Commit Graph

4569 Commits

Author SHA1 Message Date
Руслан Ижбулатов
604ca89176 registrybackend: do convert values to UTF-16
Perform conversion before writing a value out of the cache into the registry,
and convert back when reading a value into the cache out of the registry.
The registry holds UTF-8 strings.
2016-02-05 09:40:30 +01:00
Руслан Ижбулатов
c4d943186e registrybackend: fix memory leak 2016-02-05 09:39:21 +01:00
Ignacio Casal Quinteiro
b8fc289e8f registrybackend: remove useless include 2016-02-04 12:44:11 +01:00
Ignacio Casal Quinteiro
74442a0b8c registrybackend: do not leak self if there are no items 2016-02-04 11:16:45 +01:00
Ignacio Casal Quinteiro
dc97bb9b9b registrybackend: handle readability of the keys
If a key is removed or it cannot be read anymore we should
notify the backend about it so it fallbacks to the default
value.
2016-02-04 11:02:46 +01:00
Ignacio Casal Quinteiro
05dd91a0b6 registrybackend: use unicode calls intead of the ansi ones
https://bugzilla.gnome.org/show_bug.cgi?id=761504
2016-02-04 08:59:59 +01:00
Ignacio Casal Quinteiro
7161d70955 registrybackend: remove useless get_permission override
The base class does the same
2016-02-04 07:55:25 +01:00
Ignacio Casal Quinteiro
259a61ed2d registrybackend: pass the event to the cache update
This way the registry cache has more control to specify what has
changed.
2016-02-03 19:19:50 +01:00
Rico Tzschichholz
e3189527dc gio/gobject: Various introspection fixes 2016-02-03 18:13:49 +01:00
Ignacio Casal Quinteiro
c7ea434e2c registrybackend: properly propagate the partial key name 2016-02-03 13:10:09 +01:00
Ignacio Casal Quinteiro
f1a5e394b0 registrybackend: rename touched flag to readable
This is a flag used to understand if a key exists on the registry
and if it is readable. It makes more sense to rename it as readable
since anyway a key that does not exists anymore is a key that is
not readable.
2016-02-02 14:48:22 +01:00
Ignacio Casal Quinteiro
5cc997f7c4 registrybackend: close the key only if successfully opened 2016-02-02 11:08:50 +01:00
Ignacio Casal Quinteiro
bf3f827ca1 registrybackend: use ptr_array_new_with_free_func 2016-02-02 11:08:42 +01:00
Ignacio Casal Quinteiro
9098a7f927 registrybackend: minor style cleanup 2016-02-02 11:04:56 +01:00
Rico Tzschichholz
d268d9f86a socket: Fix annotation of g_socket_receive_message
https://bugzilla.gnome.org/show_bug.cgi?id=761337
2016-02-02 10:15:04 +01:00
Rico Tzschichholz
ec173eb654 application: Fix annoation of g_application_add_option_group
https://bugzilla.gnome.org/show_bug.cgi?id=761337
2016-02-02 10:11:08 +01:00
Ignacio Casal Quinteiro
169cfb250f registrybackend: avoid signed/unsigned comparison warnings 2016-02-01 14:25:05 +01:00
Ignacio Casal Quinteiro
b0776ddd18 registrybackend: use G_DECLARE_FINAL_TYPE 2016-02-01 14:24:46 +01:00
Sébastien Wilmet
95dd373024 docs: better documentation for g_file_info_copy_into()
The documentation of g_file_info_copy_into() was misleading. The
attributes are not just copied, @dest_info is also cleared at the
beginning. So any previously set attributes in @dest_info are lost.

There was a bug in gedit about this function, where some metadata were
not saved. So it might make sense to change the implementation to not
clear @dest_info, and copy one by one the attributes from @src_info to
@dest_info.

https://bugzilla.gnome.org/show_bug.cgi?id=747927
2016-01-28 20:31:53 +01:00
Sébastien Wilmet
a4ed89bf75 docs: improve doc of g_file_info_list_attributes()
The name_space can be NULL.

https://bugzilla.gnome.org/show_bug.cgi?id=747927
2016-01-28 18:49:28 +01:00
Matthias Clasen
42699e37be gio: Include filename in error message
I'm tired of seeing 'No such file or directory' in the logs without
a hint as to what is actually wrong. Including the filename here
may help me tracking down a bug in the continuous infrastructure.
2016-01-27 21:05:31 -05:00
Chun-wei Fan
d20e88fd11 gwinhttpfile.c: Fix build on Visual Studio
Visual Studio, at least the older versions, cannot use L on macros which
are defined as a constant string, plus the L must be applied to all string
literals here.  This does not look nice, but this is life...
2016-01-26 23:23:42 +08:00
Ignacio Casal Quinteiro
56b0454ba5 registrybackend: fix warning 2016-01-26 15:41:08 +01:00
Ignacio Casal Quinteiro
b57eac68e5 registrybackend: fix double-free error 2016-01-26 15:22:48 +01:00
Ignacio Casal Quinteiro
df1ffe7e27 registrybackend: remove :( from messages 2016-01-26 14:35:31 +01:00
Ignacio Casal Quinteiro
0200e4036c registrybackend: simplify g_message_win32_error 2016-01-26 13:26:41 +01:00
Ignacio Casal Quinteiro
bc85dee6b3 registrybackend: do not accept 0 as a windows error
We might end up removing from the error stack the wrong error
and this might be missleading
2016-01-26 13:19:56 +01:00
Ignacio Casal Quinteiro
bad7e4a114 registrybackend: do not leak the watch data in case of failure 2016-01-26 13:15:51 +01:00
Ignacio Casal Quinteiro
31aab1bd42 registrybackend: do not leak key and event if it cannot add the watch 2016-01-26 12:53:56 +01:00
Ignacio Casal Quinteiro
a92d97ff1f registrybackend: more style fixes
This is the never ending story
2016-01-26 12:21:48 +01:00
Ignacio Casal Quinteiro
1b5b0eff9a registrybackend: fix possible crash if cache_node is NULL 2016-01-26 12:02:18 +01:00
Ignacio Casal Quinteiro
67f6ede3e5 registrybackend: remove space before ++ 2016-01-26 11:46:51 +01:00
Ignacio Casal Quinteiro
8f7aa273de registrybackend: fix possible mem leak
If the parameters do not validate we would leak the memory.
2016-01-26 11:42:33 +01:00
Ignacio Casal Quinteiro
305a9b12c9 winhttpfile: use glib format string macro
Like this we avoid some downstream patching for msys2
2016-01-26 10:13:02 +01:00
Ignacio Casal Quinteiro
a89629db1d registrybackend: use the glib format string macro 2016-01-26 10:10:39 +01:00
Ignacio Casal Quinteiro
a159bc939d registrybackend: more cleanups 2016-01-26 09:00:35 +01:00
Ignacio Casal Quinteiro
7256f2289a registrybackend: more cleanups 2016-01-25 16:12:59 +01:00
Ignacio Casal Quinteiro
a5e819c4c3 registrybackend: style fixes 2016-01-25 16:02:03 +01:00
Philip Withnall
8c263008bb build: Calculate ABS_GLIB_RUNTIME_LIBDIR at build time
Rather than calculating it at configure time. This means it can expand
$libdir properly, and use the Make $(realpath) function rather than
invoking the non-portable `readlink -f`.

