Re: [xsl] CATCH ALL | template rule (2nd try)

[Karl Stubsjoen <kstubs@xxxxxxxxx>]

Never mind, I got it : )

On 5/17/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> Sorry, I didn't get it : (
> I am close... but picking exceptions that arent' exceptions. I changed
> the apply-templates from above to this (in a desperate attempt to make
> it work):
>
>  <xsl:apply-templates select="fx:node-set($interval)
>     [not(count(.|$scenario_a) = count($scenario_a))]
>     [not(count(.|$scenario_b) = count($scenario_b))]
>     [not(count(.|$scenario_c) = count($scenario_c))]
>     [not(count(.|$scenario_d) = count($scenario_d))]
>     [not(count(.|$scenario_e) = count($scenario_e))]"
>  mode="Exceptions"/>
>
> and then set up the following matches:
>
> <xsl:template match="fn:interval" mode="Exceptions">
> <xsl:variable name="current_interval"
> select="ancestor::fn:interval/@current_interval"/>
> <xsl:comment>EXCEPTION</xsl:comment>
>  <xsl:copy-of select="."/>
> </xsl:template>
>
> next 2 templates hide the "junk":
>
> <xsl:template match="fn:record" mode="Exceptions"/>
> <xsl:template match="node()" mode="Exceptions"/>
>
> Please help, I am not sure where to go from here.  My scenario_a data
> appears in both the applied rule (above exceptions) and in exceptions.
> Also, I have setup an ABC example which does work properly (much
> simple than my real data) but am confused as to why the real data is
> not working.
>
> Thanks,
>
> Karl..
>
>
> On 5/17/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> > I got it:
> > (without the node() match below i was getting the extra junk.. now it
> > is suppressed)
> > Thanks DIMITRE!  This is a kick ass solution!
> >
> > <xsl:apply-templates select="fx:node-set($interval)//node()
> >    [not(count(.|$scenario_a) = count($scenario_a))]
> >    [not(count(.|$scenario_b) = count($scenario_b))]
> >    [not(count(.|$scenario_c) = count($scenario_c))]
> >    [not(count(.|$scenario_d) = count($scenario_d))]
> >    [not(count(.|$scenario_e) = count($scenario_e))]"
> >  mode="Exceptions"/>
> >
> > <xsl:template match="fn:record" mode="Exceptions">
> >  <xsl:comment>EXCEPTION</xsl:comment>
> >  <xsl:copy-of select="."/>
> > </xsl:template>
> >
> > <xsl:template match="node()" mode="Exceptions"/>
> >
> >
> > On 5/17/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> > > becomes:
> > > fx:node-set($interval)[not(count(. | yourExpression) =
> > > count(yourExpression))]
> > >
> > > (lose the "whack whack")
> > >
> > > On 5/17/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> > > > I figured out the extra junk problem, we were at the wrong context of
> > > > $interval on the apply exclusions.  This context is at:  fn:interval,
> > > > so I removed the node() from the apply-templates so this:
> > > > fx:node-set($interval)//node()[not(count(. | yourExpression) =
> > > > count(yourExpression))]
> > > >
> > > > becomes:
> > > > fx:node-set($interval)//[not(count(. | yourExpression) =
> > > > count(yourExpression))]
> > > >
> > > > Junk is gone.  Does this seem correct, or is "node()" a very
important
> > > > aspect to this xpath?
> > > >
> > > >
> > > > On 5/17/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> > > > > Ahh Dimitre, quit teasing me with that XSLT 2.0 stuff!
> > > > >
> > > > > It is working great!  Except... I am getting additional junk.  Not
> > > > > sure where it is coming from.  Like an apply-templates with no
match
> > > > > and you just get the text of the elements back.  I have set up
empty
> > > > > match templates for my mode of "Exceptions", it helped some, but
I'm
> > > > > still getting back extra junk.  In my "simple example" I am not
> > > > > getting back the extra junk.  Any ideas?
> > > > >
> > > > >
> > > > > On 5/17/05, Dimitre Novatchev <dnovatchev@xxxxxxxxx> wrote:
> > > > > > On 5/18/05, Karl Stubsjoen <kstubs@xxxxxxxxx> wrote:
> > > > > > > Okay, I am going to try that!  How about this simpler example
> > > > > > > though..., would you mind providing the inverse for this one:
> > > > > > >
> > > > > > > <xsl:template match="A">
> > > > > > > <xsl:copy>
> > > > > > >  <match>
> > > > > > >    <xsl:apply-templates
select="B[X[@val=$match_v1]][X[@val=$match_v2]]"/>
> > > > > > >  </match>
> > > > > > >  <exception>
> > > > > > >    <!-- currently, copy of the match above, but would like this
to be
> > > > > > > expressed as the exception -->
> > > > > > >    <xsl:apply-templates
select="B[X[@val=$match_v1]][X[@val=$match_v2]]"/>
> > > > > > >  </exception>
> > > > > > > </xsl:copy>
> > > > > > > </xsl:template>
> > > > > >
> > > > > > N.P.
> > > > > >
> > > > > > This can be done mechanically:
> > > > > >
> > > > > > >    <xsl:apply-templates
select="B[X[@val=$match_v1]][X[@val=$match_v2]]"/>
> > > > > >
> > > > > >  <xsl:variable name="x"
select="B[X[@val=$match_v1]][X[@val=$match_v2]]"/>
> > > > > >
> > > > > >   <xsl:apply-templates select="node()[not(count(.|$x) =
count($x))]"/>
> > > > > >
> > > > > > Of course, in XPath 2.0 one will use the except operator:
> > > > > >
> > > > > >   <xsl:apply-templates select="node() except $x"/>
> > > > > >
> > > > > > Cheers,
> > > > > > Dimitre
> > > > > >
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On 5/17/05, Dimitre Novatchev <dnovatchev@xxxxxxxxx> wrote:
> > > > > > > > > <xsl:apply-templates
> > > > > > > > >
select="fx:node-set($interval)//fn:interval[//field[@fieldname='EntityID']=$F
TEInterval_Members][//field[@fieldname='EntityID']=$FTEInterval_JTED_Mains]"
> > > > > > > > > mode="Exceptions" />
> > > > > > > > >
> > > > > > > > > What is the inverse of this?  Which would suffice to be my
"CATCH
> > > > > > > > > ALL", in otherwords match everything else where the above
combination
> > > > > > > > > fails.
> > > > > > > >
> > > > > > > > fx:node-set($interval)//node()[not(count(. | yourExpression)
=
> > > > > > > > count(yourExpression))]
> > > > > > > >
> > > > > > > > generally, if there are two node-sets: $ns1 and $ns2, their
set difference
> > > > > > > >    $ns2 - $ns1
> > > > > > > > is expressed by this XPath expression:
> > > > > > > >
> > > > > > > >    $ns2[not(count(. | $ns1) = count($ns1) ) ]
> > > > > > > >
> > > > > > > > Cheers,
> > > > > > > > Dimitre