Use XSL to remove "outdated" nodes?

[Jim Bancroft]

Hi everyone,

Could someone reccomend a way to remove duplicate/outmoded nodes in an XML
document?

I didn't describe that well, so let me give an example.

For instance, if I have an XML doc like so:

<inventory>
<part>
<name="CPU"/>
<version="4"/>
</part>
<part>
<name="Hard Disk"/>
<version="3"/>
</part>
<part>
<name="CPU"/>
<version="5"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
</inventory>

I'd like to get a final document with a single CPU element whose
version="5", a Hard Disk element of version="3", and *one* Keyboard element
of version "3". If you could give me a lead, I'd be grateful. Thanks.

-Jim

[Jim Bancroft]

"Dimitre Novatchev" <dn********@yahoo.com> wrote in message
news:bd*************@ID-152440.news.dfncis.de...

This is not well-formed XML.

Sorry, let's try that again-- just something I typed quickly into the news
posting:

<inventory>
<part>
<name>CPU</name>
<version>4</version>
</part>
<part>
<name>Hard Disk</name>
<version>"3"</version>
</part>
<part>
<name>CPU</name>
<version>5</version>
</part>
<part>
<name>Keyboard</name>
<version>"3"</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
</inventory>

[Dimitre Novatchev]

In my solution that uses attributes, just replace in the xsl:sort
instructions:

"@name" with "name"
and
"@version" with "version"

=====
Cheers,

Dimitre Novatchev.
http://fxsl.sourceforge.net/ -- the home of FXSL
"Jim Bancroft" <bo*********@nospam.yahoo.com> wrote in message
news:Gs*****************@news.uswest.net...

"Dimitre Novatchev" <dn********@yahoo.com> wrote in message
news:bd*************@ID-152440.news.dfncis.de...

This is not well-formed XML.

Sorry, let's try that again-- just something I typed quickly into the news
posting:

<inventory>
<part>
<name>CPU</name>
<version>4</version>
</part>
<part>
<name>Hard Disk</name>
<version>"3"</version>
</part>
<part>
<name>CPU</name>
<version>5</version>
</part>
<part>
<name>Keyboard</name>
<version>"3"</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
</inventory>

[Marrow]

Hi Jim,

Assuming that your XML were well-formed and looked something like...

<inventory>
<part>
<name>CPU</name>
<version>4</version>
</part>
<part>
<name>Hard Disk</name>
<version>3</version>
</part>
<part>
<name>CPU</name>
<version>5</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
</inventory>

Then something like (using Muenchian technique to find distinct)...

<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml"/>
<!-- key for finding distinct part names -->
<xsl:key name="kDistinctPartName" match="part" use="name"/>
<xsl:template match="inventory">
<inventory>
<xsl:apply-templates select="part[generate-id() =
generate-id(key('kDistinctPartName',name))]"/>
</inventory>
</xsl:template>

<!-- template for handling distinct part names -->
<xsl:template match="part">
<!-- copy only the highest version of this distinct part name -->
<xsl:copy-of select="key('kDistinctPartName',name)[not(version &lt;
key('kDistinctPartName',name)/version)][1]"/>
</xsl:template>
</xsl:stylesheet>
Hope this helps
Marrow
http://www.marrowsoft.com - home of Xselerator (XSLT IDE and debugger)
http://www.topxml.com/Xselerator

"Jim Bancroft" <bo*********@nospam.yahoo.com> wrote in message
news:g4*****************@news.uswest.net...

Hi everyone,

Could someone reccomend a way to remove duplicate/outmoded nodes in an XML
document?

I didn't describe that well, so let me give an example.

For instance, if I have an XML doc like so:

<inventory>
<part>
<name="CPU"/>
<version="4"/>
</part>
<part>
<name="Hard Disk"/>
<version="3"/>
</part>
<part>
<name="CPU"/>
<version="5"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
</inventory>

I'd like to get a final document with a single CPU element whose
version="5", a Hard Disk element of version="3", and *one* Keyboard element of version "3". If you could give me a lead, I'd be grateful. Thanks.

