gtk_test_init() ends up calling gtk_init() which tries opening
the default display. CI jobs sometimes seem to trip over this
even when run through xvfb-run. Therefore we use g_test_init()
instead.
We add a small hardcoded table for now. Future versions
will improve this to parse this out of ASOPs eccdata.
If we can't determine the type of a number we add it verbatim.
It was introduced in 4b51f340c8
to prevent accessing user home during tests:
The only place where this the home is currently accessed is when loading
SIP accounts and these can already be overriden with
CALLS_SIP_ACCOUNT_FILE.
It seems that XDG_CONFIG_HOME is propagated to the flatpak runner,
even though it was only overriden for the test suite.
========================================================================
Building module calls in /builds/devrtz/calls/.flatpak-builder/build/calls-1
========================================================================
Error: module calls: Error opening directory '/builds/devrtz/calls/_build/calls': Permission denied
Reported at https://gitlab.gnome.org/Infrastructure/GitLab/-/issues/596#note_1654082
If folks is built with the 'eds' backend it will get used by default.
During testing the system address book should not be queried.
It might even be impossible if the environment is not set up to
connect to the session bus raising the following warning:
(process:701522): folks-WARNING **: 09:13:54.291: Failed to find primary PersonaStore with type ID 'eds' and ID 'system-address-book'.
Individuals will not be linked properly and creating new links between Personas will not work.
The configured primary PersonaStore's backend may not be installed. If you are unsure, check with your distribution.
Fixes: #427
Using rescan allows use to give a priority to search paths. So this way
plugins in `CALLS_PLUGIN_DIR` take precedence over plugins we ship.
This also makes sure that the plugin test searches in the same location
as CallsManager.
Previously our code assumed that g_cancellable_cancel() the async DBus
calls to libfeedback would guarantee that the underlying operation would
not be performed (i.e. triggering or ending a feedback).
However the endless ringing exhibited in #470 shows this assumption not
to hold. Therefore we avoid using g_cancellable_cancel () completely and
default to waiting for the async operation to finish.
update_ring () now sets the target state by inspecting managed calls and
the main logic will now step towards the target state:
Changing from regular/loud to soft/quiet ringing (or vice versa)
requires we first end feedback before (re)triggering it.
Additionally the "is-quiet" and "is-ringing" properties are replaced by
a new "state" property to allow changing the combination atomically.
Closes: #470
Copied from libfeedback as of 2e081602f627505e566cc0bdb6cd96c7642d8b65
and adjusted for our mocking needs.
The mocked library will be LD_PRELOADED for the moment,
but further changes to the build should allow us to simply link
to it in the future.
This will get removed soon while refactoring the ringer as cancelling
the async methods would not guarantee that the requested operation would
not be performed.
It could actually already be started (e.g. starting or stopping the
feedback), but we have not yet received the reply, so it was always a
bit flaky.
Since we will introduce another type of plugin for the policy engine
we want to have each plugin type in separate directories.
We also have to adjust:
- plugin search directories
- po file location
- update paths for calls-doc target
Bail out! CallsSipMediaPipeline-FATAL-WARNING: Error trying to setup codecs for pipeline: Could not create 'decoder' element of type avdec_g722
stderr:
(gst-plugin-scanner:196349): GLib-GObject-WARNING **: 07:29:24.149: type name '-a-png-encoder-pred' contains invalid characters
See
https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1055
