Help working with distinct records in an xml doc

[Chris Kettenbach]

Good Morning,
Sorry for xposting. Just need a liitle help.
I have an xml file that's generated from a database. How do I select
distinct values from a field in xslt and then loop through the records and
produce output. Example

<registrations>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Smith</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>Bob</first>
<last>Roberts</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Johnson</last>
</registration>
<registration>
<company>Other Company</company>
<first>Tom</first>
<last>Thomas</last>
</registration>
<registration>
<company>Other Company</company>
<first>Dan</first>
<last>Daniels</last>
</registration>
</registrations>
++++++++++++++++++++++++++++++++++++++++++++++++
I want the out put to be something like this

<table>
<tr>
<td>Company Name</td>
<td>Employees</td>
</tr>
<tr>
<td>Awesome Printers</td>
<td>John Smith<br/>
Bob Roberts<br/>
John Johnson<br/></td>
</tr>
<tr>
<td>Other Company</td>
<td>Tom Thomas<br/>
Dan Daniels<br/></td>
</tr>
</table>

Effectively writing unique company names once and then putting only the
employees from that company into the employee table cell.

Thanks for any advise.
Regards,
Chris

 

[Peter Flynn]

Good Morning,
Sorry for xposting. Just need a liitle help.
I have an xml file that's generated from a database. How do I select
distinct values from a field

XML doesn't have fields -- the database did but this isn't a database any
more. In XML they're called elements (they've got a lot in common with
fields but they ain't the same).

in xslt and then loop through the records and
produce output.

It looks like you need to process the information grouped by company name,
and there's a technique in XSLT 1.0 for doing this called Muenchian grouping
(it won't be needed in XSLT 2 because that has a built-in group-processing
command, but XSLT 2 is not a Recommendation yet).

Check the XSLT FAQ for "Muenchian" to find examples.

///Peter

[Group and followup corrected to comp.text.xml]

Example

<registrations>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Smith</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>Bob</first>
<last>Roberts</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Johnson</last>
</registration>
<registration>
<company>Other Company</company>
<first>Tom</first>
<last>Thomas</last>
</registration>
<registration>
<company>Other Company</company>
<first>Dan</first>
<last>Daniels</last>
</registration>
</registrations>
++++++++++++++++++++++++++++++++++++++++++++++++
I want the out put to be something like this

<table>
<tr>
<td>Company Name</td>
<td>Employees</td>
</tr>
<tr>
<td>Awesome Printers</td>
<td>John Smith<br/>
Bob Roberts<br/>
John Johnson<br/></td>
</tr>
<tr>
<td>Other Company</td>
<td>Tom Thomas<br/>
Dan Daniels<br/></td>
</tr>
</table>

Effectively writing unique company names once and then putting only the
employees from that company into the employee table cell.

If you scrolled down this far then you've been rewarded

The trick is to use lookup keys to test the selected group of elements
against, in order to pick only the first ocurrence in each group; then
output the group header at that point; and only then go and process all the
elements in the group to extract the detail from each of them.

The syntax is tricky because it's very compact. It selects all registration
elements, but subjects each of them to the test that [the number of nodes in
the union of (the current node and the first node with the current company
value returned by the lookup) is equal to 1] -- in effect meaning they are
one and the same node, ie the first in their company name grouping. You'll
probably need to read that 3 or 4 times -- it took me a week to grok it.

<?xml version="1.0" encoding="iso-8859-1"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0">

<xslutput method="html"/>

<xsl:key name="entry" match="registration" use="company"/>

<xsl:template match="/">
<html>
<head>
<title>Registrations</title>
</head>
<body>
<xsl:apply-templates/>
</body>
</html>
</xsl:template>

<xsl:template match="registrations">
<table>
<xsl:for-each select="registration[count(.|key('entry',company
[1])=1]">
<xsl:sort select="company"/>
<tr>
<th valign="top">
<xsl:value-of select="company"/>
</th>
<td valign="top">
<xsl:for-each select="key('entry',current()/company)">
<xsl:sort select="last"/>
<xsl:value-of select="first"/>
<xsl:text> </xsl:text>
<xsl:value-of select="last"/>
<xsl:if test="position()!=last()">
<br/>
</xsl:if>
</xsl:for-each>
</td>
</tr>
</xsl:for-each>
</table>
</xsl:template>

</xsl:stylesheet>

///Peter

 

[Stefan Ram]

XML doesn't have fields -- the database did but this isn't a
database any more. In XML they're called elements

Sometimes it might be appropriate to use attributes.

