[Refactoring] Frame::willDetachPage() shouldn't be called more than once.

[Hajime Morrita]

This is just a cleanup.

[Hajime Morrita]

Created attachment 145237 [details]
Patch

[Hajime Morrita]

Darin, Ryosuke, could you take a look? This is just a small cleanup.

[Ryosuke Niwa]

Comment on attachment 145237 [details]
Patch

This is a scary change but looks reasonable. Hopefully it won't cause any regressions.

[Ryosuke Niwa]

Comment on attachment 145237 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=145237&action=review

> Source/WebCore/page/Frame.cpp:695
>      if (Frame* parent = tree()->parent())
>          parent->loader()->checkLoadComplete();

Actually, it's not entirely clear to me why we'd have to call checkLoadComplete more than once :\

[WebKit Review Bot]

Comment on attachment 145237 [details]
Patch

Clearing flags on attachment: 145237

Committed r119219: <http://trac.webkit.org/changeset/119219>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[WebKit Review Bot]

Re-opened since this is blocked by 88088

[Adam Barth]

Comment on attachment 145237 [details]
Patch

I seem to remember trying to do this originally, but there was something I found in studying the code that didn't work properly this way...

[Hajime Morrita]

(In reply to comment #8)
> (From update of attachment 145237 [details])
> I seem to remember trying to do this originally, but there was something I found in studying the code that didn't work properly this way...
Hmm, interesting...
I'd have liked to see the failed test.
But this looks something which should keep untouched...