不区分大小写的xml和python

Question

I got this piece of code and I am trying to read all the 'ref' 'href' tags. I am not sure how to make this to be case insensitive as some of my xml files have REF or Ref or ref. Any suggestions?

    f = urllib.urlopen(url)
    tree = ET.parse(f)
    root = tree.getroot()

    for child in root.iter('ref'):
      t = child.get('href')
      if t not in self.href:
        self.href.append(t)
        print self.href[-1]

Answer 1

You can normalize tags and attributes by converting them to lowercase using the functions below as a step of preprocessing:

import xml.etree.ElementTree as ET
f = urllib.urlopen(url)
tree = ET.parse(f)
root = tree.getroot()

def normalize_tags(root):
    root.tag = root.tag.lower()
    for child in root:
        normalize_tags(child)

def normalize_attr(root):
    for attr,value in root.attrib.items():
        norm_attr = attr.lower()
        if norm_attr != attr:
            root.set(norm_attr,value)
            root.attrib.pop(attr)

    for child in root:
        normalize_attr(child)


normalize_tags(root)    
normalize_attr(root)
print(ET.tostring(root))

Answer 2

The following should help

f = urllib.urlopen(url)
tree = ET.parse(f)
root = tree.getroot()

for child in root:
  if child.tag.lower() == 'ref':
    t = child.attribute.get('href')
    if t not in self.href:
      self.href.append(t)
      print self.href[-1]

Answer 3

If you are using lxml then one option is to use XPath with regular expressions through XSLT extensions ( https://stackoverflow.com/a/2756994/2997179 ):

root.xpath("./*[re:test(local-name(), '(?i)href')]",
    namespaces={"re": "http://exslt.org/regular-expressions"})