Extract attributes and certain tag values from xml using python script

Question

I want to parse an XML content and return a dictionary which contains only the name attribute and its values as dictionary. For example:

  <ecmaarray>
   <number name="xyz1">123.456</number>  
   <ecmaarray name="xyz2">  
       <string name="str1">aaa</string>  
       <number name="num1">55</number>  
   </ecmaarray>  
   <strictarray name="xyz3">  
       <string>aaa</string>  
       <number>55</number>  
   </strictarray>  
</ecmaarray>  

The output has to be in a dictionary something like this..

Dict:{ 'xyz1': 123.456, 
       'xyz2': {'str1':'aaa', 'num1': '55'},
       'xyz3': ['aaa','55']
     }

Can any one suggest a recursive solution for this ?

Answer 1

Assuming situation like this:

<strictarray name="xyz4">
    <string>aaa</string>
    <number name="num1">55</number>
</strictarray>

is not possible, here's a sample code using lxml:

from lxml import etree


tree = etree.parse('test.xml')

result = {}
for element in tree.xpath('/ecmaarray/*'):
    name = element.attrib["name"]
    text = element.text
    childs = element.getchildren()

    if not childs:
        result[name] = text
    else:
        child_dict = {}
        child_list = []
        for child in childs:
            child_name = child.attrib.get('name')
            child_text = child.text
            if child_name:
                child_dict[child_name] = child_text
            else:
                child_list.append(child_text)

        if child_dict:
            result[name] = child_dict
        else:
            result[name] = child_list


print result

prints:

{'xyz3': ['aaa', '55'], 
 'xyz2': {'str1': 'aaa', 'num1': '55'}, 
 'xyz1': '123.456'}

You may want to improve the code - it's just a hint on where to go.

Hope that helps.