XMLParse2.py

   1 #
   2 # XMLParse2.py
   3 #
   4 # Parse arbitrary XML news streams into an object type
   5 # understandable by Planet UCC.
   6 # Now uses feedparser to parse 9 different types of RSS _and_ Atom
   7 #
   8 # (c) 2004, Davyd Madeley <[email protected]>
   9 #
  10
  11 import sys, time
  12 import CacheHandler
  13 sys.path.insert(0, 'extra')
  14 import feedparser
  15
  16 class Blog:
  17         def __init__(self):
  18                 self.blogTitle  = None
  19                 self.blogURL    = None
  20                 self.feedURL    = None
  21                 self.imageURL   = None
  22                 self.imageLink  = None
  23                 self.items      = []
  24                 self.cache      = None
  25
  26 class BlogItem:
  27         def __init__(self):
  28                 self.itemTitle  = None
  29                 self.itemDate   = None
  30                 self.itemURL    = None
  31                 self.contents   = None
  32
  33 class XMLParse:
  34         def __init__(self, URL, blogObject):
  35                 self.feedURL    = URL
  36                 self.blogObject = blogObject
  37
  38         def parse(self):
  39                 "Return a single Blog object"
  40                 item            = Blog()
  41                 if self.blogObject:
  42                         sys.stdout.write('Downloading feed %s...' % self.feedURL)
  43                         try:
  44                                 data    = feedparser.parse(self.feedURL, self.blogObject.cache.etag, self.blogObject.cache.date)
  45                                 sys.stdout.write('done.\n')
  46                         except:
  47                                 sys.stdout.write('failed.\n')
  48                                 raise
  49                                 return None
  50                         # check to see what we got returned
  51                         if data['items'] == [] and data['channel'] == {}:
  52                                 sys.stdout.write('Feed %s is upto date.\n' % self.feedURL)
  53                                 return self.blogObject
  54                 else:
  55                         sys.stdout.write('Downloading feed from %s (no cache)...' % self.feedURL)
  56                         try:
  57                                 data    = feedparser.parse(self.feedURL)
  58                                 sys.stdout.write('done.\n')
  59                         except:
  60                                 sys.stdout.write('failed.\n')
  61                                 return None
  62                 # create caching data
  63                 try:
  64                         cache           = CacheHandler.CacheObject()
  65                         cache.etag      = data['etag']
  66                         cache.date      = data['modified']
  67                         item.cache      = cache
  68                 except:
  69                         item.cache      = None
  70                 # parse the return of data into a blog
  71                 if data['channel'].has_key('title'):
  72                         item.blogTitle  = data['channel']['title']
  73                 else:
  74                         item.blogTitle  = '(Unknown)'
  75                 if data['channel'].has_key('link'):
  76                         item.blogURL    = data['channel']['link']
  77                 else:
  78                         item.blogURL    = self.feedURL
  79                 for entry in data['items']:
  80                         blogItem                = BlogItem()
  81                         if entry.has_key('title'):
  82                                 blogItem.itemTitle      = entry['title']
  83                         else:
  84                                 blogItem.itemTitle      = '(Untitled)'
  85                         if entry.has_key('link'):
  86                                 blogItem.itemURL        = entry['link']
  87                         else:
  88                                 blogItem.itemURL        = item.blogURL
  89                         if entry.has_key('date_parsed'):
  90                                 blogItem.itemDate       = time.mktime(entry['date_parsed']) + 28000
  91                         else:
  92                                 blogItem.itemDate       = 0
  93                         if entry.has_key('description'):
  94                                 blogItem.contents       = entry['description']
  95                         else:
  96                                 blogItem.contents       = '(entry could not be retrieved)'
  97                         item.items.append(blogItem)
  98                 return item