[GTK] Shadow builds are not working anymore.
Created attachment 167769 [details] Patch
The build miserably fail with a python error and even if I fix it, despite I built with jhbuild it tries to config check against system libraries.
Created attachment 167770 [details] Patch
(In reply to comment #3) > Created an attachment (id=167770) [details] > Patch Also one nice thing to do to avoid the shadow build to break regularly is to setup one of the 4 GTK bots as a shadow build :). I think we have one Qt bot setup with shadow build. It may be not a very useful feature when you only hack GTK but it is very useful for people working cross port features where you don't want the source dir to be modified. Anyway feedback on the patch. I'm not very sure of myself, my knowledge of the GTK buildsystem is very limited :)
Created attachment 167771 [details] Patch
Comment on attachment 167770 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=167770&action=review > Tools/gtk/run-with-jhbuild:26 > +wrapper = common.top_level_path('Tools', 'jhbuild', 'jhbuild-wrapper') > +os.execve(wrapper, [wrapper, '--gtk', 'run'] + sys.argv[1:], os.environ) It seems like this will break things for people who don't use the WebKit jhbuild at all.
(In reply to comment #6) > (From update of attachment 167770 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=167770&action=review > > > Tools/gtk/run-with-jhbuild:26 > > +wrapper = common.top_level_path('Tools', 'jhbuild', 'jhbuild-wrapper') > > +os.execve(wrapper, [wrapper, '--gtk', 'run'] + sys.argv[1:], os.environ) > > It seems like this will break things for people who don't use the WebKit jhbuild at all. Obvious but is it the recommended way to not use jhbuild. From the wiki it seems you have to use it. Other thing is that the trunk depends on really really recent version of libraries that for example Archlinux doesn't even provide (and Arch is know to be pretty up to date) so I wonder if the build without jhbuild is actually possible without complicated PKG_CONFIG trickery. Other thing I could do is modify the condition "os.path.exists(common.top_level_path('WebKitBuild', 'Dependencies')):" to also read WEBKITOUTPUTDIR.
(In reply to comment #7) > Obvious but is it the recommended way to not use jhbuild. From the wiki it seems you have to use it. Other thing is that the trunk depends on really really recent version of libraries that for example Archlinux doesn't even provide (and Arch is know to be pretty up to date) so I wonder if the build without jhbuild is actually possible without complicated PKG_CONFIG trickery. There are several very active WebKitGTK+ developers that do not use the internal jhbuild. This isn't likely to change any time soon. > Other thing I could do is modify the condition "os.path.exists(common.top_level_path('WebKitBuild', 'Dependencies')):" to also read WEBKITOUTPUTDIR. It looks like what we want here is to use get_dependencies_path from Tools/jhbuild/jhbuildutils.py.
Created attachment 167783 [details] Patch
Comment on attachment 167783 [details] Patch Thanks!
Comment on attachment 167783 [details] Patch Rejecting attachment 167783 [details] from commit-queue. Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 2 Last 500 characters of output: '--force', u'--reviewer', u'Martin Rob..." exit_code: 1 cwd: /mnt/git/webkit-commit-queue Parsed 2 diffs from patch file(s). patching file Tools/ChangeLog Hunk #1 succeeded at 1 with fuzz 3. patching file Tools/gtk/run-with-jhbuild Hunk #1 FAILED at 22. 1 out of 1 hunk FAILED -- saving rejects to file Tools/gtk/run-with-jhbuild.rej Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', u'--force', u'--reviewer', u'Martin Rob..." exit_code: 1 cwd: /mnt/git/webkit-commit-queue Full output: http://queues.webkit.org/results/14228700
Created attachment 167791 [details] Patch for landing
Comment on attachment 167791 [details] Patch for landing Clearing flags on attachment: 167791 Committed r130781: <http://trac.webkit.org/changeset/130781>
All reviewed patches have been landed. Closing bug.