[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: XML aggregation question?

• From: "Michael Kay" <mike@xxxxxxxxxxxx>
• To: "'Philippe Poulard'" <Philippe.Poulard@xxxxxxxxxxxxxxx>
• Date: Mon, 28 Aug 2006 21:09:01 +0100

> 
> Just a comment about the syntax :
> 
> -Saxon's syntax :
> collection('file:///c:/xslts_1_0_0/TestInputs?select=*.xsl;rec
> urse=yes')
> -Ant's syntax :
> <fileset dir="file:///c:/xslts_1_0_0/TestInputs" includes="**/*.xsl"/>
> -Why not using definitively the XPath syntax ?
> file('file:///c:/xslts_1_0_0/TestInputs')//*[ends-with(name(.)
> ,'xsl')])
> or
> file('file:///c:/xslts_1_0_0/TestInputs')//*[@extension='xsl']
> or whatever à la XPath ?

I adopted the approach of defining a URI that can be passed as a parameter
to the collection function, using syntax that's compatible with the rules
and conventions of what can go in a URI.

You seem to be looking at a more ambitious approach of modelling the
filestore directory structure as an XML instance, to which you can apply
XPath expressions. Good in theory, but I think there are a lot of practical
issues such as incompatible syntax for names.

Michael Kay
http://www.saxonica.com/

• Follow-Ups:
  • Re: XML aggregation question?
    • From: Philippe Poulard

• References:
  • Re: XML aggregation question?
    • From: Philippe Poulard