798020 - Assertion failure: "Computed width should always be constrained, so horizontal flex containers should always have a constrained main-size"

Status: RESOLVED FIXED

(Core :: Layout, defect)

Description

[Jesse Ruderman]

Attachment: testcase (asserts fatally when loaded)

With
  user_pref("layout.css.flexbox.enabled", true);
the testcase triggers:

Assertion failure: !IsAxisHorizontal(aAxisTracker.GetMainAxis()) (Computed width should always be constrained, so horizontal flex containers should always have a constrained main-size), at layout/generic/nsFlexContainerFrame.cpp:1794

Comment 1

[Jesse Ruderman]

Attachment: stack

Comment 2

[Daniel Holbert [:dholbert]]

Comment on attachment 668142 [details]
testcase (asserts fatally when loaded)

><body style="letter-spacing: 693626589697em;">

Oh, Jesse. :)

So this looks like a case where our actual computed width is large enough to overflow NS_UNCONSTRAINEDSIZE, so it looks like we've got an unconstrained computed-width, and we're not expecting that.

We should probably just change this to a NS_WARNING instead of a MOZ_ASSERT, since web content can trigger this (with ridiculously huge values) and it's not a problem, it'll just give incorrect layout. (we'll shrinkwrap instead of honoring the gigantic height)

Comment 3

[Jesse Ruderman]

Sounds reasonable.  If we ever do bug 765861, we can turn it into a conditional assert.

Comment 4

[Daniel Holbert [:dholbert]]

Attachment: fix

Comment 5

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Comment on attachment 668826 [details] [diff] [review]
fix

r=dbaron

Comment 6

[Daniel Holbert [:dholbert]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/e03f63abb826

Comment 7

[Daniel Holbert [:dholbert]]

(unprefixed testcase code before pushing, per bug 801098)

Comment 8

[(no longer active)]

https://hg.mozilla.org/mozilla-central/rev/e03f63abb826