1699812 - Perma Android /svg/text/scripted/getsubstringlength-emoji-ligatures.html | SVGTextContentElement.getSubStringLength on Emoji with ligatures - assert_approx_equals: expected 340.75 +/- 1 but got 330.3458251953125 when Gecko 88 merges to Beta on 2021-03-22

Status: NEW

(Core :: SVG, defect, P2)

Description

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

• autoland-as-beta simulation (autoland to get the latest sync of the web platform tests with upstream in, and autoland is not affected by the failure)
• How to run these simulations
• Failure log

[task 2021-03-19T15:17:47.488Z] 15:17:47     INFO - TEST-PASS | /svg/path/error-handling/bounding.svg | bounding 
[task 2021-03-19T15:17:47.488Z] 15:17:47     INFO - TEST-UNEXPECTED-FAIL | /svg/text/scripted/getsubstringlength-emoji-ligatures.html | SVGTextContentElement.getSubStringLength on Emoji with ligatures - assert_approx_equals: expected 340.75 +/- 1 but got 330.3458251953125
[task 2021-03-19T15:17:47.488Z] 15:17:47     INFO - @http://web-platform.test:8000/svg/text/scripted/getsubstringlength-emoji-ligatures.html:23:25
[task 2021-03-19T15:17:47.489Z] 15:17:47     INFO - Test.prototype.step@http://web-platform.test:8000/resources/testharness.js:2095:25
[task 2021-03-19T15:17:47.489Z] 15:17:47     INFO - test@http://web-platform.test:8000/resources/testharness.js:566:30
[task 2021-03-19T15:17:47.489Z] 15:17:47     INFO - @http://web-platform.test:8000/svg/text/scripted/getsubstringlength-emoji-ligatures.html:17:5
[task 2021-03-19T15:17:47.590Z] 15:17:47     INFO - TEST-OK | /svg/text/scripted/getsubstringlength-emoji-ligatures.html | took 351ms

Comment 1

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

Attachment: Bug 1699812 - set getsubstringlength-emoji-ligatures.html as failing on Android with Release and Beta

Test got added in bug 1697762 but build didn't get changed. Issue is pending
further investigation.

Comment 2

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/0ae2e2ece3b8
set getsubstringlength-emoji-ligatures.html as failing on Android with Release and Beta r=emilio DONTBUILD

Comment 3

[Atila Butkovits]

https://hg.mozilla.org/mozilla-central/rev/0ae2e2ece3b8

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-beta/rev/1a02f792b5fe

Comment 5

[Emilio Cobos Álvarez (:emilio)]

Jonathan, Robert, do you know why android behavior might differ here between nightly and release / beta?

Comment 6

[BugBot [:suhaib / :marco/ :calixte]]

Changing the priority to p2 as the bug is tracked by a release manager for the current nightly.
See What Do You Triage for more information

Comment 7

[Jonathan Kew [:jfkthame]]

I wonder if the test is falling victim to the asynchronous font fallback behavior. In theory I think that could affect it on any channel, but timing and/or test chunking could be different between Nightly and Release such that it shows up in one place but not the other.

Comment 8

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

The test still fails with asynchronous font fallback disabled.

Comment 9

[Jonathan Kew [:jfkthame]]

OK, scratch that idea. Is it possible to capture a screenshot of the test, so we can see how it's rendering the emoji strings? Offhand I don't see why it should behave differently on the different channels, but maybe seeing what it renders would give a clue.

Comment 10

[Intermittent Failures Robot]

2 failures in 3967 pushes (0.001 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• mozilla-beta: 2

Platform and build breakdown:

• android-em-7-0-x86_64-shippable-qr: 2
  • opt: 2

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1699812&startday=2021-03-22&endday=2021-03-28&tree=all

Comment 11

[Robert Longson [:longsonr]]

Sorry, I don't know.

Comment 12

[BugBot [:suhaib / :marco/ :calixte]]

The leave-open keyword is there and there is no activity for 6 months.
:jwatt, maybe it's time to close this bug?
For more information, please visit auto_nag documentation.