137686 - getPropertyPriority does not work for shorthand properties

Status: RESOLVED FIXED

(Core :: DOM: CSS Object Model, defect, P4)

Description

[Boris Zbarsky [:bzbarsky]]

nsDOMCSSDeclaration::GetPropertyPriority uses
nsCSSDeclaration::GetValueIsImportant, which uses nsCSSDeclaration::GetValue,
which does not handle shorthand properties.  So the priority of a shorthand
property always seems to be "normal".

Since a shorthand can fail to have a well-defined priority, I feel that this
part of the spec could use some clarification....

Comment 1

[Chris Pimlott]

Attachment: Simple testcase

Comment 2

[Boris Zbarsky [:bzbarsky]]

oops.  This was supposed to go to glazou

Comment 3

[Daniel Glazman (:glazou) (not active in Mozilla any more)]

> Since a shorthand can fail to have a well-defined priority, I feel that this
> part of the spec could use some clarification....

I disagree with that. A shorthand is "specified" is all the corresponding longhands
are set with all same priority. In that case, the priority is the common priority
of the longhands. If all corresponding longhands are not set or have different
priorities, then the shorthand is not set and its priority has no meaning.

Anyway, accepting bug.

Comment 4

[Boris Zbarsky [:bzbarsky]]

> then the shorthand is not set and its priority has no meaning

This is the case I was referring to.  In this case, returning "" for the
priority seems incorrect.  Throwing an exception seems harsh.  So the question
is, "What is reasonable behavior, and is this something the DOM spec should
address?  Or should this be implementation-dependent?"

Comment 5

[Kevin McCluskey (gone)]

Bulk moving P1-P5 un-milestoned bugs to future. 

Comment 6

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

This was fixed in https://hg.mozilla.org/mozilla-central/rev/7356c512e9e1 to what I consider a reasonable behavior -- we report "important" if all subproperties of the shorthand are set and important, and otherwise we report "".

I don't think it's worth having an open bug on this unless there's a specification that says something to the contrary.  It may, however, be worth opening a bug on the specification, which is currently http://dev.w3.org/csswg/cssom/#dom-cssstyledeclaration-getpropertypriority

Comment 7

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

I sent http://lists.w3.org/Archives/Public/www-style/2011Apr/0331.html