CSS properties taking an -epub- prefix
Edward O'Connor
I'm generating a complete list of the CSS properties which will need to take an -epub- prefix. This might be useful to other implementers, so here's what I've come up with. Did I miss anything?
Thanks,
Ted
* 3.3.2 CSS 2.0
None.
* 3.3.3 CSS 3.0 Speech
-epub-cue
-epub-cue-after
-epub-cue-before
-epub-mark
-epub-mark-after
-epub-mark-before
-epub-pause
-epub-pause-after
-epub-pause-before
-epub-phonemes
-epub-rest
-epub-rest-after
-epub-rest-before
-epub-speak
-epub-voice-balance
-epub-voice-duration
-epub-voice-family
-epub-voice-pitch
-epub-voice-pitch-range
-epub-voice-rate
-epub-voice-stress
-epub-voice-volume
* 3.3.4 CSS Fonts Level 3
None. (@font-face and its properties reached CR in CSS2.)
* 3.3.5 CSS Text Level 3
-epub-hyphens
-epub-line-break
-epub-text-align-last
-epub-text-emphasis
-epub-text-emphasis-color
-epub-text-emphasis-style
-epub-text-transform
-epub-text-trim
-epub-word-break
N.B. While in general we recommend EPUB content provide both prefixed and unprefixed rules, content using -epub-text-trim must not include an unprefixed text-trim rule.
* 3.3.6 CSS Writing Modes
-epub-writing-mode
-epub-text-orientation
-epub-caption-side
-epub-text-combine
* 3.3.7 Media Queries
None.
* 3.3.8 CSS Namespaces
@-epub-namespace
* 3.3.9 CSS Multi-Column Layout
-epub-break-after
-epub-break-before
-epub-break-inside
-epub-columms
-epub-column-count
oeb-column-number
-epub-column-fill
-epub-column-gap
-epub-column-rule
-epub-column-rule-color
-epub-column-rule-style
-epub-column-rule-width
-epub-column-width
* 3.3.10 Ruby Positioning
-epub-ruby-position
N.B. Like -epub-text-trim, content using -epub-ruby-position must not include an unprefixed ruby-position rule.
--
Edward O'Connor
Apple Inc.
Daniel Weck
On 18 Apr 2011, at 18:05, Edward O'Connor wrote:
> * 3.3.3 CSS 3.0 Speech
> -epub-mark
> -epub-mark-after
> -epub-mark-before
The "mark" feature has been removed in the latest (2011) Working Draft
which is about to be officially published by the W3C. This is one of
the discrepancies compared to the 2004 WD.
http://dev.w3.org/csswg/css3-speech
http://www.w3.org/TR/css3-speech
> -epub-speak
The 'speak' property has now been split into 2 distinct ones: 'speak'
and 'speakability'. The functionality remains essentially the same
though.
> -epub-phonemes
This feature is at risk of being removed in the latest WD, but it
effectively still features in the specification at this point in time.
Regardless, I think that the @phonetic-alphabet at-rule would need to
be prefixed as well.
Note that the current EPUB3 specification makes use of the ssml:ph
attribute (in XHTML Content Documents) as a replacement for the CSS3-
Speech "phonemes" functionality, based on the rationale that
pronunciation rules are more tightly-related to content than to style.
> -epub-cue
> -epub-cue-after
> -epub-cue-before
> -epub-pause
> -epub-pause-after
> -epub-pause-before
> -epub-rest
> -epub-rest-after
> -epub-rest-before
> -epub-voice-balance
> -epub-voice-duration
> -epub-voice-family
> -epub-voice-pitch
> -epub-voice-pitch-range
> -epub-voice-rate
> -epub-voice-stress
> -epub-voice-volume
Correct.
Note that overall, there has been very little change (if any) in terms
of the behaviors of the aforementioned properties since the 2004
Working Draft. There is a risk that the 'speakability' property gets
renamed, but the functionality is unlikely to be disputed (as far as I
can tell). The likelihood of the CSS3-Speech Module to remain stable
is high, in my opinion (touch wood).
Regards, Daniel
Peter Sorotokin
Thank you for putting this list together and sharing it with the group!
A couple of comments:
CSS Namespaces reached CR: http://www.w3.org/TR/css3-namespace/ and we
(Adobe) have already implemented a subset of it in our EPUB renderer. Also,
browsers already implemented it without prefix.
CSS Multicolumn layout also reached CR recently, so those properties should
not be prefixed.
I also wanted to reiterate that if EPUB3 defines prefixed properties, we'll
have to carry them forward. I do not envision content is being necessarily
upgraded merely because another CSS spec reaches CR.
Also, as I mentioned at the call, I think that suggestion of including both
prefixed and unprefixed versions of the properties is misleading and should
be dropped. From the minutes it seems that CSS WG does not want to
proliferate the content with unprefixed properties until a spec reaches CR.
Encouraging authors to use them would do exactly that.
Peter