seattletimes_rss_renderer.py

   1 #!/usr/bin/env python3
   2
   3 import logging
   4 from typing import Dict, List
   5 import xml
   6
   7 import generic_news_rss_renderer as gnrss
   8
   9
  10 logger = logging.getLogger(__file__)
  11
  12
  13 class seattletimes_rss_renderer(gnrss.generic_news_rss_renderer):
  14     interesting_categories = frozenset(
  15         [
  16             "Nation",
  17             "World",
  18             "Life",
  19             "Technology",
  20             "Local News",
  21             "Food",
  22             "Drink",
  23             "Today File",
  24             "Seahawks",
  25             "Oddities",
  26             "Packfic NW",
  27             "Home",
  28             "Garden",
  29             "Travel",
  30             "Outdoors",
  31         ]
  32     )
  33
  34     def __init__(
  35         self,
  36         name_to_timeout_dict: Dict[str, int],
  37         feed_site: str,
  38         feed_uris: List[str],
  39         page_title: str,
  40     ):
  41         super().__init__(
  42             name_to_timeout_dict, feed_site, feed_uris, page_title
  43         )
  44
  45     def get_headlines_page_prefix(self) -> str:
  46         return "seattletimes-nonnews"
  47
  48     def get_details_page_prefix(self) -> str:
  49         return "seattletimes-details-nonnews"
  50
  51     def should_use_https(self) -> bool:
  52         return True
  53
  54     def item_is_interesting_for_headlines(
  55         self, title: str, description: str, item: xml.etree.ElementTree.Element
  56     ) -> bool:
  57         if item.tag != "item":
  58             logger.debug(f'{title}: item.tag ({item}) isn\'t "item"?!')
  59             return False
  60         if self.is_item_older_than_n_days(item, 14):
  61             logger.info(f'{title}: too old to be interesting.')
  62             return False
  63
  64         details = {}
  65         for detail in list(item):
  66             logger.debug(
  67                 f'detail {detail.tag} => {detail.attrib} ({detail.text})'
  68             )
  69             if detail.text is not None:
  70                 details[detail.tag] = detail.text
  71         if 'category' not in details:
  72             logger.debug(f'{title}: no category in details?')
  73             return False
  74         interesting = False
  75         for x in seattletimes_rss_renderer.interesting_categories:
  76             if x in details["category"]:
  77                 logger.debug(
  78                     f'{x} looks like a good category.'
  79                 )
  80                 interesting = True
  81                 break
  82         return interesting
  83
  84     def item_is_interesting_for_article(
  85         self, title: str, description: str, item: xml.etree.ElementTree.Element
  86     ) -> bool:
  87         if self.is_item_older_than_n_days(item, 14):
  88             logger.info(
  89                 f'{title}: is too old to be interesting.'
  90             )
  91             return False
  92         return len(description) >= 65
  93
  94
  95 # Test
  96 #x = seattletimes_rss_renderer({"Test", 123},
  97 #                              "www.seattletimes.com",
  98 #                              [ "/outdoors/feed/", '/live/feed/' ],
  99 #                              "nonnews")
 100 #x.periodic_render("Fetch News")
 101 #x.periodic_render("Shuffle News")