This is the mail archive of the xsl-list@mulberrytech.com mailing list .


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: XSLT and Text Processing Languages




Paul Terray wrote:
> XSL-T advantages :
> - Much more powerful for tree querying (DOM oriented): you cannot ask for a
> node that is not in your ancestors with OmniMark, you cannot parse twice
> the same node... Just tocs are way simpler with XSL-T

Actually, OmniMark referents were made for just this sort of thing (i.e.
outputting data that you haven't encountered yet), and you can parse
twice if you want by using the do xml-parse command.

> - Very natural way of outputing XML and HTML (in OmniMark, you output the
> string of the tag, where in XSLT, you are sure the XML you output is
> well-formed)

You can always reparse the output to ensure it is well-formed, but I
agree, the template approach of XSLT is easier here.

> - XML-native : the DTD you are using with OmniMark can be SGML-tainted,
> there would be no reaction at all, whereas XSL-T is very strict with that

OmniMark 6 does away with DTDs entirely. In fact, you now have to pay if
you want DTD support. 

> Conclusion :
> - For XML->XML or XML->HTML conversion, which is more tree transformation,
> prefer XSLT

For me, just XML -> HTML, until XSLT supports XML 1.0 fully (i.e. entity
and notation support in the data model) and matures enough to handle
large documents gracefully. 

> - For any other format -> XML, Omnimark is mandatory

Actually, I prefer Java for Database -> XML conversions, though the
OmniMark folks would probably roll over and die if they heard me say
that :-).

Rick Geim
National Semiconductor
rick.geimer@nsc.com


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

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]