27637 2009-07-23 22:10:17 -0700 [Gtk] Needs a signal for didFinishDocumentLoadForFrame - fails fast/history/saves-state-after-fragment-nav.html 2010-07-30 12:36:16 -0700 1 1 1 Unclassified WebKit WebKitGTK 528+ (Nightly build) PC All RESOLVED FIXED Gtk P2 Normal --- 28823 1 jmalonzo webkit-unassigned gustavo mrobinson oldest_to_newest 134362 0 jmalonzo 2009-07-23 22:10:17 -0700 I would like to propose to that Gtk emit a signal for didFinishDocumentLoadForFrame, probably as a state of load-status. This will also allow us to pass the test above. Here's the diff of the failing test: @@ -1,3 +1 @@ -main frame - has 1 onunload handler(s) -main frame - has 1 onunload handler(s) PASS 146161 1 gustavo 2009-09-10 04:31:34 -0700 That sounds like a good idea. I feel we should study Mac's API docs for this delegate and the ones it is related to. This looks like the API used to help with DOM bindings usage, doesn't it? 146164 2 jmalonzo 2009-09-10 04:43:44 -0700 (In reply to comment #1) > That sounds like a good idea. I feel we should study Mac's API docs for this > delegate and the ones it is related to. This looks like the API used to help > with DOM bindings usage, doesn't it? I have an implementation for this but it's a bit hackish. The main issue is that there's no elegant way for us to connect to each (sub) frame. Most of the time I had to connect while the main frame is loading so I don't miss a frame (some frames get loaded dynamically). I feel that we might need this functionality once we start working on the DOM bindings so this bug can probably serve as a guinea pig. 258332 3 mrobinson 2010-07-30 12:34:21 -0700 This feature was added in https://bugs.webkit.org/show_bug.cgi?id=32170 . The test passes, so I'll unskip it now. 258333 4 mrobinson 2010-07-30 12:36:16 -0700 Committed r64368: <http://trac.webkit.org/changeset/64368>