Created attachment 522420 [details] [diff] [review] allow disabling gconf support. Configure should not automagically add gconf support if the user does not run gnome. Gconf is only used to set the default browser which is only useful for gnome-2.x users. Attached is a patch that will permit the user to disable gconf support if they do not need/want it.
Pushed: http://hg.mozilla.org/mozilla-central/rev/038340b47ab9 I assume we don't necessary need this in 2.0 if we release 2.2 in less than 3 months now.
Created attachment 543685 [details] [diff] [review] finish disabling gconf support if requested Reopened until we get a solution or current patch approved.
Comment on attachment 543685 [details] [diff] [review] finish disabling gconf support if requested This looks good ... I hope you've tested it ;-) r=me
(In reply to comment #3) > Comment on attachment 543685 [details] [diff] [review] [review] > finish disabling gconf support if requested > > This looks good ... I hope you've tested it ;-) > > r=me I have tested it :b throughly to ensure that no instance of gconf is fired up from the browser :)
If you could add author & commit-message metadata to the patch, as described here... https://developer.mozilla.org/en/Mercurial_FAQ#How_can_I_generate_a_patch_for_somebody_else_to_check-in_for_me.3f ...that will help this get checked in faster (by making things easier for whoever ends up volunteering to check it in).
Created attachment 544685 [details] [diff] [review] Finish removing gconf support when requested Carrying over review, updated summary and author as requested by Daniel
Unclear what needs to be checked in here. Please mark obsolete patches as obsolete. I assumed I should only import the third patch, but it fails to apply on mozilla-central tip.
Landed: http://hg.mozilla.org/integration/mozilla-inbound/rev/1f8010810cee (first patch was landed already in comment 1, and second patch == third patch modulo commit headers. 3rd patch applies fine with fuzz, & I verified manually that the fuzz did the right thing.)
backed out because it was compiling only on Linux
look like the #else in nsStaticXULComponents.cpp is serving the wrong ifdef?
Yup - the build failure is about missing SYSTEMPREF_MODULES in that file, which is what that ifdef/else area guards. http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla-Inbound/1310113963.1310114729.14446.gz Jory, could you fix that & rebase the patch?
(In reply to comment #11) > Yup - the build failure is about missing SYSTEMPREF_MODULES in that file, > which is what that ifdef/else area guards. > http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla-Inbound/1310113963. > 1310114729.14446.gz > > Jory, could you fix that & rebase the patch? Yeah I will fix it, be this weekend before I can get to it tho sorry for the inconvience.
Work is invalid with latest trunk builds. If the need arises I will rework everything until that time closing.