15

I need to get a list of attribute values from child elements in Python.

It's easiest to explain with an example.

Given some XML like this:

<elements>
    <parent name="CategoryA">
        <child value="a1"/>
        <child value="a2"/>
        <child value="a3"/>
    </parent>
    <parent name="CategoryB">
        <child value="b1"/>
        <child value="b2"/>
        <child value="b3"/>
    </parent>
</elements>

I want to be able to do something like:

>>> getValues("CategoryA")
['a1', 'a2', 'a3']
>>> getValues("CategoryB")
['b1', 'b2', 'b3']

It looks like a job for XPath but I'm open to all recommendations. I'd also like to hear about your favourite Python XML libraries.

7 Answers 7

7

I'm not really an old hand at Python, but here's an XPath solution using libxml2.

import libxml2

DOC = """<elements>
    <parent name="CategoryA">
        <child value="a1"/>
        <child value="a2"/>
        <child value="a3"/>
    </parent>
    <parent name="CategoryB">
        <child value="b1"/>
        <child value="b2"/>
        <child value="b3"/>
    </parent>
</elements>"""

doc = libxml2.parseDoc(DOC)

def getValues(cat):
    return [attr.content for attr in doc.xpathEval("/elements/parent[@name='%s']/child/@value" % (cat))]

print getValues("CategoryA")

With result...

['a1', 'a2', 'a3']
Sign up to request clarification or add additional context in comments.

3 Comments

Accepted because this is what I ended up using. It's a simple one-liner and I didn't need to install any extra modules. Check out the other answers too - there's some good stuff there.
python test.py Traceback (most recent call last): File "test.py", line 1, in <module> import libxml2 ImportError: No module named libxml2
@SR query: You'll probably need libxml2 to use this libxml2 example.
7

ElementTree 1.3 (unfortunately not 1.2 which is the one included with Python) supports XPath like this:

import elementtree.ElementTree as xml

def getValues(tree, category):
    parent = tree.find(".//parent[@name='%s']" % category)
    return [child.get('value') for child in parent]

Then you can do

>>> tree = xml.parse('data.xml')
>>> getValues(tree, 'CategoryA')
['a1', 'a2', 'a3']
>>> getValues(tree, 'CategoryB')
['b1', 'b2', 'b3']

lxml.etree (which also provides the ElementTree interface) will also work in the same way.

Comments

4

You can do this with BeautifulSoup

>>> from BeautifulSoup import BeautifulStoneSoup
>>> soup = BeautifulStoneSoup(xml)
>>> def getValues(name):
. . .      return [child['value'] for child in soup.find('parent', attrs={'name': name}).findAll('child')]

If you're doing work with HTML/XML I would recommend you take a look at BeautifulSoup. It's similar to the DOM tree but contains more functionality.

Comments

4

Using a standard W3 DOM such as the stdlib's minidom, or pxdom:

def getValues(category):
    for parent in document.getElementsByTagName('parent'):
        if parent.getAttribute('name')==category:
            return [
                el.getAttribute('value')
                for el in parent.getElementsByTagName('child')
            ]
    raise ValueError('parent not found')

Comments

4

My preferred python xml library is lxml , which wraps libxml2.
Xpath does seem the way to go here, so I'd write this as something like:

from lxml import etree

def getValues(xml, category):
    return [x.attrib['value'] for x in 
            xml.findall('/parent[@name="%s"]/*' % category)]

xml = etree.parse(open('filename.xml'))

>>> print getValues(xml, 'CategoryA')
['a1', 'a2', 'a3']
>>> print getValues(xml, 'CategoryB')
['b1', 'b2', 'b3]

Comments

4

In Python 3.x, fetching a list of attributes is a simple task of using the member items()

Using the ElementTree, below snippet shows a way to get the list of attributes. NOTE that this example doesn't consider namespaces, which if present, will need to be accounted for.

    import xml.etree.ElementTree as ET

    flName = 'test.xml'
    tree = ET.parse(flName)
    root = tree.getroot()
    for element in root.findall('<child-node-of-root>'):
        attrList = element.items()
        print(len(attrList), " : [", attrList, "]" )

REFERENCE:

Element.items()
Returns the element attributes as a sequence of (name, value) pairs.
The attributes are returned in an arbitrary order.

Python manual

Comments

2

I must admit I'm a fan of xmltramp due to its ease of use.

Accessing the above becomes:

  import xmltramp

  values = xmltramp.parse('''...''')

  def getValues( values, category ):
    cat = [ parent for parent in values['parent':] if parent(name) == category ]
    cat_values = [ child(value) for child in parent['child':] for parent in cat ]
    return cat_values

  getValues( values, "CategoryA" )
  getValues( values, "CategoryB" )

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.