Re: Using node-set variables in predicates (another no

Hi Chris,

Consider something like this:

<xsl:function name="mine:is-trim-element" as="xs:boolean">
  <xsl:param name="elt" as="node()"/>
  <xsl:apply-templates mode="mine:is-trim" select="$elt"/>
</xsl:function>

<xsl:template match="*" mode="mine:is-trim" as="xs:boolean">
  <xsl:sequence select="false()"/>
</xsl:template>

<xsl:template match="desc | dt | endnote | entry | younameit"
mode="mine:is-trim" as="xs:boolean">
  <xsl:sequence select="true()"/>
</xsl:template>

Essentially the function is an accessor for an apply-templates operation
that implements the filter.

Good luck, Wendell


On Tue, Jan 4, 2022 at 3:37 PM Chris Papademetrious
christopher.papademetrious@xxxxxxxxxxxx <
xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:

> Hi Dimitre,
>
>
>
> That is good advice about functions b I will keep it in mind!
>
>
>
>
>
> Hi everyone,
>
>
>
> So I ran into a limitation of using document node-set variables in
> template match expressions:
>
>
>
>   <!-- put document nodes in $trim-elements -->
>
>   <xsl:variable name="trim-elements"
> select="//(desc|dt|entry|glossterm|li|p|shortdesc|title)"/>
>
>
>
>   <!-- reference elements in $trim-elements -->
>
>   <xsl:template match="text()
>
>                        [matches(., '^\s+')]
>
>                        [ancestor::*[. intersect
> $trim-elements][not(descendant::*[. intersect $trim-elements])]]
>
>                        [not(ancestor-or-self::node()
>
>                          [ancestor::*[. intersect
> $trim-elements][not(descendant::*[. intersect $trim-elements])]]
>
>                          [preceding-sibling::node()])]">
>
>
>
> When I constructed a temporary <p> node in a variable:
>
>
>
> xsl:variable name="temporary-p">
>
>    <p><xsl:copy-of select=bb&blah blahb&b></p>
>
> </xsl:variable>
>
>
>
> then called <xsl:apply-templates> to apply the template:
>
>
>
> <xsl:apply-templates select="$temporary-p"/>
>
>
>
> the template did not get applied. This makes perfect sense b my temporary
> <p> variable does not contain any document nodes at all, so the template
> match expression would never match!
>
>
>
> But with Dimitrebs advice about functions fresh in my mind, I converted
my
> document-node variable into a node-type function:
>
>
>
>   <xsl:function name="mine:is-trim-element" as="xs:boolean">
>
>     <xsl:param name="elt" as="node()"/>
>
>     <xsl:value-of select="exists($elt[self::desc or self::dt or
> self::endnote or self::entry or self::example-title or self::glossterm or
> self::li or self::msg-severity or self::p or self::shortdesc or self::title
> or self::value-allowed or self::value-default or self::value-type])"/>
>
>   </xsl:function>
>
>
>
> Now I can use the function instead of the variable in the match expression:
>
>
>
>   <xsl:template match="text()
>
>                        [matches(., '^\s+')]
>
>
>
[ancestor::*[mine:is-trim-element(.)][not(descendant::*[mine:is-trim-element(
.)])]]
>
>                        [not(ancestor-or-self::node()
>
>
>
[ancestor::*[mine:is-trim-element(.)][not(descendant::*[mine:is-trim-element(
.)])]]
>
>
> [preceding-sibling::node()[not(mine:is-invisible(.))]])]">
>
>
>
> and the template applies as expected to my temporary variable!
>
>
>
> Follow-up question b is there some syntactic sugar that would allow me to
> write
>
>
>
> self::(a|b|c)
>
>
>
> instead of
>
>
>
> (self::a or self::b or self::c)
>
>
>
> or some other way of testing the element type in $elt altogether? I have
> another function that matches several dozen element types, and a more
> compact representation would be nice. I thought about matching local-name()
> against a list of tag names using bintersectb, but that is not much
shorter
> and it seems operationally clunkier.
>
>
>
> Thanks, this has been an enjoyable exercise for learning!
>
>
>
>    - Chris
>
>
>
>
> XSL-List info and archive <http://www.mulberrytech.com/xsl/xsl-list>
> EasyUnsubscribe <http://lists.mulberrytech.com/unsub/xsl-list/174322> (by
> email <>)
>


--
...Wendell Piez... ...wendell -at- nist -dot- gov...
...wendellpiez.com... ...pellucidliterature.org... ...pausepress.org...
...github.com/wendellpiez... ...gitlab.coko.foundation/wendell...