I have an XML file which has a lot of different nodes with the same attribute.
I was wondering if it's possible to find all these nodes using Python and any additional package like minidom or ElementTree.
-
thanks guys, i was using minidom and seems like it's not so powerful, will move to ElementTree finally.midori– midori2015-01-07 03:30:29 +00:00Commented Jan 7, 2015 at 3:30
Add a comment
|
2 Answers
You can use built-in xml.etree.ElementTree module.
If you want all elements that have a particular attribute regardless of the attribute values, you can use an xpath expression:
//tag[@attr]
Or, if you care about values:
//tag[@attr="value"]
Example (using findall() method):
import xml.etree.ElementTree as ET
data = """
<parent>
<child attr="test">1</child>
<child attr="something else">2</child>
<child other_attr="other">3</child>
<child>4</child>
<child attr="test">5</child>
</parent>
"""
parent = ET.fromstring(data)
print [child.text for child in parent.findall('.//child[@attr]')]
print [child.text for child in parent.findall('.//child[@attr="test"]')]
Prints:
['1', '2', '5']
['1', '5']
3 Comments
alex
Why assume that all of these nodes are going to be direct children of the root element? What about if the children and grandchildren, or descendants at any level of nesting, have these attributes?
parent.findall('.//[@attr]') throws an invalid descendant error in Python 2.7.
alecxe
@alex wait,
// or .// would look for node at any level of nesting. Or, I misunderstood your question..thanks.alecxe
@alex ah, if you mean to look for any node name, use
.//*[@attr].This is a good sample/start script using xpath :
# -*- coding: utf-8 -*-
from lxml import etree
fp = open("xml.xml")
tree = etree.parse(fp)
for el in tree.findall('//node[@attr="something"]'):
print(el.text)
