Re: [xslt] xsltproc memory consumption w/ large DTD / docbook

[Sebastian Rahtz <sebastian rahtz computing-services oxford ac uk>]

>  but not every time you process it. running a transformation
> > to XSLT does not need a DTD, after all. the problem
> 
>   In that case, yes it does. They make use of entities defined
> in the DTD.

which, I think, points up a non-unknown problem in XML :-}


>   DocBook stylesheets also use unparsed-entity-uri() and that
> requires knowledge of the attribute type defined only in the DTD.

surely not so? a simply well-formed document <x foo="bar"/>
can have an XSLT process it under the assumption that "bar" is
an entity (defined in the DTD subset)

> > it seems to me that anyone building a web site by defining
> > a giant single virtual document can invest a little time
> > in a Makefile, or the like.....
> 
>   That's true. But the problem is usually to maintain cross
> references over the whole set of document like indexes. Currently
> unless designing some sort of incremental database, it's impossible
> to do that directly with XSLT without requiring to process the 
> whole set at each build.

oh, sure, its one perfectly valid technique for maintaining a web site;
all I was saying was that if you work like that, you won't mind
building a small chain of tools.

It would be equally possible to do the job by using a database
of links. so you say

 <pointer to="bar"/>

and have "bar" be an entry in a lookup table in an XML file
which defines the location. but I suppose thats what you had
in mind by "incremental database".

a third technique is to have a single virtual document which
pulls in all the others using XInclude. that way the DTD only
gets read once.
-- 
Sebastian Rahtz      OUCS Information Manager
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431