ch]: template is a reserved word in
798 C++ s/template/tmpl/.
800 2000-11-11 Havoc Pennington <hp@pobox.com>
802 * gmarkup.c (g_markup_parse_context_parse): Handle a long stream
803 of bytes containing no UTF-8 character starts
805 2000-11-11 Tor Lillqvist <tml@iki.fi>
807 * glib.def: Add missing entry points.
809 * gfileutils.c (g_mkstemp): Improve chance to generate unique
810 names with less effort a bit.
812 * gfileutils.h: Add g_file_open_tmp() declaration.
814 * testglib.c: Include <io.h> on Win32.
816 * makefile.mingw.in: Correct the way to invoke sub-makes.
818 Sun Nov 5 13:20:54 2000 Owen Taylor <otaylor@redhat.com>
820 * glib-object.h: Add gtypemodule.h
822 2000-11-11 Tor Lillqvist <tml@iki.fi>
824 * gfileutils.c (g_file_open_tmp): New function, suggested by Havoc
826 (g_mkstemp): Use only one case for letters in temp file name, as
827 this will be used on systems with case-insensitive file systems.
829 * testglib.c (main): Test g_mkstemp() and g_file_open_tmp().
831 2000-11-09 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
833 * gthreadpool.c: Don't take other threads with other priorities
834 into account as changing the priority is highly
835 unportable. (Actually using it at all already is unportable, but
836 even sometimes where that works, changing priority is not
839 2000-11-05 Havoc Pennington <hp@pobox.com>
841 * gmarkup.h: rename G_MARKUP_FOO to
842 G_MARKUP_DO_NOT_USE_THIS_UNSUPPORTED_FLAG
844 * gmarkup.c: don't start doc comments with "Returns"
846 2000-11-05 Havoc Pennington <hp@pobox.com>
848 * gmarkup.c: inline docs
849 (unescape_text): properly check strtoul for failure.
851 * gerror.c (g_propagate_error): Free the src error if the dest
852 location is NULL - I'm pretty sure that's what this function was
855 2000-11-05 Havoc Pennington <hp@pobox.com>
857 * gutils.c (g_find_program_in_path): cleanup docs, sync param
858 names to those in the header
860 * gfileutils.c (g_mkstemp): clean up docs
862 * gshell.h: sync param names with param names in .c file
864 * gfileutils.h (enum GFileTest): remove trailing comma from last
865 member, confuses gtk-doc
867 * gmarkup.h: s/GMarkupErrorType/GMarkupError/g; to follow
870 2000-11-02 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
872 * gasyncqueue.c: Added documentation for asyncronous queues.
874 * gspawn.c: Include sys/select.h (some platforms need it for
877 * gspawn.c: Changed unportable __FUNCTION__ to the verbatim
880 2000-10-31 Tor Lillqvist <tml@iki.fi>
885 * gfileutils.h: Actually, g_mkstemp() is better suited in gfileutils.
887 2000-10-30 Tor Lillqvist <tml@iki.fi>
889 * configure.in: Check for mkstemp.
891 * gutils.c (g_mkstemp): New function. If HAVE_MKSTEMP, just call
892 it, otherwise use code lifted from glibc.
894 * gutils.h: Declare it.
896 * glib.def: Here, too. Plus two missing functions.
898 2000-10-30 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
900 * gcache.h, gcache.c, ghash.h, ghash.c, grel.c, grel.h, gtypes.h:
901 Introduced new function type GEqualFunc to return TRUE for equal
902 params. This is now used instead of GCompareFunc (which should
903 work akin to strcmp) here. This kind of fixes Bug #14412. Note
904 that technically GCompareFunc and GEqualFunc are still the same
905 types, as gint == gboolean.
907 * ghash.h, gutils.c: g_int_equal and g_direct_equal now return
908 gboolean to be really become GEqualFunc.
910 * gscanner.c, testglib.c, tests/hash-test.c: Some tiny changes to
911 follow the above change.
913 2000-10-27 Tor Lillqvist <tml@iki.fi>
915 * makefile.{mingw,msc}.in (glib_OBJECTS): Add gmarkup.
917 * glib.def: Add missing functions.
919 * tests/makefile.{mingw,msc}.in (TESTS): Add markup-test.
921 2000-10-24 Havoc Pennington <hp@pobox.com>
923 * gmarkup.h, gmarkup.c: New module to parse a simple
926 * Makefile.am: add gmarkup.h, gmarkup.c
928 * tests/Makefile.am: add markup-test
930 * gstring.h (g_string_new_len): new function to create a string
932 (g_string_new): avoid a gratuitous realloc
934 2000-10-26 Tor Lillqvist <tml@iki.fi>
936 * makefile.{mingw,msc}.in: Cosmetics.
938 2000-10-26 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
940 * gstrfuncs.c (g_strsplit): When the string is ended by a
941 delimiter, return an extra empty string just like for a delimiter
942 at the start of the string. This makes the function behave more
943 consistent and also fixes Bug #15026.
945 Tue Oct 24 22:09:14 2000 Tim Janik <timj@gtk.org>
947 * glib-object.h: added newly added gobject/ headers.
949 * gmesage.c: print g_message() output to stderr instead of stdout.
951 2000-10-23 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
953 * configure.in: Use one = instead of two, which is plainly wrong.
955 2000-10-19 Tor Lillqvist <tml@iki.fi>
957 * makefile.msc.in: Pass -DGSPAWN_HELPER when building it. Link
961 * gfileutils.c: Make them compile with picky MSVC.
963 * gwin32.h: New file. Move Win32-only stuff that isn't related to
964 GIOChannels here from giochannel.h.
966 * Makefile.am: Add it here.
968 * giochannel.h: Move stuff to gwin32.h.
970 * glib.h: On Win32, include gwin32.h.
972 Mon Sep 11 10:03:24 2000 Owen Taylor <otaylor@redhat.com>
974 * glib.h convert.c (g_convert_with_fallback): Change
975 ERROR_OTHER to ERROR_FAILED, add some dummy marking with _().
977 * docs/reference/glib/glib-sections.txt
978 docs/reference/glib/glib-docs.sgml Update for g_convert,
981 2000-10-16 Tor Lillqvist <tml@iki.fi>
983 * glibconfig.h.win32.in: Remove alloca stuff from here. galloca.h
984 takes care of it, correctly.
986 * giowin32.c (reader_thread): Some more debugging output.
987 (g_io_channel_win32_poll): Remove unused vars.
989 * gfileutils.c: Changes for Win32, with no unistd.h and no
992 * gspawn-win32.c: Implementation of the g_spwan_* functions for
993 Win32. Due to the general non-Unixness of Win32, much of the
994 functionality that is relatively clean to implement on Unix, is
995 hard to do on Win32. We must use a separate helper program to
996 change directory, close extra file descriptors, redirect the std
997 ones, as needed, and only then start the child process. No child
998 process pid can be returned, unfortunately. Or if we used
999 CreateProcess directly, it probably could. (Now we use the spawnv*
1000 functions from msvcrt.)
1002 * Makefile.am (EXTRA_DIST): Add gspawn-win32.c
1004 * glib.def: Add new entry points.
1007 * giowin32.c: Remove g_io_channel_win32_wait_for_condition(),
1008 g_io_channel_win32_poll() subsumes it.
1010 * gbacktrace.h: G_BREAKPOINT for MSVC (on the ix86).
1012 * gwin32.c (g_win32_getlocale): Use "sp" for
1013 LANG_CROATIAN+SUBLANG_SERBIAN_LATIN.
1015 * makefile.{mingw,msc}.in (glib_OBJECTS): Add new files.
1016 Add gspawn-win32-helper.exe rule.
1018 * tests/makefile.{mingw,msc}.in (TESTS): Add shell-test and
1021 * tests/spawn-test.c: (run_tests): On Win32, don't try to run
1022 /bin/sh, but ipconfig (no special significance in choosing that,
1023 just a program that outputs something to stdout).
1025 2000-10-15 Raja R Harinath <harinath@cs.umn.edu>
1027 Remove need for acconfig.h, and misc. cleanups.
1028 * acglib.m4 (GLIB_SIZEOF): Add 'autoheader' comment to
1030 (GLIB_BYTE_CONTENTS): Likewise.
1032 * configure.in: Add 'autoheader' comments to all AC_DEFINE(...)
1033 and AC_DEFINE_UNQUOTED(...) lines.
1034 Replace AC_MSG_CHECKING/AC_CACHE_VAL with AC_CACHE_CHECK.
1035 (AM_PROG_LIBTOOL): Move after AC_PROG_CC.
1037 * acconfig.h: Empty out.
1039 * Makefile.am (BUILT_EXTRA_DIST): New variable. List 'dist'able
1040 files that are created in the builddir.
1041 (dist-hook): Handle those files.
1042 (libglib_1_3_la_SOURCES): Remove @ALLOCA@. @ALLOCA@ should only
1043 be used in an _LDADD or _LIBADD, since it expands (if necessary)
1046 * tests/Makefile.am (BUILT_EXTRA_DIST): New variable.
1047 (dist-hook): Handle $(BUILT_EXTRA_DIST).
1049 2000-10-13 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1051 * grand.c: Added inline documentation.
1053 * gtypes.h, gnode.h, gutils.h: Readded GFreeFunc,
1054 g_node_insert_after and g_find_program_in_path resp., which
1055 mysteriously disappeared during the glib.h dissection.
1057 2000-10-12 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1059 * glibconfig.h.win32.in: Adapted accordingly to header separation
1060 and GLIB_HAVE_ALLOCA_H renaming.
1062 * Makefile.am: Added the new headers to glibinclude_HEADERS.
1064 * glib.h: Forgot to include gerror.h.
1066 * glib.h, galloca.h, garray.h, gasyncqueue.h, gbacktrace.h,
1067 gcache.h, gcompletion.h, gconvert.h, gdataset.h, gdate.h, ghash.h,
1068 ghook.h, giochannel.h, glist.h , gmacros.h, gmain.h, gmem.h,
1069 gmessages.h, gnode.h, gprimes.h, gquark.h, gqueue.h, grand.h,
1070 grel.h, gscanner.h, gslist.h, gstrfuncs.h, gstring.h, gthread.h,
1071 gthreadpool.h, gtimer.h, gtree.h, gtypes.h, gutils.h: Split glib.h
1072 into many header files mostly according to the resp. *.c-files.
1074 * gmacros.h: Added G_BEGIN_DECLS and G_END_DECLS to mean: 'in case
1075 of C++: extern "C" { ... }' analogous to glibc __BEGIN_DECLS and
1078 * configure.in, gerror.h, gfileutils.h, gshell.h, gspawn.h,
1079 gunicode.h, : Changed guard-macro names to something more
1082 * configure.in, *.h: Use G_BEGIN_DECLS and G_END_DECLS.
1084 * configure.in: Defined GLIB_HAVE_ALLOCA_H instead of including
1085 alloca.h in glibconfig.h, GLIB_HAVE_ALLOCA_H is used in glib.h.
1087 * configure.in: Removed cruft from old threading code.
1089 2000-10-09 Raja R Harinath <harinath@cs.umn.edu>
1091 Work with beta autoconf 2.50.
1092 * configure.in (GLIB_MAJOR_VERSION): Use GLIB_DIVERT_BEFORE_HELP
1093 instead of AC_DIVERT_PUSH(),AC_DIVERT_POP.
1094 (AC_CHECK_HEADERS): Remove redundant AC_DEFINE(HAVE_...).
1095 (REALLOC_0_WORKS): Move AC_DEFINE outside AC_CACHE_VAL.
1096 (dlopen): Quote nested AC_CHECK_... calls.
1098 * acglib.m4 (GLIB_TR_SH, GLIB_TR_CPP): Utility macros copied from
1100 (GLIB_DIVERT_BEFORE_HELP): New macro that works both with autoconf
1101 2.13 and beta autoconf 2.50.
1102 (GLIB_SIZEOF, GLIB_BYTECONTENTS): Use GLIB_TR_*.
1104 * Makefile.am (CONFIGURE_DEPENDENCIES): Use this to specify
1105 that 'configure' depends on acglib.m4.
1107 2000-10-09 Havoc Pennington <hp@redhat.com>
1109 * Makefile.am, tests/Makefile.am: Add new files.
1111 * tests/spawn-test.c, tests/shell-test.c: new tests for
1112 the shell/spawn stuff
1114 * gutils.c (g_find_program_in_path): convert a relative
1115 program name into an absolute pathname to an existing
1118 * gspawn.h, gspawn.c: New fork/exec API
1120 * gshell.h, gshell.c: Shell-related utilities, at the moment
1121 simply routines to parse argv and quote/unquote strings
1123 * guniprop.c (g_unichar_isspace): Return TRUE for the
1124 ASCII space characters isspace() returns TRUE for.
1126 * gfileutils.c (g_file_get_contents): Convenience function
1127 to slurp entire file into a string and return it. Partially
1128 written by Joel Becker.
1129 (g_file_test): file test function
1131 2000-10-06 Tor Lillqvist <tml@iki.fi>
1133 * makefile.msc.in: Revamp to be like makefile.mingw.in, make
1134 the MSVC build actually work again.
1136 * gmodule/makefile.msc.in
1137 * gobject/makefile.msc.in
1138 * gthread/makefile.msc.in: New files, like their mingw counterparts.
