282281 - should nsIDOMCSS2Properties.idl and nsIDOMElementCSSInlineStyle.idl be frozen?

Status: RESOLVED FIXED

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

Description

[Christian :Biesinger (don't email me, ping me on IRC)]

those two interfaces come from DOM2 Style
(http://www.w3.org/TR/DOM-Level-2-Style/css.html#CSS-ElementCSSInlineStyle)
which is a W3C REC.

so, shouldn't the interfaces be marked @status FROZEN in the moz code?

http://lxr.mozilla.org/seamonkey/source/dom/public/idl/css/nsIDOMElementCSSInlineStyle.idl
http://lxr.mozilla.org/seamonkey/source/dom/public/idl/css/nsIDOMCSS2Properties.idl

Comment 1

[Darin Fisher]

Please add nsIDOMCSS2Properties to the list for consideration, and it would
appear that nsIDOMCSSStyleDeclaration.idl is already frozen.

Comment 2

[Darin Fisher]

nevermind, looks like it was just a typo in the bug summary.

Comment 3

[Boris Zbarsky [:bzbarsky]]

This has come up before....  In brief, the DOM CSS stuff is rather badly
designed (in fact, parts of it are incompatible with CSS).

We can freeze nsIDOMElementCSSInlineStyle, I think, but I'm not sure about
nsIDOMCSS2Properties, since the DOM interface (based on CSS2) doesn't match what
we implement (CSS2.1).

Comment 4

[Darin Fisher]

> ...I'm not sure about
> nsIDOMCSS2Properties, since the DOM interface (based on CSS2) doesn't match what
> we implement (CSS2.1).

Is there an alternate version defined by CSS2.1?  Should we implement that
version of the interface instead?

Comment 5

[Boris Zbarsky [:bzbarsky]]

There isn't yet.  In general, some of the DOM CSS interfaces are due to be
withdrawn from REC status, last I checked.  When that will happen, I don't know.

Comment 6

[Hixie (not reading bugmail)]

Nobody apart from me has really indicated any interest in doing the work, and my
WHATWG work has priority. So ETA is not this year.

Comment 7

[Brian Ryner (not reading)]

(In reply to comment #3)
> We can freeze nsIDOMElementCSSInlineStyle, I think, but I'm not sure about
> nsIDOMCSS2Properties, since the DOM interface (based on CSS2) doesn't match what
> we implement (CSS2.1).

Any objections to freezing nsIDOMCSSInlineStyle as-is, then?

Comment 8

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

If it matches the DOM spec, freezing it is OK with me.

Comment 9

[Boris Zbarsky [:bzbarsky]]

OK with me too.

Comment 10

[Brian Ryner (not reading)]

Attachment: freeze it

Comment 11

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 183122 [details] [diff] [review]
freeze it

Looks good.

Comment 12

[Brian Ryner (not reading)]

Comment on attachment 183122 [details] [diff] [review]
freeze it

requesting approval to land

Comment 13

[Mike Shaver (:shaver emeritus)]

Comment on attachment 183122 [details] [diff] [review]
freeze it

a=shaver for SDK improvements.

Comment 14

[Brian Ryner (not reading)]

Comment on attachment 183122 [details] [diff] [review]
freeze it

checked in; I'm leaving the bug open since we haven't decided what to do about
nsIDOMCSS2Properties.

Comment 15

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

WONTFIX, now we unfroze everything?

Comment 16

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

Well, it was fixed, really, and we treated nsIDOMCSS2Properties as frozen despite the silliness of using it from C++ in the first place (when the whole point of the interface was shortcuts that were useful for JS).