1432773 - Intermittent /webdriver/tests/actions/mouse.py | mouse.py::test_click_element_center

Status: RESOLVED FIXED

(Remote Protocol :: Marionette, defect, P5)

Description

[Eliza Balazs [:ebalazs_]]

+++ This bug was initially created as a clone of Bug #1421878 +++

[task 2017-11-29T21:07:18.140Z] 21:07:18     INFO - TEST-PASS | /webdriver/tests/actions/mouse.py | mouse.py::test_click_at_coordinates 
2262
[task 2017-11-29T21:07:18.140Z] 21:07:18     INFO - TEST-PASS | /webdriver/tests/actions/mouse.py | mouse.py::test_context_menu_at_coordinates 
2263
[task 2017-11-29T21:07:18.140Z] 21:07:18     INFO - TEST-UNEXPECTED-FAIL | /webdriver/tests/actions/mouse.py | mouse.py::test_click_element_center 
2264
[task 2017-11-29T21:07:18.140Z] 21:07:18     INFO - session = <webdriver.client.Session object at 0x7f7e4bd73550>
2265
[task 2017-11-29T21:07:18.141Z] 21:07:18     INFO - test_actions_page = None
2266
[task 2017-11-29T21:07:18.141Z] 21:07:18     INFO - mouse_chain = <webdriver.client.ActionSequence object at 0x7f7e4bd73a50>
2267
[task 2017-11-29T21:07:18.141Z] 21:07:18     INFO - 
2268
[task 2017-11-29T21:07:18.141Z] 21:07:18     INFO -     def test_click_element_center(session, test_actions_page, mouse_chain):
2269
[task 2017-11-29T21:07:18.142Z] 21:07:18     INFO -         outer = session.find.css("#outer", all=False)
2270
[task 2017-11-29T21:07:18.142Z] 21:07:18     INFO -         center = get_center(outer.rect)
2271
[task 2017-11-29T21:07:18.142Z] 21:07:18     INFO -         mouse_chain.click(element=outer).perform()
2272
[task 2017-11-29T21:07:18.142Z] 21:07:18     INFO -         events = get_events(session)
2273
[task 2017-11-29T21:07:18.142Z] 21:07:18     INFO - >       assert len(events) == 4
2274
[task 2017-11-29T21:07:18.143Z] 21:07:18     INFO - E       assert 5 == 4
2275
[task 2017-11-29T21:07:18.143Z] 21:07:18     INFO - E        +  where 5 = len([{'altKey': False, 'button': 0, 'buttons': 0, 'ctrlKey': False, ...}, {'altKey': False, 'button': 0, 'buttons': 1, 'ct...'button': 0, 'buttons': 0, 'ctrlKey': False, ...}, {'altKey': False, 'button': 0, 'buttons': 0, 'ctrlKey': False, ...}])

Comment 1

[Maja Frydrychowicz :impossibus (was :maja_zf) (needinfo me)]

Attachment: Bug 1432773 - listen for unload in capture phase to reset click count;

The bubbling phase is too late so resetClick didn't always get called.

Review commit: https://reviewboard.mozilla.org/r/215424/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/215424/

Comment 2

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Comment on attachment 8945183 [details]
Bug 1432773 - listen for unload in capture phase to reset click count;

https://reviewboard.mozilla.org/r/215424/#review221188

::: testing/marionette/listener.js:56
(Diff revision 1)
>  
>  // Listen for click event to indicate one click has happened, so actions
>  // code can send dblclick event
>  addEventListener("click", event.DoubleClickTracker.setClick);
>  addEventListener("dblclick", event.DoubleClickTracker.resetClick);
> -addEventListener("unload", event.DoubleClickTracker.resetClick);
> +addEventListener("unload", event.DoubleClickTracker.resetClick, true);

Not directly related to this change, but should the other two event listeners also operating in the captureing phase?

Comment 3

[Maja Frydrychowicz :impossibus (was :maja_zf) (needinfo me)]

Comment on attachment 8945183 [details]
Bug 1432773 - listen for unload in capture phase to reset click count;

https://reviewboard.mozilla.org/r/215424/#review221188

> Not directly related to this change, but should the other two event listeners also operating in the captureing phase?

In the case of click and dblclick, capture versus bubble shouldn't make a difference, and that's what I confirmed in my testing as well.

Comment 4

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Comment on attachment 8945183 [details]
Bug 1432773 - listen for unload in capture phase to reset click count;

https://reviewboard.mozilla.org/r/215424/#review221188

> In the case of click and dblclick, capture versus bubble shouldn't make a difference, and that's what I confirmed in my testing as well.

Also when `preventDefault` is used by content? If all that works, lets get it in! Thanks.

Comment 5

[Maja Frydrychowicz :impossibus (was :maja_zf) (needinfo me)]

Comment on attachment 8945183 [details]
Bug 1432773 - listen for unload in capture phase to reset click count;

https://reviewboard.mozilla.org/r/215424/#review221188

> Also when `preventDefault` is used by content? If all that works, lets get it in! Thanks.

You mean `stopPropagation`? That could break things. I'll investigate and file a separate bug if needed. For now I just want to land this to fix the intermittent.

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/c0b051e8b500
listen for unload in capture phase to reset click count; r=whimboo

Comment 7

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

https://hg.mozilla.org/mozilla-central/rev/c0b051e8b500

Comment 8

[Henrik Skupin [:whimboo][⌚️UTC+2]]

We might also need it for 59. Lets wait over the weekend how it works and then we can request an uplift.

Comment 9

[Intermittent Failures Robot]

6 failures in 701 pushes (0.009 failures/push) were associated with this bug in the last 7 days.    

Repository breakdown:
* mozilla-inbound: 3
* autoland: 2
* try: 1

Platform breakdown:
* linux32: 5
* linux64-ccov: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1432773&startday=2018-01-22&endday=2018-01-28&tree=all

Comment 10

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Since Friday no more test failures. Please lets uplift to beta.

Comment 11

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

https://hg.mozilla.org/releases/mozilla-beta/rev/d791950b5abd

Comment 12

[BMO Automation]

Moving bug to Remote Protocol::Marionette component per bug 1815831.