1140 * gmodule/Makefile.am
1141 * gobject/Makefile.am
1142 * gthread/Makefile.am: Make and distribute them.
1144 * */makefile.mingw.in: Allow override of GLib version number from
1145 the build/win32/module.defs file.
1147 * glib.def: Add new entry point.
1149 * tests/gio-test.c (main): Fix the Win32-only code to use current
1150 API, g_io_channel_win32_make_pollfd() and g_io_channel_win32_poll().
1152 Fixes from Hans Breuer:
1154 * glib.h (struct DIR): Keep the last readdir result cached inside
1155 the DIR struct, to enable several DIRs being open simultaneously.
1157 * gwin32.c (g_win32_readdir): Use the above instead of static.
1159 * giowin32.c (g_io_channel_win32_make_pollfd): Insert cast to keep
1162 2000-10-05 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1164 * glib.h: Changed alloca stuff a bit: when we have a working
1165 alloca.h, we're not messing with alloca any further. Should fix a
1166 bug reported by Bernd Demian <wega@csc-dd.de>.
1168 2000-09-29 Jonathan Blandford <jrb@redhat.com>
1170 * gnode.c (g_node_insert_after): Added function to keep symmetry
1171 with g_node_insert_before.
1173 2000-09-29 Martin Baulig <baulig@suse.de>
1175 Several minor ANSI C fixes.
1177 Added missing casts:
1178 * gdate.c (g_date_fill_parse_tokens): `s = (guchar *) str'.
1179 * gmain.c (g_idle_dispatch): `func = (GSourceFunc) source_data'.
1180 (g_idle_add_full): `(gpointer) function' in call to g_source_add().
1181 * gstrfuncs.c (g_strdown): `s = (guchar *) string' and
1182 `return (gchar *) string'.
1183 (g_strup): Likewise.
1184 (g_strchug): `start = (guchar*) string' in 1st for() argument;
1185 `strlen ((gchar *) start)' in call to g_memmove().
1186 * gstring.c (g_string_down): `s = (guchar *) string->str'.
1187 (g_string_up): Likewise.
1188 * gthreadpool.c (stop_this_thread_marker):
1189 `(gpointer) &g_thread_pool_new'.
1190 * gunidecomp.h (decomp_table[]): Cast all the strings to
1193 Put text following #endif into comments:
1196 2000-09-29 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1198 * configure.in, glib.h: Added errorcheck mutexes. These are
1199 activated through the preprocessor symbol
1200 G_ERRORCHECK_MUTEXES. Need to add an extra word to StaticMutex in
1201 order to achieve this. g_(static_)mutex_* functions instrument the
1202 mutex operations with mutex name and location, when compiled with
1203 -DG_ERRORCHECK_MUTEXES. g_thread_init activates the errorcheck
1204 mutexes, when compiled with -DG_ERRORCHECK_MUTEXES.
1206 2000-09-28 Havoc Pennington <hp@pobox.com>
1208 * glib.h (GThreadPriority): fix indentation
1209 (GConvertError): generic error is conventionally called
1210 _FAILED rather than _OTHER, at least at the moment,
1211 according to GError docs in docs/reference.
1213 * gconvert.c: s/_OTHER/_FAILED/
1215 2000-09-28 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1217 * configure.in: Adjusted the test for an unimplemented
1218 getpwuid_r. Info from Michael Pruett. This is just a forward
1219 merge from glib-1-2.
1221 * configure.in: Moved determination of G_THREAD_FLAGS before
1222 G_THREAD_LIBS. Check for UnixWare systems and set the right cflags
1223 and libs there (it needs -Kthread for the native compiler and
1224 -pthread for gcc). Thanks to Boyd Lynn Gerber <gerberb@zenez.com>
1227 * configure.in: Fail immediately, when no thread library is found,
1228 instead of continuing searching for rt libs etc. Changed almost
1229 all occurances of $enable_threads to $have_threads, as that's,
1232 * tests/threadpool-test.c: Define vars inside the guard to avoid
1235 * configure.in, tests/type-test.c: Some platforms support 64 bit
1236 'long long', but you can not printf or scanf them. In that case,
1237 don't define G_G{UINT|INT}64_FORMAT. Changed the type-test program
1240 * gutils.c (g_get_current_dir): max_len can't be initialized
1241 statically as it might call a function. So do it at first call.
1243 Tue Sep 26 2000 Elliot Lee <sopwith@redhat.com>
1245 * glib.h: Add G_GNUC_PURE macro (but don't use it anywhere).
1247 Mon Sep 25 2000 Elliot Lee <sopwith@redhat.com>
1249 * garray.c, glib.h, gmem.c: Add a few missing G_GNUC_CONST's.
1251 2000-09-21 Tor Lillqvist <tml@iki.fi>
1253 * makefile.mingw.in: Add gconvert.o. Use libiconv.
1255 * config.h.win32.in: Define HAVE_GETCWD.
1257 * glib.def: Add new entry points.
1259 2000-09-21 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1261 * configure.in: The last released automake (1.4) still requires
1262 AM_PROG_LIBTOOL instead of AC_PROG_LIBTOOL, so use that for the
1265 2000-09-19 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1267 * acconfig.h, configure.in, gutils.c: Test for the existence of
1268 getcwd, and use it only when found.
1270 * glib.h: Only use the gcc-variable-macro-argument-extension for
1271 gcc >= 2.4. Both patches from Jonas Oberg <jonas@gnu.org>.
1273 Mon Sep 18 10:58:21 2000 Owen Taylor <otaylor@redhat.com>
1275 * gutf8.c: Implement g_ucs4_to_utf8 which was in
1276 the header file but not implemented.
1278 Sun Sep 17 2000 Elliot Lee <sopwith@redhat.com>
1280 * glib.h configure.in: Define g_alloca() as an
1281 alloca-that-works-anywhere.
1283 * gconvert.c: Fix warnings which could have caused problems on
1286 Sun Sep 10 12:37:40 2000 Owen Taylor <otaylor@redhat.com>
1288 * glib.h gconvert.c (g_convert): Havoc Pennington's implementation
1289 of convenient character set conversion using iconv, with
1290 the addition of GError. We probably need a fallback that
1291 just does conversions between, say UTF-8,16,32 and ISO-8859-1
1292 for targets without iconv at all.
1294 Also add g_convert_with_fallback() to take care of conversions
1295 where we accept some loss going to the target encoding.
1297 2000-09-10 Havoc Pennington <hp@redhat.com>
1299 * gutf8.c (g_utf8_validate): Add this function.
1301 Sat Sep 9 18:50:42 2000 Owen Taylor <otaylor@redhat.com>
1303 * gstrfuncs.c (g_strescape): Add a missing g_return_if_fail().
1305 Mon Aug 21 03:57:46 2000 Tim Janik <timj@gtk.org>
1307 * glib.h (G_BREAKPOINT): for non-i386 and non-alpha, or non gcc,
1308 implement BREAKPOINT() as raise (5 /* SIGTRAP */);
1310 * glib.h: provide user-definable switch G_IMPLEMENT_INLINES,
1311 to turn on compilation of inline function implementations provided
1312 in header files with extern linkage.
1313 wrap inline function implementations into ifdef __G_UTILS_C__, so we
1314 really only compile them for gutils.c and not also into arbitrary user
1315 code that wants to make use of G_IMPLEMENT_INLINES.
1316 adjusted comment apropriately.
1318 * gutils.c: to turn on compilation of inline functions, provide
1319 #define G_IMPLEMENT_INLINES 1 and #define __G_UTILS_C__.
1321 2000-09-06 Havoc Pennington <hp@redhat.com>
1325 * docs/reference/glib/tmpl/error_reporting.sgml: docs
1327 Wed Sep 6 10:28:34 2000 Owen Taylor <otaylor@redhat.com>
1329 * guniprop.c gunicode.h gutf8.c: Some inline docs fixes.
1331 2000-09-06 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1333 * glib.h, gtimer.c, tests/thread-test.c:
1334 s/G_MICROSEC/G_USEC_PER_SEC/
1336 * glib.h: Removed G_G{U}{SHORT|INT|LONG}_FORMAT from glib.h, as
1337 they are really superfluous.
1339 Tue Sep 5 20:16:27 2000 Owen Taylor <otaylor@redhat.com>
1341 * configure.in docs/Makefile.am: Add gtk-doc checks
1342 for newly added docs/reference/ subdir.
1344 2000-09-05 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1346 * gthread.c (g_thread_error_quark): Don't use a G_LOCK, as it
1349 2000-09-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1351 * gstring.c (g_string_free): Use g_return_val_if_fail instead of
1352 g_return_if_fail, as the function now is supposed to return
1355 * gerror.c, gerror.h (g_propagte_error): Added function
1356 g_propagte_error to hand over local errors to the calling
1359 * glib.h: Include gerror.h before it is used for some g_thread_*
1362 * gthread.c, gthreadpool.c, glib.h: Enable error reporting for
1363 thread creation, namely for g_thread_create, g_thread_pool_new,
1364 g_thread_pool_push and g_thread_pool_set_max_threads.
1366 * tests/thread-test.c, tests/threadpool-test.c: Adapted
1369 2000-08-31 Tor Lillqvist <tml@iki.fi>
1373 * giowin32.c (g_io_channel_win32_make_pollfd): New function, to
1374 make a GPollFD from a GIOChannel. Creates the events and starts
1375 the reader thread if necessary.
1378 * giowin32.c (g_io_channel_win32_poll): No use for separate
1379 condition parameter.
1381 * gmain.c (g_get_current_time): (Win32): Simplify, use
1382 GetSystemTimeAsFileTime().
1384 2000-08-27 Tor Lillqvist <tml@iki.fi>
1386 * giowin32.c (g_io_channel_win32_poll): New function, otherwise
1387 like g_io_channel_win32_wait_for_condition(), but accept several
1389 (g_io_channel_win32_wait_for_condition): Call
1390 g_io_channel_win32_poll().
1392 * glib.h: Declare g_io_channel_win32_poll().
1394 * gwin32.c (g_win32_error_message): Don't believe return value
1397 2000-08-25 Elliot Lee <sopwith@redhat.com>
1399 * glib.h, gunicode.h, gmodule/gmodule.h:
1400 Mark the following functions G_GNUC_CONST (to allow optimization)
1401 because their results are a function of only their parameters:
1402 g_int_hash, g_int_equal, g_direct_hash, g_direct_equal,
1403 g_quark_to_string, g_date_is_leap_year, g_date_days_in_month,
1404 g_date_monday_weeks_in_year, g_date_sunday_weeks_in_year,
1405 g_spaced_primes_closest, g_unichar_is*, g_unichar_to*,
1406 g_unichar_*digit_value, g_unichar_type
1408 2000-08-21 Elliot Lee <sopwith@redhat.com>
1410 * gobject/Makefile.am, gobject/gobject-query.c,
1411 gobject/gparamspecs.c: Fix inclusion of config.h
1413 Mon Aug 21 14:46:23 2000 Owen Taylor <otaylor@redhat.com>
1415 * tests/gio-test.c: Fix a couple of trivial bugs that
1416 were causing warnings.
1418 Mon Aug 21 14:39:36 2000 Owen Taylor <otaylor@redhat.com>
1420 * glib.h: Use C99 varargs macros where possible
1421 (check __STDC_VERSION__), otherwise, on gcc, use an alternate
1422 form of gcc varargs which is more likely
1423 to be supported going forward. (Based on some code
1426 2000-08-17 Darin Adler <darin@eazel.com>
1429 * garray.c: (g_array_free), (g_ptr_array_free),
1430 (g_byte_array_free): Return the data left behind.
1431 * gstring.c: (g_string_free): Return the data left behind.
1433 Changed the free calls that leave data behind so they
1434 return a pointer to the left-behind data, NULL if told not
1435 to leave anything behind. This makes these calls easier
1436 to use correctly, without any incompatible API change for
1437 callers that don't know about the return value. Of course,
1438 it would be even clearer if the free calls weren't dual-purpose
1441 2000-08-12 Tor Lillqvist <tml@iki.fi>
1443 * giowin32.c: Some indentation and spacing fixes. Add some more
1445 (g_io_win32_add_watch): New function, with common code from
1446 g_io_win32_fd_add_watch and g_io_win32_sock_add_watch. Don't start
1447 more than one reader thread for a GIOChannel. We should obviously
1448 have just one reader thread reading a file descriptor or socket.
1450 2000-08-10 Havoc Pennington <hp@redhat.com>
1452 * gthread-2.0.pc.in (Cflags): don't duplicate glib Cflags
1454 * gmodule-2.0.pc.in (Cflags): don't duplicate glib Cflags
1456 * gobject-2.0.pc.in (Cflags): don't duplicate Cflags from glib
1459 2000-08-10 Havoc Pennington <hp@redhat.com>
1461 * glib-2.0.pc.in (Cflags): Look in glib-2.0/include for
1464 2000-08-07 Tor Lillqvist <tml@iki.fi>
1466 * tests/gio-test.c (shutdown_source): New function, that calls
1467 g_source_remove(). Check return value of g_source_remove(), and
1468 decrement running subprocess counter only if g_source_remove()
1469 actually did remove the source.
