1344423 - Click outside event listener popup doesn't work sometimes

Status: RESOLVED DUPLICATE of bug 1344504

(DevTools :: Inspector, defect)

Description

[684sigma]

I have a problem with Firefox Beta 52. It doesn't happen in older versions like Firefox ESR 45.
Sometimes click outside event listener popup doesn't work.
It happens unpredictably, however, I noticed one specific scenario when it happens

1. Open https://bugzilla.mozilla.org/show_bug.cgi?id=1344309
2. Open inspector, click on event listener button near html node
3. Click on "Computed" tab or "Console" tab, or any button on toolbar

Result: event listener popup disappears, and clicked button becomes focused (focus outline appears), but click doesn't happen
Expected: clicked tab should be selected / clicked button should be activated. To understand it better, try good scenarios:
- click "Search HTML" or click on another node in markup while event listener popup is opened
- click on any element while normal popups like color picker are opened

Comment 1

[Loic]

regression range:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=9ee862a030f350e481e3fa3a09d6dc28d4dbf591&tochange=ef5f932101e5b833b2429407cb0873471b4d764e

Jan Odvarko — Bug 1259819 - HTML Sidebar component. r=pbro

Comment 2

[Loic]

Well, it's not only the tabs of the sidebar widget but also the tabs of the dev widget.

Comment 3

[Gerry Chang [:gchang]]

Too late for 51 and 52 will be released this week. Mark 51 won't fix.

Comment 4

[Julian Descottes [:jdescottes]]

I think this bug report is actually a duplicate of Bug 1344504. 

The problem described by the logger is not that the tab gets the focus or not (which might be related to the HTML sidebar), but that the click does not select the tab (which has the same root cause as Bug 1344504).

By configuration, the event listener popup is consuming outside clicks, which is why clicks performed while the tooltip is opened are just closing the tooltip and not performing any additional action.

For now, I will just block this one on Bug 1344504, and we will check if the issue is fixed when Bug 1344504 is resolved.

Comment 5

[684sigma]

(In reply to Julian Descottes [:jdescottes] from comment #4)
> I think this bug report is actually a duplicate of Bug 1344504. 

In fact, Bug 1344504 is a part of this bug, reported by Loic 2 hours after reading this bug, which was was filed as "Click outside event listener popup doesn't work". It appears that you noticed bug 1344504 first because of Loic's actions. I don't know why it was necessary to re-upload my bug; I'll discuss it with Loic.


(In reply to Loic from comment #2)
> Well, it's not only the tabs of the sidebar widget but also the tabs of the
> dev widget.

It was pretty obvious from the start that the problem is popup itself, blocking click outside of it, yet you decided to file a part of this bug as bug 1344504, without even mentioning it here, as if it was you who detected and investigated it. In this bug, you changed summary to irrelevant.
It devalues my time and effort spent to find and upload this bug report. Is re-uploading valid bug reports a normal practice? If it is, should I report bugs in a very sloppy manner, in order to save my time? If it's not,
1) Please provide me with instructions on how to avoid situations like this from now on. I want to fix my mistake
2) In particular, this bug. How could I describe the problem more clear than "Click outside event listener popup doesn't work"? Was there a way to convince a person reading my bug report not to re-upload it?

Comment 6

[Loic]

I filed bug 1344504 after triaging yours, because I thought it was another bug, both have different regression ranges even if the underlying cause is probably the same.

I triaged hundred bugs on BMO for 5 years and it's not rare to find new bugs when triaging.
As many reporters are just one-time reporters, it's common to file new bug oneself rather than ask reporter to do it.

If it makes you happy, we can close mine as duplicate, I don't really care. :)

Comment 7

[684sigma]

(In reply to Loic from comment #6)
> I filed bug 1344504 after triaging yours, because I thought it was another
> bug, both have different regression ranges even if the underlying cause is
> probably the same.

Seems reasonable. Sorry, finding all those regression ranges is likely a lot of work too.


> If it makes you happy, we can close mine as duplicate, I don't really care.

It would make me happy, but it won't help much, since your reply means that it may happen any time. The only thing that matters is long term. Please do not change anything, also hide last 3 comments if you can.

Comment 8

[Liz Henry (:lizzard) (relman/hg->git project)]

Looks like we have a good chance of getting this fixed in Firefox 53 or 54. Thanks 684sigma, loic, and julian, I do think it's a good idea to leave this open and check back to verify this issue gets fixed once bug 1344504 lands.

Comment 9

[Patrick Brosset <:pbro>]

I tested this in Firefox 55 and could not reproduce. So I think bug 1344504 did fix the issue. Closing as duplicate for this reason.

Comment 10

[Liz Henry (:lizzard) (relman/hg->git project)]

Fixed in the duplicate bug in 55.