reporter: can you clarify whether this is a regression on Windows (i.e. did it ever work there?). It's unlikely that CL changed things significantly on Windows.

(elly/vasilii: 2 busy sheriff shifts have put me way behind on things - I won't be able to look at this immediately.)

It's possible the anchoring is not working properly in popups (or maybe just extension popups).

Deleted: Screenshot from 2018-03-14 11-50-02.png 91.8 KB

	Screenshot from 2018-03-14 11-50-02.png 91.8 KB View Download

MacViews triage: vasilii@, how is this looking? Let's target a fix at M67.

It's not a high priority because the window is gone quickly anyway.

Are we still targeting this for M67 based on comment #4?

I'll analyse it for M67.

** Bulk Edit **

There are only two M67 dev releases left on 04/03 & 04/10 before M67 branch on 04/12. Please try to land the fix ASAP to trunk so we can move forward with 50%-50% experiment on M67 Canary/Dev (if possible at all). Thank you.

** Bulk Edit **

There is only one dev release left 04/10 before M67 branch on 04/12. Please try to land the fix ASAP to trunk so we can move forward with 50%-50% experiment on M67 Canary/Dev (if possible at all). Thank you.

FYI: Change/Fix has to be landed in trunk latest by 4:00 PM PT, Friday (04/06) so we can pick it up for next week dev release. 

The password manager works on this site. But the bubble is incorrectly anchored.

Deleted: Screen Shot 2018-04-04 at 16.53.11.png 71.1 KB

	Screen Shot 2018-04-04 at 16.53.11.png 71.1 KB View Download

New positioning.

Deleted: Screen Shot 2018-04-06 at 14.13.52.png 58.9 KB

	Screen Shot 2018-04-06 at 14.13.52.png 58.9 KB View Download

I have a pending fix for alignment here https://chromium-review.googlesource.com/c/chromium/src/+/998163

I verified that the bubble appears if the window stays in the screen. In the original reproduction case it doesn't appear because the window closes before we show it. Probably the macviews bubble has longer animation or it's just slower. Anyway, it doesn't make sense to debug the race condition because we can't assume that 0.5sec is just enough to click 'Save' in the bubble. We should handle this case somehow differently. Thus, I'm leaving the bug open.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/9d05ec002dd34b7cb4b9d661aed08bbc70a7959d

commit 9d05ec002dd34b7cb4b9d661aed08bbc70a7959d
Author: Vasilii Sukhanov <vasilii@chromium.org>
Date: Mon Apr 09 10:39:24 2018

Align password bubble properly on Mac.

The case when there was no omnibox wasn't handled properly. The bubble should be top-center aligned with the window.

Bug: 821347
Change-Id: I6af3395da3cbf6f399625c4020015941c5e0b092
Reviewed-on: https://chromium-review.googlesource.com/998163
Reviewed-by: Michael Wasserman <msw@chromium.org>
Reviewed-by: Elly Fong-Jones <ellyjones@chromium.org>
Commit-Queue: Vasilii Sukhanov <vasilii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#549143}
[modify] https://crrev.com/9d05ec002dd34b7cb4b9d661aed08bbc70a7959d/chrome/browser/ui/cocoa/tab_dialogs_views_mac.mm
[modify] https://crrev.com/9d05ec002dd34b7cb4b9d661aed08bbc70a7959d/ui/views/bubble/bubble_border.cc
[modify] https://crrev.com/9d05ec002dd34b7cb4b9d661aed08bbc70a7959d/ui/views/bubble/bubble_border_unittest.cc

Can this be marked as fixed if nothing else is pending?

I didn't really fix it. We need to rethink the UX flow for such a use case. It's not high priority and doesn't block anything.