Iterate xml using namespace prefix

Question

I have an xml file with a default namespace, like this:

<?xml version="1.0" encoding="utf-8"?>
<root xmlns="somelongnamespace">
 <child>...</child>
</root>

I starting using lxml to iterate and query this file, but I would like to use a namespace prefix, like this:

from lxml import etree
xml = etree.parse("myfile.xml")
root = xml.getroot()
c = root.findall('ns:child')

What do I need to do for this to work? I cannot change the file, but I could change the xml object after loading.

I read the relevant lxml documentation, searched and tried all kinds of suggestions, but got none of them to work unfortunately. This does sound like a very common question...?

Answer 1

You could map the ns prefix to your default namespace by creating a Python dictionary.

from lxml import etree
xml = etree.parse("myfile.xml")
root = xml.getroot()
# a dictionary where the key "ns" is the prefix you want to use, and the value "somelongnamespace" is the namespace URI from your XML
namespace = {"ns": "somelongnamespace"}
children = root.findall('ns:child', namespaces=namespace)
if children:
 print("Found children:")
 for child in children:
 print(f"{etree.tostring(child, encoding='unicode').strip()}")
else:
 print("No children found.")

---

If you need to perform many operations on the XML, you can create a class that encapsulates the namespace and provides methods for querying the XML.

class XmlNamespaceHandler:
 def __init__(self, xml_file, namespace):
 self.xml = etree.parse(xml_file)
 self.root = self.xml.getroot()
 self.nsmap = {'ns': namespace}
 def findall(self, xpath):
 return self.root.findall(xpath, namespaces=self.nsmap)
xml_handler = XmlNamespaceHandler("myfile.xml", "somelongnamespace")
children = xml_handler.findall('ns:child')

• Michel de Ruiter: Sure. But is there no way around having to pass namespaces=mynamespaces every time? :-(
• Lewis: @MicheldeRuiter I added an alternative where the namespace and XML data are encapsulated within a class.