Don't update URL bar or SSL icon for pending history navs until they commit.

Don't update URL bar or SSL icon for pending history navs until they commit.

BUG=89564
TEST=Go back or forward to a slow URL.  No omnibox change until it commits.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=98853

[Charlie Reis, 2011-08-30 01:34:46]

This is one approach for the important part of the fix for issue 89564, which
corrects the URL bar and SSL icons.  I thought about changing GetActiveEntry
itself, but that confused things when multiple back/forwards were involved.

I'd like to consider also updating Browser::NavigationStateChanged to have it
call UpdateCommandsForTabState on any changed_flags (and not just
INVALIDATE_URL).  That would make the back/forward buttons update immediately
rather than waiting for either a commit or the user to switch tabs and back. 
That would need a little extra work in some corner cases, but it might be worth
it.

I'll address the part of 89564 where the navigation in the repro steps never
commits in a separate CL.

[brettw, 2011-08-30 16:15:18]

LGTM

http://codereview.chromium.org/7790018/diff/6/content/browser/tab_contents/na...
File content/browser/tab_contents/navigation_controller.h (right):

http://codereview.chromium.org/7790018/diff/6/content/browser/tab_contents/na...
content/browser/tab_contents/navigation_controller.h:81: // this is the method
you will typically want to call.
Can you update this comment to say when you should call GetVisibleEntry instead?
A similar comment on GetVisibleEntry might also be nice.

[Charlie Reis, 2011-08-30 17:04:58]

http://codereview.chromium.org/7790018/diff/6/content/browser/tab_contents/na...
File content/browser/tab_contents/navigation_controller.h (right):

http://codereview.chromium.org/7790018/diff/6/content/browser/tab_contents/na...
content/browser/tab_contents/navigation_controller.h:81: // this is the method
you will typically want to call.
On 2011/08/30 16:15:18, brettw wrote:
> Can you update this comment to say when you should call GetVisibleEntry
instead?
> A similar comment on GetVisibleEntry might also be nice.

Sure-- fixed.

[commit-bot: I haz the power, 2011-08-30 19:01:29]

Change committed as 98853