Matthias Clasen
00151ec0e1
Fix a locking problem in g_main_context_iterate()
...
We failed to ensure that the context is locked on every exit of
the function. This fixes bug 583324.
2009-05-27 12:04:35 -04:00
Hans Breuer
209a662c2f
Correct reference and implement close_fn ( #578769 )
...
Do proper referencing and unreferencing of
GWinHttpFileInputStream::file and
GWinHttpFileInputStream::file::vfs. Implement
GWinHttpFileInputStream::close_fn.
2009-05-27 15:12:31 +03:00
Alexander Larsson
623f99dc3d
Add required defines on solaris for GSocket to build ( #582856 )
...
Turns out that the msg_control and related things are not defined in the
system headers unless you define some _XOPEN_SOURCE things.
2009-05-27 11:11:37 +02:00
Alexander Larsson
f99be75ff9
Remove mention of non-existing calls from docs
...
The docs mentions a separate seekable API for the various file streams
which don't actually exists. Change this to refer to the generic
GSeekable calls.
2009-05-25 15:46:35 +02:00
Alexander Larsson
3c0feca7f1
Don't end enum with comma ( #583663 )
...
This is valid C but breaks C++, so don't put it in headers.
2009-05-25 12:47:12 +02:00
Alexander Larsson
c93c98732a
Don't return something from a void function ( #583408 )
2009-05-25 12:46:03 +02:00
Alexander Larsson
ce171195d9
Add g_inet_address_get_native_size ( #583205 )
...
This can be used to get the size of g_inet_address_to_bytes().
2009-05-25 11:33:56 +02:00
Miquel Esplà
fb70c6420b
Added Valencian-Catalan translation
2009-05-25 00:24:08 +02:00
Jorge Gonzalez
bde3e9c354
Updated Spanish translation
2009-05-23 20:00:22 +02:00
Christian Persch
37bd225d7b
Fix grammar in error message
...
Bug #583198 .
2009-05-23 18:00:06 +02:00
Christian Persch
76ebe6c069
Plug a mem leak
...
Bug #583196 .
2009-05-23 17:59:54 +02:00
Christian Persch
fc9a888b8c
Add G_G[U]INTPTR_FORMAT and G_GINTPTR_MODIFIER
...
Add macros for printf formatting for g[u]intptr. Bug 569376.
2009-05-23 17:58:10 +02:00
Christian Persch
d046bfe577
Document g[u]intptr
...
Bug 569375.
2009-05-23 17:56:10 +02:00
Christian Persch
54b43229f4
Use g_error_new_valist instead of a private copy of it
2009-05-23 17:53:01 +02:00
Yeti
01b79d67ef
Make g_error_new_valist public
...
Bug #569024 .
2009-05-23 17:52:58 +02:00
Alexander Larsson
14d53dcfa8
Fix connect on win32
...
We should not wait before calling connect for non-blocking connect.
Also, use the right error code for the nonblocking pending case on win32.
2009-05-20 16:40:55 +02:00
Alexander Larsson
bb66fdcce3
Fix build on win32
2009-05-20 16:40:54 +02:00
Alexander Larsson
c897cca601
Don't add unix specific APIs on win32
2009-05-20 16:40:54 +02:00
Alexander Larsson
5a46e4d140
Clarify g_cancellable_push_current docs wrt cancellable being NULL ( #575013 )
2009-05-20 14:55:14 +02:00
Alexander Larsson
ab29e09dac
Don't dereference identifier_type if it is NULL ( #579558 )
2009-05-20 14:46:51 +02:00
Robert Bragg
d0d10e847f
Take a reference on the gio file monitors while signaling pending file changes
...
It was possible for a signal handler to remove the last reference and
dispose the monitor. If there were remaining pending_file_changes they
tried to dereference the disposed monitor.
This patch simply calls g_object_{ref,unref} around the loop that signals
the changes.
2009-05-20 14:23:53 +02:00
Alexander Larsson
6a9df8256a
Use G_GSSIZE_FORMAT where needed ( #577884 )
2009-05-20 14:17:27 +02:00
Sjoerd Simons
7498049a16
Propagate errors from g_network_service_address_enumerator_next_async in its _finish function
2009-05-20 13:59:50 +02:00
Jonathon Jongsma
2be66c0458
Typedef GResolverClass in the standard way
...
It's a bit lame, but some of our C++ wrapping scripts expect objects to be
typedefed like:
typedef struct _FooClass FooClass;
struct {} _FooClass;
Rather than:
typedef struct {} FooClass;
Functionally they're the same, but the former makes our lives easier in the
short term
2009-05-20 13:58:13 +02:00
Alexander Larsson
cb7a300e31
Don't return something from void function ( #583229 )
...
g_async_initable_init_async is void, don't return something from it.