1471 (recv_message): Call shutdown_source() on EOF condition. Return
1472 FALSE on G_IO_HUP and G_IO_ERR condition. Fix printf format typo.
1474 Sun Aug 6 20:06:02 2000 Tim Janik <timj@gtk.org>
1476 * gmessages.c (g_log_domain_check_free): keep *last updated while
1477 running through the domain list, so we don't screw up the removal,
1478 patch provided by Gady Kozma <gadykozma@hotmail.com>.
1480 Sun Aug 6 20:03:41 2000 Tim Janik <timj@gtk.org>
1482 * gmessages.c (g_log_remove_handler): keep *last updated while running
1483 through the handler list, so we don't screw up the removal.
1485 Sun Jul 30 16:54:13 2000 Owen Taylor <otaylor@redhat.com>
1487 * gunicode.h: Fix stray character
1489 * gutf8.c (g_unichar_to_utf8): Allow outbuf to be NULL, in
1490 which case we just compute the length.
1492 2000-07-31 Havoc Pennington <hp@redhat.com>
1494 * Makefile.am (EXTRA_DIST): forgot to add .pc.in to EXTRA_DIST
1496 2000-07-31 Havoc Pennington <hp@redhat.com>
1498 * glib-2.0.pc.in, gobject-2.0.pc.in, gmodule-2.0.pc.in,
1499 gobject-2.0.pc.in: pkg-config data files
1501 * Makefile.am: Install/dist the .pc files
1503 * configure.in: Output the .pc files
1505 2000-07-31 Tor Lillqvist <tml@iki.fi>
1507 * giowin32.c (buffer_read): The code didn't compile (must
1508 have been sleepy when committing). "return" instead of "break"
1510 (g_io_win32_fd_add_watch): Cannot check if the file descriptor is
1511 readable by calling ReadFile to read zero bytes. ReadFile blocks
1512 on NT even if trying to read nothing at all. So, don't check if
1513 file descriptor is readable; assume this function isn't called
1516 Sun Jul 30 10:44:16 2000 Tim Janik <timj@gtk.org>
1518 * gmain.c (g_get_current_time): fix tor's recent changes which
1519 got rid of a required variable in the non-windows path.
1521 2000-07-30 Tor Lillqvist <tml@iki.fi>
1523 Finally, a new and improved IO Channel and condition watch
1524 implementation for Win32. Based on code provided by Craig Setera.
1526 When watching file descriptors, for which there is no select()
1527 like functionality on Win32 that would work on all Win32 platforms
1528 for all types of file descriptors (including anonymous pipes), we
1529 start a new thread that blocks while trying to read from the file
1530 descriptor. When the read returns, a Win32 Event is signalled that
1531 the polling routine eventually notices. Meanwhile, the data being
1532 read is stored in a circular buffer, from where the IO channel's
1533 read() method picks it up.
1535 If the buffer fills up the reading thread has to wait for space
1536 becoming available. For this another Win32 Event is used. The IO
1537 Channel's read() method signals this when it has read some data
1540 The separate reader thread(s), and the circular buffer(s) with
1541 associated events mean lots of possibilities for fun parallelism
1542 errors. But it seems to work OK, i.e. GIMP runs.
1544 * gmain.c: Small changes to the Win32 polling function.
1545 (g_main_win32_get_poll_func): New function. Perhaps it would be a
1546 good idea to provide this on all platforms.
1548 * giowin32.c: The bulk of the new implementation.
1549 (g_io_channel_win32_wait_for_condition): New function. To be used
1550 where on Unix one does a select() on the channel's fd, like
1551 libgimp's gimp_extension_process(). Could be provided on all
1554 * glib.h: Update documentation for IO Channels on Win32. Remove
1555 the declarations for the as of now obsolete old functions related
1556 to IO Channels for pipes with "wakeup" messages.
1558 * glib.def: Some new functions.
1560 * tests/gio-test.c: New file, to test GIOChannel and main loop.
1563 * tests/makefile.mingw.in: Add it.
1565 (Later the same night:)
1567 * giowin32.c: Compile in the debugging code all the time, but only
1568 output debug messages if told so. Add (unadvertised) function to
1569 turn on/off debug messages for a channel.
1571 (buffer_read): Don't loop. It is expected behaviour to return a
1572 short read occasionally, for instance when reading from
1573 pipes. It's the calling code that should loop if it *knows* how
1574 much the writer has written.
1576 * tests/gio-test.c: Correct the program's name in the output.
1577 (recv_message): Loop calling g_io_channel_read() (in a new
1578 function read_all()) until we have all the bytes we want (that we
1579 know the writer has written/will write).
1581 Thu Jul 27 05:15:11 2000 Tim Janik <timj@gtk.org>
1583 * gstrfuncs.c (g_strlcpy, g_strlcat): completed tor's fix
1584 to cover both #ifdef branches.
1586 2000-07-26 Tor Lillqvist <tml@iki.fi>
1588 * gstrfuncs.c (g_strlcpy, g_strlcat): Return 0 on error, not NULL.
1590 * glib.def: Add g_strlcpy, g_strlcat.
1592 * glibconfig.h.win32.in: Add gsize and gssize.
1594 Wed Jul 26 12:59:31 2000 Tim Janik <timj@gtk.org>
1596 * *.[hc]: applied patch from Andreas Persenius <ndap@swipnet.se> that
1597 updates the license headers to the GNU Lesser General Public License,
1598 as well as updating the copyright year to 2000.
1600 Wed Jul 26 05:47:48 2000 Tim Janik <timj@gtk.org>
1605 * glib.h: added g_strlcat() and g_strlcpy() wrappers, supplied by
1606 David Wheeler <dwheeler@ida.org>:
1608 * glib.h, gstrfuncs.c: added g_strlcpy and g_strlcat to support
1609 safe manipulation of fixed-length string buffers.
1610 These functions were originally developed by Todd Miller to simplify
1611 development of security-related programs, and
1612 are available on many (but not all) Unix-like systems,
1613 including OpenBSD, FreeBSD, and Solaris. See
1614 ftp://ftp.openbsd.org/pub/OpenBSD/src/lib/libc/string/strlcpy.3
1615 and http://www.openbsd.org/security.html.
1616 If there's a strlcpy/strlcat on the system, it's called, otherwise
1617 an implementation is provided.
1619 * testglib.c: Added tests for g_strlcpy, g_strlcat.
1621 Wed Jul 26 05:03:24 2000 Tim Janik <timj@gtk.org>
1623 * acglib.m4 (GLIB_SIZEOF): include <stdlib.h> and <stddef.h> if
1624 STDC_HEADERS is defined.
1627 * glibconfig.h: define gsize and gssize in terms of GLIB_SIZEOF_SIZE_T
1629 * glib.h (g_return_if_reached): applied darin's fix for copy'n
1630 paste error in the macro implementation.
1632 Wed Jul 26 00:46:03 2000 Tim Janik <timj@gtk.org>
1634 * glib.h: applied patch from Darin Adler <darin@eazel.com> which
1635 supplies g_return_if_reached(), g_return_val_if_reached() and
1638 2000-07-22 Tor Lillqvist <tml@iki.fi>
1640 * build-dll: Fix resource handling, the resource file got left out
1641 from the DLL after all... Remove the WIN32APIHEADERS, not needed
1642 with current windres.
1644 * glib.def: Add new functions.
1646 2000-07-20 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1648 * gutils.c, glib.h: Mark the functions g_basename and g_dirname
1649 deprecated. They will issue an warning once, when compiled with
1650 G_ENABLE_DEBUG, but continue to work as before. Instead the
1651 functions g_path_get_basename and g_path_get_dirname should be
1652 used, which BOTH return newly allocated memory, that has to freed
1653 by g_free. The new g_path_get_basename now strips trailing slashes
1654 from the path. This fixes #5097. For discussion see
1655 http://mail.gnome.org/pipermail/gtk-devel-list/2000-April/003139.html
1657 * gwin32.c, testglib.c, tests/dirname-test.c: Use the new
1658 functions instead of the old ones.
1660 * ghash.c, gscanner.c, glib.h: Mark the functions
1661 g_hash_table_freeze, g_hash_table_thaw and thus
1662 g_scanner_freeze_symbol_table and g_scanner_thaw_symbol_table
1663 deprecated. They will issue an warning once, when compiled with
1664 G_ENABLE_DEBUG. This fixes Bug #3883. For discussion see
1665 http://mail.gnome.org/pipermail/gtk-devel-list/2000-April/003139.html
1667 2000-07-19 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1669 * configure.in, glib.h: glibconfig.h and glib.h now include files
1670 outside of the extern "C" block. Makes some C++ compiler
1671 happy. Reported by Denis Vakatov <vakatov@peony.nlm.nih.gov>.
1673 Sat Jul 15 23:49:03 2000 Owen Taylor <otaylor@redhat.com>
1675 * glib/glib.texi: Remove incomplete start of info file -
1676 real docs are in RDP.
1678 Sat Jul 15 22:44:22 2000 Owen Taylor <otaylor@redhat.com>
1680 * configure.in: Add build/Makefile and
1681 build/win32/Makefile to AC_OUTPUT() so things build
1684 Sat Jul 15 09:11:46 2000 Tim Janik <timj@gtk.org>
1686 * gstrfuncs.c (g_strncasecmp): fixed an off by 0 error (yeah,
1687 the function went off when the while (n--) loop failed due to
1688 n==0 ;), reported by Jean-Louis HAMEL <jlhamel@club-internet.fr>.
1690 2000-07-15 Tor Lillqvist <tml@iki.fi>
1692 * Makefile.am (SUBDIRS): Include the "build" module in GLib, too,
1693 to make it more self-contained. If your CVS client doesn't
1694 automatically get it, do a cvs get build in glib.
1696 * */makefile.mingw.in: Include make.mingw from build in the glib
1699 Fri Jul 14 16:26:35 2000 Owen Taylor <otaylor@redhat.com>
1703 Fri Jul 14 12:22:49 2000 Owen Taylor <otaylor@redhat.com>
1705 * configure.in Makefile.am glib-config.m4 glib.m4: Move
1706 glib-config to glib-config-2.0 move glib.m4 to
1709 * Makefile.am gobject/Makefile.am gmodule/Makefile.am
1710 gthread/Makefile.am tests/Makefile.am: Change
1711 library names to libglib-1.3.la, etc, so that we
1712 can distinguish glib-1.2 and glib-2.0 on the linkline.
1714 * Makefile.am gobject/Makefile.am gmodule/Makefile.am:
1715 Move include files into /usr/include/glib-2.0.
1717 Thu Jul 6 18:54:49 2000 Owen Taylor <otaylor@redhat.com>
1719 * docs/Makefile.am (EXTRA_DIST): Remove info files
1722 2000-07-14 Tor Lillqvist <tml@iki.fi>
1724 * glib.def: Add g_error functions.
1726 * makefile.mingw.in: Add gbacktrace.o.
1728 * gbacktrace.c: No need to include <process.h>.
1730 2000-07-12 Havoc Pennington <hp@redhat.com>
1732 * glib.h: #include <gerror.h>
1734 * Makefile.am (include_HEADERS): Add gerror.h
1735 (libglib_la_SOURCES): Add gbacktrace.c
1737 * gbacktrace.c: Move g_on_error_query() in here (moved on the
1738 server, so history is preserved)
1740 * gerror.h: GError interface
1742 * gerror.c: GError implementation replaces stuff that's now in
1745 Sun Jul 9 21:20:45 2000 Owen Taylor <otaylor@redhat.com>
1747 * gunicode.h: Include stddef.h instead of stdlib.h
1749 2000-07-08 Tor Lillqvist <tml@iki.fi>
1751 * glib.h (GLIB_VAR): Rename the GUTILS_C_VAR macro to GLIB_VAR.
1753 * gunicode.h: Mark the g_utf8_skip array with GLIB_VAR.
1755 * glib.def: Add two missing entry points.
1757 Thu Jul 6 15:35:28 2000 Owen Taylor <otaylor@redhat.com>
1761 * Makefile.am (EXTRA_DIST): Dist fixes.
1763 * configure.in: Moderate the warnings just a little bit.
1765 2000-07-05 Tor Lillqvist <tml@iki.fi>
1767 * README.win32: Update.
1769 Mon Jul 3 17:58:02 2000 Owen Taylor <otaylor@redhat.com>
1771 * gutf8.c (g_utf8_get_charset_internal): Fix up
1772 to correspond to configure.in checks.
1774 Mon Jul 3 17:18:19 2000 Owen Taylor <otaylor@redhat.com>
1776 * glib.h: Comment g_get_codeset() out of the header file
1777 temporarily. (Very similar to g_get_charset(), need
1778 to resolve the two.)
1780 2000-07-01 Tor Lillqvist <tml@iki.fi>
1782 * glib.def: Add new entry points.
1784 * makefile.{mingw,msc}.in: Add the new Unicode object files.
1786 Thu Jun 29 15:57:28 2000 Owen Taylor <otaylor@redhat.com>
1790 * Makefile.am: added snapcheck target to go along with snapshot
1792 * gstring.c glib.h (g_string_hash): Add g_string_hash to
1793 go along with g_string_equal.
