X-Git-Url: https://wannabe.guru.org/gitweb/?a=blobdiff_plain;f=generic_news_rss_renderer.py;h=798c06c90962d001a87ed93f1a9a2114a3ccbc88;hb=72938579d41b01fe93f18b97605bcb7dfd147d4b;hp=b87ab05caa49ed646ecff87266cfca482576aab5;hpb=4b1f3d8a8b278ca6d62f461ea80c8ea21080c301;p=kiosk.git

diff --git a/generic_news_rss_renderer.py b/generic_news_rss_renderer.py
index b87ab05..798c06c 100644
--- a/generic_news_rss_renderer.py
+++ b/generic_news_rss_renderer.py
@@ -1,7 +1,9 @@
+import datetime
+from dateutil.parser import parse
 import file_writer
 import grab_bag
 import renderer
-import httplib
+import http.client
 import page_builder
 import profanity_filter
 import random
@@ -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
@@ -28,6 +31,12 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
     def get_details_page_prefix(self):
         pass
 
+    def get_headlines_page_priority(self):
+        return "4"
+
+    def get_details_page_priority(self):
+        return "6"
+
     def should_use_https(self):
         pass
 
@@ -50,14 +59,37 @@ 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 find_pubdate(self, item):
+        return item.findtext('pubDate')
+
+    def munge_pubdate(self, pubdate):
+        return pubdate
+
     def item_is_interesting_for_headlines(self, title, description, item):
-        pass
+        return True
+
+    def is_item_older_than_n_days(self, item, n):
+        pubdate = self.find_pubdate(item)
+        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":
@@ -77,13 +109,50 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
             return False
         for msg in subset:
             headlines.add_item(msg)
-        f = file_writer.file_writer('%s_4_none.html' % (
-            self.get_headlines_page_prefix()))
+        headlines.set_custom_html("""
+<STYLE>
+a:link {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+a:visited {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+a:active {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+</STYLE>""")
+        f = file_writer.file_writer('%s_%s_none.html' % (
+            self.get_headlines_page_prefix(),
+            self.get_headlines_page_priority()))
         headlines.render_html(f)
         f.close()
 
         details = page_builder.page_builder()
         details.set_layout(page_builder.page_builder.LAYOUT_ONE_ITEM)
+        details.set_custom_html("""
+<STYLE>
+a:link {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+a:visited {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+a:active {
+  color: black;
+  text-decoration: none;
+  font-weight: bold;
+}
+</STYLE>""")
         details.set_title("%s" % self.page_title)
         subset = self.details.subset(1)
         if subset is None:
@@ -91,10 +160,11 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
             return False
         for msg in subset:
             blurb = msg
-            blurb += "</TD>\n"
+            blurb += u'</TD>'
             details.add_item(blurb)
-        g = file_writer.file_writer('%s_6_none.html' % (
-            self.get_details_page_prefix()))
+        g = file_writer.file_writer('%s_%s_none.html' % (
+            self.get_details_page_prefix(),
+            self.get_details_page_priority()))
         details.render_html(g)
         g.close()
         return True
@@ -107,10 +177,10 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
         for uri in self.feed_uris:
             if self.should_use_https():
                 self.debug_print("Fetching: https://%s%s" % (self.feed_site, uri))
-                self.conn = httplib.HTTPSConnection(self.feed_site)
+                self.conn = http.client.HTTPSConnection(self.feed_site)
             else:
                 self.debug_print("Fetching: http://%s%s" % (self.feed_site, uri))
-                self.conn = httplib.HTTPConnection(self.feed_site)
+                self.conn = http.client.HTTPConnection(self.feed_site)
             self.conn.request(
                 "GET",
                 uri,
@@ -118,8 +188,8 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
                 {"Accept-Charset": "utf-8"})
             response = self.conn.getresponse()
             if response.status != 200:
-                print("%s: RSS fetch_news error, response: %d" % (self.page_title,
-                                                                  response.status))
+                print(("%s: RSS fetch_news error, response: %d" % (self.page_title,
+                                                                  response.status)))
                 self.debug_print(response.read())
                 return False
 
@@ -132,8 +202,12 @@ 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,
@@ -148,41 +222,39 @@ class generic_news_rss_renderer(renderer.debuggable_abstaining_renderer):
                     self.debug_print('Found bad words in item "%s"' % title)
                     continue
 
-                #print u"Title: %s\nDescription: %s\nLink: %s\nImage: %s\n" % (
-                #    title, description, link, image)
-
                 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)
+
+                pubdate = self.find_pubdate(item)
+                if pubdate is not None:
+                    pubdate = self.munge_pubdate(pubdate)
+                    ts = parse(pubdate)
+                    blurb += u" <FONT COLOR=#cccccc>%s</FONT>" % (
+                        ts.strftime("%b %d"))
 
                 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()