Saving each node of a xml document as a file with XProc

Question

I got a xml document with follwing structure

<listOfNodes>
    <node name="file1">content1</node>
    <node name="file2">content2</node>
    ...
    <node name="fileN">contentN</node>
</listOfNodes>

I want to create a xproc pipeline with the result:

file1.txt with content1
file2.txt with content2
...
fileN.txt with contentN

My first approach is:

<p:declare-step name="step_1" type="ts:Extract">
        <p:input port="source" />
        <p:filter select="nodes" name="step_1.1" />
         <p:for-each name="step_1.2">
            <p:iteration-source><p:pipe port="result" step="step_1.1"/></p:iteration-source>
            <p:store method="text" >
                 <p:with-option name="href" select="concat('file:/', 'step_1_',iteration-position(),'.txt')"/>
            </p:store>
        </p:for-each>
</p:declare-step>

But I get a DX0006 error on this.

Can anybody help me, please?

Your approach should essentially have worked. It is only lacking namespace declarations, and iteration-position() needs p: prefix. See my answer below.. — grtjn, Mar 27 '14 at 21:13

score 2 · Accepted Answer · answered Mar 19 '14 at 16:24

Provided content1, ..., contentN are well-formed XML, the following pipeline does the trick:

<p:declare-step version="1.0" xmlns:p="http://www.w3.org/ns/xproc">
  <p:input port="source"/>
  <p:for-each>
    <p:iteration-source select="/*/node"/>
    <p:variable name="filename" select="concat(/node/@name, '.txt')"/>
    <p:unwrap match="/*"/>
    <p:store>
      <p:with-option name="href" select="$filename"/>
    </p:store>
  </p:for-each>
</p:declare-step>

If content1, ..., contentN are not well-wormed XML (plain text, multiple sibling elements etc.), then you can wrap them in an XML wrapper element before you apply p:store (or simply don't apply the p:unwrap step). If you don't want to use a wrapper element, then standard p:store will not work (some XProc implementations may support extension attributes on p:store to store only the contents of the document element; but that is another story). One possibility is to use p:xslt instead.

This looks like the perfect solution in pure xproc, thank you — Oliver, Mar 20 '14 at 13:42

Oliver · Answer 2 · 2014-03-19T12:31:36.187

Following Pipeline does the job:

<p:declare-step xmlns:p="http://www.w3.org/ns/xproc" xmlns:c="http://www.w3.org/ns/xproc-step" version="1.0">
    <p:input port="source" />
    <p:xslt name="create-document">
        <p:input port="source" />
        <p:input port="stylesheet">
            <p:document href="splitter.xsl" />
        </p:input>
        <p:input port="parameters"><p:empty/></p:input>
    </p:xslt>
    <p:for-each>
        <p:iteration-source>
            <p:pipe step="create-document" port="secondary" />
        </p:iteration-source>
            <p:store>
                <p:with-option name="method" select="'xml'" />
                <p:with-option name="href" select="p:base-uri(/*)" />
            </p:store>
        </p:for-each>

<p:sink>
    <p:input port="source"><p:pipe port="result" step="create-document"/></p:input>
</p:sink>
</p:declare-step>

But I also had to Split the file in the xslt file:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">

    <xsl:template match="node">
        <xsl:variable name="filename" select="concat(@name,'.txt')" />
        <saveTo><xsl:value-of select="$filename" /></saveTo>
        <xsl:result-document href="{$filename}">
            <node><xsl:value-of select="current()" /></node>
        </xsl:result-document>
    </xsl:template>

    <!-- standard copy template -->
    <xsl:template match="@*|node()">
        <xsl:copy>
            <xsl:apply-templates select="@*"/>
            <xsl:apply-templates/>
        </xsl:copy>
    </xsl:template> 
</xsl:stylesheet>

score 0 · Answer 3 · answered Mar 27 '14 at 21:18

Just to understand why your code was throwing an error, I tried it myself. But couldn't reproduce your error. This ran just fine in XMLCalabash:

<p:declare-step name="step_1" type="ts:Extract" version="1.0" xmlns:p="http://www.w3.org/ns/xproc" xmlns:ts="ts">
    <p:input port="source" />
    <p:filter select="/*/node" name="step_1.1" />
     <p:for-each name="step_1.2">
        <p:iteration-source><p:pipe port="result" step="step_1.1"/></p:iteration-source>
        <p:store method="text" >
             <p:with-option name="href" select="concat('step_1_', p:iteration-position(), '.txt')"/>
        </p:store>
    </p:for-each>
</p:declare-step>

Note that I had to add p and ts namespace declarations, and I also added p: before interation-position(). I removed 'file:/' for the practical reason that the files were being written in the root of my file-system.

I could find no record of an DX0006 error. Maybe it tried writing files in the root at your machine as well, but was not allowed to?

HTH!

Saving each node of a xml document as a file with XProc

3 Answers3