1795 Tue Jun 27 12:40:23 EDT 2000 David A. Wheeler <dwheeler@dwheeler.com>
1797 * glib.h: Added g_string_equal for comparing GStrings;
1798 changed g_str_equal so it returns gboolean (instead of gint).
1800 * gstring.c: Modified GString implementation to support embedded
1801 ASCII NUL ('\0') characters, and implemented g_string_equal.
1803 * testglib.c tests/string-test.c: Added tests for g_string_equal
1804 and tests for proper handling of embedded ASCII NUL characters.
1806 Wed Jun 28 22:52:00 2000 Owen Taylor <otaylor@redhat.com>
1808 * Makefile.am (libglib_la_SOURCES): Fix
1809 gunichartable.h => gunichartables.h. (From Eric Limings)
1811 Fri Jun 23 17:20:26 2000 Tim Janik <timj@gtk.org>
1813 * glib.h: define gstring in terms of gchar*. this typedef reflects
1814 the type name of the primitive G_TYPE_STRING in the gobject module.
1816 Wed Jun 21 12:09:03 2000 Owen Taylor <otaylor@redhat.com>
1818 * gunicode.h gutf8.c guniprop.c gunidecomp.[ch] gunichartables.h
1819 Makefile.am glib.h: Initial pass at adding unicode support
1820 functions. A few things still need to be implemented, a bit
1821 of cleanup needs to be done, tests need to be added, and
1822 the docs need to be finished, but this should allow replacing
1823 most or all use of libunicode.
1825 2000-06-06 Tor Lillqvist <tml@iki.fi>
1827 * giowin32.c (g_io_channel_win32_pipe_readable): If we are
1828 watching the same pipe for different conditions (with different
1829 callbacks), check them all. Only call the callback for G_IO_IN
1830 from here. (This bug popped up when a watch for G_IO_ERR|G_IO_HUP
1831 was added to gimplib.)
1833 2000-05-30 Tor Lillqvist <tml@iki.fi>
1835 * gutils.c (g_locale_get_codeset): Implement on Win32.
1837 * glib.def: Add g_get_codeset.
1839 * tests/Makefile.am (EXTRA_DIST): makefile.cygwin* has been
1840 renamed to makefile.mingw*.
1842 Tue May 30 16:01:32 2000 Owen Taylor <otaylor@redhat.com>
1844 * glib.h gutils.c: Move the g_locale_get_codeset() up in the
1845 header file to correspond to to comments about memory
1846 management. Rename to g_get_codeset() to avoid
1847 polluting the g_locale_* namespace, which probably
1848 would have g_locale_get_codeset (GLocale *locale).
1851 Mon May 29 14:10:35 2000 Owen Taylor <otaylor@redhat.com>
1853 * gutils.c (g_locale_get_codeset): Add function to get the
1854 codeset name for the current locale.
1856 * configure.in acconfig.h: Add check for nl_langinfo(CODESET);
1858 Fri May 19 11:39:29 2000 Tim Janik <timj@gtk.org>
1860 * gutils.c (g_snprintf):
1861 (g_vsnprintf): added argument assertments.
1863 * gstring.c (g_string_assign): added argument assertments.
1864 (g_string_truncate): make len a guint.
1866 Fri May 19 09:00:44 2000 Tim Janik <timj@gtk.org>
1868 * gmem.c (g_free): fixed SIZEOF_LONG==4 assumption with
1869 ENABLE_MEM_CHECK, from Art Haas <ahaas@neosoft.com>.
1871 * gslist.c (g_slist_reverse): shut up compiler.
1873 * gscanner.c (g_scanner_get_token_ll): removed inline assignment.
1875 * garray.c: remove index>=0 checks for unsigned indices.
1877 * gmain.c (g_idle_prepare): timeout assignment fix.
1879 * gtree.c (g_tree_node_rotate_right): shut up compiler.
1881 2000-05-13 Tor Lillqvist <tml@iki.fi>
1884 * tests/makefile.mingw.in
1885 * build-dll: Rename makefile.cygwin(.in) to
1886 makefile.mingw(.in), which better describes what it is. Move the
1887 build of gmodule, gthread and gobject DLLs to makefiles in those
1888 directories. Move resource file handling and build number bump to
1889 build-dll, where it sits much cleaner.
1892 * Makefile.am (EXTRA_DIST): Update accordingly.
1894 * glib.h: Add G_PI, G_PI_2, G_PI_4, G_E, G_LN2, G_LN10 and
1895 G_SQRT2. M_PI etc aren't necessarily in <math.h> in strict ISO C
1898 * glib.def: Add g_strcanon.
1900 * gtree.c (g_tree_node_rotate_left): Remove unused variables.
1902 * gwin32.c (g_win32_opendir): Remove unneeded statement.
1904 Thu May 4 02:04:46 2000 Tim Janik <timj@gtk.org>
1906 * configure.in (STRIP_DUMMY): some Make 3.79 $(strip ) versions are
1907 broken and require an empty arg, give it to them.
1909 Fri Apr 28 23:54:35 2000 Tim Janik <timj@gtk.org>
1911 * setup things for a new sub-library libgobject:
1913 * Makefile.am (SUBDIRS): added gobject
1915 * glib-config.in: feature -lgobject.
1917 * configure.in (AC_OUTPUT): generate gobject/Makefile.
1919 * glib.m4 (AM_PATH_GLIB): feature gobject module.
1921 * glib.spec.in: added %{prefix}/lib/libgobject-1.3.so.*
1923 Fri Apr 28 21:41:49 2000 Tim Janik <timj@gtk.org>
1925 * glib.h: added G_STRLOC macro.
1926 G_STRUCT_OFFSET(): signedness corrections.
1927 (G_CSET_DIGITS): list 0-9.
1928 * gscanner.c (g_scanner_config_template): use G_CSET_DIGITS.
1934 (g_strreverse): return the modified string instead of void, so
1935 calls to these functions can be nested.
1936 (g_strcanon): new function, canonicalizes string according to
1937 a given character set.
1939 Fri Apr 28 19:45:16 2000 Tim Janik <timj@gtk.org>
1941 * gasyncqueue.c (g_async_queue_unref): get rid of an unused variable.
1943 Wed May 10 19:52:44 2000 Owen Taylor <otaylor@redhat.com>
1945 * glib.m4: Print found version when test succeeds.
1947 2000-05-04 Tor Lillqvist <tml@iki.fi>
1949 * makefile.cygwin.in
1950 * tests/makefile.cygwin.in: Include the common makefile snippet
1951 from ../build/win32.
1953 Maybe CVSROOT/modules should be changed so that the 'build' module
1954 is included within the glib module (and gtk+, and gimp, and maybe
1955 others later), in the same way as the 'macros' module is included
1956 in lots of GNOME CVS modules?
1958 2000-05-02 Tor Lillqvist <tml@iki.fi>
1960 * glib.def: Add new functions.
1962 * makefile.{cygwin,msc}.in (glib_OBJECTS): Add new object files.
1964 * tests/makefile.{cygwin,msc}.in: Add threadpool-test.
1966 2000-04-28 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1968 * gasyncqueue.c: New File implementing an asynchronous queue to be
1969 used for asynchronous inter-thread communication.
1971 * gthreadpool.c: New File implementing a thread pool to be used
1972 for distributing work among several threads.
1974 * glib.h: Added the type and function declarations for these two
1977 * tests/threadpool-test.c: New File implementing a test for the
1978 thread pool. This also checks the asynchronous queue underlying
1981 * tests/Makefile.am: Changed accordingly.
1983 2000-04-26 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
1985 * configure.in: Look for both pthread_create and pthread_join in
1986 the thread library. Some systems define one of them, but not both
1987 in libc. Arghh. Now we really start a thread and join it later and
1988 check, whether the thread to actually ran.
1990 * glib.h, gcache.c, gtree.c: Changed the 'value' parameter of
1991 g_cache_remove from gpointer to gconstpointer. Dito for the 'key'
1992 parameter of g_tree_lookup and g_tree_remove and the 'data'
1993 parameter of g_tree_search. This function now takes a function of
1994 type GCompareFunc instead of GSearchFunc. This fixes Bug
1995 #8267. Thanks to Juan Toledo <toledo@users.sourceforge.net> for
1998 * glib.h: Removed declaration of GSearchFunc.
2000 * gmem.c: s/GSearchFunc/GCompareFunc/.
2002 2000-04-19 Tor Lillqvist <tml@iki.fi>
2004 * glib.def: Update entry point list.
2006 2000-04-19 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2008 * glib.h (G_TRYLOCK): Made the debugging G_TRYLOCK call also work
2009 for compilers with funny G_STMT_(START|END) macros.
2011 * tests/thread-test.c: Implemented a check for that.
2013 * gutils.c (g_getenv): Changed the win32 part of this function to
2014 be thread safe and to make the returned environment string
2015 persistent to match the UN*X behavior. This is again a response to
2018 * glib.h (G_LOCK_NAME): Removed parentheses around the lock name,
2019 as that seems to cause problems for some compilers and really
2022 Wed Apr 19 08:32:32 2000 Tim Janik <timj@gtk.org>
2024 * gscanner.c (g_scanner_new): make sure that
2025 scanner->config->cset_skip_characters is "" instead of NULL, so we
2026 don't segfault further on.
2028 2000-04-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2030 * glib.h, glist.h, gslist.h: Changed the 'data' parameters from
2031 gpointer to gconstpointer for the functions
2032 g_(list|slist)_(remove|find|find_custom|index), as they do not
2033 change this parameter. This fixes bug #4836.
2035 * glib.h: Changed comment for g_getenv to reflect, that the
2036 returned memory must not be freed. Fixes bug #8983.
2038 2000-04-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2040 * configure.in, acconfig.h: Add configure test for garbage
2041 collector friendliness for GLib. If enabled, ENABLE_GC_FRIENDLY
2044 * garray.c, ghash.c, glist.c, gmain.c, gmem.c, gnode.c, gqueue.c,
2045 gslist.c, gtree.c: If ENABLE_GC_FRIENDLY is defined, NULLify all
2046 memory released by the user, but cached by GLib. This lets a
2047 garbage collector have a more correct view of the actually used
2050 * garray.c, glib.h: Added g_(array|ptr_array|byte_array)_sized_new
2051 functions, that reserve a certain amount of memeory for the array
2052 at creation time to avoid reallocation. Fixes bug #6707 from
2053 Charles Kerr <ckerr@osserver1.nssl.noaa.gov>.
2055 * glib.h, gqueue.c, tests/queue-test.c (main): Renamed
2056 g_queue_create to g_queue_new in conformance to all other GLib
2059 2000-04-07 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2061 * grand.c (g_rand_new): Fixed bug. Thanks to Marko Kreen
2062 <marko@l-t.ee> for reporting that.
2064 2000-03-26 Tor Lillqvist <tml@iki.fi>
2066 * README.win32: Tell about using the mingw-based gcc, which is
2067 much easier than modifying the cygwin gcc to product mingw code
2068 for the msvcrt runtime.
2070 * makefile.cygwin.in (WIN32APIHEADERS): Kludge to make it work
2071 with a "pure" mingw gcc, too.
2073 2000-03-24 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2075 * garray.c: Made GArray behave correctly. Now zero_terminated
2076 really means, that the element array->data[array->len] exists and
2077 is zeroed, and clear means that any unassigned elements obtained
2078 through g_array_set_size (the only way to get unassigned elements
2079 AFAICT) are zeroed. Added some macros to make the code more
2080 obvoius. Also made GPtrArray zero elements after
2081 g_ptr_array_set_size. This is done in a portbale way (assignment
2082 of NULL instead of just memsetting it to zero), though that might
2083 be more portability than we actually want.
2085 * Makefile.am, gthread/Makefile.am, gmodule/Makefile.am,
2086 tests/Makefile.am: Added various win32 related *.in files to
2087 EXTRA_DIST to let 'make distcheck' procude all the corresponding
2088 files, which it silently fails to do currently.
2090 2000-03-23 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2092 * configure.in: After finding the right thread library (containing
2093 e.g. pthread_create) we now search for the right realtime library
2094 (containing e.g. sched_get_priority_max). Makes the output of the
2095 thread related libraries correct.
2097 * gtimer.c (g_usleep): The current implementation of g_usleep
2098 (simply calling select) doesn't work reliable for multi-threaded
2099 programs on some platforms (bad omen for the main loop....), so I
2100 changed the implementation for thread-using programs to wait for a
2101 GCond for the specified amount of time (NB: sleep and usleep are
2102 not MT-safe in general, because they often use signals).
2104 Wed Mar 22 16:49:57 2000 Owen Taylor <otaylor@redhat.com>
2106 * gmem.c (g_mem_chunk_area_compare): Fix indentation.
2108 2000-03-22 Elliot Lee <sopwith@redhat.com>
2110 * gmem.c (g_mem_chunk_area_compare): Fix 64-bitness bug in
2111 comparing two pointers more than 4G apart.
2113 2000-03-22 Tor Lillqvist <tml@iki.fi>
2115 * gutils.c: Move Win32-only includes after inclusion of glib.h, so
2116 that G_OS_WIN32 is defined.
2118 * glibconfig.h.win32.in: Add GSystemThread.
