Re: Question on implementation of the language property

Paul,

> We would like to hear what the various XSL-FO implementations 
> accept for values of the language property, specifically, 
> whether 2 character language codes are accepted or rejected.

In RenderX XEP, there is no control on the syntax of language property.
The property value is taken as a key to select a correspondent 
hyphenation table in a configuration file. If a user decides to define 
a hyphenation table for Klingon or Elvish, it's up to him to choose 
a proper language tag for it. Needless to say, two-byte codes are accepted.

> At least some WG members believe the XSL spec should require
> the use of 3 character codes for the language property [skip] ...
>
> Others believe 2 character values (allowed by RFC 3066 and
> allowed as values for the xml:lang shorthand [4]) should also 
> be allowed as values for the language property.
> 
> Note that, in any case, both 2 and 3 character values are
> allowed for values of the xml:lang shorthand -- [skip]...

I would like to join the supporters of two-byte codes -
just because they are already allowed in xml:lang. 

A very serious problem of the current XSL FO spec is its lack
of coherency: things designed for XSL FO and CSS legacy
follow different logic. We have:

  "margin-top" but not "margin-before";
  "space-before" but not "space-top";
  "border-top" but not "border-before";
  "padding-before.conditionality" but not "padding-top.conditionality"

etc. etc. In each case, the user should learn by heart which restrictions
apply to each single group of properties. Excluding two-byte language
codes would add yet another item to this list. IMHO, the WG should 
either permit the same set of values in both xml:lang and language, or 
exclude xml:lang at all.

Best regards,
Nikolai Grigoriev
RenderX

Received on Tuesday, 23 July 2002 18:18:04 UTC