Change feed URL.

[planet-ucc.git] / update-planet

diff --git a/update-planet b/update-planet
index c6f04dd..7f31e52 100755 (executable)
--- a/update-planet
+++ b/update-planet
@@ -7,34 +7,78 @@
 # (c) 2004, Davyd Madeley <[email protected]>
 #
 
-import sys, urllib2, codecs
-import XMLParse, XMLWriter
+# standard python modules
+import sys, codecs
+# planetUCC modules
+import XMLParse2 as XMLParse, XMLWriter, CacheHandler
+# planetUCC output plugins
+import XHTMLWriter, RSS2Writer, RSS1Writer, FOAFWriter, OPMLWriter
 
-# step 1: read in the config and download the feeds
+# step 1: read in the config and check each object from cache
+cache  = CacheHandler.CacheHandler()
 feeds  = []
+
 for feed in open('feedlist').readlines():
        if feed.strip()[0] != '#':
                storage         = feed.strip().split('\t')
                name, feed      = storage[0], storage[-1]
-               sys.stdout.write('Downloading feed "%s" from %s... ' % (name, feed))
                try:
-                       # XXX: might want to consider some good caching code in here
-                       feeds.append((name, feed, urllib2.urlopen(feed).read()))
-                       sys.stdout.write('done.\n')
+                       feeds.append((name, feed, cache.getBlog(name, feed)))
                except:
-                       sys.stdout.write('failed.\n')
+                       sys.stderr.write('DEBUG: update-planet: something went wrong retrieving feed\n')
 
 # step 2: process each feed
+tainted        = False
 blogs  = []
 for feed in feeds:
-       xml     = XMLParse.XMLParse(feed[2]).parse()
-       for blog in xml:
-               blog.blogTitle  = feed[0]
-       blogs   += xml
+       # XMLParse2 takes two paramaters, a URL and a CacheObject
+       blog    = XMLParse.XMLParse(feed[1], feed[2]).parse()
+       if blog:
+               blog.blogName   = feed[0]
+               blog.feedURL    = feed[1]
+               blogs.append(blog)
+               # check the old copy of the cache, vs the new copy
+               if not feed[2] or not feed[2].cache or not blog.cache or feed[2].cache != blog.cache:
+                       tainted = True
+               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):
+                       tainted = True
+               # write the cache back down to disk
+               cache.storeBlog(blog)
+       else:
+               pass
+
+# step 3: sift the feeds
+xmlwriter      = XMLWriter.XMLWriter(blogs)
 
-# step 3: write feed to disk
+# step 4: write feed to disk
 try:
-       codecs.open('planet.html', 'wb', 'utf-8').write(XMLWriter.XMLWriter(XMLWriter.XHTMLWriter, blogs).write())
+       codecs.open('planet.html', 'wb', 'utf-8').write(xmlwriter.write(XHTMLWriter.XHTMLWriter))
 except:
        sys.stderr.write('DEBUG: update-planet: could not write planet.html, aborting\n')
        raise
+
+try:
+       codecs.open('rss2.xml', 'wb', 'utf-8').write(xmlwriter.write(RSS2Writer.RSS2Writer))
+except:
+       sys.stderr.write('DEBUG: update-planet: could not write rss2.xml, aborting\n')
+
+try:
+       codecs.open('rss1.xml', 'wb', 'utf-8').write(xmlwriter.write(RSS1Writer.RSS1Writer))
+except:
+       sys.stderr.write('DEBUG: update-planet: could not write rss1.xml, aborting\n')
+
+try:
+       codecs.open('foaf.xml', 'wb', 'utf-8').write(xmlwriter.write(FOAFWriter.FOAFWriter))
+except:
+       sys.stderr.write('DEBUG: update-planet: could not write foaf.xml, aborting\n')
+
+try:
+       codecs.open('opml.xml', 'wb', 'utf-8').write(xmlwriter.write(OPMLWriter.OPMLWriter))
+except:
+       sys.stderr.write('DEBUG: update-planet: could not write opml.xml, aborting\n')
+
+
+if not tainted:
+       sys.stdout.write('PlanetUCC: no objects have changed in the cache, not updating\n')
+       sys.exit(1)
+