2120 2000-03-22 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2122 * gutils.c (g_get_current_dir): Ok, Marcus Brinkmann
2123 <Marcus.Brinkmann@ruhr-uni-bochum.de> convinced me, that 128 KB
2124 path length might not be enough for the HURD. So I changed the
2125 loop to at least avoid an integer overflow, which could happen at
2128 * configure.in: Test for sched_yield as the native yield function
2129 first. Corrected typo g_thread_sleep -> g_usleep. Corrected
2130 message for the pthread_create test. Negative Priorities are
2131 allowed (and used on Solaris), so consider
2132 sched_get_priority_min failed only if it returns -1, not <0. Check
2133 for sched_get_priority_min also in -lrt, if not found in -lpthread
2134 alone and add -lrt to G_THREAD_LIBS then. Remove special case
2135 handling of priorities for older solaris versions and posix
2136 threads. Thanks to Wan-Teh Chang <wtc@netscape.com> for suggesting
2137 some of those changes.
2139 * config.guess, config.sub, ltconfig, ltmain.sh:
2140 Upgrade to libtool 1.3.4.
2142 2000-03-21 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2144 * glib.h, configure.in, gutils.h: always define G_GNUC_EXTENSION,
2145 even when not needed by GLib. That's actually also the way, the
2146 GLib reference manual describes that macro. Therefore I had to
2147 remove the lonesome #include <glibconfig.h> in gutils.c, which
2148 doesn't seem to be needed there however. This change should make
2149 Ben Gertzfield <che@debian.org> happy.
2151 * gutils.c: Furthermore two warnings in gutils.c were voided,
2152 which crept in due to my last change.
2154 * gutils.c (g_get_current_dir): Allocate only up to 128KB for a
2155 pathname. While this is an arbitrary value just like 2048, it
2156 seems to be enough (after all, even 4GB is an arbitrary value).
2158 2000-03-20 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2160 * gmain.c (g_main_poll): Warn in case of an error during the call
2161 to poll(2). Closes Bug#7564 as reported by David Helder
2162 <dhelder@umich.edu>.
2164 * gutils.c (g_get_current_dir): Make g_get_current_dir work on
2165 systems with unlimited pathname length like the HURD (It worked
2166 there before, but only for pathes shorter than 2048). Closes
2167 Bug#4525 as reported by Marcus Brinkmann
2168 <Marcus.Brinkmann@ruhr-uni-bochum.de>.
2170 2000-03-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2172 * giounix.c (g_io_unix_write, g_io_unix_read): Interpret EINTR as
2175 2000-03-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2177 * configure.in: Added the missing POSIX_NO_YIELD and
2178 POSIX_NO_PRIORITIES warning messages.
2180 * configure.in: Use AC_TRY_RUN instead of AC_TRY_LINK, to test for
2181 real thread support. On solaris pthread_create can be linked to
2182 even in -lc, but it doesn't work then.
2184 * configure.in: Don't use priorities for threads, when the
2185 minimal/maximal priorities couldn't be determined at configure
2188 * configure.in, gthread.c: Always define GSystemThread in
2189 glibconfig.h to represent a system thread.
2191 * configure.in: Do not use native recursive threads, when
2192 possibe. We use some features, that they do not expose (namely the
2195 * glib.h, gthread.c: Redefined GStaticRecMutex. The functions are
2196 now implemented in a different way, which should be way
2197 faster. Alsothere are now functions g_static_rec_mutex_unlock_full
2198 and g_static_rec_mutex_lock_full to leave/enter a recursive mutex
2201 * gthread.c (g_thread_self): Do not test the system_thread to be
2202 non-zero to speed things up.
2204 * gthread.c (g_mutex_init): Therefore set the system_thread of the
2207 * tests/thread-test.c: Rerun all tests once again, but this time
2208 we fool the system into thinking, that the available thread system
2209 is not native, but userprovided.
2211 2000-03-13 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2213 * gqueue.c (g_queue_push_tail_link, g_queue_push_head_link): We
2214 want the next and prev pointer of the inserted link to be NULL.
2216 2000-03-06 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2218 * configure.in: Another small change to the pthread_.. search
2219 pattern. Should work *now* for AIX.
2221 2000-03-04 Tor Lillqvist <tml@iki.fi>
2223 * gwin32.c (g_win32_error_message): New function that returns the
2224 message string for a Win32 error code.
2226 * glib.h: Declare it.
2228 * glib.def: Export it, plus g_node_copy.
2230 2000-03-03 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2232 * configure.in: Make the search for pthread_attr_... prototypes
2233 find names at the start of a line also, like it is on AIX. Thanks
2234 to Valdis Kletnieks <Valdis.Kletnieks@vt.edu> for the info.
2236 Wed Mar 1 10:39:39 2000 Tim Janik <timj@gtk.org>
2238 * gslist.c (g_slist_reverse): minor optimization.
2240 * testglib.c (g_node_test): added a couple of tests for
2244 * gnode.c (g_node_copy): new function to copy subtrees,
2245 supplied by dbsears@ix.netcom.com.
2246 changed iterator to walk the children list backwards, so
2247 we get down from O(n^2) to O(n).
2249 * gnode.c (g_node_first_sibling): applied patch from
2250 dbsears@ix.netcom.com to optimize access if node->parent
2253 * gutils.c (g_get_any_init): backed out HAVE_PW_GECOS check around
2254 assignment of g_real_name, sicne HAVE_PW_GECOS is never defined and
2255 thus breaks the original code.
2257 * merged changes from 1.2.7.
2259 Fri Jan 28 11:37:41 2000 Owen Taylor <otaylor@redhat.com>
2261 Bug #4156 - Changes vaguely modelled after Scott Gifford's patch
2263 * gtimer.c (g_timer_elapsed): Never report negative times -
2266 * gmain.c (g_timeout_prepare): Guard against unexpected
2267 clock shifts by never setting a timeout of more than
2268 data->interval msecs.
2270 2000-02-27 Tor Lillqvist <tml@iki.fi>
2272 * glib.def: Add new functions.
2274 2000-02-23 Tor Lillqvist <tml@iki.fi>
2276 * README.win32: Add a missing step to the setup instructions for
2277 gcc-2.95.2. Thanks to Arnaud Charlet.
2279 * glib.def: Add missing entry point.
2281 2000-02-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2283 * configure.in: Changed GCC version test to also accept major
2284 versions > 2. Thanks to Ben Gertzfield <che@debian.org> for
2287 Thu Feb 17 12:53:44 2000 Tim Janik <timj@gtk.org>
2289 * gstring.c: changed g_str_hash() to a 31 bit version based on
2290 a submission by Karl Nelson and hand optimized ad absurdum by
2293 * gstring.c: applied patch from havoc for new gstring functions,
2294 added some more sanity checks, coding style fixups.
2296 2000-02-13 Havoc Pennington <hp@pobox.com>
2298 * tests/string-test.c (main): Add tests for the new GString
2301 * testglib.c (main): Add tests for the new GString features
2303 * gstring.c (g_string_insert_len): New function; insert
2304 a given length of string at a given position.
2305 (g_string_append): reimplement in terms of g_string_insert_len
2306 (g_string_append_len): new function
2307 (g_string_insert_c): accept -1 for "pos" arg to mean "append"
2308 (g_string_append_c): reimplement in terms of g_string_insert_c
2309 (g_string_prepend): reimplement in terms of g_string_insert_len
2310 (g_string_prepend_len): new function
2311 (g_string_prepend_c): reimplement in terms of g_string_insert_c
2312 (g_string_insert): reimplement in terms of g_string_insert_len
2314 * glib.h: Declare g_string_insert_len, g_string_append_len,
2315 g_string_prepend_len
2317 Sun Feb 13 08:16:47 2000 Tim Janik <timj@gtk.org>
2319 * configure.in: wtf??? someone destroyed the configure.in, reverting to
2320 an older version from Feb 4 which apears to still work.
2322 2000-02-07 Tor Lillqvist <tml@iki.fi>
2324 * gmodule.rc.in gthread.rc.in: Move to corresponding subdirectories.
2327 * gmodule/Makefile.am
2328 * gthread/Makefile.am: Change accordingly.
2330 * makefile.cygwin: Corresponding changes, some cleanup.
2332 2000-02-05 Tor Lillqvist <tml@iki.fi>
2334 * glib.rc.in gmodule.rc.in gthread.rc.in: New files, for putting
2335 version info in the DLLs on Win32.
2337 * Makefile.am: Generate corresponding *.rc files and distribute
2340 * makefile.cygwin.in: Add rules to automatically bump a "build
2341 number" in the version info in the rc files each time the DLL is
2342 built. But do this only for the person who releases binaries. If
2343 others build the DLLs, the build number is set to zero.
2345 Fri Feb 4 19:36:05 2000 Tim Janik <timj@gtk.org>
2348 * gdataset.c: return stolen data from g_datalist_id_remove_no_notify()
2349 and g_dataset_id_remove_no_notify() to avoid second lookup for common
2352 2000-02-01 Tor Lillqvist <tml@iki.fi>
2355 * gstrfuncs.c (g_filename_to_utf8, g_filename_from_utf8): New
2356 functions for conversion between UTF-8 and the encoding expected
2357 by C runtime functions like open() and stat(), and returned by
2360 Implement them on Win32 where we use the system "ANSI" codepage,
2361 which might be single-byte or double-byte. On Unix, just skip the
2362 issue for now and provide dummy implementations that return a copy
2367 * glib.def: Minor updates.
2369 Wed Jan 26 05:24:38 2000 Tim Janik <timj@gtk.org>
2372 * gmain.c: s/current_time/dispatch_time/ for the dispatch() handlers.
2373 refetch the current time after invocation of poll() to cover up for
2374 the time spent in that function call.
2376 Fri Jan 21 10:18:24 2000 Owen Taylor <otaylor@redhat.com>
2378 * glib.h (G_N_ELEMENTS): Added G_N_ELEMENTS macro to determine
2379 the number of elements in an array.
2381 Sun Jan 9 13:28:36 2000 Tim Janik <timj@gtk.org>
2383 * gstrfuncs.c (g_strtod): correctly fetch the current locale,
2386 1999-12-16 Tor Lillqvist <tml@iki.fi>
2388 * gmodule/gmodule-win32.c: Use FormatMessage to translate system
2389 error codes into textual messages.
2391 1999-11-25 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2393 * glib.h (G_TRYLOCK): This of course should return TRUE in a
2394 program with a thread-disabled GLib.
2396 1999-11-18 Tor Lillqvist <tml@iki.fi>
2398 * glib.def: g_strjoin was missing.
2400 1999-11-16 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2402 * acconfig.h, config.h.win32.in, configure.in: Renamed
2403 GLIB_SIZEOF_PTHREAD_T to GLIB_SIZEOF_SYSTEM_THREAD to reflect
2406 * configure.in: Cope with systems, that have a pthread_t type,
2407 that is not a pointer. Hint from Karl Nelson
2408 <kenelson@ece.ucdavis.edu>. Define GLIB_SIZEOF_SYSTEM_THREAD to 4
2409 for Solaris. Cope with systems, that have no default mutex
2410 initialize, like obviously most DCE systems.
2412 * glib.h, gthread.c: Changed the prototype of thread_create and
2413 thread_self to return the system thread into provided memory
2414 instead of a return value. This is necessary, as HPUX has a
2415 pthread_t, that is bigger than the biggest integral type there.
2417 * gthread.c: system_thread is no longer a pointer, but an memory
2418 area of size GLIB_SIZEOF_SYSTEM_THREAD. Changed the
2419 zeroinitialization and the tests for zeroness accordingly.
2421 1999-11-09 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2423 * configure.in: Create docs/glib-config.1 from
2424 docs/glib-config.1.in. Makes 'make distcheck' happy (and me too).
2426 * glib-config.1: Removed from CVS, as it is a generated file.
2428 1999-11-08 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2430 * configure.in: Make the test for getpwuid_r work on newer AIX
2431 versions, too. Still works on Solaris and Linux. Patch from Craig
2432 Rodrigues <rodrigc@mediaone.net>.
2434 1999-11-08 Tor Lillqvist <tml@iki.fi>
2436 * gwin32.c (g_win32_getlocale): Look at env vars LC_ALL, LC_CTYPE
2437 and LANG first. Some refinements to the sublanguage logic.
2439 1999-11-04 Tor Lillqvist <tml@iki.fi>
2441 * makefile.{cygwin,msc}.in: Add gwin32 object. Add rule to make .i
2442 (preprocessed source) files.
2444 1999-11-01 Tor Lillqvist <tml@iki.fi>
2447 * glib.def: Rename Win32-only functions from gwin_* to g_win32_*
2448 to match the GLib naming conventions.
2453 * Makefile.am: Move the Win32-only functions to the new
2456 1999-10-31 Tor Lillqvist <tml@iki.fi>
2458 * gutils.c (gwin_getlocale): New Win32-specific function, returns
2459 a Unixish current locale string (en, zh_TW etc).
2461 * glib.h: Declare it.
2463 * glib.def: Export it.
2465 * testglib.c: Test it.
2467 * gmessages.c (Win32: ensure_stdout_valid): Some improvements,
2468 make sure we don't call AllocConsole several times, which I think
2471 Sun Oct 31 18:55:01 1999 ape@spacetec.no (Asbjorn Pettersen)
2473 * gcache.c (g_cache_remove): Test if node is NULL.
