Bug 25817 – svg/dom/SVGScriptElement/script-load-and-error-events.svg is flakey

• Status: RESOLVED
• Resolution: FIXED
• Component: SVG

[Eric Seidel (no email)]

svg/dom/SVGScriptElement/script-load-and-error-events.svg is flakey

We need to fix the test so it stops sometimes failing on the bots.   We can also skip it if necessary.

Example failure: http://build.webkit.org/results/Windows%20Debug%20(Tests)/r43761%20(1061)/results.html

[Eric Seidel (no email)]

*** Bug 29700 has been marked as a duplicate of this bug. ***

[Nikolas Zimmermann]

*** Bug 46998 has been marked as a duplicate of this bug. ***

[Nikolas Zimmermann]

*** Bug 76558 has been marked as a duplicate of this bug. ***

[Alexey Proskuryakov]

Created attachment 211597 [details]
improve failure output

This test fails very often.

I cannot reproduce locally, and cannot find anything wrong through code inspection. Let's see if this additional information helps.

[Tim Horton]

Comment on attachment 211597 [details]
improve failure output

View in context: https://bugs.webkit.org/attachment.cgi?id=211597&action=review

> LayoutTests/svg/dom/SVGScriptElement/script-load-and-error-events.svg:-112
> -    scriptElement.href.baseVal = "resources/not-existant.js";

Are you going to move this file, too?

[Alexey Proskuryakov]

Comment on attachment 211597 [details]
improve failure output

Committed <http://trac.webkit.org/r155727>, clearing review flag.

[Alexey Proskuryakov]

OK, so it says "FAIL: The following tests failed: 6 (actual result: )". 

This means that we are not done running tests 1-5 when we start test 6. One of those gets into errorEventHandler, and since lastTest variable is set, we get into endTest().

[Alexey Proskuryakov]

Oh, but of course. Test 5 is expected to generate a load event first, and then an error event. We start test 6 as soon as we get the load event from test 5.

[Alexey Proskuryakov]

Created attachment 211603 [details]
proposed fix

[Alexey Proskuryakov]

Committed <http://trac.webkit.org/r155733>.