At 08:42 2003 04 04 +0100, David(_dot_)Pawson(_at_)rnib(_dot_)org(_dot_)uk
wrote:
Ken Holman said:
At 2003-04-03 18:51 +0200, you wrote:
<fo:region-before display-align="before" extent="10mm"/>
<fo:region-body margin-top="10mm"/>
BTW, tools may accept the order above, but they do so
improperly. Another
tool might reject the above because the order implied by the
Recommendation
is that the body is specified first.
Section 6.2 states the content is described using
content-model syntax, but
doesn't say that the order as expressed using that syntax can be
violated. Section 6.4.12 explicitly orders region-body first.
I queried this the other day;
Question to the WG, is the order <emph>Required</emph>
as per 6.4.12
Contents:
(region-body,region-before?,region-after?,region-start?,region-end?)
Where did you send this query? I can't find it on the xsl-editors
archive at http://lists.w3.org/Archives/Public/xsl-editors/2003JanMar/.
However, see http://www.w3.org/Style/XSL/2003/01/FO-DoC#02OctDec-0063
for the answer to Eliot Kimber who basically asked the same question.
It is an error if an XSL-FO result tree it doesn't conform
to the content model restrictions in the XSL spec.
"It is an error" means that an implementation may (but need not)
issue a message or warning and may (but need not) recover in some
implementation dependent way.
In this case, I understand many (most?) implementations choose to
recover in the obvious way. But Ken is right that you may someday
come across an implementation that rejects such a document/stylesheet,
and it would be well within its right to do so.
paul