2474 If not tested, GIMP's script-fu will crash.
2476 Sun Oct 17 18:11:40 1999 Tim Janik <timj@gtk.org>
2478 * gdataset.c (g_data_set_internal): remove g_dataset_global_lock around
2479 destroy() notification here as well.
2481 1999-10-15 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2483 * gdataset.c (g_datalist_clear_i): Avoid Freezing, when g_datalist
2484 is called recursivly. Reported by Ola Andersson <rand@ling.umu.se>.
2486 Tue Oct 12 14:17:12 1999 Tim Janik <timj@gtk.org>
2488 * glib.h: removed useless g_string(x) macro that cluttered the namespace
2489 and was just a poor wrapper around the cpp '#' symbol, use #x if you
2490 need to work around this.
2491 added new macro G_STRINGIFY(arg) that will convert arg to a string,
2492 no matter whether it contains macros or not.
2494 1999-10-12 Tor Lillqvist <tml@iki.fi>
2496 * config.h.win32.in: Define the new GLIB_SIZEOF_* constants here,
2499 * glib.h: Small Win32 comments improvement.
2501 Tue Oct 12 12:16:12 1999 Tim Janik <timj@gtk.org>
2503 * gmessages.c (g_printf_string_upper_bound): completly new
2504 implementation for printf string upper bounds calculation.
2505 we handle all glibc 2.1 format specifiers now, except for positional
2506 parameters (%nn$...) and wide char strings, plus some obscure upper
2507 case variants of the standard conversions. this fixes a lot of
2508 bugs in the old code, i.e.
2509 - NULL format strings
2510 - floats with exponents >+24
2512 - precision specifications in general
2513 - negative field widths
2514 - %p for SIZEOF_VOID_P > 4 platforms
2515 we now issue warnigns in places where the old code would have
2516 caused buffer overruns anyways. warnings are suppressed when invoked
2517 from glogv(), to avoid infinite recursions if someone passes a log
2518 message that comes with really obscure format specifications.
2520 Tue Oct 12 11:49:00 1999 Tim Janik <timj@gtk.org>
2522 * gstrfuncs.c: nuked old g_printf_string_upper_bound() version.
2524 Tue Oct 12 03:34:40 1999 Tim Janik <timj@gtk.org>
2526 * glib.h: added GFloatIEEE754 and GDoubleIEEE754 unions to access sign,
2527 mantissa and exponent of IEEE floats and doubles (required by the new
2528 version of g_printf_string_upper_bound). the unions are endian specific,
2529 we handle G_LITTLE_ENDIAN and G_BIG_ENDIAN as of currently. ieee floats
2530 and doubles are supported (used for storage) by at least intel, ppc and
2532 http://twister.ou.edu/workshop.docs/common-tools/numerical_comp_guide/ncg_math.doc.html
2534 Mon Oct 11 18:01:49 1999 Tim Janik <timj@gtk.org>
2536 * configure.in: added additional checks to figure sizes of size_t,
2537 ptrdiff_t and intmax_t (required by g_printf_string_upper_bound).
2539 Wed Oct 6 12:44:23 PDT 1999 Manish Singh <yosh@gimp.org>
2541 * configure.in: blah. use G_WITH_CYGWIN instead of G_HAVE_CYGWIN
2543 1999-10-05 Tor Lillqvist <tml@iki.fi>
2545 * glib.h: (Win32) Drop the mapping of POSIX function names to the
2546 underscored versions, it's unnecessary after all. With MSVC we get
2547 them from oldnames.lib, with gcc-2.95 and mingw32 from
2548 -lmoldname-msvc. Add comment about what headers to include for
2551 * glibconfig.h.win32.in: Don't define WIN32 and NATIVE_WIN32.
2553 * gerror.c (g_on_error_query): (Win32) Slightly increased verbosity.
2555 * build-dll: Don't strip.
2557 * tests/string-test.c tests/dirname-test.c: Use G_OS_WIN32.
2559 * glib.def: Add g_thread_use_default_impl.
2561 Sun Oct 3 19:46:55 PDT 1999 Manish Singh <yosh@gimp.org>
2563 * configure.in: use G_HAVE_CYGWIN instead of G_OS_FEATURE_CYGWIN
2565 Sun Oct 3 19:25:42 PDT 1999 Manish Singh <yosh@gimp.org>
2569 * glibconfig.h.win32: G_OS_FOO #defines. I *think* I got the cygwin
2570 and beos stuff right, but I haven't tested it. The respective
2571 porters should fix any screwups
2581 * testglib.c: use G_OS stuff
2583 Wed Sep 22 01:53:18 1999 Tim Janik <timj@gtk.org>
2585 * glib.h (NULL): define NULL as (0L) if __cplusplus is defined, to
2586 avoid "ANSI C++ forbids implicit conversion from `void *' in argument
2587 passing" errors upon NULL usage in C++ programs (gcc-2.95 is on crack
2588 for erroring out on this, instead of just issueing a warning).
2590 * glib.h (g_trash_stack_pop): use uncasted NULL again.
2592 Fri Sep 17 10:24:45 1999 Tim Janik <timj@gtk.org>
2594 * gmem.c (g_mem_chunk_compute_size) (g_mem_chunk_new): applied
2595 patch from Soeren Sandmann <sandmann@daimi.au.dk>, to force mem
2596 chunk's area sizes to be a multitiple of atom_size, and to
2597 eliminate the MAX_MEM_AREA restriction of 65536 bytes. we also
2598 catch cases where users pass an area size < atom size with a
2599 return_if_fail statement now (which is ok, because previously this
2600 lead to memory corruption anyways).
2602 Thu Sep 16 13:19:54 1999 Tim Janik <timj@gtk.org>
2604 * glib.h (g_trash_stack_pop): add explicit (GTrashStack*) cast for NULL
2605 pointer to cure ANSI C++ error.
2607 Mon Sep 13 23:25:59 1999 Tim Janik <timj@gtk.org>
2609 * gmessages.c (g_logv): in case we have to abort the program,
2610 debugging is enabled and we are not called recursively, try
2611 to abort with raise (SIGTRAP) first, so developers may ignore
2612 certain failure conditions during debugging stage.
2614 Thu Aug 26 15:09:36 1999 Tim Janik <timj@gtk.org>
2617 * gmodule/Makefile.am:
2618 * gthread/Makefile.am: added --export-dynamic so we can load dynmic
2619 modules, (required, according to the libtool 1.3.3 docu).
2621 1999-07-23 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2623 * grand.c (g_rand_new): Use /dev/urandom, as it doesn't block,
2624 which /dev/random might do. Do not XOR the time, when getting the
2625 seed form /dev/urandom, as this is good itself. Prevent the
2626 initial seed from being zero, which causes the PRNG to produce
2627 only zeros. Hints from Colin Plumb <colin@pgp.com>.
2629 1999-08-17 Tor Lillqvist <tml@iki.fi>
2631 * glib.h (g_trash_stack_push): Add a cast.
2634 * glist.c: Make the inline functions static inline, and add
2635 separate extern wrappers. Not all compilers produce callable entry
2636 points for inline functions, even if gcc does.
2638 Sun Aug 15 02:47:14 1999 Tim Janik <timj@gtk.org>
2640 * glib.h (g_trash_stack_pop): eliminate memset() call, since string.h
2641 has not neccessarily been included prior to glib.h.
2643 Mon Aug 2 21:03:10 1999 Tim Janik <timj@gtk.org>
2645 * configure.in: added --enable-msg-prefix option.
2647 * gmessages.c (g_log_default_handler): feature "prg_name (pid:%u): "
2648 if --enable-msg-prefix was selected (use "(process:%u): " if
2649 g_get_prgname () returns NULL, along the lines of g_on_error_query).
2651 1999-08-03 Tor Lillqvist <tml@iki.fi>
2655 * tests/strfunc-test.c: Rename g_strccpy to g_strcompress and
2656 g_strecpy to g_strescape per Tim Janik's suggestion. Dropped the
2657 destination parameter, always g_malloc a new string. Fix bug in
2658 g_strcompress, octal digits were gobbled up without limit, should
2661 Sources that use g_strescape must have ifdefs to be compilable
2662 both with GLib 1.2 and 1.3.
2664 Sat Jul 31 17:52:03 PDT 1999 Manish Singh <yosh@gimp.org>
2667 * gstrfuncs.c: the #define for g_strescape interfered with the
2668 compilation of the function, so just remove the function and
2669 note that it's deprecated in the header
2671 1999-08-01 Tor Lillqvist <tml@iki.fi>
2673 * gstrfuncs.c (g_strccpy, g_strecpy): New functions.
2675 * glib.h: Declare and document them. Define the deprecated
2676 g_strescape as a macro that calls g_strecpy.
2678 * tests/strfunc-test.c (main): Test them.
2680 * makefile.{cygwin,msc}.in
2681 * tests/makefile.{cygwin,msc}.in: Remove gstack and its test
2684 * glib.def: Additions and removals.
2686 * README.win32: Improve gcc build instructions.
2688 * build-dll: Also build import library for MSVC.
2690 Sat Jul 24 20:11:35 1999 Tim Janik <timj@gtk.org>
2692 * merged GLib 1.3.0 with glib-1.2.3 from Fri Jul 16 22:18:36.
2693 * incorporated proposed cleanups from gtk-devel-list.
2695 * bumped version number to GLib-1.3.1
2701 removed string tokenisation (we got g_strsplit() and g_strjoin()
2702 already) and readline functions.
2703 s/g_list_delete/g_list_delete_link.
2704 implemented g_slist_delete_link.
2705 removed notion of g_ATEXIT() macro in glib.h, this is an *internal*
2706 macro, g_atexit() is provided for public consumption.
2707 added GTrashStack inline utility functions.
2708 reimplement double eneded queues.
2709 removed GStack implementation, people can use a queue or a (singly)
2710 linked list for this task.
2711 deprecated g_strescape(), we need the SunOS variants here.
2713 * gdate.c: added DEBUG_MSG() macro to wrap old messages.
2717 * upgrade to libtool 1.3.3.
2719 1999-07-21 Tor Lillqvist <tml@iki.fi>
2721 Win32: With the latest gcc (2.95, pre-release), we can have binary
2722 compatibility with MSVC by using the switch -fnative-struct. No
2723 longer build DLLs with .gcc in the name when using gcc.
2725 * README.win32: Renew gcc build instructions.
2727 * build-dll: Comments change, handle also .a files.
2729 * makefile.cygwin.in
2730 * tests/makefile.cygwin.in: Remove .gcc from DLL name.
2732 1999-07-13 Tor Lillqvist <tml@iki.fi>
2734 * README.win32: Correct URL for mingw runtime sources.
2736 * build-dll: Combine commands with &&.
2738 * glib.h: Map also rmdir() and hypot() for MSVCRT library.
2740 * makefile.cygwin.in
2741 * tests/makefile.cygwin.in: New DLL naming style. GCC-compiled DLLs are
2742 now called *.gcc.dll, to avoid binary incompatibilities with
2743 MSVC-compiled versions.
2745 * makefile.msc.in: Cosmetics.
2747 1999-07-07 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2749 * configure.in: Test for pthread_join rather than for
2750 pthread_create to determine the right thread-lib. Makes it work on
2751 mips-sgi-irix6.5. Hitn from to Jari Vuoksenranta
2752 <javu@piano.ux.phys.jyu.fi>.
2754 1999-07-02 Tor Lillqvist <tml@iki.fi>
2756 * README.win32: Note about need to fix another bug in the mingw32
2759 * makefile.msc.in: Debugging turned on via an nmake variable,
2760 no need to edit the makefile.
2762 1999-07-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2764 * configure.in, acconfig.h, gutils.c: Added a g_memmove
2765 replacement for platforms without memmove, where bcopy can't
2766 handle overlapping copies and the corresponding checks, which is
2767 taken form the PERL Configure routine.
2769 * glib.h: Updated the commentary about g_memmove to be right and
2772 * configure.in: Removed test for rand_r, as it isn't used anymore.
2774 1999-06-30 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2776 * glib.h, grand.c: Finally removed the g_random_normal and
2777 g_rand_normal functions.
2779 1999-06-28 Tor Lillqvist <tml@iki.fi>
2781 * glib.def: Add missing export of g_strncasecmp.
2783 1999-06-21 Jose Mercado <jmercado@mit.edu>
2785 * glib.spec.in: Changed version number (1.1->1.3) in files section
2786 to allow rpm to build packages again.
2788 1999-06-21 Tor Lillqvist <tml@iki.fi>
2790 * README.win32: Update the pthreads snapshot version we want.
2791 Advice how to hand-expand the makefile.*.in files.
2793 * config.h.win32.in: Define values needed by Sebastian Wilhelmi's
2796 * glib.def: Add new functions.
2798 * glibconfig.h.win32.in: Update the pthreads snapshot version.
2801 * gthread.c: Include config.h, guard inclusion of unistd.h. When
2802 using gcc on Win32, g_thread_functions_for_glib_use must be marked
2803 for export here, too.
2805 * gtimer.c: Implement g_usleep on native Win32 using Sleep (which
2806 only has millisecond granularity, though).
2808 * makefile.cygwin.in
2809 * makefile.msc.in: Update pthreads snapshot version. File
2810 name changes. Remove testgthread.
