[Bug 76828] [next] TpClientFactory should be the top level singleton

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=76828

--- Comment #1 from Simon McVittie <simon.mcvittie at collabora.co.uk> ---
> + self = g_weak_ref_get (&singleton);

[No action required]

You get bonus penguin points for preemptively using a thread-safe GWeakRef, but
to be fully thread-aware, all reads and explicit writes to @singleton would
need to be protected by a lock. I might implement that, just for completeness.

>   * @must_have_unique_name: If set %TRUE by a subclass, the #TpProxy
> - * constructor will fail if a well-known bus name is given
> + * constructed will fail if a well-known bus name is given

[smcv will fix]

I'd prefer this to say something vague about it being a programming error.

> - bus_daemon = tp_tests_dbus_daemon_dup_or_die ();
> - tp_proxy_invalidate ((TpProxy *) bus_daemon, &invalidation_reason);
> -
> - has_props = tp_proxy_check_interface_by_id ((TpProxy *) bus_daemon,
> - TP_IFACE_QUARK_DBUS_DAEMON, &error);
> -
> - /* Borrowing the interface should fail because the proxy is invalidated. */
> - g_assert (!has_props);
> - g_assert (error != NULL);
> - g_assert_cmpuint (error->domain, ==, invalidation_reason.domain);
> - g_assert_cmpint (error->code, ==, invalidation_reason.code);
> - g_assert_cmpstr (error->message, ==, invalidation_reason.message);

[smcv will fix]

We shouldn't just delete test coverage, but this can hopefully be stuck onto an
existing test.

> --- a/telepathy-glib/introspection.am
> +++ b/telepathy-glib/introspection.am
> @@ -36,7 +36,6 @@ TelepathyGLib_1_gir_FILES = \
...
> - $(srcdir)/dbus-daemon.c $(srcdir)/dbus-daemon.h \

[smcv will fix]

This is an introspection regression, but hopefully if we move the GValue and
a{sv} utilities out of dbus.h, we can introspect it OK.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.