2009-05-20 13:49:47 +02:00
Christian Persch
1cbdd2495d
Preserve errno when using g_set_error with _() and g_strerror()
2009-05-20 13:47:55 +02:00
Christian Persch
80cfd099f3
Use g_set_error_literal
...
Bug #583206 .
2009-05-20 13:46:39 +02:00
Alexander Larsson
c20b8d4d53
Check that close_fn is not %NULL before calling ( #578499 )
...
Some streams have no close function, so this caused a crash.
2009-05-20 13:41:25 +02:00
Stefan Kost
0a280dadec
docs: fix typo
2009-05-20 14:29:49 +03:00
Sjoerd Simons
9033b37589
Add helper functions for connecting to service ( #583061 )
2009-05-20 12:41:50 +02:00
Dan Winship
25800ed4a3
Ignore SIGPIPE when using GSocket
...
http://bugzilla.gnome.org/show_bug.cgi?id=583001
2009-05-20 12:32:58 +02:00
Alexander Larsson
2ea22d8f46
Remove non-existing parameter from docs
...
g_unix_socket_address_abstract_names_supported has no address argument
2009-05-20 12:14:55 +02:00
Alexander Larsson
9d36c86f8a
Add missing symbols to gio.symbols
2009-05-20 12:14:55 +02:00
Alexander Larsson
9346f461f3
Add missing types to gio.types
2009-05-20 12:14:55 +02:00
Alexander Larsson
5cd86fbda6
Remove protocol names, instead use an enum with common protocols
...
The whole protocol name thing is pretty weird. The getprotobyname functions
seem to only specify one mapping for name <-> ids, so all families/types
must use the same values. Plus the values used for the protocols are
standardized by IANA, so are always the same.
So, we drop using names for protocols, intead introducing an enum with
a few commonly availible and used protocols.
2009-05-20 12:14:50 +02:00
Alexander Larsson
6d01593b06
Clarify "current condition" in g_socket_create_source docs
2009-05-20 11:30:43 +02:00
Alexander Larsson
8088182644
Actually implement max_threads for GThreadedSocketService
2009-05-20 11:28:27 +02:00
Alexander Larsson
053f9e72b1
Add support for graceful disconnect to GTcpConnection
2009-05-20 11:19:47 +02:00
Alexander Larsson
f061765e54
Call sync close function directly in async implementation
...
The g_io_stream_wrapper fails since there is already an outstanding
operation (the async close).
2009-05-20 11:18:34 +02:00
Alexander Larsson
fdfdec36d0
Add send-data, a g_socket_client test case
2009-05-19 13:44:11 +02:00
Alexander Larsson
18373cfbe8
Set the listen backlog before calling listen.
...
If we set it after it won't be used.
2009-05-19 12:06:29 +02:00
Alexander Larsson
64383fa3d9
Fix g_unix_socket_address_abstract_names_supported docs
...
The previous description was a cut-n-paste from another function.
2009-05-19 11:53:42 +02:00
Alexander Larsson
0ffe722193
Add g_socket_shutdown
2009-05-19 11:52:33 +02:00
Alexander Larsson
03441e724a
Document GOutputVector and GInputVector
2009-05-19 11:27:07 +02:00
Alexander Larsson
e1a4389cbc
Rename g_socket_check_pending_error to g_socket_check_connect_result
...
This is only used for connect anyway, and this describes the operation
better.
2009-05-19 10:57:58 +02:00
Alexander Larsson
bcba61c951
Update docs on listen backlog
...
Mention g_socket_set_listen_backlog in g_socket_listen.
Explain that listen backlock needs to be set before calling
listen. Also verify this with a g_return_if_fail.
2009-05-19 10:47:55 +02:00
Alexander Larsson
69130db81a
Read socket state in g_socket_get_local/remote_address
...
Previously we saved the location in various places which is unnecessary
and sometimes even wrong. For instance, we saved the address we bound to
which may not have the final port set.
2009-05-19 10:40:27 +02:00
Alexander Larsson
f8cd1c5304
Clean up refereces to @protocol_id in g_socket_new docs
...
This was not fully updated from the protocol to protocol_id change.
2009-05-19 10:16:32 +02:00
Alexander Larsson
3756ddb0ee
Ignore error when setting SO_REUSEADDR
...
The main error would be "not supported" which could happen for e.g.
unix domain sockets, we don't really care, as this is mainly something
for TCP to help out a bit.
2009-05-18 23:24:13 +02:00
Alexander Larsson
f24c7fa9cb
Add support for abstract unix socket addresses
2009-05-18 21:31:28 +02:00
