+2001-01-06 Hans Breuer <hans@breuer.org>
+
+ * glib.def : don't try to export g_strcpy, it is g_stpcpy;
+ updated and added some tweaking for functions which got
+ renamed recently, to avoid the update hassle if it can
+ simply be done compatible. Should vanish if there is a
+ stable version for win32.
+
+Fri May 4 11:49:18 2001 Owen Taylor <otaylor@redhat.com>
+
+ * Released 1.3.5
+
+ * NEWS: Updated
+
+ * configure.in (GLIB_MICRO_VERSION): Up version to 1.3.5,
+ interface/binary age 0.
+
+2001-05-04 Sven Neumann <sven@convergence.de>
+
+ * ghash.c: fixed a typo in a comment.
+
+ * gtree.[ch]: added new functions g_tree_new_full(), g_tree_replace(),
+ g_tree_steal() and g_tree_foreach() to adapt GTree to the GHashTable
+ API. Moved comments into the C file.
+
+ * docs/reference/glib/glib-sections.txt
+ * docs/reference/glib/tmpl/glib-unused.sgml
+ * docs/reference/glib/tmpl/hash_tables.sgml
+ * docs/reference/glib/tmpl/linked_lists_double.sgml
+ * docs/reference/glib/tmpl/linked_lists_single.sgml
+ * docs/reference/glib/tmpl/macros_misc.sgml
+ * docs/reference/glib/tmpl/trees-binary.sgml: updated documentation
+
+Thu May 3 06:38:28 2001 Owen Taylor <otaylor@redhat.com>
+
+ * g[s]list.c (g_[s]list_foreach) docs/Changes-2.0.txt: Make
+ foreach() safe against removal of the _current_ element. While
+ this could break some code, the new behavior is consistent with
+ the rest of GLib/GTK+ and probably is what people expect in most
+ cases. (Suggested by Paul Kuykendall, #50071)
+
+Wed May 2 11:10:22 2001 Owen Taylor <otaylor@redhat.com>
+
+ * gutf8.c (g_utf8_to_ucs4_fast): Fix read past end of the string.
+ (#50404, fix from Jonas Borgström)
+
+Sun Apr 29 00:37:34 2001 Tim Janik <timj@gtk.org>
+
+ * ghook.[hc]: made hook ids a gulong.
+
+2001-04-20 Dan Winship <danw@ximian.com>
+
+ * configure.in: Add a check for the Darwin dynamic linker. Use
+ AC_TRY_LINK when checking for "nonposix getpwuid_r" so it notices
+ "no getpwuid_r" correctly.
+
+ * testglib.c (main): Make template[] bigger to prevent an overrun.
+ Remove an unused variable. Initialize error to NULL.
+
+ * tests/gio-test.c (main): Add a cast to prevent a warning when
+ size_t is a long.
+
+ * tests/type-test.c (main): Add an #ifdef to prevent a warning
+ when G_HAVE_GINT64 is defined and G_GINT64_FORMAT isn't.
+
+2001-04-19 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
+
+ * ghash.c, ghash.h: Remove definition of g_hash_table_freeze and
+ g_hash_table_thaw. Instead added G_DISABLE_DEPRECATED-guarded
+ macros to ghash.h to go along the lines of the standard.
+
+ * gscanner.c, gscanner.h: Dito for g_scanner_freeze_symbol_table
+ and g_scanner_thaw_symbol_table.
+
+ * gutils.c, gutils.h: Dito for g_dirname. g_basename is still
+ defined in gutils.c, but declared ing gutils.h only
+ G_DISABLE_DEPRECATED-guarded.
+
+ * configure.in: Removed bashism in test for the pkg-config
+ version.
+
+ * configure.in: Rewrote test for multithread flag. Now uses
+ localtime_r, which hopefully has a consistent prototype across
+ different platforms. Also it uses a clever double EGREP trick
+ instead of compiling, which could give false positives. Thanks to
+ Dan Winship <danw@ximian.com> for the hint.
+
+Wed Apr 18 17:35:38 2001 Owen Taylor <otaylor@redhat.com>
+
+ * gutils.c (_glib_gettext): Add missing static pointed
+ out by Michael Meeks.
+
+Wed Apr 18 09:37:07 2001 Owen Taylor <otaylor@redhat.com>
+
+ * MAINTAINERS: Removed. Keeping README, README.cvs-commits
+ HACKING, and AUTHORS up to date is plenty without extra
+ random files that someone thought a module should have.
+ (Actually, I believe this was used for debbugs in the past.)
+
+Tue Apr 17 11:47:07 2001 Owen Taylor <otaylor@redhat.com>
+
+ * Released 1.3.4
+
+ * NEWS: Updated
+
+Tue Apr 17 10:43:36 2001 Owen Taylor <otaylor@redhat.com>
+
+ * gstrfuncs.c: Define _GNU_SOURCE for stpcpy
+
+ * tests/mainloop-test.c (main): Wait for all threads
+ to start before beginning tests.
+
+2001-04-17 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
+
+ * gthreadpool.c (g_thread_pool_thread_proxy): Until now every
+ thread pool always had at least one tread waiting to avoid
+ switching overhead in case a new task would be added soon after
+ one finished. This however means a big waste of threads, if many
+ mostly inactive thread pools are involved. Now such a waiting
+ thread will only wait for half a second (This value is of course
+ very randomly picked) and go to the global threadpool afterwards.
+
Mon Apr 16 12:04:52 2001 Owen Taylor <otaylor@redhat.com>
* configure.in: Remove warnings about conflicts with the