 

[Chris Kettenbach]

Hi Peter,
I get error when processing the stylesheet. It errors here.

<xsl:for-each select="registration[count(.|key('entry',company
[1])=1]">

specifically:

Expression does not return a DOM node.
registration[-->count(.|key('entry',company[1])=1]<--

Ideas?


Thanks,
Chris

Good Morning,
Sorry for xposting. Just need a liitle help.
I have an xml file that's generated from a database. How do I select
distinct values from a field

XML doesn't have fields -- the database did but this isn't a database any
more. In XML they're called elements (they've got a lot in common with
fields but they ain't the same).

in xslt and then loop through the records and
produce output.

It looks like you need to process the information grouped by company name,
and there's a technique in XSLT 1.0 for doing this called Muenchian
grouping
(it won't be needed in XSLT 2 because that has a built-in group-processing
command, but XSLT 2 is not a Recommendation yet).

Check the XSLT FAQ for "Muenchian" to find examples.

///Peter

[Group and followup corrected to comp.text.xml]

Example

<registrations>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Smith</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>Bob</first>
<last>Roberts</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Johnson</last>
</registration>
<registration>
<company>Other Company</company>
<first>Tom</first>
<last>Thomas</last>
</registration>
<registration>
<company>Other Company</company>
<first>Dan</first>
<last>Daniels</last>
</registration>
</registrations>
++++++++++++++++++++++++++++++++++++++++++++++++
I want the out put to be something like this

<table>
<tr>
<td>Company Name</td>
<td>Employees</td>
</tr>
<tr>
<td>Awesome Printers</td>
<td>John Smith<br/>
Bob Roberts<br/>
John Johnson<br/></td>
</tr>
<tr>
<td>Other Company</td>
<td>Tom Thomas<br/>
Dan Daniels<br/></td>
</tr>
</table>

Effectively writing unique company names once and then putting only the
employees from that company into the employee table cell.

If you scrolled down this far then you've been rewarded

The trick is to use lookup keys to test the selected group of elements
against, in order to pick only the first ocurrence in each group; then
output the group header at that point; and only then go and process all
the
elements in the group to extract the detail from each of them.

The syntax is tricky because it's very compact. It selects all
registration
elements, but subjects each of them to the test that [the number of nodes
in
the union of (the current node and the first node with the current company
value returned by the lookup) is equal to 1] -- in effect meaning they are
one and the same node, ie the first in their company name grouping. You'll
probably need to read that 3 or 4 times -- it took me a week to grok it.

<?xml version="1.0" encoding="iso-8859-1"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0">

<xslutput method="html"/>

<xsl:key name="entry" match="registration" use="company"/>

<xsl:template match="/">
<html>
<head>
<title>Registrations</title>
</head>
<body>
<xsl:apply-templates/>
</body>
</html>
</xsl:template>

<xsl:template match="registrations">
<table>
<xsl:for-each select="registration[count(.|key('entry',company
[1])=1]">
<xsl:sort select="company"/>
<tr>
<th valign="top">
<xsl:value-of select="company"/>
</th>
<td valign="top">
<xsl:for-each select="key('entry',current()/company)">
<xsl:sort select="last"/>
<xsl:value-of select="first"/>
<xsl:text> </xsl:text>
<xsl:value-of select="last"/>
<xsl:if test="position()!=last()">
<br/>
</xsl:if>
</xsl:for-each>
</td>
</tr>
</xsl:for-each>
</table>
</xsl:template>

</xsl:stylesheet>

///Peter

 

[toudidel]

hint: from performed elements you have to create attributtes in yor new
element, because there isn't posibility to being few the same attributtes in
one xml element

td

 

[Peter Flynn]

Hi Peter,
I get error when processing the stylesheet. It errors here.

