Re: [xsl] Page-break/Static content

[Arved Sandstrom <Arved_37 at chebucto dot ns dot ca>]

Although solutions to the problem presented by Tobi Kamke have already been
introduced to this discussion (by Alexey Gokhberg), I have a few extra points.

One, the error message from FOP. This was doubtless confusing under the
circumstances, but indeed, according to the spec, you cannot have two flows in
any one page-sequence with the same flow name. In this context fo:flow and
fo:static-content are both flows, incidentally.

Two, Tobi tripped up on a FOP feature (that I must take the blame for) that I
hope people don't actually use until things in the spec have been resolved some
more. Namely, it is my position that the XSL spec is ambiguous as to whether a
page-sequence can only contain one flow, or more.

Here is an excerpt from section 6.4.5 of the spec:

<excerpt>
The default ordering constraint of section 4.7.1 (General Ordering Constraints)
does not apply to the fo:page-sequence. The default ordering constraints apply
to the child fo:flow objects; special ordering constraints apply to the child
fo:static-content objects.

Contents: (title?,static-content*,flow)
</excerpt>

The BNF grammar says one fo:flow; the paragraph before the BNF says multiple.
Which one is the typo? Who knows? There is no language anywhere else in the
spec that treats the multiplicity of fo:flow within a page-sequence.

Normally I would treat the BNF like the gospel itself - in this case I don't.
In particular I can think of situations (practical ones, not just pathological
examples) where the ability to have 2 flows in a page sequence would be pretty
handy.

In any case, the effect of this is - IF you want to experiment, you can
currently use multiple fo:flows in FOP, within one page-sequence. I would not
do production stuff with any more than one fo:flow per page sequence, since
none of the other XSL-FO implementations support this viewpoint.

Regards,
Arved Sandstrom

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list