1196887 - display:inline should compute to inline-block any time the block-direction differs from the containing block's block-dir

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Jonathan Kew [:jfkthame]]

In bug 1134849, we made display:inline compute to inline-block for elements whose writing-mode is orthogonal to their containing block's mode.

However, this isn't quite right, according to the CSS Writing Modes draft: we should do this in any case where the block directions differ, not only when they're orthogonal. (In particular, this behavior should apply to the vertical-rl / vertical-lr pair of modes.)

(If sideways-lr/sideways-rl are accepted by the WG, I expect this will also apply to vertical-* vs sideways-* modes; i.e. any time the computed writing-mode changes, we should map inline to inline-block.)

Comment 1

[Jonathan Kew [:jfkthame]]

Attachment: Compare the writing-mode property, not only whether it is horizontal or vertical, when deciding whether to compute display:inline as inline-block.

Comment 2

[Daniel Holbert [:dholbert]]

Comment on attachment 8650704 [details] [diff] [review]
Compare the writing-mode property, not only whether it is horizontal or vertical, when deciding whether to compute display:inline as inline-block.

Looks good! Could you include a reftest? (with a vertical-lr inline element inside of vertical-rl, or something like that. I expect you can set a large 'width' and/or 'height' on the inline element, and if we honor them, then that means we're treating it as an inline-block and we should pass the test.)

r=me with that

Comment 3

[Jonathan Kew [:jfkthame]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/d4a9f12d3b9f540c23254e4df5c14308d4589d59
Bug 1196887 - Compare the writing-mode property, not only whether it is horizontal or vertical, when deciding whether to compute display:inline as inline-block. r=dholbert

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/d4a9f12d3b9f