2812 * tests/makefile.cygwin.in
2813 * tests/makefile.msc.in: Add thread-test. Link with gthread lib.
2815 1999-06-18 Jeff Garzik <jgarzik@pobox.com>
2817 * tests/Makefile.am: Re-order tests in alpha order.
2819 1999-06-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2821 * configure.in: Changed test for pthread_attr_setstacksize from
2822 AC_TRY_COMPILE to AC_TRY_LINK.
2824 1999-06-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2826 * configure.in, acglib.m4, acconfig.h, glib.h, gthread.c:
2827 Completed the thread support in GLib. Thread creation,
2828 prioritizing threads, yielding, joining threads as well as
2829 reader/writer locks and recursive mutexes are now in place. Please
2830 test heavily on your platform. It is so far tested on
2831 Linux/i386/pthreads, Solaris/Sparc/pthreads and
2832 Solaris/Sparc/solaristhreads.
2834 * gtimer.c, glib.h: Implement g_usleep (gulong microseconds) for
2835 thread safe sleeping. (sleep() is not MT-safe at all!)
2837 * gutils.c: Avoid compiler warning.
2839 * tests/Makefile.am, tests/thread-test.c: New program to test some
2840 aspects of the thread implementation.
2842 * gthread.c, Makefile.am: Renamed from gmutex.c to reflect the
2845 * configure.in: Purged all appearances of nspr.
2847 Wed Jun 2 11:42:46 PDT 1999 Manish Singh <yosh@gimp.org>
2853 * ltmain.sh: upgrade to libtool 1.3.2 (BeOS changes merged)
2855 1999-05-29 Tor Lillqvist <tml@iki.fi>
2857 * gstrfuncs.c (g_strescape): Backslashify also '"' characters.
2859 * glib.h: Document g_strescape.
2861 1999-05-12 Tor Lillqvist <tml@iki.fi>
2863 * glib.h (Win32): Map fileno to _fileno for mingw32. Map fstat to
2866 * README.win32: Advice also to remove -lmoldname in the
2867 patch to the egcs-1.1.2 spec file.
2869 Wed May 12 00:23:55 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
2871 * gmodule/Makefile.am: Another small fix.
2873 1999-05-08 Tor Lillqvist <tml@iki.fi>
2875 * Makefile.am tests/Makefile.am: Correct rules for making the
2876 win32-related files that are made from corresponding .in files.
2877 Is there a cleaner way than explicitly writing rules that invoke
2880 Fri Jul 16 22:18:36 PDT 1999 Manish Singh <yosh@gimp.org>
2883 * ltmain.sh: upgrade to libtool 1.3.3
2885 1999-06-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2887 * gdate.c, gstrfuncs.c, gstring.c: Fixed the use of the
2888 is..... and to..... macros, which take unsigned chars, not chars!
2889 Thanks to Morten Welinder <terra@diku.dk> for pointing this out.
2891 Thu Jun 3 16:30:31 PDT 1999 Manish Singh <yosh@gimp.org>
2893 * gerror.c (g_on_error_query): check isatty() before querying so
2894 we don't loop endlessly
2896 Sat May 29 11:16:29 PDT 1999 Manish Singh <yosh@gimp.org>
2902 * ltmain.sh: upgrade to libtool 1.3.2
2904 1999-05-26 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
2906 * gmain.c: provide a poll prototype for SunOS, as they do not do
2907 it self. Hint from Christian Parg <cparg@fs-design.de>.
2909 Tue May 25 12:23:07 1999 Owen Taylor <otaylor@redhat.com>
2911 * gstrfuncs.c (g_strchug): Use g_memmove() not memmove().
2912 (Reported by Charles Levert <charles@comm.polymtl.ca>)
2914 Mon May 10 22:03:52 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
2916 * Released GLib 1.2.3
2918 1999-05-08 Tor Lillqvist <tml@iki.fi>
2920 * Makefile.am tests/Makefile.am: Correct rules for making the
2921 win32-related files that are made from corresponding .in files.
2922 Is there a cleaner way than explicitly writing rules that invoke
2925 Sat May 1 10:18:01 PDT 1999 Manish Singh <yosh@gimp.org>
2931 * ltmain.sh: upgrade to libtool 1.3
2933 1999-04-30 Tor Lillqvist <tml@iki.fi>
2935 * Makefile.am: Don't distribute glibconfig.h.win32.in, but
2936 glibconfig.h.win32. Generate it when making a dist. Also generate
2937 makefile.msc and config.h.win32 from corresponding .in files when
2940 * configure.in: Also substitute @GLIB_INTERFACE_AGE@ and
2941 @GLIB_BINARY_AGE@ (needed in config.h.win32).
2943 * glibconfig.h.win32.in: Use static mutex structure and initial
2944 value corresponding to the 1999-04-07 snapshot of pthreads-win32.
2946 * tests/Makefile.am: Distribute makefile.msc. Generate it when
2949 * tests/makefile.msc.in: New file.
2951 * tests/node-test.c: Include <stdlib.h> for exit().
2953 Thu Apr 29 02:16:36 1999 Tim Janik <timj@gtk.org>
2955 * gstrfuncs.c: minor code cleanups.
2957 Tue Apr 27 13:11:29 1999 Owen Taylor <otaylor@redhat.com>
2959 * gmain.c (g_main_poll): Mask out ERR HUP and NVAL from
2960 the events field so we don't give IRIX fits.
2962 Tue Apr 20 08:42:22 1999 Tim Janik <timj@gtk.org>
2964 * gscanner.c (g_scanner_unexp_token): behave conservative with
2965 G_TOKEN_IDENTIFIER_NULL and always assume scanner->value.v_string
2966 to be "null" in that case.
2968 1999-04-20 Havoc Pennington <hp@pobox.com>
2970 * gutils.c (g_vsnprintf): When using the vsnprintf()
2971 implementation, '\0'-terminate the resulting string
2972 and return its length rather than -1.
2974 Mon Apr 19 13:42:21 1999 Owen Taylor <otaylor@redhat.com>
2976 * gmain.c (g_main_iterate): Added missing
2977 #ifdef G_THREADS_ENABLED. (I never liked G_THREADS_ENABLED in
2980 1999-04-18 Havoc Pennington <hp@pobox.com>
2982 * gutils.c (g_snprintf): When using the vsnprintf()
2983 implementation, '\0'-terminate the resulting string
2984 and return its length rather than -1.
2986 Fri Apr 16 06:52:07 1999 Tim Janik <timj@gtk.org>
2988 * gscanner.c (g_scanner_unexp_token): feature G_TOKEN_EOF as a valid
2989 expected token as well, so we get "- expected end of file" instead of
2990 "- expected (unknown) token <0>".
2992 Tue Apr 13 16:16:14 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
2994 * Released GLib 1.2.2
2996 1999-04-12 Elliot Lee <sopwith@mh69.mh.cuc.edu>
2998 * g_strchug(): s/strcpy/memmove/
3000 1999-04-12 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3002 * configure.in: Adjusted the test for an unimplemented
3003 getpwuid_r. Info from Michael Pruett <mikep@ugcs.caltech.edu>.
3005 Sun Apr 11 15:07:34 1999 Tim Janik <timj@gtk.org>
3007 * configure.in: bumped versin number to GLib 1.2.2, interface 2,
3012 Sun Apr 11 14:37:06 1999 Tim Janik <timj@gtk.org>
3014 * gstrfuncs.c (g_strcasecmp): always check for s1, s2 != NULL.
3016 Sat Apr 10 19:30:50 1999 Tim Janik <timj@gtk.org>
3018 * glib.h: removed braces around inline strings for the G_GNUC_FUNCTION
3019 and G_GNUC_PRETTY_FUNCTION macros, so the macros can be used for
3020 compile time string concatenation.
3022 Thu Apr 8 19:53:19 1999 Owen Taylor <otaylor@redhat.com>
3024 * gmain.c (g_main_iterate): Check for two threads
3025 calling g_main_iterate at once.
3027 * gmain.c: If the set of poll file descriptors changes
3028 during a call to poll(), abort that call, and start
3029 a new poll. My test program still segfaults
3030 obscurely on glibc 2.0 (in read()!!!), but now it works on
3031 glibc 2.1, so I'll blame something else for the other segfault.
3033 1999-03-31 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3035 * configure.in: Fixed slight bug, that made configure hang on some
3036 systems. Please do not merge this into 1.3 branch. It's taken care
3037 of differently there. Info from J. Rhett Aultman
3038 <cuplan@alley.gator.net>
3040 Wed Mar 24 21:23:47 CST 1999 Shawn T. Amundson <amundson@gtk.org>
3042 * Released GLib 1.2.1
3047 sanity_check: updated
3049 * glibconfig.h.win32.in:
3051 docs/glib-config.1.in:
3052 docs/Makefile.am: Added files used to generate new files.
3054 * glibconfig.h.win32:
3055 docs/glib-config.1: Removed, now generated.
3057 * configure.in: Added to output now-generated files.
3059 Tue Mar 23 13:43:39 PST 1999 Manish Singh <yosh@gimp.org>
3061 * giounix.c: add user_data param to check and prepare functions
3063 Mon Mar 22 03:54:43 1999 Tim Janik <timj@gtk.org>
3066 * gmain.c: add user_data to the GSource ->check and ->prepare
3067 functions, so it can be used to e.g. pass a GPollFd.
3068 (g_main_poll): only add poll records with an events mask != 0 to the
3069 fd_array. don't even bother calling poll_func() if fds=timeout=0.
3070 added debugging printouts around poll_func() invokation that can be
3071 enabled with #define G_MAIN_POLL_DEBUG.
3073 Fri Mar 19 16:29:50 PST 1999 Manish Singh <yosh@gimp.org>
3079 * ltmain.sh: upgrade to libtool 1.2f
3081 * autogen.sh: libtool is not required to autogen glib
3083 * acconfig.h: remove WITH_SYMBOL_UNDERSCORE (not explictly
3086 1999-03-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3088 * gmem.c: Fixed another stupid fault of mine: Did
3089 s/g_static_/g_private_/g
3091 Wed Mar 17 03:17:42 1999 Tim Janik <timj@gtk.org>
3093 * configure.in bumped versin number to GLib 1.2.1, interface 1,
3098 * glib.h: added GLIB_CHECK_VERSION() macro similar to
3099 GTK_CHECK_VERSION().
3101 Sun Mar 14 17:50:35 1999 Tim Janik <timj@gtk.org>
3103 * gmem.c (g_mem_chunk_*): changed a bunch of g_assert() statements
3104 to g_return_if_fail().
3106 (g_mem_chunk_print):
3107 (g_mem_chunk_info): removed some extraneous "\n"s at the end of the log
3110 * gtimer.c (g_timer_*): changed a bunch of g_assert() statements
3111 to g_return_if_fail().
3112 * grel.c (g_*): changed a bunch of g_assert() statements to
3113 g_return_if_fail() and added some extra ones to check relation != NULL.
3115 1999-03-12 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3117 * configure.in: Also accept _Pctime_r instead of ctime_r, while
3118 seraching for the right `_REENTRANT' flag. This is for Digital
3119 UNIX 4.0d. Thanks to Sascha Brawer <sb@adasys.ch>.
3121 Tue Mar 9 23:25:50 1999 Tim Janik <timj@gtk.org>
3123 * configure.in: check for working realloc (NULL,).
3124 * gmem.c (g_realloc): use malloc() for initial allocation on systems
3125 where realloc(NULL,) will not work (this is the case on SunOS, reported
3128 Mon Mar 8 07:42:08 1999 Tim Janik <timj@gtk.org>
3130 * ghook.c (g_hook_unref): when !hook_list->is_setup, wrap the
3131 flag around the call to g_hook_free() to avoid spurious
3132 warnings (happens during destruction phase).
3134 1999-03-03 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3136 * glibconfig.h.win32, config.h.win32: Moved G_THREADS_IMPL_POSIX
3137 from config.h.win32 to glibconfig.h.win32
3139 * acconfig.h, configure.in, config.h.win32: Added test for DCE
3140 versions of mutex_trylock and cond_timedwait. The win32 versions
3141 are posix, aren't they?
3143 1999-03-02 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3145 * gmem.c: Fixed a stupid cut'n'paste error of mine. Thanks to
3146 Friedrich Dominicus <Friedrich.Dominicus@inka.de>
3148 1999-03-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3150 * gutils.c (g_get_any_init): Fixed yet another bloody
3151 implementation of getpwuid_r on AIX. Thanks to Olaf Dietsche
3152 <olaf.dietsche+list.gtk@netcologne.de>. I would like a configure
3153 test better than that, but have no idea, how to do that easily.
3155 Sun Feb 21 22:11:51 CST 1999 Shawn T. Amundson <amundson@gtk.org>
3157 * Released GLib 1.2.0
3161 Wed Feb 24 00:08:42 CST 1999 Shawn T. Amundson <amundson@gtk.org>
3163 * *.[ch]: inserted additional note to look for ChangeLog and
3164 AUTHORS file for a log of modifications.
3166 Sun Feb 21 14:01:00 1999 Dr Mike <drmike@redhat.com>
3168 * Made specfile generated, tweaked slightly
3170 Sat May 8 06:00:17 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
3174 gthread/Makefile.am: Better testing reveals better
3175 methods. Fixes for BeOS.
