+Sun Nov 4 20:29:31 2001 Owen Taylor <otaylor@redhat.com>
+
+ * glib/gdir.[ch]: Indentation fixes, some rewriting of docs to
+ conform to gtk-doc standard.
+
+ * glib/gdir.[ch] (g_dir_close): Remove the boolean
+ return value. What would you do if closing failed?
+ What would the user do if you printed a warning
+ message "closing directory %d failed"?
+
+2001-11-04 Hans Breuer <hans@breuer.org>
+
+ * glib/gdir.[hc] : (new files) simplified wrapper around dirent
+ functions to improve portability of downstream libs/apps
+ * glib/makefile.msc.in : use them
+ * glib/glib.def : export them
+ * config.h.win32.in : needing HAVE_DIRENT_H defined
+
+2001-11-04 Tor Lillqvist <tml@iki.fi>
+
+ * tests/Makefile.am: On Win32, create separate .exp file for
+ module-test.o and link with that. Otherwise the GNU linker doesn't
+ export the g_clash_func.
+
+2001-11-03 Hans Breuer <hans@breuer.org>
+
+ * glib/giowin32.c : static correctness
+
+ * glib/gwin32.c : some more #if defined (SUBLANG_* ...
+ (g_win32_get_package_installation_subdirectory) use g_build_filename
+ instead of duplicating its logic
+
+ * glib/glib.def : removed duplicates, added mising, removed
+ compat cruft
+
+Fri Nov 2 19:54:16 2001 Tim Janik <timj@gtk.org>
+
+ * glib/gbacktrace.h (G_BREAKPOINT): remove public signal.h include.
+
+Thu Nov 1 21:48:43 2001 Owen Taylor <otaylor@redhat.com>
+
+ * tests/mainloop-test.c (recurser_idle): Recurse
+ with may_block = FALSE, so we don't get into
+ the pathology where the recurser_idle recurses
+ for 10 iterations, and the only thing that is
+ running is the recurser idle, which adds
+ another recursion for each of those 10 iterations
+ and....
+
+ * tests/mainloop-test.c (create_crawler): Fix race
+ condition where a crawler source could be destroyed
+ before it was added to the crawler array.
+
+ * test/Makefile.am: Add mainloop-test back.
+
+2001-11-01 Marius Andreiana <mandreiana@yahoo.com>
+
+ * configure.in: Added ro (Romanian) to ALL_LINGUAS
+
+2001-10-31 Havoc Pennington <hp@redhat.com>
+
+ * glib/gmain.c (g_main_loop_run): fix to the locking so we don't
+ hang
+
2001-10-31 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
+ * configure.in: Make --with-threads=none mean, that we want thread
+ support, but no default thread implementation. This also was the
+ original intention, but disapeared around 1998... So it doesn't
+ seem to be the most requested feature, but we went far to make it
+ possible in the code, so I resurrected this feature.
+
+ * INSTALL.in: Describe the --enable-gc-friendly, --disable-threads
+ and --with-threads options of configure.
+
* glib/gmain.c (g_main_context_acquire, g_main_context_release,
g_main_context_wait): Use the right conditional (G_THREADS_ENABLED
instead of G_THREAD_ENABLED). Also remove wrong
projects / platform / upstream / glib.git / blobdiff