Question

I have some tagged data that I am processing using lxml. When I open a file I do not know before the file is opened if I have one or more of three types of elements (I could have one, two or three different elements and multiple instances of any type I have)

I need some information about these elements that is contained in child tags of the element

<element_type_1>
        <name>joe smith</name>
</element_type_1>
<element_type_2>
        <name>mary smith</name>
</element_type_2>
<element_type_3>
        <name>patrick smith</name>
</element_type_3>

So in this case I have all three types but only one of each type however there could be up to some arbitrary large number of any type.

I am getting the elements by using cssselect 3 times in my function

def get_types(myTree):
    type_dict=defaultdict(list)
    type_dict['type_1']=myTree.cssselect('element_type_1')
    type_dict['type_2']=myTree.cssselect('element_type_2')
    type_dict['type_3']=myTree.cssselect('element_type_3')
    ret type_dict

This seems overly redundant

Am I missing something that would clean up this a bit?

FYI I am doing this because for each type I have to match some other data from a related document

The early answers suggest I need to clarify a bit – I want to avoid running through the tree three times

Answer 1

You could do this:

for i in range(1, 4):
    type_dict['type_%d' % i] = myTree.cssselect('element_type_%d' % i)