Re: Using an XSL Formatter as an XSL-FO Web Browser

At 16:24 2001 02 13 GMT, David Carlisle wrote:
>[Someone else had written]:
>> It also allows shorthand properties
>
>It's a shame that these were inherited from CSS. I can understand why
>XSL should share a model with CSS, but I don't really see why it should
>share this particular bit of syntax. Which rather complicates the
>inheritence model as you can not (conceptually) just look up the
>ancestor:: axis to find the place where the property was set. You have to
>decode shorthands, each with some ad hoc syntax. (supporting this in
>xmltex would be a major complication).

The XSL spec explains how shorthands map to XSL properties.
I would expect any implementation to do the mapping first,
and then the inheritance happens with the (non-shorthand)
XSL properties as always.  So the only implementation overhead 
would be to support the mapping, not to do any more during
inheritance.

That said, CSS shorthands are in the "complete" conformance 
level (that's even more than the "extended" level).  I would 
think an XSL-FO stylesheet writer would be foolish to assume 
their stylesheet would be interoperable if they used shorthands.
I certainly don't expect my XSL-FO implementation will support
most shorthands in the near future (though, of course, user
demand will be the final determinant of that).

paul

Received on Tuesday, 13 February 2001 11:48:08 UTC