i_love_and_hate_xpath.py

from postneo.category.models import Category
from postneo.blog.models import Entry
from lxml import etree

tree = etree.parse(open('../../../wordpress-20090526.xml', 'r'))

def field_data(tree, name):
    """Return the text for a field element with a particular name"""
    return tree.find("field[@name='%s']" % name).text

for row in tree.find("//table_data[@name='wp_categories']").getchildren():
    """Handle category import"""
    c, created = Category.objects.get_or_create(
        pk = field_data(row, 'cat_ID'),
        name = field_data(row, 'cat_name'),
        slug = field_data(row, 'category_nicename'),
        description = field_data(row, 'category_description') or '',
        parent__pk=field_data(row, 'category_parent') or None)
    print c