Re: [xsl] best practices for using XSLT modes

["Dimitre Novatchev dnovatchev@xxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

I agree with Martin.

I also use modes when performing multi-pass processing -- typically the XML
document that is the result of pass-N-1 and is to be processed by pass-N
has different structure and meaning than any of the documents produced by
previous passes , or future documents, to be produced by future passes.

So we have M sets of transformations, each possibly having template(s)
matching the root node or identically named/typed items and processing
these in a different, pass-specific way.

Without using modes I wonder if it would be possible at all to do this
processing in a single transformation -- and even if this is somehow
(crazy) possible, the code would be a torture to read, understand and
maintain.

Cheers,
Dimitre

On Tue, Dec 3, 2019 at 10:51 PM Martin Honnen martin.honnen@xxxxxx <
xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:

> Am 04.12.2019 um 07:42 schrieb Mukul Gandhi gandhi.mukul@xxxxxxxxx:
> > Hi all,
> >    I imagine that, using XSLT modes is useful. I've been trying
> > different XSLT approaches for solving a class of XML transformation
> > problems.
> >
> > Below is an example of what I've tried (I present an XML document, two
> > different XSLT stylesheets [non schema aware] to process the XML
> > document, and an identical transformation output with both the
> > presented stylesheets):
> >
> > XML document:
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> > <root>
> >    <a val="-1"/>
> >    <a val="-4"/>
> >    <a val="5"/>
> >    <a val="3"/>
> >    <a val="2"/>
> > </root>
> >
> > Stylesheet 1:
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> > <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform";
> >                          version="3.0">
> >
> >     <xsl:output method="xml" indent="yes"/>
> >
> >     <xsl:template match="root">
> >        <result>
> >           <xsl:apply-templates select="a[number(@val) gt 0]" mode="gt0"/>
> >           <xsl:apply-templates select="a[number(@val) lt 0]" mode="lt0"/>
> >        </result>
> >     </xsl:template>
> >
> >     <xsl:template match="a" mode="gt0">
> >       <val><xsl:value-of select="@val"/>: positive</val>
> >     </xsl:template>
> >
> >     <xsl:template match="a" mode="lt0">
> >       <val><xsl:value-of select="@val"/>: negative</val>
> >     </xsl:template>
> >
> > </xsl:stylesheet>
> >
> > Stylesheet 2:
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> > <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform";
> >                          version="3.0">
> >
> >     <xsl:output method="xml" indent="yes"/>
> >
> >     <xsl:template match="root">
> >        <result>
> >           <xsl:apply-templates select="a[number(@val) gt 0]"/>
> >           <xsl:apply-templates select="a[number(@val) lt 0]"/>
> >        </result>
> >     </xsl:template>
> >
> >     <xsl:template match="a[number(@val) gt 0]">
> >       <val><xsl:value-of select="@val"/>: positive</val>
> >     </xsl:template>
> >
> >     <xsl:template match="a[number(@val) lt 0]">
> >       <val><xsl:value-of select="@val"/>: negative</val>
> >     </xsl:template>
> >
> > </xsl:stylesheet>
> >
> > Both of above stylesheets, achieve the same thing and generate
> > following output:
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> > <result>
> >    <val>5: positive</val>
> >    <val>3: positive</val>
> >    <val>2: positive</val>
> >    <val>-1: negative</val>
> >    <val>-4: negative</val>
> > </result>
> >
> > The intent of mentioned transformations, is that the result is little
> > reorganization of the input.
> >
> > My questions are following,
> > Which of above mentioned XSLT transformations, is better over the
> > other, particularly considering the use of modes (conceptually &
> > possibly wrt to any other factors)?
>
>
> That example seems to be too simple or artificial to show the value of
> modes. In general I think modes have their value if you need to process
> the same type of nodes twice e.g. once for generating a table of
> contents and the second time for splitting into result documents. Or, in
> the context of XSLT 3, if you need to separate processing steps working
> with streamed input nodes from ones using grounded ones.