This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 29594 - Clarification of font-variant-ligatures CSSOM serialized read values
Summary: Clarification of font-variant-ligatures CSSOM serialized read values
Status: NEW
Alias: None
Product: CSS
Classification: Unclassified
Component: Fonts (show other bugs)
Version: unspecified
Hardware: PC Linux
: P2 normal
Target Milestone: ---
Assignee: John Daggett
QA Contact: public-css-bugzilla
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-04-25 12:01 UTC by Dominik Röttsches
Modified: 2016-04-25 15:20 UTC (History)
3 users (show)

See Also:


Attachments

Description Dominik Röttsches 2016-04-25 12:01:42 UTC
The CSSOM describes the CSS wide serialization steps, including the serialization of shorthand properties.

https://drafts.csswg.org/cssom/#serialize-a-css-value

In it:

"To serialize a CSS component value depends on the component, as follows:"

This list does not contain any font, or font variant specific rules, so it seems like we are missing a clarification for what should be the canonical serialization of font-variant-ligatures.

When writing to the fontVariantLigatures style property, or when specifying a font-variant-ligatures style attribute in a CSS style declaration, and specifying "none" as the value, the serialization for the read value could be either 
* "none" or 
* "no-common-ligatures no-discretionary-ligatures no-historical-ligatures no-contextual".

Which should be the canonical serialized value? How can we incorporate this clarification into the spec? Thanks.
Comment 1 Tab Atkins Jr. 2016-04-25 15:20:43 UTC
The general (sadly, unstated) rule is to prefer shorter representations when possible.  Nobody's taken the time to go thru implementations and document the exceptions to that policy (and figure out if they can be changed), so we haven't specced it normatively.

But the point stands - definitely prefer "none".