Hi all,

    I have an xml file...

    <xml version="1.0">
    <Sample>
        <Data ref="#strt1">This is Data</Data>

        <Start refer="strt1" childref="#traverse1">This is to traverse1</Start>
        <Start refer="strt2" childref="#traverse2">This is to traverse2</Start>
        <Start refer="strt3" childref="#traverse3">This is to traverse3</Start>

        <End id="traverse1" name="Final"> This is the End</End>
        <End id="traverse2" name="Final Text"> This is the End</End>

    </Sample>

I need to parse it this way...

1. Get the value of the attribute "ref" for the element "Data" i.e. '#strt1"

2. Remove the symbol "#" i.e. "strt1"

3. Get the value of attribute "childref" for the element Start whose "refer" attribute value matches "strt1" i.e. "#traverse1"

4. Get the value of attribute "name" for the element End whose "id" attribute value matches "traverse1" i.e. "Final".


How can I do this??? Plz help. This is urgent.

I usually don't reply when people say the problem is urgent, because I prefer giving answers if I think people will have time to study them properly. But I'll make an exception.

1. Data/@ref

2. translate(Data/@ref, '#', '')

3. key('Start', translate(Data/@ref, '#', '')) where you have defined

<xsl:key name="Start" match="Start" use="@refer"/>

4. This just repeats steps 2 and 3 so you already know how to do it.

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

Thanx a lot Michael

I have another problem... If there are multiple values for any attribute, how do I do this??

The scenario is like this

        <Start refer="strt1" childref="#traverse1 #traverse2">Traverse</Start>

        <End id="traverse1" name="Final"> This is the End</End>
        <End id="traverse2" name="Final Text"> This is the End</End>
        <End id="traverse3" name="End"> End</End>

Plz help...

XPath 2.0 has a built-in function tokenize() to split the attribute value into a sequence of tokens. If you're stuck with 1.0, there's an XSLT template str:tokenize at www.exslt.org that you can copy into your stylesheet (use the template rather than the extension function for portability).

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

Thanx a lot...

Everything works fine if tested seperately. But combined, it doesnt serve my purpose. My actual need is I have an xml file like this:

<xml version="1.0">
    <Sample>
        <Data ref="#strt1">This is Data</Data>

        <Start refer="strt1" anotherRef="#another" childref="#traverse1 #traverse2">This is to traverse</Start>
        <Start refer="strt2" childref="#traverse3">This is to traverse</Start>

        <Another id="another" endRef="#end1" name="Final"> This is another Traversal</Another>

        <Mid id="traverse1" endRef="#end1" name="Final"> This is the Traversal1</Mid>
        <Mid id="traverse2" endRef="#end2" name="Final Text"> This is the Traversal2</Mid>
        <Mid id="traverse3" endRef="#end3" name="End"> This is the Traversal3</Mid>

        <End id="end1" name="Final"> This is the End1</End>
        <End id="end2" name="Final Text"> This is the End2</End>
        <End id="end3" name="End"> This is the End3</End>

    </Sample>

The output should be a txt file like this:

This is another Traversal,This is the Traversal1,This is the End1
This is another Traversal,This is the Traversal2,This is the End2

The logic of traversing the data is the same as mentioned earlier...

Need comments plz...