Re: [xsl] Custom node-set in a variable

[Wendell Piez <wapiez@xxxxxxxxxxxxxxxx>]

Leonid,

At 09:42 AM 2/17/2005, you wrote:
> Now I need to exclude certain xml elements from the transformations by
> name.
> The list of such elements is long.
> Can I use a variable inside my XSL as a custom node-set, something like:
>    <xsl:variable name="bypass-names">
>       <element1 />
>       <element2 />
> ...
>    </xsl:variable>

You can do this, but it requires a little extra effort to do so in XSLT 
1.0. This is because a variable declared this way is not bound to a true 
node-set, but rather to a result tree fragment -- which is much like a node 
set with the caveat that you can't query into it -- which is, 
unfortunately, exactly what you need to do.

The XSLT 1.0 workaround to this problem takes two forms:

* use an extension function such as exslt:node-set() to turn the result 
fragment into a node set
* use the document() function to query the document where this node set 
appears, instead of simply using the variable as bound. Since document('') 
returns the stylesheet as a (parsed) document, this can be done like so:

   <xsl:variable name="bypass-names-RTF">
      <element1 />
      <element2 />
...
   </xsl:variable>

   <xsl:variable name="bypass-names"
   select="document('')/*/xsl:variable[@name='bypass-names-RTF]/*"/>

So now $bypass-names is actually a node set, which you can traverse.

Now as for your real question....

> And what would be the good way to handle the exclusion, keeping in mind
> performance?

Actually, not as you have it, like

<xsl:variable name="bypass-names-RTF">
   <element1 />
   <element2 />
   ...
</xsl:variable>

but rather something like

<xsl:variable name="bypass-names-RTF">
   <name>element1</name>
   <name>element2</name>
   <name>element3</name>
   ...
</xsl:variable>

This is nice because then the names you want to bypass can be collected as 
a set of nodes whose values, not names, can be checked against. That is, 
since $bypass-names (using the declaration described above) will be a set 
of <name> elements, the names you want to exclude is available as the 
*values* of the node-set $bypass-names.

The equality operator "=" in XPath is very handy for the test you then need 
to do ... for any element in your input, when

test="name() = $bypass-names"

you know its name is in the set.

(You might want to research the name() and local-name() functions, since it 
could matter which one you use, if you have namespaces in play.)

This is a special application of a "lookup table", which you will find 
documented in the XSLT FAQ and elsewhere.

Cheers,
Wendell


======================================================================
Wendell Piez                            mailto:wapiez@xxxxxxxxxxxxxxxx
Mulberry Technologies, Inc.                http://www.mulberrytech.com
17 West Jefferson Street                    Direct Phone: 301/315-9635
Suite 207                                          Phone: 301/315-9631
Rockville, MD  20850                                 Fax: 301/315-8285
----------------------------------------------------------------------
  Mulberry Technologies: A Consultancy Specializing in SGML and XML
======================================================================