| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | | |
|
|/ / |
|
|\ \
| | |
| | |
| | |
| | | |
Declare libdconf_service as a dependency
See merge request GNOME/dconf!22
|
|/ /
| |
| |
| | |
This reverts commit b799b08ec4e5978ab8db233a9da1532bf7918324.
|
| |
| |
| |
| | |
This reverts commit bf103302332603a637f1c33e603c93b82aa1dc0b.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The dconf-service internal static library depends on generated files,
and we need to make sure that those files are available by the time we
build the tests that depend on them, as Meson builds everything in
parallel. If we don't do that, in cases of massive parallelised
builders, we will end up with errors like:
```
ccache cc -Itests/tests@@writer@exe -Itests -I../tests -I. -I../ -Itests/../service -I../tests/../service -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -O0 -g '-DSYSCONFDIR="/etc"' -fno-common -Wmissing-prototypes -Wwrite-strings -O2 -g -m64 -mtune=generic '-DSRCDIR="/ostbuild/source/dconf/tests"' -MD -MQ 'tests/tests@@writer@exe/writer.c.o' -MF 'tests/tests@@writer@exe/writer.c.o.d' -o 'tests/tests@@writer@exe/writer.c.o' -c ../tests/writer.c
../tests/writer.c:24:37: fatal error: service/dconf-generated.h: No such file or directory
```
Reviewed-by: nobody
Signed-off-by: Emmanuele Bassi <ebassi@gnome.org>
|
|\ \
| | |
| | |
| | |
| | | |
service: Allow opening corrupt GVDB files when writing
See merge request GNOME/dconf!8
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
It’s a little counterintuitive that a function called ‘read’ could
sometimes move a file (which could be considered a write).
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Otherwise the tests fail now that we’re compiling with
G_LOG_USE_STRUCTURED.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Use g_test_subprocess() and g_test_trap_subprocess() instead. This
requires splitting up some of the unit tests, since subprocesses are
handled at a per-test level, and you can’t trivially fork multiple
different subprocesses from a single test (like you could with fork()).
While fork was safe to use on POSIX (and dconf only targets POSIX), the
fact that we had to redefine GLIB_VERSION_MIN_REQUIRED to hide the
deprecation errors was not doing wonders for the maintainability of the
tests.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This enables structured logging for all parts of dconf, which will make
it easier to check for certain log messages in the unit tests.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | |
| | |
| | | |
This will make unit testing the code easier.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This doesn’t bump our GLib dependency, since we already depend on GLib
2.44.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If a GVDB file cannot be opened due to being corrupt, move it out of the
way, warn, and open a new blank database instead.
This prevents the situation where a corrupt database stops the entire
desktop session from loading.
Note that the dconf_gvdb_utils_read_file() code path is only taken
inside DConfWriter. The DConf engine sources (such as
dconf-engine-source-system.c) open the GVDB tables separately, and
already all handle errors gracefully.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
https://gitlab.gnome.org/GNOME/glib/issues/1454
|
|\ \
| | |
| | |
| | |
| | | |
Replace all hard-coded /etc path with sysconfdir
See merge request GNOME/dconf!21
|
|/ /
| |
| |
| |
| |
| |
| | |
This is useful for JHBuild environments and systems that don't want
to use /etc/dconf.
