Namespace-Aware Parsing in Practice

When working with namespace-aware XML programmatically, it is important to use APIs that expose the expanded name — the combination of namespace URI and local name — rather than the raw prefixed name. In Python's xml.etree.ElementTree, element tags are returned in Clark notation: {namespace-uri}localname. When writing XPath expressions against namespace-aware parsers you must register a prefix-to-URI mapping even if the document uses a different prefix. In Java, the DocumentBuilderFactory and SAXParserFactory classes must have namespace awareness explicitly enabled. Once enabled, the DOM's getNamespaceURI() and getLocalName() methods return the correct values, and SAX callbacks receive separate namespace URI, local name, and qualified name arguments. Failing to enable namespace awareness is a common bug that causes namespace-prefixed elements to appear as if they have no namespace. A related pitfall is assuming that a prefix is stable. Because two documents describing the same vocabulary may use different prefixes, code that compares element.tagName to a prefixed string like "dc:title" is fragile. Always compare against the expanded name. Schema validators, XSLT processors, and XPath engines all operate on expanded names internally, which is why namespace declarations matter even when you control both producer and consumer.

Example

# Python: namespace-aware parsing with ElementTree

import xml.etree.ElementTree as ET

xml_data = """<?xml version="1.0"?>
<root xmlns="http://example.com/default"
      xmlns:dc="http://purl.org/dc/elements/1.1/">
    <item id="1">
        <dc:title>XML Namespaces</dc:title>
        <description>A guide to XML namespaces.</description>
    </item>
</root>"""

tree = ET.fromstring(xml_data)

# Clark notation: {namespace-uri}localname
ns = {
    'ex': 'http://example.com/default',
    'dc': 'http://purl.org/dc/elements/1.1/',
}

for item in tree.findall('ex:item', ns):
    # Attribute id is in null namespace — no prefix needed
    print('ID:', item.get('id'))

    title = item.find('dc:title', ns)
    desc  = item.find('ex:description', ns)

    print('Title:', title.text if title is not None else 'N/A')
    print('Desc: ', desc.text  if desc  is not None else 'N/A')

# Accessing the tag directly returns Clark notation
first_item = tree.find('{http://example.com/default}item')
print('Tag:', first_item.tag)
# Output: {http://example.com/default}item

HTML

Full Editor

Related Resources

XML Reference

Complete tag & property list

XML How-To Guides

Step-by-step practical guides

XML Exercises

Practice what you've learned