treecutter/directory.py

   1 #!/usr/bin/python
   2 import os
   3 import fnmatch
   4 from lxml import etree
   5 import treecutter.constants as const
   6 import re
   7
   8 class Directory():
   9     """Class containing the state of the directory with articles"""
  10     def __init__(self):
  11         self._cwd = u'.'
  12         self._tree = []
  13         self._basepath = re.compile('[/\w\._-]*/\w+',re.UNICODE)
  14
  15     def scan(self):
  16         for dirname, dirnames, filenames in os.walk(self._cwd):
  17             for filename in filenames:
  18                 if fnmatch.fnmatch(filename, '*.xml'):
  19                     file_ = os.path.join(dirname,filename)
  20                     doc = etree.parse(file_)
  21                     title = doc.xpath(u'/db:article/db:info/db:title',namespaces=const.XPATH)
  22                     menu  = doc.xpath(u'/db:article/db:info/db:titleabbrev',namespaces=const.XPATH)
  23                     if title and menu:
  24                         base = self._basepath.match(file_).group()
  25                         link = base.replace('index','')[1:]
  26                         self._tree.append(link)
  27
  28     def set(self):
  29         return set(self._tree)