Reading XML using Python minidom and iterating over each node

Question

I have an XML structure that looks like the following, but on a much larger scale:

<root>
<conference name='1'>
<author>Bob</author>
<author>Nigel</author>
</conference>
<conference name='2'>
<author>Alice</author>
<author>Mary</author>
</conference>
</root>

For this, I used the following code:

dom = parse(filepath)
conference=dom.getElementsByTagName('conference')
for node in conference:
    conf_name=node.getAttribute('name')
    print conf_name
    alist=node.getElementsByTagName('author')
    for a in alist:
        authortext= a.nodeValue
        print authortext

However, the authortext that is printed out is 'None.' I tried messing around with using variations like what is below, but it causes my program to break.

authortext=a[0].nodeValue

The correct output should be:

1
Bob
Nigel
2
Alice
Mary

But what I get is:

1
None
None
2
None
None

Any suggestions on how to tackle this problem?

Answer 1

I played around with it a bit, and here's what I got to work:

# ...
authortext= a.childNodes[0].nodeValue
print authortext

leading to output of:

C:\temp\py>xml2.py
1
Bob
Nigel
2
Alice
Mary

I can't tell you exactly why you have to access the childNode to get the inner text, but at least that's what you were looking for.

Answer 2

your authortext is of type 1 (ELEMENT_NODE), normally you need to have TEXT_NODE to get a string. This will work

a.childNodes[0].nodeValue

Answer 3

Element nodes don't have a nodeValue. You have to look at the Text nodes inside them. If you know there's always one text node inside you can say element.firstChild.data (data is the same as nodeValue for text nodes).

Be careful: if there is no text content there will be no child Text nodes and element.firstChild will be null, causing the .data access to fail.

Quick way to get the content of direct child text nodes:

text= ''.join(child.data for child in element.childNodes if child.nodeType==child.TEXT_NODE)

In DOM Level 3 Core you get the textContent property you can use to get text from inside an Element recursively, but minidom doesn't support this (some other Python DOM implementations do).