This fixes problems where `readlink` would be called on an invalid path
(due to a variable not being expanded) and would evaluate to "", which
would then cause things to be installed in the wrong place.

https://bugzilla.gnome.org/show_bug.cgi?id=744772
2016-01-24 14:44:44 +00:00
Philip Withnall
3c0cddfe80 gdbusobjectmanagerserver: Clarify recommended ObjectManager paths
Otherwise people might try to export the object manager at ‘/’, which
doesn’t work. And I have no intention of making it work.

https://bugzilla.gnome.org/show_bug.cgi?id=760852
2016-01-19 17:10:40 +00:00
Philip Withnall
1725580f26 gdbusobjectmanagerserver: Convert a DocBook link to Markdown 2016-01-19 17:01:14 +00:00
Philip Withnall
c3d6934f18 gio: Add DTLS interfaces
Add a new GDtlsConnection interface, plus derived GDtlsClientConnection
and GDtlsServerConnection interfaces, for implementing Datagram TLS
support in glib-networking.

A GDtlsConnection is a GDatagramBased, so may be used as a normal
datagram socket, wrapping all datagrams from a base GDatagramBased in
DTLS segments.

Test cases are included in the implementation in glib-networking.

https://bugzilla.gnome.org/show_bug.cgi?id=752240
2016-01-18 14:25:06 +00:00
Allison Ryan Lortie
86c5d8978d GDBusMethodInvocation: document behaviour change
We changed the behaviour of this API to adapt to a change in the D-Bus
specification.  Document the new behaviour, along with the time of the
change.

https://bugzilla.gnome.org/show_bug.cgi?id=755421
2016-01-13 10:51:44 -05:00
Lars Uebernickel
dbea81b02d gdbus: don't send unexpected replies
gdbus sets NO_REPLY_EXPECTED when no callback is given to
g_dbus_connection_call(). It makes sense that it also handles the server
side correctly by discarding replies to clients that don't want one.

https://bugzilla.gnome.org/show_bug.cgi?id=755421
2016-01-13 10:44:49 -05:00
Philip Withnall
3add5e2837 gio: Document thread safety of the streams API
Specifically, GIOStream and the TLS connection streams.

Includes wording adapted from suggestions by Dan Winship
<danw@gnome.org>.

https://bugzilla.gnome.org/show_bug.cgi?id=735754
2016-01-11 15:58:42 +00:00
Javier Jardón
4e78a0a9df Revert "Use upstream gettext instead the glib one"
This causes several problems:
- Compilation in FreeBSD with --enable-gtk-doc broke
- Modules that still use the AM_GLIB_GNU_GETTEXT macro
  doesnt compile anymore because /usr/share/glib-2.0/gettext
  is not filled with the correct files, as this was done in
  the glib custom po/Makefile.in.in

See https://bugzilla.gnome.org/show_bug.cgi?id=622991

This reverts commit e5c752371c.
2016-01-10 22:44:24 +00:00
Javier Jardón
e5c752371c Use upstream gettext instead the glib one
https://bugzilla.gnome.org/show_bug.cgi?id=622991
2016-01-09 18:49:22 +00:00
Matt Watson
c1e2a8d727 resource file: add cancel to dummy monitor
gfilemonitor has a cancel vfunc and will call into the in dispose.
If we don't stub it out we get a segfault.
2016-01-08 18:08:20 -08:00
Daiki Ueno
50645b724a gsettings: Install gettext ITS rules
Recent gettext has a feature to allow consumer projects to supply their
own string extraction rules for XML files, in ITS format.

Gettext still ships the rule for *.gschema.xml, but it would be better
maintained in the upstream project.

See the gettext documentation for details:
http://www.gnu.org/software/gettext/manual/html_node/Preparing-ITS-Rules.html

https://bugzilla.gnome.org/show_bug.cgi?id=760199
2016-01-08 12:31:18 +09:00
Christophe Fergeau
ea5ca11761 gio: Document that g_inet_address_new_* return value must be g_object_unref'ed 2016-01-07 15:21:08 +01:00