- From: Jirka Kosek <jirka@kosek.cz>
- Date: Wed, 10 Oct 2012 14:38:13 +0200
- To: Yves Savourel <ysavourel@enlaso.com>
- CC: public-multilingualweb-lt@w3.org
- Message-ID: <50756C35.3050408@kosek.cz>
On 10.10.2012 14:25, Yves Savourel wrote:
> I think it's a good idea. But it seems a bit "un-formal" as there
> seems to be nothing in http://www.w3.org/TR/xmlschema-2/#regexs that
> requires implementations to have such flag. So there is no guaranty
> that . can be used like that.
XML schema solves this by three string types -- xs:string,
xs:normalizedString and xs:token -- each type has different level of
whitespace normalization applied before xs:pattern is evaluated. So
effectively if you want . to include \n and \r you can simply use
xs:token instead of xs:string and you are done.
But such flag should be available in all sensible implementations, see:
http://www.w3.org/TR/xquery-operators/#flags
> Another option would be to use a range with all Unicode characters,
> but obviously we can't either because we can't specify the lower
> bound as it would be an invalid XML character.
Yes, that's another possibility.
> So I think your solution ("." + a bit about . including \r\n) is the
> best Jirka.
+1
Yves, will you please implement this change in spec as you were
authoring this data category?
Jirka
--
------------------------------------------------------------------
Jirka Kosek e-mail: jirka@kosek.cz http://xmlguru.cz
------------------------------------------------------------------
Professional XML consulting and training services
DocBook customization, custom XSLT/XSL-FO document processing
------------------------------------------------------------------
OASIS DocBook TC member, W3C Invited Expert, ISO JTC1/SC34 member
------------------------------------------------------------------
Received on Wednesday, 10 October 2012 12:38:41 UTC