yet again

[planet-ucc.git] / XMLParse.py

diff --git a/XMLParse.py b/XMLParse.py
index a50925f..8ce8e40 100644 (file)
--- a/XMLParse.py
+++ b/XMLParse.py
@@ -46,8 +46,11 @@ class XMLParse:
                elif self.dom.documentElement.tagName == 'rdf:RDF':
                        # this is an RDF document
                        self.news       = RDFParse(self.dom)
+               elif self.dom.documentElement.tagName == 'feed':
+                       # this seems to be an Atom feed
+                       self.news       = AtomParse(self.dom)
                else:
-                       sys.stderr.write('DEBUG: XMLParse: Unknown XML document %s\n' % dom.documentElement.tagname)
+                       sys.stderr.write('DEBUG: XMLParse: Unknown XML document \'%s\'\n' % self.dom.documentElement.tagName)
                        
        def parse(self):
                "Return a list of Blog objects from the XML file we parsed"
@@ -58,19 +61,58 @@ class XMLParse:
                        self.bloglist   = self.news.parse()
                        return self.bloglist
 
-class RDFParse:
+class Parse:
+       "Generic class for parsing XML feeds"
        def __init__(self, dom):
                self.dom        = dom
                self.root       = dom.documentElement
-       
+
        def __retrieve_value__(self, fromNode):
+               "Retrieve a value from between two nodes"
                for node in fromNode.childNodes:
                        if node.nodeType == 3:
                                return node.nodeValue
                        else:
-                               sys.stderr.write('DEBUG: RDFParse: Asked to retrieve value from wrong part of tree\n')
+                               sys.stderr.write('DEBUG: Parse: Asked to retrieve value from wrong part of tree\n')
                                return None
-       
+
+class AtomParse(Parse):        
+       def parse(self):
+               channel = Blog()
+               for node in self.root.childNodes:
+                       if node.nodeType == 1 and node.tagName == 'title':
+                               channel.blogTitle       = self.__retrieve_value__(node)
+                       elif node.nodeType == 1 and node.tagName == 'link' and node.attributes.has_key('rel') and node.attributes['rel'].value == "alternate":
+                               if node.attributes.has_key('href'):
+                                       channel.blogURL         = node.attributes['href'].value
+                               else:
+                                       sys.stderr.write('DEBUG: AtomParse: Could not find href for link, ignoring\n')
+                       elif node.nodeType == 1 and node.tagName == 'entry':
+                               # create an item and add it to the list
+                               item    = BlogItem()
+                               channel.items.append(item)
+                               # handlers for tags
+                               for node2 in node.childNodes:
+                                       if node2.nodeType == 1 and node2.tagName == 'created':
+                                               date    = self.__retrieve_value__(node2)
+                                               try:
+                                                       item.itemDate   = time.mktime(time.strptime(date, '%Y-%m-%dT%H:%M:%SZ')) + 28800
+                                               except:
+                                                       sys.stderr.write("DEBUG: AtomParse: time string %s is unparseable\n" % date)
+                                       elif node2.nodeType == 1 and node2.tagName == 'link' and node2.attributes.has_key('rel') and node2.attributes['rel'].value == 'alternate':
+                                               if node2.attributes.has_key('href'):
+                                                       item.itemURL    = node2.attributes['href'].value
+                                               else:
+                                                       sys.stderr.write('DEBUG: AtomParse: Could not find href for link, ignoring\n')
+                                       elif node2.nodeType == 1 and node2.tagName == 'title':
+                                               item.itemTitle  = self.__retrieve_value__(node2)
+                                       elif node2.nodeType == 1 and node2.tagName == 'summary':
+                                               for node3 in node2.childNodes:
+                                                       if node3.nodeType == 1 and node3.tagName == 'div':
+                                                               item.contents   = self.__retrieve_value__(node3)
+               return [channel]
+
+class RDFParse(Parse):
        def parse(self):
                channel = Blog()
                for node in self.root.childNodes:
@@ -90,7 +132,7 @@ class RDFParse:
                                        elif node2.nodeType == 1 and node2.tagName == 'dc:date':
                                                date                    = self.__retrieve_value__(node2)
                                                try:
-                                                       item.itemDate   = time.mktime(time.strptime(date, '%Y-%m-%dT%H:%M:%S+07:00')) + 3600
+                                                       item.itemDate   = time.mktime(time.strptime(date, '%Y-%m-%dT%H:%M:%S+08:00'))
                                                except:
                                                        sys.stderr.write("DEBUG: RDFParse: time string %s unparseable\n" % date)
                                        elif node2.nodeType == 1 and node2.tagName == 'description':
@@ -98,21 +140,7 @@ class RDFParse:
                                channel.items.append(item)
                return [channel]
 
-class RSS2Parse:
-       def __init__(self, dom):
-               self.dom        = dom
-               self.root       = dom.documentElement
-
-       def __retrieve_value__(self, fromNode):
-               "Returns the value from between two nodes, ie <node>text</node>"
-               for node in fromNode.childNodes:
-                       if node.nodeType == 3:
-                               # this is the information contained within our node
-                               return node.nodeValue
-                       else:
-                               sys.stderr.write('DEBUG: RSS2Parse: Asked to retrieve value from wrong part of tree\n')
-                               return None
-       
+class RSS2Parse(Parse):
        def __parse_item__(self, fromNode):
                "Returns a BlogItem collected from fromNode"
                item    = BlogItem()