Summary: | [GTK] Split libWebCore into two libWebCore and libWebCoreGtk | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Carlos Garcia Campos <cgarcia> | ||||||||
Component: | WebKit2 | Assignee: | Nobody <webkit-unassigned> | ||||||||
Status: | RESOLVED FIXED | ||||||||||
Severity: | Normal | CC: | eric, gustavo, mrobinson, xan.lopez | ||||||||
Priority: | P2 | Keywords: | Gtk | ||||||||
Version: | 528+ (Nightly build) | ||||||||||
Hardware: | PC | ||||||||||
OS: | Linux | ||||||||||
Bug Depends on: | |||||||||||
Bug Blocks: | 60546 | ||||||||||
Attachments: |
|
Description
Carlos Garcia Campos
2011-05-10 01:40:52 PDT
Created attachment 92930 [details]
Patch
Comment on attachment 92930 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=92930&action=review This looks fine to me, but I'd like to get Xan and Kov to look at this one as well. > Source/WebCore/GNUmakefile.am:762 > +libWebCoreGtk_la_CPPFLAGS = \ > + -DBUILDING_WEBKIT \ > + $(global_cppflags) \ Would it make sense to remove GTK_CLFAGS here? (In reply to comment #2) > (From update of attachment 92930 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=92930&action=review > > This looks fine to me, but I'd like to get Xan and Kov to look at this one as well. > > > Source/WebCore/GNUmakefile.am:762 > > +libWebCoreGtk_la_CPPFLAGS = \ > > + -DBUILDING_WEBKIT \ > > + $(global_cppflags) \ > > Would it make sense to remove GTK_CLFAGS here? global_cppflags contains some flags for GTK (deprecation guards, etc.), but I think it doesn't contain GTK_CFLAGS. Anyway, as long as it doesn't link to GTK, it's not a problem. Created attachment 95903 [details]
Updated patch
Patch updated to current git master. I also moved FileChooserGtk.cpp from webcoregtk to webcore since it doesn't actually use gtk and all freetype files except FontPlatformDataFreeType.cpp which is the only one using gtk
Comment on attachment 95903 [details] Updated patch Attachment 95903 [details] did not pass gtk-ews (gtk): Output: http://queues.webkit.org/results/8763088 Comment on attachment 95903 [details]
Updated patch
Cannot find anything weird here. At some point it would be interesting to add a comment explaining why we have the libraries we have here, otherwise I suppose it's not easy to figure out our makefiles.
The patch works for me on a clean build, I think the bot doesn't rebuild libWebCore, it only builds libWebCoreGtk so that all symbols in libWebCoreGtk are already in libWebCore. Created attachment 97777 [details]
Patch updated to apply on current git master
I've also added a comment explaining why we need this split, as suggested by Xan.
Comment on attachment 97777 [details]
Patch updated to apply on current git master
Looks good to me.
Committed r89251: <http://trac.webkit.org/changeset/89251> |