550500 - synthesizeMouse function needs to be adjusted to more reflect reality when double/triple-clicking

Status: REOPENED

(Testing :: Mochitest, defect)

Description

[Martijn Wargers (dead)]

Attachment: testcase

In the sendMouseEvent function:
http://mxr.mozilla.org/mozilla-central/source/testing/mochitest/tests/SimpleTest/EventUtils.js#18

When we do a double-click for instance, we only generate 1 click with clickCount 2 and dblclick event, which is not what happens, when you do a real double-click.
With a real double-click, you first get a click with clickCount 1, then a click with clickCount 2, then a dblclick event.
This is probably the cause for bug 537968.
So EventUtils.js should do something similar, to more reflect reality.

Tests that could be affected by this change are:
http://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/downloads/tests/chrome/test_bug_429247.xul
http://mxr.mozilla.org/mozilla-central/source/layout/style/test/test_pointer-events.html?force=1
http://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/downloads/tests/chrome/test_bug_429247.xul
Stuff that use tree_shared.js:
http://mxr.mozilla.org/mozilla-central/source/toolkit/content/tests/widgets/tree_shared.js
http://mxr.mozilla.org/mozilla-central/source/toolkit/content/tests/widgets/test_richlist_direction.xul
http://mxr.mozilla.org/mozilla-central/source/toolkit/content/tests/browser/browser_bug295977_autoscroll_overflow.js
http://mxr.mozilla.org/mozilla-central/source/browser/base/content/test/browser_pageInfo.js
http://mxr.mozilla.org/mozilla-central/source/browser/base/content/test/browser_bug432599.js
http://mxr.mozilla.org/mozilla-central/source/browser/base/content/test/browser_overflowScroll.js

Comment 1

[Martijn Wargers (dead)]

Attachment: testcase

Comment 2

[Martijn Wargers (dead)]

Attachment: patch

This takes into effect what Neal wants with the function, commented in bug 537968, comment 9.
I haven't tested this patch yet.

Comment 3

[Martijn Wargers (dead)]

I uploaded a patch to the tryserver, but for some reason it failed on winmo or something:
http://tinderbox.mozilla.org/showbuilds.cgi?tree=MozillaTry

I tested the patch with test_pointer-events.html, that worked at least.

Comment 4

[Martijn Wargers (dead)]

Ok, builds are here:
https://build.mozilla.org/tryserver-builds/[e-mail]-1267815144/
But I'm more interested what Tinderbox says, regarding the tests.

Comment 5

[Martijn Wargers (dead)]

Comment on attachment 430666 [details] [diff] [review]
patch

Tryserver tinderbox seems to be all green, so I guess this patch is ready for review.

Comment 6

[Martijn Wargers (dead)]

Sorry, tryserver just mailed me that some tests were failing:
Summary of unittest results:
check: 577/0
xpcshell-tests: 746/0
reftest: 4211/0/242
crashtest: 1484/0/10
mochitest-plain: 156662/<em class="testfail">4</em>/1345
mochitest-chrome: 8759/0/109
mochitest-browser-chrome: 4371/<em class="testfail">2</em>/3
mochitest-a11y: 13726/0/8

http://tinderbox.mozilla.org/showlog.cgi?log=MozillaTry/1267815208.1267824397.8599.gz
159687 ERROR TEST-UNEXPECTED-FAIL | /tests/toolkit/content/tests/widgets/test_tree.xul | cycleHeader column sortDirection - got "ascending", expected ""
160561 ERROR TEST-UNEXPECTED-FAIL | /tests/toolkit/content/tests/widgets/test_tree_hier.xul | cycleHeader column sortDirection - got "ascending", expected ""
161321 ERROR TEST-UNEXPECTED-FAIL | /tests/toolkit/content/tests/widgets/test_tree_hier_cell.xul | cycleHeader column sortDirection - got "ascending", expected ""
161858 ERROR TEST-UNEXPECTED-FAIL | /tests/toolkit/content/tests/widgets/test_tree_single.xul | cycleHeader column sortDirection - got "ascending", expected ""
PROCESS-CRASH | automation.py | application crashed (minidump found)
Thread 1 (crashed)
PROCESS-CRASH | automation.py | application crashed (minidump found)
Thread 1 (crashed)
PROCESS-CRASH | automation.py | application crashed (minidump found)
Thread 1 (crashed)
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/browser/base/content/test/browser_overflowScroll.js | Scrolled one page of tabs with a double click - Got 114, expected 16
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/browser/base/content/test/browser_pageInfo.js | When the lock is clicked twice there should be only one page info dialog - Got 2, expected 1

At least some of those failures are files that I already mentioned, that could possibly cause test failures, so I have to look into these.

Comment 7

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Martijn, have you had time in the last two weeks to check those failures?