2006-03-13 Anders Carlsson <andersca@luminoth.local>
* configure.in: Revert fix for #322476, it breaks module loading
since libtool on darwin makes shared modules use .so and shared
libraries use .dylib. The fix breaks shared module loading everywhere
in GTK+.
2006-03-02 Marcus Brinkmann <mb@g10code.de>
Implement watches for GIOChannels for write file descriptors on
Win32 (#333098).
* glib/giowin32.c (GIOWin32Channel): Add a new direction field.
(read_thread): Initialize direction.
(write_thread): New function.
(buffer_write): New function.
(g_io_win32_prepare): Handle the G_IO_WIN32_FILE_DESC case for the
write direction.
(g_io_win32_fd_write): Call buffer_write() if there is a writer
thread.
(g_io_win32_fd_close): Set space_avail_event for writer threads.
(g_io_win32_fd_create_watch): Create the writer thread if
condition is G_IO_OUT.
(g_io_channel_win32_make_pollfd): Likewise here.
2006-03-11 Tor Lillqvist <tml@novell.com>
* gthread-win32.c: #define _WIN32_WINDOWS as 0x0401 to get
declaration for IsDebuggerPresent() when using MSVC6. (#333879,
Kazuki Iwamoto)
2006-03-09 Matthias Clasen <mclasen@redhat.com>
* glib/templ/timers.sgml: Mention that the second
parameter of g_timer_elapsed can be NULL (#333916,
Christian Neumair)
2006-03-07 Matthias Clasen <mclasen@redhat.com>
* glib/gutf8.c (_g_charset_get_aliases): Match the prototype
in gconvert.c to fix build problems in NetBSD. (#333651,
Julio M. Merino Vidal)
Tue Mar 7 17:09:07 2006 Tim Janik <timj@gtk.org>
* gobject.c (g_object_compat_control): added case3 to hand out the
original floating flag handler. required by gtk+ >= 2.10.
2006-03-06 Anders Carlsson <andersca@imendio.com>
* configure.in:
Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
(#314794, Gregor Riepl)
2006-03-02 Tor Lillqvist <tml@novell.com>
* gthread-win32.c (G_PRIVATE_MAX): Increase to 100. 16 was rather
low.
(g_private_new_win32_impl): Can't use g_error() here as
g_private_new() is called a few times by GLib internally before
the messaging system that g_error() requires is ready. Thanks to
Tim Janik for noticing. Just display a MessageBox() and abort()
instead.
2006-03-01 Tor Lillqvist <tml@novell.com>
* glib/gutils.c (g_listenv): Separate implementation on Win32: Use
the wide character API on NT-based Windows. Return UTF-8 strings.
* glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
meant it wasn't present in the import library. PRIVATE is used
only for the backwards-compatibility DLL ABI stability hacks.
* gobject/tmpl/gtype.sgml:
add @since: for _add_private, _GET_PRIVATE
* gobject/tut_gobject.xml:
fix example to use ->priv and not ->private
* gobject/tut_howto.xml:
fix g_type_class_add_private example
2006-02-14 Tor Lillqvist <tml@novell.com>
* glib/tmpl/iochannels.sgml: Document some Windows-specific issues.
* glib/glib-sections.txt: Move three Windows-specific functions
that now are documented from being Private to the correct section.
2006-02-20 Tor Lillqvist <tml@novell.com>
* gthread-win32.c (g_thread_exit_win32_impl): Make the
implementation of GPrivate behave more closely as in POSIX
threads: The value associacted with a GPrivate must be set to NULL
before calling the destructor. (The destructor gets the original
value as argument.) A destructor might re-associate a non-NULL
value with some GPrivate. To deal with this, if after all
destructors have been called, there still are some non-NULL
values, the process is repeated. (#331367)
* tests/asyncqueue-test.c:
* tests/list-test.c:
* tests/slist-test.c: Updated to test _sort, _sort_with_data,
_insert_sorted and _insert_sorted_with_data API.
2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
* glib/gthreadpool.c: Fix deadlock when signalling the thread
which freed a thread pool (#331110, Chris Wilson).
Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
* glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
posix_memalign() in case we actually are going to use it, because
we detected a compliant implementation (#328997).
* configure.in (enable_included_printf): don't include malloc.h when
testing for posix_memalign() funcitonality, since this may break the
test on some systems (#328997).
