Add didCreateNewDocument() notification

Add didCreateNewDocument() notification

BUG=416907
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185074

[robwu, 2014-11-06 15:19:42]

Patchset #3 (id:40001) has been deleted

[robwu, 2014-11-06 15:19:46]

Patchset #2 (id:20001) has been deleted

[robwu, 2014-11-06 15:20:14]

Patchset #1 (id:1) has been deleted

[robwu, 2014-11-06 15:28:06]

rob@robwu.nl changed reviewers:
+ japhet@chromium.org

[robwu, 2014-11-06 15:28:07]

Added the willReplaceDocumentWhileExecutingJavaScriptURL() notification, to be
used in https://codereview.chromium.org/684143002/.

[Nate Chapin, 2014-11-06 17:27:37]

https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
File Source/core/loader/FrameLoaderClient.h (right):

https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
Source/core/loader/FrameLoaderClient.h:163: virtual void
dispatchWillReplaceDocumentWhileExecutingJavaScriptURL() = 0;
This feels hacky. I'd feel much better about a didCreateNewDocument() or
similar, called from FrameLoader::didBeginDocument(). That should cover all
cases when the document changes, which would allow the code in the chromium repo
to only InvalidateFrame() in one place (doing so in DidStartProvisionalLoad
feels hacky as well).

[robwu, 2014-11-06 18:13:08]

https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
File Source/core/loader/FrameLoaderClient.h (right):

https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
Source/core/loader/FrameLoaderClient.h:163: virtual void
dispatchWillReplaceDocumentWhileExecutingJavaScriptURL() = 0;
On 2014/11/06 17:27:37, Nate Chapin wrote:
> This feels hacky. I'd feel much better about a didCreateNewDocument() or
> similar, called from FrameLoader::didBeginDocument(). That should cover all
> cases when the document changes, which would allow the code in the chromium
repo
> to only InvalidateFrame() in one place (doing so in DidStartProvisionalLoad
> feels hacky as well).

Done. We will still use two InvalidateFrame() calls though: the new
DidCreateNewDocument plus FrameDetached().

[Nate Chapin, 2014-11-10 21:50:14]

On 2014/11/06 18:13:08, robwu wrote:
>
https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
> File Source/core/loader/FrameLoaderClient.h (right):
> 
>
https://codereview.chromium.org/710443002/diff/60001/Source/core/loader/Frame...
> Source/core/loader/FrameLoaderClient.h:163: virtual void
> dispatchWillReplaceDocumentWhileExecutingJavaScriptURL() = 0;
> On 2014/11/06 17:27:37, Nate Chapin wrote:
> > This feels hacky. I'd feel much better about a didCreateNewDocument() or
> > similar, called from FrameLoader::didBeginDocument(). That should cover all
> > cases when the document changes, which would allow the code in the chromium
> repo
> > to only InvalidateFrame() in one place (doing so in DidStartProvisionalLoad
> > feels hacky as well).
> 
> Done. We will still use two InvalidateFrame() calls though: the new
> DidCreateNewDocument plus FrameDetached().

Ah, yeah. That's a lot more reasonable than at provisional load start. LGTM,
thanks!

[robwu, 2014-11-10 21:54:50]

The CQ bit was checked by rob@robwu.nl

[commit-bot: I haz the power, 2014-11-10 21:55:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/710443002/80001

[commit-bot: I haz the power, 2014-11-10 23:01:34]

Committed patchset #2 (id:80001) as 185074