When committing a provisional load that didn't change the RVH, still abort pending navigations.

When committing a provisional load that didn't change the RVH, still abort pending navigations.

Also, when sending error events for a frame, mark them as failed, so we don't touch them again in the future

BUG=139117
TEST=none


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=149632

[jochen (gone - plz use gerrit), 2012-08-02 12:29:14]

there's still one crash occurring that seems to come from old frames being
tracked although the RVH is gone long ago.

I can't reproduce it, but I hope this fixes it

plz review

[Matt Perry, 2012-08-02 12:59:31]

lgtm

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
File chrome/browser/extensions/api/web_navigation/frame_navigation_state.cc
(right):

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/frame_navigation_state.cc:188:
DCHECK(frame_ids_.size() == 1);
nit: use DCHECK_EQ(1u, size)

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
File chrome/browser/extensions/api/web_navigation/web_navigation_api.cc (right):

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/web_navigation_api.cc:383: if
(is_main_frame)
Did you mean to add this check for is_main_frame? The old code sent errors for
subframe frame loads too.

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/web_navigation_api.cc:564:
web_contents,
nit: indent + 2

[jochen (gone - plz use gerrit), 2012-08-02 14:09:49]

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
File chrome/browser/extensions/api/web_navigation/frame_navigation_state.cc
(right):

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/frame_navigation_state.cc:188:
DCHECK(frame_ids_.size() == 1);
On 2012/08/02 12:59:31, Matt Perry wrote:
> nit: use DCHECK_EQ(1u, size)

Done.

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
File chrome/browser/extensions/api/web_navigation/web_navigation_api.cc (right):

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/web_navigation_api.cc:383: if
(is_main_frame)
On 2012/08/02 12:59:31, Matt Perry wrote:
> Did you mean to add this check for is_main_frame? The old code sent errors for
> subframe frame loads too.

The old check said "the pending RVH' commits, and only main frame navigations
can happen in a pending RVH.

Now I want to drop old navigations no matter what RVH commits its main frame

https://chromiumcodereview.appspot.com/10834125/diff/5002/chrome/browser/exte...
chrome/browser/extensions/api/web_navigation/web_navigation_api.cc:564:
web_contents,
On 2012/08/02 12:59:31, Matt Perry wrote:
> nit: indent + 2

Done.

[commit-bot: I haz the power, 2012-08-02 14:10:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jochen@chromium.org/10834125/13001

[commit-bot: I haz the power, 2012-08-02 15:32:31]

Change committed as 149632