Hi,

I am using this select statement to return unique clients from my xml file and put them into a drop down list.
//portfolio/campaign[@client_name][not(@client_name=preceding-sibling::campaign/@client_name)]

This line works perfect, but I also need another select statement to return unique creative types from my xml file.
Tried this:
//portfolio/campaign/creative[@client_type][not(@client_type=preceding-sibling::creative/@client_type)]

This is the struture of my xml

<portfolio>
    <campaign client_name="client1">
        <creative creative_type = "banner"></creative>
        <creative creative_type = "island"></creative>
        <creative creative_type = "banner"></creative>
    </campaign>
    <campaign client_name="client2">
        <creative creative_type = "banner"></creative>
        <creative creative_type = "island"></creative>
        <creative creative_type = "banner"></creative>
    </campaign>
    <campaign client_name="client3">
        <creative creative_type = "banner"></creative>
        <creative creative_type = "island"></creative>
        <creative creative_type = "banner"></creative>
    </campaign>
</portfolio>

The select statement for creative types should return only two results:
banner
island

But mine returns:
banner
island
banner
island
banner
island

It returns unique types for creatives with the same parent but ignores other campaigns.

I hope this is clear. I need a select statement that will return the proper result as shown above. Can anyone help?

In 2.0 of course you can simply use the distinct-values() function.

In 1.0, try

portfolio/campaign/creative/@creative_type[not(. = ../preceding::creative/@creative-type)]

or for better performance, use Muenchian grouping.

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

Thanks, that worked perfect.

portfolio/campaign/creative/@creative_type[not(. = ../preceding::creative/@creative_type)]