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(__name__)
  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__(name_to_timeout_dict, feed_site, feed_uris, page_title)
  42
  43     def get_headlines_page_prefix(self) -> str:
  44         return "seattletimes-nonnews"
  45
  46     def get_details_page_prefix(self) -> str:
  47         return "seattletimes-details-nonnews"
  48
  49     def should_use_https(self) -> bool:
  50         return True
  51
  52     def item_is_interesting_for_headlines(
  53         self, title: str, description: str, item: xml.etree.ElementTree.Element
  54     ) -> bool:
  55         if item.tag != "item":
  56             logger.debug(f'{title}: item.tag ({item}) isn\'t "item"?!')
  57             return False
  58         if self.is_item_older_than_n_days(item, 14):
  59             logger.info(f"{title}: too old to be interesting.")
  60             return False
  61
  62         details = {}
  63         for detail in list(item):
  64             logger.debug(f"detail {detail.tag} => {detail.attrib} ({detail.text})")
  65             if detail.text is not None:
  66                 details[detail.tag] = detail.text
  67         if "category" not in details:
  68             logger.debug(f"{title}: no category in details?")
  69             return False
  70         interesting = False
  71         for x in seattletimes_rss_renderer.interesting_categories:
  72             if x in details["category"]:
  73                 logger.debug(f"{x} looks like a good category.")
  74                 interesting = True
  75                 break
  76         return interesting
  77
  78     def item_is_interesting_for_article(
  79         self, title: str, description: str, item: xml.etree.ElementTree.Element
  80     ) -> bool:
  81         if self.is_item_older_than_n_days(item, 14):
  82             logger.info(f"{title}: is too old to be interesting.")
  83             return False
  84         return len(description) >= 65
  85
  86
  87 # Test
  88 # x = seattletimes_rss_renderer({"Test", 123},
  89 #                              "www.seattletimes.com",
  90 #                              [ "/outdoors/feed/", '/live/feed/' ],
  91 #                              "nonnews")
  92 # x.periodic_render("Fetch News")
  93 # x.periodic_render("Shuffle News")