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 from treecutter.docbook import Docbook
   7 import re
   8
   9 class Directory():
  10     """Class containing the state of the directory with articles"""
  11     def __init__(self):
  12         self._cwd = u'.'
  13         self._tree = []
  14         self._basepath = re.compile('[/\w\._-]*/[\w-]+',re.UNICODE)
  15
  16     def scan(self):
  17         for dirname, dirnames, filenames in os.walk(self._cwd):
  18             for filename in filenames:
  19                 if fnmatch.fnmatch(filename, '*.xml'):
  20                     file_ = os.path.join(dirname,filename)
  21                     doc = Docbook(file_)
  22                     (title, menu) = doc.title()
  23                     draft = doc.status() == "draft"
  24                     level = doc.userlevel()
  25
  26 #                    doc = etree.parse(file_)
  27 #                    title = doc.xpath(u'/db:article/db:info/db:title',namespaces=const.XPATH)
  28 #                    menu  = doc.xpath(u'/db:article/db:info/db:titleabbrev',namespaces=const.XPATH)
  29 #                    draft = doc.xpath(u'/db:article[@status="draft"]',namespaces=const.XPATH)
  30                     if draft and draftflag:
  31                         draft = False
  32                     if title and menu and not draft and level <= levelflag:
  33                         base = self._basepath.match(file_).group()
  34                         link = base.replace('index','')[1:]
  35                         self._tree.append(link)
  36
  37     def set(self):
  38         return set(self._tree)