xml containing 1 child

Consider the following code

xml = """<?xml version="1.0" encoding="utf-8" ?>
<root>
    <children>
        <child>
            <name>A</name>
        </child>
    </children>
</root>"""

xmltodict.parse(xml)['root']['children']['child']

Wouldn’t you expect to have an iterable object even when there is only 1 child?

[acjackson]

How can I use the force_list option for ALL elements? Do I really have to list every element?

[candlerb]

What you could do is to nominate particular elements which should always generate lists, e.g.

xmltodict.parse(xml, force_list=set('child'))

This is the approach taken by Perl’s XML::Simple with the ForceArray setting.

A problem with this is that if the <children> element is empty, it’s not going to give you an empty list of {"child": []}. This is due to how xmltodict groups elements at a lower level, i.e. it’s not children which is a list, but child which is a list.

So to work with xmltodict, I think you need to specify the enclosing element as well:

xmltodict.parse(xml, force_list={'children': 'child'})

Implies: “if you see a <children> element then always build a <child> with empty list underneath it”. Any actual <child> elements will then be added to that list.

Aside: there is a related but different problem, which is dealing with a heterogenous sequence of repeating elements. For example:

<stuff>
  <foo>.1.</foo>
  <bar>.2.</bar>
  <bar>.3.</bar>
  <foo>.4.</foo>
  <bar>.5.</bar>
</stuff>

At the moment you get:

OrderedDict([(u'stuff', OrderedDict([(u'foo', [u'.1.', u'.4.']), (u'bar', [u'.2.', u'.3.', u'.5.'])]))])

which loses the ordering relationship between the foo and bar elements.

This would have to be dealt with differently. For example:

xmltodict.parse(xml, force_list={'stuff': True})

could mean that <stuff> is forced to have a list of children, and that this list should contain repeated elements rather than grouped elements. I don’t know how practical that is to implement though. And it would probably have implications for unparse.