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 feedparser.USER_AGENT = "PlanetUCC/1.0b +http://planet.ucc.asn.au/ %s" % feedparser.USER_AGENT
  17
  18 class Blog:
  19         def __init__(self):
  20                 self.blogTitle  = None
  21                 self.blogURL    = None
  22                 self.feedURL    = None
  23                 self.imageURL   = None
  24                 self.imageLink  = None
  25                 self.items      = []
  26                 self.cache      = None
  27
  28 class BlogItem:
  29         def __init__(self):
  30                 self.itemTitle  = None
  31                 self.itemDate   = None
  32                 self.itemURL    = None
  33                 self.contents   = None
  34
  35 class XMLParse:
  36         def __init__(self, URL, blogObject):
  37                 self.feedURL    = URL
  38                 self.blogObject = blogObject
  39
  40         def parse(self):
  41                 "Return a single Blog object"
  42                 item            = Blog()
  43                 if self.blogObject and self.blogObject.cache:
  44                         sys.stdout.write('Downloading feed %s...' % self.feedURL)
  45                         try:
  46                                 data    = feedparser.parse(self.feedURL, self.blogObject.cache.etag, self.blogObject.cache.date)
  47                                 sys.stdout.write('done.\n')
  48                         except:
  49                                 sys.stdout.write('failed.\n')
  50                                 return None
  51                         # check to see what we got returned
  52                         if data['items'] == [] and data['channel'] == {}:
  53                                 sys.stdout.write('Feed %s is upto date.\n' % self.feedURL)
  54                                 return self.blogObject
  55                 else:
  56                         sys.stdout.write('Downloading feed from %s (no cache)...' % self.feedURL)
  57                         try:
  58                                 data    = feedparser.parse(self.feedURL)
  59                                 sys.stdout.write('done.\n')
  60                         except:
  61                                 sys.stdout.write('failed.\n')
  62                                 return None
  63                 # create caching data
  64                 try:
  65                         cache           = CacheHandler.CacheObject()
  66                         try:
  67                                 cache.etag      = data['etag']
  68                         except:
  69                                 cache.etag      = None
  70                         try:
  71                                 cache.date      = data['modified']
  72                         except:
  73                                 cache.date      = None
  74                         item.cache      = cache
  75                 except:
  76                         sys.stderr.write('DEBUG: XMLParse2: cache item generation failed\n')
  77                         item.cache      = None
  78                 # parse the return of data into a blog
  79                 if data['channel'].has_key('title'):
  80                         item.blogTitle  = data['channel']['title']
  81                 else:
  82                         item.blogTitle  = '(Unknown)'
  83                 if data['channel'].has_key('link'):
  84                         item.blogURL    = data['channel']['link']
  85                 else:
  86                         item.blogURL    = self.feedURL
  87                 for entry in data['items']:
  88                         blogItem                = BlogItem()
  89                         if entry.has_key('title'):
  90                                 blogItem.itemTitle      = entry['title']
  91                         else:
  92                                 blogItem.itemTitle      = '(Untitled)'
  93                         if entry.has_key('link'):
  94                                 blogItem.itemURL        = entry['link']
  95                         else:
  96                                 blogItem.itemURL        = item.blogURL
  97                         if entry.has_key('date_parsed'):
  98                                 blogItem.itemDate       = time.mktime(entry['date_parsed']) + 28800
  99                         else:
 100                                 blogItem.itemDate       = 0
 101                         if entry.has_key('description'):
 102                                 blogItem.contents       = entry['description']
 103                         else:
 104                                 blogItem.contents       = '(entry could not be retrieved)'
 105                         item.items.append(blogItem)
 106                 return item