seattletimes_rss_renderer.py

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