3177 Sat May 8 01:52:29 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
3188 gmodule/gmoduleconf.h.in
3189 gmodule/gmodule-beos.c
3190 gthread/Makefile.am: Port to BeOS by myself and Richard Offer.
3192 1999-05-06 Tor Lillqvist <tml@iki.fi>
3194 * makefile.msc.in makefile.cygwin.in glibconfig.h.win32.in
3195 config.h.win32.in tests/makefile.msc.in tests/makefile.cygwin.in:
3196 New files, used to generate corresponding non-.in files when
3197 making a dist. This is just so the version numbers will be kept in
3198 synch automatically.
3200 * configure.in: Also substitute @GLIB_MAJOR_VERSION@,
3201 @GLIB_MINOR_VERSION@, and @GLIB_INTERFACE_AGE@.
3203 * Makefile.am tests/Makefile.am: Also distribute makefile.cygwin.
3205 * gerror.c (g_on_error_query): On Win32, put up a MessageBox and
3208 * glib.def: Add a couple of functions.
3210 Sat May 1 10:26:20 PDT 1999 Manish Singh <yosh@gimp.org>
3216 * ltmain.sh: upgrade to libtool 1.3
3218 1999-04-25 Tor Lillqvist <tml@iki.fi>
3220 * README.win32: More editing.
3221 * build-dll: Use gcc, not ld to link.
3222 * glib.h: On native Win32 use _unlink().
3223 * gscanner.c: Use corrent NATIVE_WIN32 feature test macro,
3225 * gstring.c: Include <io.h> on Win32 for _read prototype.
3226 * gutils.c: Remove old IO channel code (was in #if 0).
3227 * makefile.cygwin: Don't need to link with kernel32 and msvcrt
3228 explicitly, they are included anyway.
3230 1999-04-24 Tor Lillqvist <tml@iki.fi>
3232 Support added for building using a GNU toolchain on Win32,
3233 i.e. gcc -mno-cygwin on cygwin (a.k.a. mingw32, using egcs-1.1.2).
3235 * README.win32: Updated.
3236 * build-dll makefile.cygwin tests/makefile.cygwin: New files.
3237 * glib.h glib.def glibconfig.h.win32: Slight updates.
3238 * gmain.c: No need to include <fcntl.h> and <io.h> on Win32.
3239 * gmain.c gutils.c testglib.c tests/string-test.c: Test for
3240 NATIVE_WIN32, not _MSC_VER.
3241 * gmutex.c: Must declare g_thread_functions_for_glib_use as
3242 exported (using the GUTILS_C_VAR macro).
3243 * gutils.c gmodule/libgplugin_[ab].c: LibMain not needed.
3244 * gmodule/gmoduleconf.h.win32: Need underscore with gcc.
3245 * gthread/gthread.c: With gcc on Win32, must use memcpy to assign
3246 value of g_thread_functions_for_glib_use (?).
3247 * makefile.msc tests/makefile.msc: Cosmetics.
3249 Fri Apr 23 14:29:25 BST 1999 Tony Gale <gale@gtk.org>
3251 * glib.h: Fix typo in g_string_ncasecmp macro (by me).
3252 Add b_string_strncasecmp macro.
3254 1999-04-22 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3256 * gutils.c (g_get_any_init): use sysconf (_SC_GETPW_R_SIZE_MAX) as
3257 the new initinal bufsize for getpwuid_r on systems, that support
3258 this. Hint from Holger Duerer <H.Duerer@zait.uni-bremen.de>.
3260 Sat Apr 17 20:55:13 BST 1999 Tony Gale <gale@gtk.org>
3262 * glib.h, gstring.c: Add new g_string functions for reading
3263 from file/socket descriptors, and tokenising strings.
3265 Added various g_string macros.
3267 Tue Apr 13 23:28:32 1999 Tor Lillqvist <tml@iki.fi>
3269 * README.win32: Mention the tests directory.
3271 * glib.def: Add the functions from grand.c.
3273 * glibconfig.h.win32: Add unsigned max values, and the format
3276 * makefile.msc: Add grand.
3278 * tests/{date-test,node-test}.c: Include <stdlib.h> for exit().
3280 * tests/makefile.msc: New file.
3282 1999-04-12 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3284 * glib.h: Moved struct declaration up. Style fixes.
3286 * grand.c: Style fixes. Only try to open /dev/random once.
3288 * tests/rand-test.c (main): New tests; Slight bug fix.
3290 1999-04-09 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3292 * grand.c, tests/rand-test.c: New files to implement the Mersenne
3293 Twister Pseudo Random Number Generator.
3295 * glib.h, AUTHORS, Makefile.am, tests/Makefile.am: Changed
3298 Thu Apr 8 21:12:30 CDT 1999 Shawn T. Amundson <amundson@gtk.org>
3300 * Released GLib 1.3.0
3302 1999-03-30 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3304 * configure.in: Added a check for the right format to printf and
3305 scanf long longs. It is %qi instead of %lli on FreeBSD for
3308 1999-03-28 Raja R Harinath <harinath@cs.umn.edu>
3310 * Makefile.am (glibconfig.h): Make sure `glibconfig.h' exists
3311 after the rule is fired.
3312 (install-exec-local): Install glibconfig.h only if the contents
3313 are different from the currently installed glibconfig.h.
3315 1999-03-26 Raja R Harinath <harinath@cs.umn.edu>
3317 * Makefile.am (configexecincludedir): Rename from
3318 configincludedir so that glibconfig.h will be installed
3319 as part of `make install-exec'.
3321 Thu Mar 25 22:45:47 1999 Tor Lillqvist <tml@iki.fi>
3323 * config.h.win32: Update version numbers.
3325 * glibconfig.h.win32: Update version numbers and pthreads-win32-
3326 related magic values.
3328 * README.win32: Some improvements.
3330 * makefile.msc: Add gqueue and gstack. Correct version number.
3332 Fri Mar 19 16:29:50 PST 1999 Manish Singh <yosh@gimp.org>
3338 * ltmain.sh: upgrade to libtool 1.2f
3340 * autogen.sh: libtool is not required to autogen glib
3342 * acconfig.h: remove WITH_SYMBOL_UNDERSCORE (not explictly
3345 1999-03-18 Jeff Garzik <jgarzik@pobox.com>
3347 * glib.def: Add new g_list, g_stack, g_queue functions.
3349 1999-03-18 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3351 * configure.in: added new AC_SUBST(GTHREAD_COMPILE_IMPL_DEFINES)
3352 to hold various defines to get the right thread implementation on
3353 different platforms. Also look in -ldce for pthread_create. Should
3354 make it work on HP-UX 10.x. Information from "D. Emilio Grimaldo
3355 Tunon" <emilio_tunon@nl.compuware.com>.
3357 1999-03-17 Jeff Garzik <jgarzik@pobox.com>
3359 * gstack.c, gqueue.c:
3360 Add copyright, clean up code a bit.
3362 1999-03-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3364 * configure.in: Added missing values for G_MAXU(SHORT|INT|LONG) on
3365 platforms with only /usr/include/values.h.
3367 * acconfig.h: Removed unnecessary macros.
3369 * glibconfig.h.win32, config.h.win32: Moved G_THREADS_IMPL_POSIX
3370 from config.h.win32 to glibconfig.h.win32. Taken from glib 1.2
3373 * configure.in: Also accept _Pctime_r instead of ctime_r, while
3374 seraching for the right `_REENTRANT' flag. This is for Digital
3375 UNIX 4.0d. Taken from glib 1.2 branch.
3377 Wed Mar 17 03:14:56 1999 Tim Janik <timj@gtk.org>
3379 * glib.h: added GLIB_CHECK_VERSION() macro similar to
3380 GTK_CHECK_VERSION().
3382 Wed Mar 17 01:46:28 1999 Tim Janik <timj@gtk.org>
3384 * merges from glib-1-2:
3386 Sun Mar 14 17:50:35 1999 Tim Janik <timj@gtk.org>
3388 * gmem.c (g_mem_chunk_*): changed a bunch of g_assert() statements
3389 to g_return_if_fail().
3391 (g_mem_chunk_print):
3392 (g_mem_chunk_info): removed some extraneous "\n"s at the end of the log
3395 * gtimer.c (g_timer_*): changed a bunch of g_assert() statements
3396 to g_return_if_fail().
3397 * grel.c (g_*): changed a bunch of g_assert() statements to
3398 g_return_if_fail() and added some extra ones to check relation != NULL.
3400 Tue Mar 9 23:25:50 1999 Tim Janik <timj@gtk.org>
3402 * configure.in: check for working realloc (NULL,).
3403 * gmem.c (g_realloc): use malloc() for initial allocation on systems
3404 where realloc(NULL,) will not work (this is the case on SunOS, reported
3407 Mon Mar 8 07:42:08 1999 Tim Janik <timj@gtk.org>
3409 * ghook.c (g_hook_unref): when !hook_list->is_setup, wrap the
3410 flag around the call to g_hook_free() to avoid spurious
3411 warnings (happens during destruction phase).
3413 1999-03-02 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3415 * gmem.c: Fixed a stupid cut'n'paste error of mine. Thanks to
3416 Friedrich Dominicus <Friedrich.Dominicus@inka.de>
3418 1999-03-16 Timur Bakeyev <mc@bat.ru>
3420 * configure.in: Fix problem with pthread_create in libc, as running
3421 "gcc test.c -l " is not legal.
3423 1999-03-16 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3425 * tests/type-test.c: Added a test for the
3426 G_(U)?INT(16|32|64)_FORMAT and G_(MIN|MAX|MAXU)(SHORT|INT|LONG)
3429 * configure.in: Removed G_(U)?INT8_FORMAT again, as it can't be
3432 * configure.in: Added the macros G_MAXU(SHORT|INT|LONG). I do not
3433 know how to handle these on platforms with /usr/include/values.h,
3434 but without /usr/include/limits.h. Please someone add this.
3437 1999-03-15 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3439 * configure.in: Added the macros G_(U)?INT(8|16|32|64)_FORMAT to
3440 use for printf and (much more important) scanf format strings for
3441 the corresponding GLib types.
3443 * glib.h Added G_(U)?(SHORT|INT|LONG)_FORMAT for consistency. It
3444 however makes no sense to also provide G_(FLOAT|DOUBLE)_FORMAT, as
3445 they are different for printf (f for both) and scanf (f for float,
3446 lf for double). Defining G_INT_FORMAT makes sense however, as we
3447 might want to define gint to something different than int someday
3448 in the future. Idea from Sascha Brawer <sb@adasys.ch>.
3450 1999-03-14 Jeff Garzik <jgarzik@pobox.com>
3453 Commented out debugging output.
3455 * tests/Makefile.am, tests/date-test.c:
3456 Added test of the GDate module, based closely on testgdate.c.
3458 * tests/Makefile.am:
3459 Bugfix - compile tests with @GLIB_DEBUG_FLAGS@.
3461 1999-03-14 Raja R Harinath <harinath@cs.umn.edu>
3463 * configure.in (glibconfig.h): Remove widechar tests and defines.
3464 (fd_set): Change the grep for `fd_mask' to search for `fd_set'.
3465 * gerror.c (fd_mask): Remove conditional typedef. It is not used
3466 elsewhere in the file.
3467 * gmain.c (fd_mask): Likewise.
3469 1999-03-12 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3471 * configure.in: Test for posix threads first, then for dce threads.
3473 1999-03-11 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3475 * configure.in: Revamped the thread configure stuff. Now dce
3476 threads (old posix draft) are recogniced. This is necessary,
3477 because dce threads are in fact working quite differently from
3478 posix threads. Also changed the conditions for checking for MT
3479 safe functions a bit, because G_THREADS_IMPL_NONE still have to
3480 compile thread safe. NOTE: Please do not commit my change to
3481 glib-1-2/{acconfig.h,configure.in,config.h.win32} from 1999-03-03,
3482 as the current change will take care of that too.
3484 Tue Mar 9 14:37:32 1999 Jeff Garzik <jgarzik@pobox.com>
3486 * Makefile.am, glib.h, gstack.c, gqueue.c,
3487 tests/Makefile.am, tests/queue-test.c, tests/stack-test.c:
3488 Added stack, queue ADTs and related tests.
3491 New g_list_delete() function.
3493 Sat Mar 6 11:03:08 1999 Asbjorn Pettersen <ape@lrdpf.spacetec.no>
3495 * gutils.c (g_get_any_init): add OS/2 changes.
3496 change '\\' in HOME to '/'.
3498 1999-03-03 Josh MacDonald <jmacd@spin.dsl.pacbell.net>
3500 * glib.def: g_spaced_primes_closest was omitted here, so I
3501 couldn't build Xdelta on Windows.
3503 1999-03-01 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
3505 * gutils.c (g_get_any_init): Fixed yet another bloody
3506 implementation of getpwuid_r on AIX. Thanks to Olaf Dietsche
3507 <olaf.dietsche+list.gtk@netcologne.de>. I would like a configure
3508 test better than that, but have no idea, how to do that easily.
3510 Sat Feb 27 01:18:47 1999 Tim Janik <timj@gtk.org>
3512 * ChangeLog: moved old ChangeLog to ChangeLog.pre-1-2, and started
3515 * configure.in: set glib version to 1.3.0.