2001-01-25 Tor Lillqvist <tml@iki.fi>
* giowin32.c: Socket support rewritten. It was utterly broken, and
untested in fact. We still do use a thread for each socket being
watched, but instead of blocking in recv() (which of course was
plain stupid for sockets being liste()ed on), we block in
select(). The read method for sockets calls recv(). It is now
possible for the application to call accept(), recv() or send() in
the callback, just like on Unix. Tested with code kindly provided
by Andrew Lanoix.
Rename g_io_channel_win32_new_stream_socket() to
g_io_channel_win32_new_socket() as it isn't restricted to stream
sockets.
* gmain.c (g_poll): Related changes in the Win32 version of
g_poll(). When polling for messages, always do a PeekMessage()
first. We used to miss messages if several were posted between
calls to g_poll().
* giochannel.h: Improve Win32-related comments.
* gutf8.c: (Win32) Include <stdio.h> for sprintf.
* tests/gio-test.c: (Win32) Add tests for polling for Windows
messages.
* tests/makefile.mingw.in: Remove superfluous compilation command
line.
Mon Jan 15 21:12:49 2001 Owen Taylor <otaylor@redhat.com>
* configure.in acconfig.h glibintl.h gutils.c
po/{Makefile.in.in,POTFILES.in,po2tbl.in}: Add gettext
support.
* glib-gettext.m4 acinclude.m4: Clean up the GTK+ gettext macros
some more and put them in this file, though they also need
to be included in acinclude.m4 due to the brokeness of
aclocal.
* gspawn.c gspawn-win32.c gutf8.c gconvert.c gfileutils.c
gshell.c: Remove dummy _() #defines, include glibintl.m4.
Fri Jan 5 11:25:42 2001 Owen Taylor <otaylor@redhat.com>
* configure.in (PACKAGE): move $enable_debug down below
checks for GCC to avoid setting CFLAGS prematurely,
change checks to avoid adding -g twice.
* gutf8.c (g_ucs4_to_utf8): Support len < 0 to mean
0 termination.
* gutf8.c (g_utf8_to_ucs4): Terminate result with 0.
* tests/mainloop-test.c (main): Fix uses of
g_main_loop_destroy().
* tests/unicode-encoding.c tests/Makefile.am tests/utf8.txt:
Tests for unicode-conversion code.
* gconvert.c (g_convert, g_convert_with_fallback): work around
a couple of GNU libc bugs.
* gconvert.[ch] (g_{locale,filename}_{to,from}_utf8): Standardize
arguments to match g_convert(). Document.
* gunicode.[ch]:
- Implement conversion functions to and from UTF-16
- Standardize unicode conversion functions on prototype like
g_convert.
- Add a lot of error checking to unicode conversion functions.
* gunicode.[ch] (g_utf8_to_ucs4_fast): Add fast, non-checking
variant of g_utf8_to_ucs4.
* gutf8.c (g_utf8_validate):
- add g_return_if_fail (str != NULL).
- add checks for overlong strings, non-valid Unicode characters (>= 110000)
and single surrogates.
2000-12-17 Tor Lillqvist <tml@iki.fi>
* gutf8.c (g_utf8_get_charset_internal): (Win32) Use GetACP to get
the current ANSI codepage.
* gunicode.h: Add comment that the static string g_get_charset
sets the parameter to point to should be copied in case the
charset might be changed later in the program.
2000-12-13 Havoc Pennington <hp@redhat.com>
* gconvert.c (open_converter): make static
* gutf8.c (g_utf8_validate): Simplify logic a bit, maybe
speeding it up - now we just return FALSE if we had to bail out
for any reason before getting to the end of the string, as defined
by a nul byte if len was -1, defined by the len otherwise. This
also fixes a bug where nul bytes were not treated as invalid
when the length was specified.
Sun Jul 30 16:54:13 2000 Owen Taylor <otaylor@redhat.com>
* gunicode.h: Fix stray character
* gutf8.c (g_unichar_to_utf8): Allow outbuf to be NULL, in
which case we just compute the length.
Wed Jul 26 12:59:31 2000 Tim Janik <timj@gtk.org>
* *.[hc]: applied patch from Andreas Persenius <ndap@swipnet.se> that
updates the license headers to the GNU Lesser General Public License,
as well as updating the copyright year to 2000.
Wed Jun 21 12:09:03 2000 Owen Taylor <otaylor@redhat.com>
* gunicode.h gutf8.c guniprop.c gunidecomp.[ch] gunichartables.h
Makefile.am glib.h: Initial pass at adding unicode support
functions. A few things still need to be implemented, a bit
of cleanup needs to be done, tests need to be added, and
the docs need to be finished, but this should allow replacing
most or all use of libunicode.