<xsl:for-each select="registration[count(.|key('entry',company
[1])=1]">

specifically:

Expression does not return a DOM node.
registration[-->count(.|key('entry',company[1])=1]<--

Ideas?

What processor are you using? I ran it through Saxon without errors, and the
output was:

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Registrations</title>
</head>
<body>
<table>
<tr>
<th valign="top">Awesome Printers</th>
<td valign="top">John Johnson<br>Bob Roberts<br>John Smith
</td>
</tr>
<tr>
<th valign="top">Other Company</th>
<td valign="top">Dan Daniels<br>Tom Thomas
</td>
</tr>
</table>
</body>
</html>

///Peter

Good Morning,
Sorry for xposting. Just need a liitle help.
I have an xml file that's generated from a database. How do I select
distinct values from a field

XML doesn't have fields -- the database did but this isn't a database any
more. In XML they're called elements (they've got a lot in common with
fields but they ain't the same).

in xslt and then loop through the records and
produce output.

It looks like you need to process the information grouped by company
name, and there's a technique in XSLT 1.0 for doing this called Muenchian
grouping
(it won't be needed in XSLT 2 because that has a built-in
group-processing command, but XSLT 2 is not a Recommendation yet).

Check the XSLT FAQ for "Muenchian" to find examples.

///Peter

[Group and followup corrected to comp.text.xml]

Example

<registrations>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Smith</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>Bob</first>
<last>Roberts</last>
</registration>
<registration>
<company>Awesome Printers</company>
<first>John</first>
<last>Johnson</last>
</registration>
<registration>
<company>Other Company</company>
<first>Tom</first>
<last>Thomas</last>
</registration>
<registration>
<company>Other Company</company>
<first>Dan</first>
<last>Daniels</last>
</registration>
</registrations>
++++++++++++++++++++++++++++++++++++++++++++++++
I want the out put to be something like this

<table>
<tr>
<td>Company Name</td>
<td>Employees</td>
</tr>
<tr>
<td>Awesome Printers</td>
<td>John Smith<br/>
Bob Roberts<br/>
John Johnson<br/></td>
</tr>
<tr>
<td>Other Company</td>
<td>Tom Thomas<br/>
Dan Daniels<br/></td>
</tr>
</table>

Effectively writing unique company names once and then putting only the
employees from that company into the employee table cell.

If you scrolled down this far then you've been rewarded

The trick is to use lookup keys to test the selected group of elements
against, in order to pick only the first ocurrence in each group; then
output the group header at that point; and only then go and process all
the
elements in the group to extract the detail from each of them.

The syntax is tricky because it's very compact. It selects all
registration
elements, but subjects each of them to the test that [the number of nodes
in
the union of (the current node and the first node with the current
company value returned by the lookup) is equal to 1] -- in effect meaning
they are one and the same node, ie the first in their company name
grouping. You'll probably need to read that 3 or 4 times -- it took me a
week to grok it.

<?xml version="1.0" encoding="iso-8859-1"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0">

<xslutput method="html"/>

<xsl:key name="entry" match="registration" use="company"/>

<xsl:template match="/">
<html>
<head>
<title>Registrations</title>
</head>
<body>
<xsl:apply-templates/>
</body>
</html>
</xsl:template>

<xsl:template match="registrations">
<table>
<xsl:for-each select="registration[count(.|key('entry',company
[1])=1]">
<xsl:sort select="company"/>
<tr>
<th valign="top">
<xsl:value-of select="company"/>
</th>
<td valign="top">
<xsl:for-each select="key('entry',current()/company)">
<xsl:sort select="last"/>
<xsl:value-of select="first"/>
<xsl:text> </xsl:text>
<xsl:value-of select="last"/>
<xsl:if test="position()!=last()">
<br/>
</xsl:if>
</xsl:for-each>
</td>
</tr>
</xsl:for-each>
</table>
</xsl:template>

</xsl:stylesheet>

///Peter