[OSX] Allow one zoom bubble to override another.

[OSX] Allow one zoom bubble to override another.

At present, if a user changes zoom on OSX while the zoom bubble is
already being shown, the new zoom bubble (with the most up to date
information) is not shown. This CL changes this behaviour by allowing
a new zoom bubble to be created, and replace the existing one.

Example: load a PDF and repeatedly click the zoom +/- controls in the
lower-right corner of the viewer.

BUG=444995
Committed: https://crrev.com/84567ceada9f89c8e962d18ab494b1d1bb5a3830
Cr-Commit-Position: refs/heads/master@{#314147}

[wjmaclean, 2015-01-29 22:45:17]

wjmaclean@chromium.org changed reviewers:
+ erikchen@chromium.org, rohitrao@chromium.org

[wjmaclean, 2015-01-29 23:36:44]

Here's a revised version of the reverted CL from
https://codereview.chromium.org/849693006/.

This version places checks in zoom_bubble_controller.mm to avoid dereferencing
delegate_ if it is null. It also closes any existing bubble at the start of
ShowBubble().

Question: since we will now close the bubble_ at the start of ShowBubble(), is
there any advantage to having ZoomDecoration set the bubble's delegate to null?
If not, we wouldn't need to make the delegate a property, in which case
ZoomBubbleController's assumptions about when to check delegate_ might be OK.

[erikchen, 2015-01-30 18:40:30]

On 2015/01/29 23:36:44, wjmaclean wrote:
> Here's a revised version of the reverted CL from
> https://codereview.chromium.org/849693006/.
> 
> This version places checks in zoom_bubble_controller.mm to avoid dereferencing
> delegate_ if it is null. It also closes any existing bubble at the start of
> ShowBubble().
> 
> Question: since we will now close the bubble_ at the start of ShowBubble(), is
> there any advantage to having ZoomDecoration set the bubble's delegate to
null?
> If not, we wouldn't need to make the delegate a property, in which case
> ZoomBubbleController's assumptions about when to check delegate_ might be OK.

The bubble's delegate definitely needs to be set to null before |bubble_| is
nil-ed or ZoomBubbleController's is destroyed, since it's a raw pointer, and it
could otherwise outlive ZoomBubbleController.

[erikchen, 2015-01-30 18:40:43]

lgtm

[wjmaclean, 2015-02-02 14:14:54]

The CQ bit was checked by wjmaclean@chromium.org

[commit-bot: I haz the power, 2015-02-02 14:15:33]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/886883002/1

[commit-bot: I haz the power, 2015-02-02 14:20:36]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-02-02 14:20:38]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[rohitrao (ping after 24h), 2015-02-02 15:08:21]

LGTM

[wjmaclean, 2015-02-02 15:13:27]

The CQ bit was checked by wjmaclean@chromium.org

[commit-bot: I haz the power, 2015-02-02 15:13:45]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/886883002/1

[commit-bot: I haz the power, 2015-02-02 15:16:40]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-02-02 15:17:44]

Patchset 1 (id:??) landed as
https://crrev.com/84567ceada9f89c8e962d18ab494b1d1bb5a3830
Cr-Commit-Position: refs/heads/master@{#314147}

[erikchen, 2015-02-04 19:32:20]

A revert of this CL (patchset #1 id:1) has been created in
https://codereview.chromium.org/897923002/ by erikchen@chromium.org.

The reason for reverting is: Causes window flickering:
https://code.google.com/p/chromium/issues/detail?id=455184.