- From: mozer <xmlizer@gmail.com>
- Date: Mon, 3 Nov 2008 09:33:31 +0100
- To: "Florent Georges" <fgeorges@fgeorges.org>
- Cc: "XProc Comments" <public-xml-processing-model-comments@w3.org>
Florent, I would say "overkill" If you need to keep track of whitespace subtleties you probably have to put the content in another file and use <p:document or <p:load Regards, Xmlizer On Mon, Nov 3, 2008 at 1:27 AM, Florent Georges <fgeorges@fgeorges.org> wrote: > > Hi, > > I don't remember to have seen this question before. Did you > consider for p:inline a mechanism like the whitespaces handling in > XSLT? I can imagine use of p:inline to store input of a step directly > into the pipeline definition, while wanting to properly indent the > inline document, even if the document should not contain those > whitespace. > > For instance, if I give the following input: > > <p:inline> > <dummy/> > </p:inline> > <p:inline> > <foobar/> > </p:inline> > > to an XSLT step that contains the following template rule: > > <xsl:template match="/"> > <xsl:copy-of select="collection()"/> > </xsl:template> > > the result is (the '|' character shows the column 0): > > | > | <dummy/> > | > | <foobar/> > | > > That is, an text node, then an element node, then a text node... I > wonder if it is really what is wanted, instead of simply a sequence of > two element nodes: > > |<dummy/><foobar/> > > Regards, > > -- > Florent Georges > http://www.fgeorges.org/ > >
Received on Monday, 3 November 2008 13:09:07 UTC