Add a Google News RSS-based renderer. Minor improvements to all

[kiosk.git] / generic_news_rss_renderer.py

diff --git a/generic_news_rss_renderer.py b/generic_news_rss_renderer.py
index ab59888cdedeae2d23575e48215f365d2d326692..21f9afea3e630003f60f5e6f105d71b7548a19e6 100644 (file)
--- a/generic_news_rss_renderer.py
+++ b/generic_news_rss_renderer.py
@@ -1,3 +1,5 @@
+import datetime
+from dateutil.parser import parse
 import file_writer
 import grab_bag
 import renderer
@@ -10,7 +12,8 @@ import xml.etree.ElementTree as ET
 
 class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
     def __init__(self, name_to_timeout_dict, feed_site, feed_uris, page_title):
-        super(generic_news_rss_renderer, self).__init__(name_to_timeout_dict, False)
+        super(generic_news_rss_renderer, self).__init__(name_to_timeout_dict,
+                                                        False)
         self.debug = 1
         self.feed_site = feed_site
         self.feed_uris = feed_uris
@@ -50,14 +53,31 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
     def find_link(self, item):
         return item.findtext('link')
 
+    def munge_link(self, link):
+        return link
+
     def find_image(self, item):
         return item.findtext('image')
 
+    def munge_image(self, image):
+        return image
+
     def item_is_interesting_for_headlines(self, title, description, item):
-        pass
+        return True
+
+    def is_item_older_than_n_days(self, item, n):
+        pubdate = item.findtext('pubDate')
+        if pubdate is not None:
+            pubdate = parse(pubdate)
+            tzinfo = pubdate.tzinfo
+            now = datetime.datetime.now(tzinfo)
+            delta = (now - pubdate).total_seconds() / (60 * 60 * 24)
+            if (delta > n):
+                return True
+        return False
 
     def item_is_interesting_for_article(self, title, description, item):
-        pass
+        return True
 
     def periodic_render(self, key):
         if key == "Fetch News":
@@ -91,7 +111,7 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
             return False
         for msg in subset:
             blurb = msg
-            blurb += b'</TD>\n'
+            blurb += u'</TD>'
             details.add_item(blurb)
         g = file_writer.file_writer('%s_6_none.html' % (
             self.get_details_page_prefix()))
@@ -132,8 +152,12 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
                 description = item.findtext('description')
                 if description is not None:
                     description = self.munge_description(description)
+                image = self.find_image(item)
+                if image is not None:
+                    image = self.munge_image(image)
                 link = item.findtext('link')
-                image = item.findtext('image')
+                if link is not None:
+                    link = self.munge_link(link)
 
                 if (title is None or
                     not self.item_is_interesting_for_headlines(title,
@@ -151,38 +175,31 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
                 #print u"Title: %s\nDescription: %s\nLink: %s\nImage: %s\n" % (
                 #    title, description, link, image)
 
-                blurb = """<DIV style="padding:8px;
+                blurb = u"""<DIV style="padding:8px;
                                  font-size:34pt;
                                  -webkit-column-break-inside:avoid;">"""
                 if image is not None:
-                    blurb += '<IMG SRC="%s" ALIGN=LEFT HEIGHT=115 style="padding:8px;">\n' % image
-                blurb += '<P><B>%s</B>' % title
+                    blurb += u'<IMG SRC="%s" ALIGN=LEFT HEIGHT=115 ' % image
+                    blurb += u'style="padding:8px;">'
+
+                if link is None:
+                    blurb += u'<P><B>%s</B>' % title
+                else:
+                    blurb += u'<P><B><A HREF="%s">%s</A></B>' % (link, title)
 
                 if (description is not None and
-                    self.item_is_interesting_for_article(title, description, item)):
+                    self.item_is_interesting_for_article(title,
+                                                         description,
+                                                         item)):
                     longblurb = blurb
-                    longblurb += "<BR>"
+                    longblurb += u"<BR>"
                     longblurb += description
-                    longblurb += "</DIV>"
+                    longblurb += u"</DIV>"
                     longblurb = longblurb.replace("font-size:34pt",
                                                   "font-size:44pt")
-                    self.details.add(longblurb.encode('utf-8', errors='ignore'))
+                    self.details.add(longblurb)
 
-                blurb += "</DIV>"
-                self.news.add(blurb.encode('utf-8', errors='ignore'))
+                blurb += u"</DIV>"
+                self.news.add(blurb)
                 count += 1
         return count > 0
-
-# Test
-#x = generic_news_rss_renderer(
-#    {"Fetch News" : 1,
-#     "Shuffle News" : 1},
-#    "rss.cnn.com",
-#    [ "/rss/generic_news_topstories.rss",
-#      "/rss/money_latest.rss",
-#      "/rss/generic_news_tech.rss",
-#    ],
-#    "Test" )
-#if x.fetch_news() == 0:
-#    print "Error fetching news, no items fetched."
-#x.shuffle_news()