void WebFrameProxy::didFailLoad() { ........ ..... m_title = String(); // causing title to empty even though Page(and title) is displayed } When Page loading is stopped before load is finished. we should not set the title to empty string We still have the content to display on screen(we have full Frame & dom tree present). So if page is stopped the UI process always get the empty title even though the page related to that title exists.
Created attachment 202657 [details] Patch
One thing that's very helpful to do in cases like this is svn blame, which may explain rationale behind the code. Here, it points to <http://trac.webkit.org/changeset/74282>. As you can see, the behavior was chosen conscientiously, and even mentioned in ChangeLog .
Comment on attachment 202657 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=202657&action=review But I still agree that the behavior is wrong. > Source/WebKit2/ChangeLog:3 > + [WebFrameProxy::didFailLoad] title becomes empty even though Page(and title) is Please update the title in ChangeLog to match Bugzilla. We use brackets in title for a very specific purpose (marking patches that are of limited interest only).
Created attachment 202757 [details] Patch
(In reply to comment #3) > (From update of attachment 202657 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=202657&action=review > > But I still agree that the behavior is wrong. > > > Source/WebKit2/ChangeLog:3 > > + [WebFrameProxy::didFailLoad] title becomes empty even though Page(and title) is > > Please update the title in ChangeLog to match Bugzilla. > > We use brackets in title for a very specific purpose (marking patches that are of limited interest only). Hi I updated the changLog as requested. please give +1 for this patch
cq+
Comment on attachment 202757 [details] Patch Clearing flags on attachment: 202757 Committed r150689: <http://trac.webkit.org/changeset/150689>
All reviewed patches have been landed. Closing bug.