It's common sense that we should rename this API (to statusMessageChanged?!). Maybe there should also be a statusMessage property, but this needs to be discussed.
Created attachment 119086 [details] Patch
LGTM.
I'm not too fond of having the property there without any proof that this would actually be useful. We're not even using it ourselves in the mini browser, so how can we be sure it's useful? Even with the signal I have to wonder: Do other browsers actually use this right now? (It's not used for link hovering) (I know it's in the WK2 C API, but does anyone know of an example web app/site that uses window.status that works?)
(In reply to comment #3) > I'm not too fond of having the property there without any proof that this would actually be useful. We're not even using it ourselves in the mini browser, so how can we be sure it's useful? > > Even with the signal I have to wonder: Do other browsers actually use this right now? (It's not used for link hovering) > > (I know it's in the WK2 C API, but does anyone know of an example web app/site that uses window.status that works?) Interesting point. It seems that "window.status" itself is not expected to work in HTML5 (even though WebKit supports it). http://dev.w3.org/html5/spec/Overview.html#dom-window-status
(In reply to comment #4) > (In reply to comment #3) > > I'm not too fond of having the property there without any proof that this would actually be useful. We're not even using it ourselves in the mini browser, so how can we be sure it's useful? > > > > Even with the signal I have to wonder: Do other browsers actually use this right now? (It's not used for link hovering) > > > > (I know it's in the WK2 C API, but does anyone know of an example web app/site that uses window.status that works?) > > Interesting point. It seems that "window.status" itself is not expected to work in HTML5 (even though WebKit supports it). > > http://dev.w3.org/html5/spec/Overview.html#dom-window-status To be honest, I discovered this feature when I've worked on this bug. Never heard about it before and I'm afraid this can't be very useful as well. If we want to at least support this feature, then maybe we should just have the signal as we had before? The test case could be modified a bit.
(In reply to comment #5) > (In reply to comment #4) > > (In reply to comment #3) > > > I'm not too fond of having the property there without any proof that this would actually be useful. We're not even using it ourselves in the mini browser, so how can we be sure it's useful? > > > > > > Even with the signal I have to wonder: Do other browsers actually use this right now? (It's not used for link hovering) > > > > > > (I know it's in the WK2 C API, but does anyone know of an example web app/site that uses window.status that works?) > > > > Interesting point. It seems that "window.status" itself is not expected to work in HTML5 (even though WebKit supports it). > > > > http://dev.w3.org/html5/spec/Overview.html#dom-window-status > > To be honest, I discovered this feature when I've worked on this bug. Never heard about it before and I'm afraid this can't be very useful as well. If we want to at least support this feature, then maybe we should just have the signal as we had before? The test case could be modified a bit. This is a great reason for removing the API altogether.
Comment on attachment 119086 [details] Patch r- as it turns out that we should rather remove the status message API altogether.
Adjusting title to latest conclusions ;-)
Created attachment 120020 [details] Patch
Comment on attachment 120020 [details] Patch Clearing flags on attachment: 120020 Committed r103347: <http://trac.webkit.org/changeset/103347>
All reviewed patches have been landed. Closing bug.