seattletimes_rss_renderer.py

   1 import datetime
   2 import generic_news_rss_renderer as gnrss
   3 import sets
   4
   5 class seattletimes_rss_renderer(gnrss.generic_news_rss_renderer):
   6     interesting_categories = sets.ImmutableSet([
   7         'Nation',
   8         'World',
   9         'Life',
  10         'Technology'
  11         'Local News',
  12         'Food',
  13         'Drink',
  14         'Today File',
  15         'Seahawks',
  16         'Oddities',
  17         'Packfic NW',
  18         'Home',
  19         'Garden',
  20         'Travel',
  21         'Outdoors',
  22     ])
  23
  24     def __init__(self, name_to_timeout_dict, feed_site, feed_uris, page_title):
  25         super(seattletimes_rss_renderer, self).__init__(
  26             name_to_timeout_dict,
  27             feed_site,
  28             feed_uris,
  29             page_title)
  30         self.oldest = datetime.datetime.now() - datetime.timedelta(14)
  31         self.debug_print("oldest story we'll keep: %s" % self.oldest)
  32
  33     def debug_prefix(self):
  34         return "seattletimes"
  35
  36     def get_headlines_page_prefix(self):
  37         return "seattletimes-nonnews"
  38
  39     def get_details_page_prefix(self):
  40         return "seattletimes-details-nonnews"
  41
  42     def should_use_https(self):
  43         return True
  44
  45     def item_is_interesting_for_headlines(self, title, description, item):
  46         if item.tag != "item":
  47             self.debug_print("Item.tag isn't item?!")
  48             return False
  49
  50         details = {}
  51         for detail in item.getchildren():
  52             self.debug_print("detail %s => %s (%s)" % (detail.tag,
  53                                                        detail.attrib,
  54                                                        detail.text))
  55             if detail.text != None:
  56                 details[detail.tag] = detail.text
  57         if "category" not in details:
  58             self.debug_print("No category in details?!")
  59             self.debug_print(details)
  60             return False
  61
  62         interesting = False
  63         for x in seattletimes_rss_renderer.interesting_categories:
  64             if x in details["category"]:
  65                 self.debug_print("%s looks like a good category." % x)
  66                 interesting = True
  67         if not interesting:
  68             return False
  69
  70         if 'enclosure' in details:
  71             if 'pubDate' in details:
  72                 x = details['pubDate']
  73                 x = x.rsplit(' ', 1)[0]
  74                 # Fri, 13 Nov 2015 10:07:00
  75                 dt = datetime.datetime.strptime(x, '%a, %d %b %Y %H:%M:%S')
  76                 if dt < self.oldest:
  77                     self.debug_print("%s is too old." % (
  78                         details["pubDate"]))
  79                     return False
  80         return True
  81
  82     def item_is_interesting_for_article(self, title, description, item):
  83         return len(description) >= 65
  84
  85 #x = seattletimes_rss_renderer({"Test", 123},
  86 #                              "www.seattletimes.com",
  87 #                              [ "/life/feed/" ],
  88 #                              "nonnews")
  89 #x.periodic_render("Fetch News")
  90 #x.periodic_render("Shuffle News")