get_xpath.py

def get_xpath(element):
    """Get the XPath string to element.

    This code is a port of a C# snippet published at
    http://stackoverflow.com/a/9221904
    """
    path = '/' + element.tag
    parent = element.getparent()

    if parent is not None:
        siblings = parent.findall(element.tag)

        if len(siblings) > 1:
            position = 1
            for sibling in siblings:
                if sibling is element:
                    break
                else:
                    position += 1
            path += '[%s]' % position

        path = get_xpath(parent) + path

    return path

tests.py

import unittest
from StringIO import StringIO

from lxml import etree


class GetXpathTests(unittest.TestCase):
    def test_root_level(self):
        sample = StringIO("""
            <article>
              <front>
                Hey!
              </front>
            </article>
         """)
        sample = etree.parse(sample)
        element = sample.getroot()

        self.assertEqual(utils.get_xpath(element), '/article')

    def test_first_level(self):
        sample = StringIO("""
            <article>
              <front>
                Hey!
              </front>
            </article>
         """)
        sample = etree.parse(sample)
        element = sample.xpath('front')[0]

        resulting_xpath = utils.get_xpath(element)
        self.assertEqual(resulting_xpath, '/article/front')
        self.assertIn(element, sample.xpath(resulting_xpath))

    def test_first_level_many(self):
        sample = StringIO("""
            <article>
              <front>
                Hey!
              </front>
              <front>
                You!
              </front>
            </article>
         """)
        sample = etree.parse(sample)
        element = sample.xpath('front')[1]

        resulting_xpath = utils.get_xpath(element)
        self.assertEqual(resulting_xpath, '/article/front[2]')
        self.assertIn(element, sample.xpath(resulting_xpath))

    def test_second_level(self):
        sample = StringIO("""
            <article>
              <front>
                Hey!
                <title>
                  Foobar
                </title>
              </front>
            </article>
         """)
        sample = etree.parse(sample)
        element = sample.xpath('front/title')[0]

        resulting_xpath = utils.get_xpath(element)
        self.assertEqual(resulting_xpath, '/article/front/title')
        self.assertIn(element, sample.xpath(resulting_xpath))

    def test_first_level_many(self):
        sample = StringIO("""
            <article>
              <front>
                Hey!
              </front>
              <front>
                You!
                <title>
                  Foobar
                </title>
                <title title-type="actual">
                  Baz
                </title>
              </front>
            </article>
         """)
        sample = etree.parse(sample)
        element = sample.xpath('front/title[@title-type="actual"]')[0]

        resulting_xpath = utils.get_xpath(element)
        self.assertEqual(resulting_xpath, '/article/front[2]/title[2]')
        self.assertIn(element, sample.xpath(resulting_xpath))