Question

I am experiencing some trouble using the Python 2.6.5 xml.etree.ElementTree library. In particular, if I setup a simple xml element like the following

>>> import xml.etree.ElementTree as etree
>>> xml = etree.fromstring("<a><b><c>xy</c></b></a>")

i have no problems with the library when accessing the inner element nodes, e.g.:

>>> etree.tostring(xml.find('b'))
'<b><c>xy</c></b>'
>>> xml.find('b') == None
False
>>> bool(xml.find('b'))
True

However, I am encountering a strange boolean interpretation of leaf element nodes, see:

>>> etree.tostring(xml.find('b/c'))
'<c>xy</c>'
>>> xml.find('b/c') == None
False
>>> bool(xml.find('b/c'))
False

Note that in the last command, the element xml.find(‘b/c’), which is clearly non-None, evaluates to False. This is especially annoying since i cannot use the idiom

>>> leaf = xml.find('b/c'):
>>> if leaf:
>>>     do_stuff(leaf)

to check whether a leaf element exists. (I have to check explicitly for ‘xml.find(‘b/c’) != None’.)

Can someone please explain this (for me unexpected) behavior?

Answer 1

Ok, sorry, after submitting this question I found the explicit statement of the solution in the Python documentation: http://docs.python.org/release/2.6.5/library/xml.etree.elementtree.html#the-element-interface , at the end of the section.

Caution: Because Element objects do not define a nonzero() method, elements with no subelements will test as False.

Sorry for the inconvenience.