Andrew McCreight [:mccr8] Reporter
	Description • 8 years ago

This bug was filed from the Socorro interface and is 
report bp-c151cf56-673d-49b4-8d61-199ee2170322.
=============================================================

I only see two crashes from one installation, but it looks like a null deref in code landed in bug 1333990, so maybe there's something that can be fixed here.

	Kris Maglione [:kmag] Assignee
	Comment 1 • 8 years ago

Hm. I don't think it's possible for mScript to be null there. Maybe we're passing in a nuked sandbox and failing to enter its compartment, though.

	Kris Maglione [:kmag] Assignee
	Comment 2 • 8 years ago

Ah. JS::FinishOffThreadScript returns null if the script has a syntax error.

	Shu-yu Guo [:shu]
	Comment 4 • 8 years ago
mozreview-review

Comment on attachment 8850848 [details]
Bug 1349989: Correctly handle errors for OMT-compiled scripts.

https://reviewboard.mozilla.org/r/123370/#review125754

::: js/xpconnect/tests/unit/test_compileScript.js:64
(Diff revision 1)
> +  // compilation.
> +  let scriptUrl = `data:,${";".repeat(1024 * 1024)}(`;
> +
> +  yield Assert.rejects(
> +    ChromeUtils.compileScript(scriptUrl),
> +    /SyntaxError: expected expression, got end of script/);

I think SyntaxError itself suffices, rather not have tests depend on the explicit wording of errors.

	Kris Maglione [:kmag] Assignee
	Comment 5 • 8 years ago

https://hg.mozilla.org/integration/mozilla-inbound/rev/518cd94b4580e11a76724b70c37ee0cf476de3b6
Bug 1349989: Correctly handle errors for OMT-compiled scripts. r=shu

	Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)
	Comment 6 • 8 years ago
bugherder

https://hg.mozilla.org/mozilla-central/rev/518cd94b4580