Remove XML node using Java SE

Question

How to remove XML nodes in Java SE? I am using org.apache.xerces. Thanks. The below code isn't working.

DOMParser parser = new DOMParser();
System.out.println(DIR_STRING + "/" + jmsFileNameString);
parser.parse(DIR_STRING + "/" + jmsFileNameString);
Document doc = parser.getDocument();
NodeList list = doc.getElementsByTagName("*");
for (int i = 0; i < list.getLength(); i++) {
    if (list.item(i).getNodeName().matches(HEADER_REGEXP)) {
        list.item(i).getParentNode().removeChild(list.item(i)));
    }
}

G_H · Accepted Answer · 2011-08-25 12:14:57Z

Creating an entire DOM for this is rather overkill. You'll have the entire XML tree in memory, which can be rather heavy for large documents. I suggest one of the following:

Parse with SAX or StAX, simply copying stuff to output unless you want it filtered out.
Apply an XSLT transformation that copies everything by default, but has one or more templates that don't do anything with their input, thus filtering it out.

Option 2 is the easiest and in my experience XSLT in Java is fast and memory-efficient, especially for a simple use-case like this.

These two templates will be what you need:

The default copy

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

The "filter":

<xsl:template match="//*[your predicate here]">
    <!-- Don't do a thing -->
</xsl:template>

EDIT: I just noticed that you don't just filter out specific names, but those that match a regular expression. XPath functions could be enough to make the predicate selecting the target nodes. But if needed, Java String functions can be used in XSLT through extension. It does make this solution slightly more complicated, but still worth it for taking the XML parsing out of your hands.

Shumon Saha · Accepted Answer · 2011-08-25 12:40:31Z

0

It does work; however, it doesn't save to the XML file. The changes are saved in the Document object.

To save to the Document object in a file:

    OutputFormat of = new OutputFormat("XML","UTF-8",true);
    XMLSerializer serializer = new XMLSerializer();
    serializer.setOutputFormat(of);
    serializer.setOutputByteStream(new FileOutputStream(PATH));
    serializer.serialize(doc);

edited Aug 25, 2011 at 12:40

answered Aug 25, 2011 at 12:13

Shumon Saha

1,4456 gold badges17 silver badges33 bronze badges

1 Comment

forty-two Over a year ago

Ok, then you need to save the tree back to XML. You can use LSSerializer from org.w3c.dom.ls package

Collectives™ on Stack Overflow

Remove XML node using Java SE

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related