- From: C. M. Sperberg-McQueen <cmsmcq@blackmesatech.com>
- Date: Fri, 09 Aug 2024 09:16:12 -0600
- To: iXML Community Group <public-ixml@w3.org>
Some of the talks and discussions at Balisage last week have made me wonder: would it be a good idea to set about adding material to our existing collection of sample iXML grammars? - a Relax NG schema (and optionally a DTD and XSD schema) for the XML output of the parser? - an XSLT stylesheet (or an iXML grammar) to re-serialize the XML in the original notation? - rudimentary documentation of the XML vocabulary used in the parser output produced by the sample iXML grammar? All of these may be helpful when people use iXML to move data in some non-XML notation into XML to work with it, especially if they want to use XML tools to modify the original data and write it back in the original notation. That may not always be terribly likely for the sample grammars, but it may be worth while to make people aware that these are things they may want to do for their own grammars. A Relax NG schema can be generated from the vXML form of a grammar; the ixml-to-rng.xsl stylesheet in my Gingersnap library provides a proof of concept, and I have recently worked out ways to improve it (a discussion of what is and is not feasible may be found at [1]). I believe several people are working on automatic generation of transformations to go from vXML back to the notation described by an iXML grammar, but I believe that most existing transforms of that class have been constructed manually. Automatic generation of skeleton documentation is on my to-do list (but it has been there for a while, so I make no promises -- real soon now). Does it make sense to use the sample directory to show not just iXML grammars but also tools for working with the output produced by the iXML processor? Michael [1] https://github.com/cmsmcq/gingersnap/blob/main/doc/ixml-to-rng.org -- C. M. Sperberg-McQueen Black Mesa Technologies LLC http://blackmesatech.com
Received on Friday, 9 August 2024 15:46:34 UTC