- From: John E. Simpson <simpson@polaris.net>
- Date: Fri, 14 Jan 2000 09:25:21 -0500
- To: xsl-list@mulberrytech.com, "'xsl-list@mulberrytech.com'" <xsl-list@mulberrytech.com>
- Cc: "'xsl-editors@w3.org'" <xsl-editors@w3.org>
At 03:41 PM 01/13/2000 -0700, Mike Brown wrote:
>[I wrote, earlier:]
> > Okay, I'll bite: If there's a root node but no "data source
> > from which the source tree was constructed," then what is
> > the thing being transformed?
>
>Data from documents referred to via document(), or nothing.
>
>Transformation is not really what's happening, anyway. It's
>creation of a new tree based on the root node of the "main"
>source tree, and *possibly* other nodes in that source tree
>or other source trees obtained via document().
Okay, I understand about using document() (at least with one argument, as
somebody -- Mike Kay? -- just said :). But then doesn't that document
constitute a source tree, even if it consists of just a single root element?
>...I "transform" tons of data from various source trees.
>I just don't always need to have anything in the primary
>source tree except for a root node that allows the processor
>to begin processing. Examples:
>
><xsl:stylesheet ...>
> <xsl:template match="/">
> <p>Hello World</p>
> </xsl:template>
></xsl:stylesheet>
This is kind of what I meant by the term "generating data." I know you were
simplifying for the sake of example; without wanting to flog the issue too
much, and without necessarily asking for copious examples, could you
describe a situation in which it makes sense to have XSLT do this? It seems
like swatting flies with a hammer. Do you have some process hooked up at
the back end which must take all its input from an XSLT processor?
><xsl:stylesheet ...>
> <xsl:template match="/">
> <xsl:apply-templates select="document('foo.xml')" mode="bar"/>
> </xsl:template>
> <xsl:template match="/" mode="bar">
> <p>There are <xsl:value-of select="count(//node())"/> nodes in
>foo.xml</p>
> </xsl:template>
></xsl:stylesheet>
Yes, I can see this. Maybe I'm reading the spec too casually, but using the
document() function to provide your sole source of "input" doesn't
radically change the notion of having a source tree, does it? (Actually,
even the word "radically" seems superfluous.)
Somewhat less fuddled but still scratching my head... thanks if you can
clarify further.
===================================================================
John E. Simpson | Last night I stayed up late playing
simpson@polaris.net | poker with Tarot cards. I got a full
http://www.flixml.org | house and four people died.
| (Stephen Wright)
Received on Friday, 14 January 2000 09:24:37 UTC