786946 - text-decoration: line-through should not be painted on a <select> child

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Deleted User]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:15.0) Gecko/20100101 Firefox/15.0
Build ID: 20120821170930

Steps to reproduce:

Outlined in this StackOverflow post: 
http://stackoverflow.com/questions/12191753/css-overriding-an-inherited-text-decorationline-through#comment16324982_12191753  

Demo at this fiddle: http://jsfiddle.net/Y7ZVp/


Actual results:

Line through remains on child element with parent text-decoration: line-through even if child has text-decoration: none


Expected results:

The child should return to normal text.

Comment 1

[(mostly gone) XtC4UaLL [:xtc4uall]]

Attachment: Reporter's Testcase

Comment 2

[(mostly gone) XtC4UaLL [:xtc4uall]]

I see no Difference to other Browsers.

Comment 3

[Ioana (away)]

Reproducible on the 09/04 Firefox 18 Nightly on Mac and Linux (didn't try on Windows).

As XtC4UaLL says, the same issue reproduces in Safari, Chrome, and Opera.

Comment 4

[Boris Zbarsky [:bzbarsky]]

And that's because the spec says to do this.

Marking invalid, but if someone has the time to find the oldest report on this and mark duplicate, feel free!

Comment 5

[Boris Zbarsky [:bzbarsky]]

Though wait.  I don't see the strike-through going through the select on Chrome on Mac.  Does it do so on Windows?

Comment 6

[Boris Zbarsky [:bzbarsky]]

Resummarizing to cover the actual issue, not the incorrect initial description.

David, I think we've considered not propagating text-decoration to "replaced" elements in the past...  Thoughts?

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

(In reply to Boris Zbarsky (:bz) from comment #6)
> Resummarizing to cover the actual issue, not the incorrect initial
> description.
> 
> David, I think we've considered not propagating text-decoration to
> "replaced" elements in the past...  Thoughts?

In some cases we've changed their default display (or that of an anonymous child) from 'inline' to 'inline-block' to cause this (https://hg.mozilla.org/mozilla-central/rev/3a73e65c935e ; https://hg.mozilla.org/mozilla-central/rev/c9dff8b867f0).  If we need to do it without that corresponding change in 'display' it might be reasonable as well.

Comment 8

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

(And I can't think of anything left other than select, after those two changes, actually.)

Comment 9

[Boris Zbarsky [:bzbarsky]]

OK, well UA behavior for this is as follows:

Default display:
  WebKit: No decorations
  Presto: No decorations
  Trident: No decorations
  Gecko: Decorations

display:inline:
  WebKit: Decorations on combobox, not listbox
  Presto: No decorations
  Trident: Decorations in standards mode and IE8 mode.  
           No decorations in IE7 and quirks modes.
  Gecko: Decorations

display:inline-block:
  WebKit: No decorations
  Presto: No decorations
  Trident: No decorations
  Gecko: No decorations

So just flipping the default display will give us the IE8/IE9 behavior and at least match WebKit on combobox.  Sounds like we should do that.

Comment 10

[Boris Zbarsky [:bzbarsky]]

Attachment: Don't propagate text-decoration to <select> by default.

Comment 11

[Boris Zbarsky [:bzbarsky]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/8d12416121a6

Comment 12

[Boris Zbarsky [:bzbarsky]]

And backed out, because it breaks the font-inflation tests for <select>.  David, is that a bug in the patch, or in the tests?

Comment 13

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Probably the patch, since those tests were written based on desired behavior.

Try adding the single line content->IsHTML(nsGkAtoms::select) to the function IsFontSizeInflationContainer in nsFrame.cpp (next to option and optgroup).

Comment 14

[Boris Zbarsky [:bzbarsky]]

Yeah, that works.  I'm going to reland with that change.

Comment 15

[Boris Zbarsky [:bzbarsky]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/175c768112a4

Comment 16

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/175c768112a4