-Jim

[Jim Bancroft]

Dimitre, Marrow:

Thanks for the assistance-- appreciate it.

-Jim

[Marrow]

Hi Jim,

Assuming that your XML were well-formed and looked something like...

<inventory>
<part>
<name>CPU</name>
<version>4</version>
</part>
<part>
<name>Hard Disk</name>
<version>3</version>
</part>
<part>
<name>CPU</name>
<version>5</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
</inventory>

Then something like (using Muenchian technique to find distinct)...

<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml"/>
<!-- key for finding distinct part names -->
<xsl:key name="kDistinctPartName" match="part" use="name"/>
<xsl:template match="inventory">
<inventory>
<xsl:apply-templates select="part[generate-id() =
generate-id(key('kDistinctPartName',name))]"/>
</inventory>
</xsl:template>

<!-- template for handling distinct part names -->
<xsl:template match="part">
<!-- copy only the highest version of this distinct part name -->
<xsl:copy-of select="key('kDistinctPartName',name)[not(version &lt;
key('kDistinctPartName',name)/version)][1]"/>
</xsl:template>
</xsl:stylesheet>
Hope this helps
Marrow
http://www.marrowsoft.com - home of Xselerator (XSLT IDE and debugger)
http://www.topxml.com/Xselerator

"Jim Bancroft" <bo*********@nospam.yahoo.com> wrote in message
news:g4*****************@news.uswest.net...

Hi everyone,

Could someone reccomend a way to remove duplicate/outmoded nodes in an XML
document?

I didn't describe that well, so let me give an example.

For instance, if I have an XML doc like so:

<inventory>
<part>
<name="CPU"/>
<version="4"/>
</part>
<part>
<name="Hard Disk"/>
<version="3"/>
</part>
<part>
<name="CPU"/>
<version="5"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
</inventory>

I'd like to get a final document with a single CPU element whose
version="5", a Hard Disk element of version="3", and *one* Keyboard element of version "3". If you could give me a lead, I'd be grateful. Thanks.

-Jim

[Marrow]

Hi Jim,

Assuming that your XML were well-formed and looked something like...

<inventory>
<part>
<name>CPU</name>
<version>4</version>
</part>
<part>
<name>Hard Disk</name>
<version>3</version>
</part>
<part>
<name>CPU</name>
<version>5</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
<part>
<name>Keyboard</name>
<version>3</version>
</part>
</inventory>

Then something like (using Muenchian technique to find distinct)...

<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml"/>
<!-- key for finding distinct part names -->
<xsl:key name="kDistinctPartName" match="part" use="name"/>
<xsl:template match="inventory">
<inventory>
<xsl:apply-templates select="part[generate-id() =
generate-id(key('kDistinctPartName',name))]"/>
</inventory>
</xsl:template>

<!-- template for handling distinct part names -->
<xsl:template match="part">
<!-- copy only the highest version of this distinct part name -->
<xsl:copy-of select="key('kDistinctPartName',name)[not(version &lt;
key('kDistinctPartName',name)/version)][1]"/>
</xsl:template>
</xsl:stylesheet>
Hope this helps
Marrow
http://www.marrowsoft.com - home of Xselerator (XSLT IDE and debugger)
http://www.topxml.com/Xselerator

"Jim Bancroft" <bo*********@nospam.yahoo.com> wrote in message
news:g4*****************@news.uswest.net...

Hi everyone,

Could someone reccomend a way to remove duplicate/outmoded nodes in an XML
document?

I didn't describe that well, so let me give an example.

For instance, if I have an XML doc like so:

<inventory>
<part>
<name="CPU"/>
<version="4"/>
</part>
<part>
<name="Hard Disk"/>
<version="3"/>
</part>
<part>
<name="CPU"/>
<version="5"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
<part>
<name="Keyboard"/>
<version="3"/>
</part>
</inventory>

I'd like to get a final document with a single CPU element whose
version="5", a Hard Disk element of version="3", and *one* Keyboard element of version "3". If you could give me a lead, I'd be grateful. Thanks.

-Jim