See Changelog

[planet-ucc.git] / XMLParse2.py

diff --git a/XMLParse2.py b/XMLParse2.py
new file mode 100644 (file)
index 0000000..1cec668
--- /dev/null
+++ b/XMLParse2.py
@@ -0,0 +1,98 @@
+#
+# XMLParse2.py
+#
+# Parse arbitrary XML news streams into an object type
+# understandable by Planet UCC.
+# Now uses feedparser to parse 9 different types of RSS _and_ Atom
+#
+# (c) 2004, Davyd Madeley <[email protected]>
+#
+
+import sys, time
+import CacheHandler
+sys.path.insert(0, 'extra')
+import feedparser
+
+class Blog:
+       def __init__(self):
+               self.blogTitle  = None
+               self.blogURL    = None
+               self.feedURL    = None
+               self.imageURL   = None
+               self.imageLink  = None
+               self.items      = []
+               self.cache      = None
+
+class BlogItem:
+       def __init__(self):
+               self.itemTitle  = None
+               self.itemDate   = None
+               self.itemURL    = None
+               self.contents   = None
+
+class XMLParse:
+       def __init__(self, URL, blogObject):
+               self.feedURL    = URL
+               self.blogObject = blogObject
+               
+       def parse(self):
+               "Return a single Blog object"
+               item            = Blog()
+               if self.blogObject:
+                       sys.stdout.write('Downloading feed %s...' % self.feedURL)
+                       try:
+                               data    = feedparser.parse(self.feedURL, self.blogObject.cache.etag, self.blogObject.cache.date)
+                               sys.stdout.write('done.\n')
+                       except:
+                               sys.stdout.write('failed.\n')
+                               raise
+                               return None
+                       # check to see what we got returned
+                       if data['items'] == [] and data['channel'] == {}:
+                               sys.stdout.write('Feed %s is upto date.\n' % self.feedURL)
+                               return self.blogObject
+               else:
+                       sys.stdout.write('Downloading feed from %s (no cache)...' % self.feedURL)
+                       try:
+                               data    = feedparser.parse(self.feedURL)
+                               sys.stdout.write('done.\n')
+                       except:
+                               sys.stdout.write('failed.\n')
+                               return None
+               # create caching data
+               try:
+                       cache           = CacheHandler.CacheObject()
+                       cache.etag      = data['etag']
+                       cache.date      = data['modified']
+                       item.cache      = cache
+               except:
+                       item.cache      = None
+               # parse the return of data into a blog
+               if data['channel'].has_key('title'):
+                       item.blogTitle  = data['channel']['title']
+               else:
+                       item.blogTitle  = '(Unknown)'
+               if data['channel'].has_key('link'):
+                       item.blogURL    = data['channel']['link']
+               else:
+                       item.blogURL    = self.feedURL
+               for entry in data['items']:
+                       blogItem                = BlogItem()
+                       if entry.has_key('title'):
+                               blogItem.itemTitle      = entry['title']
+                       else:
+                               blogItem.itemTitle      = '(Untitled)'
+                       if entry.has_key('link'):
+                               blogItem.itemURL        = entry['link']
+                       else:
+                               blogItem.itemURL        = item.blogURL
+                       if entry.has_key('date_parsed'):
+                               blogItem.itemDate       = time.mktime(entry['date_parsed'])
+                       else:
+                               blogItem.itemDate       = 0
+                       if entry.has_key('description'):
+                               blogItem.contents       = entry['description']
+                       else:
+                               blogItem.contents       = '(entry could not be retrieved)'
+                       item.items.append(blogItem)
+               return item