update-planet

   1 #!/usr/bin/python
   2 #
   3 # update-planet
   4 #
   5 # Downloads feeds from the URLs specified and generates the XHTML files.
   6 #
   7 # (c) 2004, Davyd Madeley <[email protected]>
   8 #
   9
  10 # standard python modules
  11 import sys, codecs
  12 # planetUCC modules
  13 import XMLParse2 as XMLParse, XMLWriter, CacheHandler
  14 # planetUCC output plugins
  15 import XHTMLWriter, RSS2Writer, RSS1Writer, FOAFWriter, OPMLWriter
  16
  17 # step 1: read in the config and check each object from cache
  18 cache   = CacheHandler.CacheHandler()
  19 feeds   = []
  20
  21 for feed in open('feedlist').readlines():
  22         if feed.strip()[0] != '#':
  23                 storage         = feed.strip().split('\t')
  24                 name, feed      = storage[0], storage[-1]
  25                 try:
  26                         feeds.append((name, feed, cache.getBlog(name, feed)))
  27                 except:
  28                         sys.stderr.write('DEBUG: update-planet: something went wrong retrieving feed\n')
  29
  30 # step 2: process each feed
  31 tainted = False
  32 blogs   = []
  33 for feed in feeds:
  34         # XMLParse2 takes two paramaters, a URL and a CacheObject
  35         blog    = XMLParse.XMLParse(feed[1], feed[2]).parse()
  36         if blog:
  37                 blog.blogName   = feed[0]
  38                 blog.feedURL    = feed[1]
  39                 blogs.append(blog)
  40                 # check the old copy of the cache, vs the new copy
  41                 if not feed[2] or not feed[2].cache or not blog.cache or feed[2].cache != blog.cache:
  42                         tainted = True
  43                 elif len(blog.items) > 0 and len(feed[2].items) > 0 and (blog.items[0].itemTitle != feed[2].items[0].itemTitle or blog.items[0].contents != feed[2].items[0].contents):
  44                         tainted = True
  45                 # write the cache back down to disk
  46                 cache.storeBlog(blog)
  47         else:
  48                 pass
  49
  50 # step 3: sift the feeds
  51 xmlwriter       = XMLWriter.XMLWriter(blogs)
  52
  53 # step 4: write feed to disk
  54 try:
  55         codecs.open('planet.html', 'wb', 'utf-8').write(xmlwriter.write(XHTMLWriter.XHTMLWriter))
  56 except:
  57         sys.stderr.write('DEBUG: update-planet: could not write planet.html, aborting\n')
  58         raise
  59
  60 try:
  61         codecs.open('rss2.xml', 'wb', 'utf-8').write(xmlwriter.write(RSS2Writer.RSS2Writer))
  62 except:
  63         sys.stderr.write('DEBUG: update-planet: could not write rss2.xml, aborting\n')
  64
  65 try:
  66         codecs.open('rss1.xml', 'wb', 'utf-8').write(xmlwriter.write(RSS1Writer.RSS1Writer))
  67 except:
  68         sys.stderr.write('DEBUG: update-planet: could not write rss1.xml, aborting\n')
  69
  70 try:
  71         codecs.open('foaf.xml', 'wb', 'utf-8').write(xmlwriter.write(FOAFWriter.FOAFWriter))
  72 except:
  73         sys.stderr.write('DEBUG: update-planet: could not write foaf.xml, aborting\n')
  74
  75 try:
  76         codecs.open('opml.xml', 'wb', 'utf-8').write(xmlwriter.write(OPMLWriter.OPMLWriter))
  77 except:
  78         sys.stderr.write('DEBUG: update-planet: could not write opml.xml, aborting\n')
  79
  80
  81 if not tainted:
  82         sys.stdout.write('PlanetUCC: no objects have changed in the cache, not updating\n')
  83         sys.exit(1)
  84