6 Commits

Author	SHA1	Message	Date
Matthias Clasen	03a48e1ade	Fix the closure test in continuous ... This test has the same problem as the mapping-test - it uses SIGUSR1 without checking the signal mask. Apply the same fix here.	2014-05-30 10:22:35 -04:00
Colin Walters	d9e01e0c37	tests/closure: Send SIGUSR1 only to our pid ... Otherwise in e.g. the gnome-ostree integrationtest system, we end up sending SIGUSR1 to the *entire session*, which triggers various badness in untested debugging paths from gnome-session.	2013-07-22 15:14:34 +01:00
Colin Walters	00f5c63621	tests/closure: Use correct prototype for signal callback ... This test worked when compiled without optimization, but fails with -O2. Presumably we just happened to find the GMainLoop off the stack somewhere. https://bugzilla.gnome.org/show_bug.cgi?id=704267	2013-07-22 09:34:24 -04:00
Dan Winship	88ab63155d	gobject/tests/closure: fix on win32 ... (The g_closure_unref() was wrong, but was not causing errors on linux for some reason.) https://bugzilla.gnome.org/show_bug.cgi?id=704447	2013-07-21 20:50:11 +01:00
Colin Walters	35e331b869	tests/closure: Fix invalid unref ... The closure is floating, g_source_set_closure() will claim the ref, so we do not need to unref it.	2013-07-21 19:42:34 +01:00
Dan Winship	72a7e824d6	gsourceclosure: fix idle/timeout/signal closures, add child watch support ... And add a test for all source types. https://bugzilla.gnome.org/show_bug.cgi?id=704267	2013-07-17 21:41:45 -04:00