1663867 - Implement margin layout for children of nsMathMLContainerFrame

Status: RESOLVED DUPLICATE of bug 1908069

(Core :: MathML, enhancement)

Description

[Delan Azabani]

+++ This bug was initially created as a clone of Bug #1658121 +++

Comment 1

[Delan Azabani]

Attachment: Bug 1663867 - Implement margin layout for children of nsMathMLContainerFrame.

This patch doesn’t try to handle width/height or similar properties ([[ https://github.com/mathml-refresh/mathml/issues/45 | mathml-refresh/mathml#45 ]]) or non-horizontal writing modes ([[ https://bugzilla.mozilla.org/show_bug.cgi?id=1077525 | bug 1077525 ]]).

• [[ https://phabricator.services.mozilla.com/D87594?id=328558 | Diff 1 ]] is based on [[ https://phabricator.services.mozilla.com/D86471?id=327743 | Diff 5 ]] + [[ https://phabricator.services.mozilla.com/D86471?id=327965 | Diff 6 ]] of D86471
  • [[ https://phabricator.services.mozilla.com/D86471?id=327743 | Diff 5 ]] implements margin well enough for <mspace> in nsMathMLContainerFrame::Place
  • [[ https://phabricator.services.mozilla.com/D86471?id=327965 | Diff 6 ]] fixes the incorrect handling of <percentage> + auto values
  • 6/6 [[ http://wpt.live/mathml/relations/css-styling/padding-border-margin/margin-001.html | margin-001.html ]] + 0/73 [[ http://wpt.live/mathml/relations/css-styling/padding-border-margin/margin-002.html | margin-002.html ]] + 5/16 [[ http://wpt.live/mathml/relations/css-styling/padding-border-margin/margin-003.html | margin-003.html ]] (was 0 + 0 + 0)
  • would be 1/73 [[ http://wpt.live/mathml/relations/css-styling/padding-border-margin/margin-002.html | margin-002.html ]] if not for [[ https://bugzilla.mozilla.org/show_bug.cgi?id=1658135 | bug 1658135 ]] (see [[ https://bucket.daz.cat/d9f15b512873442b.patch | this WPT patch ]] for workaround)

Comment 2

[Delan Azabani]

My internship with Igalia is coming to an end, so I won’t have any more time to work on this bug. My unfinished patch (D87594) is now welcome to anyone interested in picking it up.

Comment 3

[Frédéric Wang (:fredw)]

I'm experimenting a different approach on bug 1908069 that relies on the existing ReflowInput.ComputedPhysicalMargin(), so hopefully that can address the concerns raised during review.