xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrzej Dmoch <...@step.pl>
Subject Re: AW: Reverse XSL:FO
Date Wed, 26 Apr 2000 17:26:59 GMT
Guy Hulbert wrote:

> On Wed, 26 Apr 2000, Andrzej Dmoch wrote:
>         <snip>
> AD> I don't agree with you. Coding in XML is one thing. Transforming to PDF
> What do you mean "agree".  Postscript (and PDF, I presume) is a programming
> language.  XML is not.  To get something equivalent to postscript you would
> need XML, a DTD, a stylesheet PLUS something else.  For example, you can
> write a postscript file which will display a mandelbrot set by computing
> all the points in it.  Possibly XSL is this powerful.  I understand that CSS
> is not (Jon Bosak published a commentary to this effect).

I mean syntactically isomorfic. Refer to TeXML at http://www.alphaworks.ibm.com/
for the XML format syntactically isomorfic with TeX files (which is also a kind of
programming language). If you have such format you can write one program to
convert data from "PDF isomorfic XML format" to PDF, (just one parsers
serialization method) and use all XML/XSLT power to convert any XML data to PDF.
One of transformation steps can be XSL:FO.

> AD> isomorfic format from XSL:FO using XSLT written once by someone is second
> So, it's not a simple matter of "transforming to PDF".  It can't be isomorphic
> in the mathematical sense because the functionality is not equivalent.

Syntactically - it can be, just like for example there can be XML format
syntactically isomorfic with Java source files.

> But within the limitations of the particular stylesheet specification.  I
> understand that CSS can't even number sections.

I meant XSLT stylesheet. It can do very advanced things (it is in fact a nice
declarative programming language).


View raw message