Re: [xsl] Diferent behaviour using document() in IE / FF

Subject: Re: [xsl] Diferent behaviour using document() in IE / FF
From: António Mota <amsmota@xxxxxxxxx>
Date: Thu, 1 Sep 2005 12:14:44 +0100
So you're saying the IE behaviour is the behaviour expected from a
conformant processor, and that this is not open to implementation,
right?

I'll try to look at FF related foruns for more info, but it seems
strange no one complaint about it until now.

However, i seem to remember some similar problem from some months ago,
but i can't figure out what it was, nor the solution...


On 8/31/05, Michael Kay <mike@xxxxxxxxxxxx> wrote:
> This looks prima facie like an FF bug.
>
> Michael Kay
> http://www.saxonica.com/
>
> > -----Original Message-----
> > From: Antsnio Mota [mailto:amsmota@xxxxxxxxx]
> > Sent: 31 August 2005 18:59
> > To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> > Subject: [xsl] Diferent behaviour using document() in IE / FF
> >
> > I have this strange behaviour using XSLT in IE and FF.
> >
> > I have a XML like this:
> > <doc>
> > <aaa>
> > <bbb>ccc</bbb>
> > </aaa>
> > <ccc>
> > <ddd>eee</eee>
> > </ccc>
> > </doc>
> >
> > In the client (browser) i load this documento and apply a xslt so it
> > results in this:
> >
> > <doc>
> > <aaa>
> > <bbb>ccc</bbb>
> > </aaa>
> > </doc>
> >
> > Sometime later in the application, i use another xslt that uses that
> > document (call it "doc.xml")
> >
> > <xsl:variable name="xmldoc" select="document('doc.xml')"/>
> > <xsl:apply-templates select="$xmldoc"/>
> >
> > what happens is that teh content of the variable xmldoc is, in IE, the
> > original document, as i expected
> >
> > <doc>
> > <aaa>
> > <bbb>ccc</bbb>
> > </aaa>
> > <ccc>
> > <ddd>eee</eee>
> > </ccc>
> > </doc>
> >
> > but in FF i get the first transformed doc
> >
> > <doc>
> > <aaa>
> > <bbb>ccc</bbb>
> > </aaa>
> > </doc>
> >
> > Since i want to access the nodes on the original doc (in this example
> > <ccc>), in FF my app won't work.
> >
> > I'm wondering  what the correct behaviour should be, or is this
> > implementation dependent?
> >
> > Thanks.

Current Thread