Created attachment 127874 [details] [review] Don't test X11 autolaunching if it was disabled at compile time If dbus was configured with --disable-x11-autolaunch, then test/test-dbus-launch-x11.sh is not going to see the behaviour it was meant to. The first symptom seen is that the assertion test -n "$DBUS_SESSION_BUS_WINDOWID" fails, during the test test_exit_with_x11 --exit-with-session; but running dbus-launch --exit-with-x11 is also not going to work, and terminating the X11 display is not going to make dbus-daemon exit.
Comment on attachment 127874 [details] [review] Don't test X11 autolaunching if it was disabled at compile time Review of attachment 127874 [details] [review]: ----------------------------------------------------------------- ::: test/Makefile.am @@ +190,3 @@ > test-dbus-launch-x11.sh \ > $(NULL) > +endif DBUS_ENABLE_X11_AUTOLAUNCH Shouldn’t there be an ``` else EXTRA_DIST += \ test-dbus-launch-x11.sh \ $(NULL) ``` ?
(In reply to Philip Withnall from comment #1) > Shouldn’t there be an > ``` > else > EXTRA_DIST += \ > test-dbus-launch-x11.sh \ > $(NULL) > ``` > ? No need - Automake manages to work out that this script *could be* in a dist_ variable, which is sufficient to get it included in the dist tarball. (I have tested this, it does work.) You're thinking of glib(-tap).mk, which goes behind Automake's back to a sufficient extent that yes you do sometimes need to resort to this technique.
Comment on attachment 127874 [details] [review] Don't test X11 autolaunching if it was disabled at compile time Review of attachment 127874 [details] [review]: ----------------------------------------------------------------- r+
Fixed in git for 1.10.14, 1.11.8
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.