Subject: RE: CSS, XSL and MathML. Some questions. From: Dan Hable <DHable@xxxxxxxxxxxxx> Date: Thu, 2 Jul 1998 15:32:46 -0500 |
<snip> >Note 2. The actual transformation of the trees requires some sort of >non-trivial scripting / programming capability. This is what I mean. MathML will require more programming features than ECMAScript will be able to provide. Does this mean we should change the specs for this one application? IMHO, I wouldn't do it. I'm not so sure if XSL does (or will) provide processor instructions. A processor instruction, or PI, is a feature from SGML that is ignored by the parser but can be understood by another program. For example, If I had this SGML document: <heading>This is my heading</heading> <math-statment><?MathML format = "Trig"> x = sine( theta)</math-statment> The parser would parse all of the data but where <?MathML format="Trig"> appears, it would dump this command into the document itself. Another program, like a view, would then come by and pick out the tag and do the processing. XML does incorporate this idea of PI's but the format is slightly different. The same PI in XML format would be: <?MathML format = "Trig" ?>. You may remember this from somewhere. How about <?XML version="1.0"?>? That's right. XML requires an XML processor instruction. This allows XML documents to be parsed using a SGML parser. It is quite clever and does make everything compatable. Using a set of processor instructions in XSL would: 1) Keep the same style for writting the stylesheets. Everything would still look like HTML lags instead of Schema. 2) Introduce a new set of more complex processing options. This is the goal of all this bandwidth, right? 3) Keep the additional commands in the document in such a way that if a viewer didn't understand the commands it could just skip the commands. Point 3 would be the major benfit I would see to using PI's in the stylesheet. If everyone started complaining about XSL not being powerful enough, the W3C, or Microsoft and Netscape, could add additional commands without breaking the system. My thoughts on the issue. ~~ Dan ~~ XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
CSS, XSL and MathML. Some questions, Stan Devitt | Thread | Re: CSS, XSL and MathML. Some quest, Martin Bryan |
Re: `High-level' format specificati, Brandon Ibach | Date | RE: XSL and IE5, Jonathan Marsh |
Month |