1038929 - Add Mochitests for CSS 2.1 spec section 9.7

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[sgbowen]

Section 9.7 of the spec for 2.1 indicates that display values must be set as shown in a table whenever position is absolute/fixed or float is left/right. We don't currently have a mochitest to ensure this happens correctly.

Comment 1

[sgbowen]

bz: As far as I can gather from your comment in bug 1021952, you think the row of the table for "other/same as specified" should be handled on a case-by-case basis, depending on the spec for the value?

In the case of ruby, this is currently unclear, but you think it should be block, correct?

I see no way around hardcoding a mapping from display values to what they should be converted to on float/position. If new display values are added, I can make it fail, so the person knows to update the test...is this okay?

Comment 2

[Boris Zbarsky [:bzbarsky]]

> you think the row of the table for "other/same as specified" should be handled on a
> case-by-case basis, depending on the spec for the value?

In the spec, yes.

> In the case of ruby, this is currently unclear, but you think it should be block, correct?

Modulo the good point Tab raised in the www-style thread....

> I see no way around hardcoding a mapping from display values to what they should be
> converted to on float/position.

Indeed.

> If new display values are added, I can make it fail, so the person knows to update the
> test...is this okay?

Yes, that's perfect.

Comment 3

[sgbowen]

Attachment: mochitest.patch v1

This should do it, assuming I understand everything correctly. The test passes for me locally.

Comment 4

[sgbowen]

Attachment: mochitest.patch v2

(assuming I remember to qrefresh the patch :) )

Comment 5

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8457436 [details] [diff] [review]
mochitest.patch v2

r=me.  Thanks!

Comment 6

[sgbowen]

Arguably successful try run:
https://tbpl.mozilla.org/?tree=Try&rev=bbd7ddb3a0d7

The failure is a known intermittent which popped back up recently (see bug 947565).

Comment 7

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

This landed with an incorrect bug number (1038928).

Landed in https://hg.mozilla.org/integration/mozilla-inbound/rev/08c5f5e266c5

Comment 8

[Daniel Holbert [:dholbert]]

https://hg.mozilla.org/mozilla-central/rev/08c5f5e266c5