Re: [XPath20] Decoding URIs

> Scenario: I get an XML document that links relatively to images besides
> it using xlink:href. The images' names may contain diacritics.

within the document, the URI's should not be %hh escaped, they should
(or at least can) just contain the uniocde characters. The xlink
processor (like an xsl processor) is supposed to %hh-encode the strings
before passing the genererated URI down the wire.

Of course if you don't have control over your input documents this may
not help, and as you say the dull part would be writing the utf8 decoder
in xsl2, unless you use an extension function of course, in which case
you could probably just escape to the host languages uri-decoding
library...


David

________________________________________________________________________
This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________