https://bugzilla.gnome.org/show_bug.cgi?id=739299
|
|\ \
| | |
| | |
| | |
| | | |
Engine: Change overflow thresholds in subscription counts from GMAXUINT32 to GMAXUINT
See merge request GNOME/dconf!20
|
|/ /
| |
| |
| | |
GMAXUINT
|
|\ \
| | |
| | |
| | |
| | | |
Revert "Merge branch 'fix/build-gtk-doc' into 'master'"
See merge request GNOME/dconf!19
|
|/ /
| |
| | |
This reverts merge request !18
|
| | |
|
|\ \
| | |
| | |
| | |
| | | |
Prepare 0.29.1
See merge request GNOME/dconf!4
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Engine: account for multiple simultaneous subscriptions to the same path
Closes #2
See merge request GNOME/dconf!5
|
| | | |
| | | |
| | | |
| | | | |
Use g_assert_false instead of g_assert in unit tests
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
state transition is atomic
Update comment about threading, documenting the new lock
Add documentation comments for new utility functions
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
subscriptions to the same path
Remove accidental whitespace change
Simplify branching in watch_fast and unwatch_fast
Indentation fixes
Store the subscription counts directly in the hash table pointer instead of mallocing ints
Add documentation comments for new utility functions
|
| | | | |
|
|\ \ \ \
| |/ / /
|/| | |
| | | |
| | | | |
Build: revert to the old "enable-gtk-doc" flag since there is no reason to change it
See merge request GNOME/dconf!18
|
|/ / /
| | |
| | |
| | | |
change it
|
|\ \ \
| |/ /
|/| |
| | |
| | | |
Update GVDB from gvdb/master
See merge request GNOME/dconf!17
|
| | |
| | |
| | |
| | |
| | | |
Signed-off-by: Philip Withnall <withnall@endlessm.com>
Reviewed-by: nobody
|
| |\ \
|/ / /
| | _ |
|
| |
| |
| |
| | |
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| |
| |
| |
| |
| |
| | |
This is based on the commit 59a24ab5a3 in GLib.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| |
| |
| |
| |
| |
| | |
(This is based on commit 892fc2e4 from dconf.)
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
| |
| |
| |
| | |
https://bugzilla.gnome.org/show_bug.cgi?id=776504
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
glibc string.h declares memcpy() with attribute(nonnull(1,2)), causing
calls with NULL arguments to be treated as undefined behaviour.
This is consistent with ISO C99 and C11, which state that passing 0
to string functions as an array length does not remove the requirement
that the pointer to the array is a valid pointer.
gcc -fsanitize=undefined catches this while running OSTree's test suite.
Similarly, running the GLib test suite reports similar issues for
qsort(), memmove(), memcmp().
(This is a partial cherry-pick of commit e5ed410c8c0fe823883 from GLib.)
Signed-off-by: Simon McVittie <smcv@debian.org>
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=775510
Reviewed-by: Colin Walters
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
GVDB is essentially part of GLib, so should have the same maintainer
list.
At least this way, it’s not just maintained by one absentee maintainer.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
|
|\ \
| | |
| | |
| | |
| | | |
dconf_gdbus_get_worker_context(): improve GObject deadlock workaround
See merge request GNOME/dconf!15
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The existing workaround for
https://bugzilla.gnome.org/show_bug.cgi?id=674885
doesn't go far enough, and deadlocks can occur, for example, with the
GSocket type. Extend the workaround logic to all the types from
glib/gio/gdbusprivate.c:ensure_required_types().
|
| | |
| | |
| | |
| | | |
that the coverage can be reported
|
|\ \ \
| |/ /
|/| |
| | |
| | | |
Correct the meson option to build gtk docs, also enable man pages and gtk docs…
See merge request GNOME/dconf!16
|
|/ /
| |
| |
| | |
docs in the build and test stage to catch errors earlier
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The GSettings backend loadable module should not expose any symbol
except for the g_io_module_* ones. By using the internal dependency
objects, we end up using `link_whole` on libdconf-common and
libdconf-gdbus-thread, which means that we now inherit all the symbols
exposed by those static libraries.
We still want to use `link_whole` every time we depend on those static
libraries, but for the GSettings backend module we make an exception,
and use `link_with` with the build targets, instead of the internal
dependencies.
This fixes the `abicheck` test unit.
Reviewed-by: nobody
Signed-off-by: Emmanuele Bassi <ebassi@gnome.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The build currently, unintentionally, depends on side effects, namely:
that a variable defined in a meson.build file will be valid for any
other included meson.build file, until it gets re-defined.
We use the same `deps` variable in many places, and we end up depending
on the inclusion order when we get to the client library, which depends
on the gsettings backend defining the `deps` variable.
Reviewed-by: nobody
Signed-off-by: Emmanuele Bassi <ebassi@gnome.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Some of the internal static libraries in dconf implement public symbols.
This means that they must be linked using `link_whole`, to prevent the
linker from hiding unused symbols in there when building the shared
libdconf.so.
This fixes the build of projects using the dconf API directly, like
dconf-editor, in GNOME Continuous.
Reviewed-by: nobody
Signed-off-by: Emmanuele Bassi <ebassi@gnome.org>
|
|\ \
| | |
| | |
| | |
| | | |
Various meson related improvements
See merge request GNOME/dconf!11
|