Thu Sep 25 15:43:08 2003 Owen Taylor <otaylor@redhat.com>
* tests/gobject/testmodule.[ch] test/gobject/Makefile.am:
Dummy dynamic type module for testing type plugin code
and dynamic types.
* test/gobject/defaultiface.c: Test of g_default_interface_ref,
etc.
Mon Sep 29 10:51:01 2003 Owen Taylor <otaylor@redhat.com>
* gtype.[ch]: Add g_type_default_interface_ref/peek/unref
for accessing the default vtable of an interface.
Fri Sep 26 17:24:53 2003 Owen Taylor <otaylor@redhat.com>
* gtype.c (g_type_instance_get_private): You can
have instance_real_class_bsa be non-NULL, but still
the class not be in the bsa. (Found by Kris Rietveld)
Fri Sep 12 16:31:40 2003 Owen Taylor <otaylor@redhat.com>
* gsignal.[ch]: Add g_signal_accumulator_true_handled(), to
do TRUE-stops-emit signals.
* Makefile.am: Move testoverride.c and testifaceinit.c to
tests/gobject.
2003-09-12 Matthias Clasen <maclas@gmx.de>
Make the g_value_set_x_take_ownership() functions "official"
part of the API (#100948):
* gvaluetypes.[hc]: Add g_value_take_string() (synonym to the
now deprecated g_value_set_string_take_ownership()).
* gparam.[hc]: Add g_value_take_param() (synonym to the
now deprecated g_value_set_param_take_ownership()).
* gobject.[hc]: Add g_value_take_object() (synonym to the
now deprecated g_value_set_object_take_ownership()).
* gboxed.[hc]: Add g_value_take_boxed() (synonym to the
now deprecated g_value_set_boxed_take_ownership()).
* gobject/gobject-sections.txt: Add new g_value_take_x() functions.
* gobject/tmpl/param_value_types.sgml: Document new g_value_take_x()
functions. (#100948)
2003-09-12 Noah Levitt <nlevitt@columbia.edu>
* glib/gen-unicode-tables.pl: Take a directory where to look for the
unicode files on the command line instead of 7 individual files.
Tue Sep 2 19:37:21 2003 Tim Janik <timj@gtk.org>
* gtype.[hc]: added support for a "default vtable" per interface,
that interface vtables are initialized from.
the default vtable is initialized and finalized through class_init,
class_finalize and class_data from the interfaces GTypeInfo struct.
(type_data_last_unref_Wm): unload child plugin before unreffing
parent type.
testifaceinit.c: minor fixups. fixed up base_init() assertions, since
with a default vtable, base_init() may be called multiple times.
added default initializer to iface1.
Tue Sep 2 14:53:41 2003 Tim Janik <timj@gtk.org>
* gobject-query.c (main): fix iterating over fundamental types.
* gtype.c: applied patch from owen which keeps internal
class initialization state to maintain class and interface
initialization happen in the order of:
1. class' base_init
2. interface' base_init
= interfaces added after here are immediately base_init-ialized
3. class_init
4. Interface_init
= interfaces added here are immediately Interface_init-ialized
Wed Aug 27 01:25:40 2003 Owen Taylor <otaylor@redhat.com>
* Makefile.am testifaceinit.c: Add a detailed test case
for interface initialization, testing the ability to
add interfaces during class initialization and the ordering
of interface base_init, class init, and interface_init.
(Expected to fail at the moment.)
Mon Aug 25 14:51:46 2003 Manish Singh <yosh@gimp.org>
* gtypemodule.c (g_type_module_register_type): fix typo in last
commit, cast to GTypeValueTable * to get rid of const warning.
2003-08-25 Tor Lillqvist <tml@iki.fi>
* glib/giowin32.c (read_thread): Avoid UNLOCKing the critical
section twice, which might cause a hang. (#120653)
* glib/giowin32.c (g_io_channel_unix_new): Warn if fd is both a
valid file descriptor and socket.
Mon Aug 25 14:16:48 2003 Owen Taylor <otaylor@redhat.com>
* gtypemodule.c (g_type_module_register_type): Clarify
docs on the return. Fix a memory leak if a type with a
value table is reloaded.
Mon Aug 25 12:17:20 2003 Owen Taylor <otaylor@redhat.com>
* glib/gmain.c (g_main_context_unref_and_unlock):
When DISABLE_MEM_POOLS is set, loop through and free
the poll records explicitely, since g_mem_chunk_destroy()
won't do it. (#118121, Morten Welinder)
2003-08-19 Tor Lillqvist <tml@iki.fi>
* glib/gmessages.c (g_logv): [Win32] Make the fatal error message
box easier to notice with MB_SETFOREGROUND. Also use MB_ICONERROR
to get a nice stop-sign icon.
