Re: [xsl] compare two node sets

Hi Wendell,

Thank you for your solution. I have a question about it, which has to do 
with the fact that, in my case, the <author>/<director> elements also 
may have a <middle> child element, which I did not mention in my 
original post. Now, if I understand correctly, your solution does not 
differentiate between
        <director>
           <first>J.</first>
           <middle>R.</first>
           <last>Doe</last>
        </director>

and

        <director>
           <first>J. R.</first>
           <last>Doe</last>
        </director>

Is that right? (I am not saying that this is necessarily a problem. It 
may even be seen as a feature rather than a bug. I just want to be sure 
that I understand the solution correctly.)
Best,
Wolfhart


On 21.01.20 12:42, Piez, Wendell A. (Fed) 
wendell(_dot_)piez(_at_)nist(_dot_)gov wrote:
> Hi,
>
> Another solution not yet suggested in this thread is to avoid deep-equal() and 
> rely on a signature function.
>
> So for example,
>
> <xsl:function name="f:signature" as="xs:string">
>    <xsl:param name="who" as="node()"/>
>    <xsl:value-of select="string-join($who/(first, last), ' ')"/>
> </xsl:function>
>
>
> Then your list of author-directors is 
> //author[f:signature(.)=parent::film/child::director/f:signature(.)]
>
> (Or the other way around.)
>
> Wrap the expression in the filter in not() to retrieve authors who did not 
> direct, etc.
>
> One advantage of this approach is it's easy to adjust the signature logic to 
> real-world exigencies.
>
> Cheers, Wendell
>
> -----Original Message-----
> From: Wolfhart Totschnig wolfhart(_dot_)totschnig(_at_)mail(_dot_)udp(_dot_)cl 
> <xsl-list-service(_at_)lists(_dot_)mulberrytech(_dot_)com>
> Sent: Sunday, January 19, 2020 4:22 PM
> To: xsl-list(_at_)lists(_dot_)mulberrytech(_dot_)com
> Subject: Re: [xsl] compare two node sets
>
> Thank you, David, Michael, and Liam for the prompt replies! Michael's solution 
> seems to be the simplest to implement. I use Saxon 9 HE, so XPath 2.0 should be 
> okay. And, indeed, quadratic performance should not be an issue. However, Saxon 
> throws the following error:
>
> XPST0003: Unexpected token "every" at start of expression
>
> Is there a typo in the expression? I used the expression as given:
>
> <xsl:when test="count(//director) eq count(//author) and every $d in //director 
> satisfies some $a in //author satisfies deep-equal($d/*, $a/*)">
>
> To clarify, the context node is the <film> element.
>
> Wolfhart
>
>
> On 19.01.20 17:55, Liam R. E. Quin liam(_at_)fromoldbooks(_dot_)org wrote:
> > On Sun, 2020-01-19 at 20:37 +0000, Wolfhart Totschnig
> > wolfhart(_dot_)totschnig(_at_)mail(_dot_)udp(_dot_)cl wrote:
> > > Hello,
> > >
> > > I have an XSL/XPath problem to which I cannot find the solution. I
> > > have an xml file with data about films, in the following form
> > > (simplified):
> > >
> > > [..]
> > > . By contrast, in the following example the test should return
> > > <false>:
> > >
> > >       <film>
> > >          <title>M</title>
> > >          <director>
> > >             <first>Fritz</first>
> > >             <last>Lang</last>
> > >          </director>
> > >          <author>
> > >             <first>Thea von</first>
> > >             <last>Harbou</last>
> > >          </author>
> > >          <author>
> > >             <first>Fritz</first>
> > >             <last>Lang</last>
> > >          </author>
> > >       </film>
> > Why?
> >
> > As stated,
> >    <xsl:mode on-no-match="shallow-copy" />
> >
> >     <xsl:template match="/">
> >       <xsl:apply-templates select="/films/film[
> >           some $a in author satisfies
> >           (
> >               ($a/first = director/first)
> >               and ($a/last = director/last)
> >           )
> >       ]"/>
> >     </xsl:template>
> >
> > Liam
--~----------------------------------------------------------------
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
EasyUnsubscribe: http://lists.mulberrytech.com/unsub/xsl-list/1167547
or by email: xsl-list-unsub(_at_)lists(_dot_)mulberrytech(_dot_)com
--~--