Subject: Re: [xsl] How can I compare XSLT with Programming languages for parsing XML? From: David Carlisle <davidc@xxxxxxxxx> Date: Mon, 24 Jul 2006 15:27:36 +0100 |
> Hi, > > I am developing XSLT stylesheets to parse > MathML(Mathematical markup language) and convert it to > text. I would like to justify my design decision of > selecting XSLT and not another programming language, > what benefits XSLT has over others for parsing XML > documents? > > Dima > XSLT (which as others have said, isn't a parser) is ideally suited for this. It is of course a good general purpose XML transformation language but as it happens transforming MathML was one of the explicit use cases looked at when designing XSLT. MathML 1 and XSLT 1 were designed around the same time frame and we had at least one joint phone conference between the XSL and Math Working groups where requirements for processing XML documents like MathML were discussed. (One notable feature MathML shares with some other vocabularies is that processing is often controlled by the first child rather than the parent, it's <apply><sin/><mi>π</mi></apply> rather than <sin><mi>π</mi></sin> some other transformation/querying languages (including early drafts of xpath which didn't have the sibling axes in quite the same form) make the first form rather harder to deal with than the second. Incidentally a common reason for transforming MathML to text is as a first stage in an audio rendering of mathematical expressions, to generate input for a standard speech system. If that's your motivation you may want to look at some similar projects in that area, many of which do use XSLT. http://www.w3.org/Math/Software/mathml_software_cat_converters.html David
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: [xsl] How can I compare XSLT wi, Michael Kay | Thread | [xsl] How to get innermost ancestor, J. S. Rawat |
[xsl] listing elements & sub-elemen, Karl | Date | Re: [xsl] listing elements & sub-el, David Carlisle |
Month |