Tue Aug 19 09:42:06 2003 Owen Taylor <otaylor@redhat.com>
* m4macros/glib-gettext.m4: Add $INTLLIBS to $LIBS
temporarily, not -lintl. (Problem with fix for
#119171, pointed out by James Henstridge)
Tue Aug 19 05:21:04 2003 Tim Janik <timj@gtk.org>
* testgobject.c (main): check private instance data after
initialization.
* gtype.c: for instances with private data, store the real class
pointer in a bsearch array during class initialization.
(g_type_instance_get_private): fetch the real class of
an instance from the bsearch array if necessary.
Tue Aug 19 03:55:29 2003 Tim Janik <timj@gtk.org>
* glib/gbsearcharray.h: license change, no restrictions are made
in the new license at all, the implmentation is simply provided
"as is" to allow copy-pasting the code into any third-party app.
naturally, the implementation is self-contained within the header
file to allow this.
default to non-shrinking behaviour, allow users to request shrinking
via G_BSEARCH_ARRAY_AUTO_SHRINK.
creation, deletion functions are now named g_bsearch_array_create()
and g_bsearch_array_free().
fixed const in prototypes, removed cruft.
(g_bsearch_array_insert): take only three arguments, do nothing if the
node to insert is already there.
(g_bsearch_array_replace): insert or replace if the node is already
there.
(g_bsearch_array_remove): remove nodes by index, the index of a node
can be found via g_bsearch_array_get_index().
removed other g_bsearch_array_remove*() variants.
(g_bsearch_array_lookup): minor optimizations.
(g_bsearch_array_lookup_sibling): return nodes on mismatches.
* glib/gbsearcharray.c: removed.
Tue Aug 19 04:08:14 2003 Tim Janik <timj@gtk.org>
* gvalue.c: adapt to new gbsearcharray.h code.
(g_value_register_transform_func): turn transform function
replacement into a valid operation.
* gsignal.c: adapt to new gbsearcharray.h code.
* gboxed.c: adapt to new gbsearcharray.h code.
2003-08-16 Tor Lillqvist <tml@iki.fi>
Fix#117925 (Dov Grobgeld):
* glib/gutils.c (g_find_program_in_path, g_basename,
g_path_get_basename, g_path_is_absolute, g_path_skip_root,
g_path_get_dirname, g_get_any_init): On Win32, look also for
slashes ('/') as pathname separators.
* glib/gfileutils.c (g_file_open_tmp): Ditto. If the template
contains a pathname separator, include the actual one in the error
message, instead of always the canonical one.
(g_build_filename): Separate implementation on Win32 that looks
for either slash or backslash. Document Unix/Windows differences.
* tests/testglib.c
* tests/strfunc-test.c: Test above functionality on Win32.
2003-08-15 Tor Lillqvist <tml@iki.fi>
* glib/gmain.c (g_poll): [Win32] Don't exceed handle array
bounds. Warn if there would be too many handles to wait
for. (WaitForMultipleObjects() has a relatively low limit of 64
handles. The Win32 IO channel code should be fixed not to need to
wait for one handle per file or socket being watched. Later.)
2003-08-13 Tor Lillqvist <tml@iki.fi>
* glib/Makefile.am
* gmodule/Makefile.am
* gobject/Makefile.am
* gthread/Makefile.am: Use srcdir also in references to the .def
files. (#118885, Jeff Bonggren)
2003-08-12 Tor Lillqvist <tml@iki.fi>
* glib/gconvert.c (g_locale_from_utf8): Drop the Win32-specific
implementation. It used WideCharToMultiByte(), which is broken as
it stores unconvertable characters as fallback characters
(question marks) in the destination string without being able to
tell how much of the conversion succeeded. Using g_convert() like
on Unix is better and simpler. (#117872)
(g_locale_to_utf8): No need for the Win32-specific implementation
here, either.
(have_broken_filenames): Define as TRUE on Win32.
(g_filename_to_utf8, g_filename_from_utf8): Drop Win32 ifdefs. As
have_broken_filenames() now is defined TRUE on Win32, works as
before.
* acinclude.m4: Copy newer versions of JH_CHECK_XML_CATALOG and JH_PATH_XML_CATALOG from
gtk-doc to enable configuring without xmlcatalog in PATH. (#119115)
2003-08-10 Tor Lillqvist <tml@iki.fi>
* glib/gutils.c (g_getenv): Don't use a cache of variable name to
value mappings on Win32, as that breaks g_setenv() and
g_unsetenv(). Only call ExpandEnvironmentStrings() if necessary,
and in that case return a quarkified string. It is still
questionable how necessary expanding embedded environment variable
references is. Possibly the whole Win32-specific g_getenv()
implementation could be removed. (#119520)
