X-Git-Url: https://wannabe.guru.org/gitweb/?a=blobdiff_plain;f=bellevue_reporter_rss_renderer.py;h=4e1ff6261f31650c609ec153f3fcf3f878573809;hb=477c84bcdefc573278ec8ed8974d5be8baac562d;hp=2776ca0bbf52f95dd755a5e08b4e395c03e65830;hpb=c06bfef53f70551e7920bc4facce27f47b89e2ba;p=kiosk.git

diff --git a/bellevue_reporter_rss_renderer.py b/bellevue_reporter_rss_renderer.py
index 2776ca0..4e1ff62 100644
--- a/bellevue_reporter_rss_renderer.py
+++ b/bellevue_reporter_rss_renderer.py
@@ -1,12 +1,17 @@
 #!/usr/bin/env python3
 
+import logging
 import re
 from typing import List, Dict
 import xml
+import xml.etree.ElementTree as ET
 
 import generic_news_rss_renderer as gnrss
 
 
+logger = logging.getLogger(__file__)
+
+
 class bellevue_reporter_rss_renderer(gnrss.generic_news_rss_renderer):
     """Read the Bellevue Reporter's RSS feed."""
 
@@ -17,13 +22,9 @@ class bellevue_reporter_rss_renderer(gnrss.generic_news_rss_renderer):
         feed_uris: List[str],
         page_title: str,
     ):
-        super(bellevue_reporter_rss_renderer, self).__init__(
+        super().__init__(
             name_to_timeout_dict, feed_site, feed_uris, page_title
         )
-        self.debug = True
-
-    def debug_prefix(self) -> str:
-        return "bellevue_reporter(%s)" % (self.page_title)
 
     def get_headlines_page_prefix(self) -> str:
         return "bellevue-reporter"
@@ -34,42 +35,74 @@ class bellevue_reporter_rss_renderer(gnrss.generic_news_rss_renderer):
     def should_use_https(self) -> bool:
         return True
 
-    def munge_description(self, description: str) -> str:
+    def munge_description(self, description: str, item: ET.Element) -> str:
         description = re.sub("<[^>]+>", "", description)
         description = re.sub(
             "Bellevue\s+Reporter\s+Bellevue\s+Reporter", "", description
         )
         description = re.sub("\s*\-\s*Your local homepage\.\s*", "", description)
+        description = re.sub("[Ww]ire [Ss]ervice", "", description)
         return description
 
     @staticmethod
     def looks_like_football(title: str, description: str) -> bool:
         return (
             title.find("NFL") != -1
-            or re.search("[Ll]ive [Ss]tream", title) != None
-            or re.search("[Ll]ive[Ss]tream", title) != None
-            or re.search("[Ll]ive [Ss]tream", description) != None
+            or re.search("[Ll]ive [Ss]tream", title) is not None
+            or re.search("[Ll]ive[Ss]tream", title) is not None
+            or re.search("[Ll]ive [Ss]tream", description) is not None
+        )
+
+    @staticmethod
+    def looks_like_review(title: str, description: str) -> bool:
+        return "review" in title or "Review" in title
+
+    @staticmethod
+    def looks_like_spam(title: str, description: str) -> bool:
+        return (
+            'marketplace' in description
+            or 'national-marketplace' in description
+            or re.search('[Ww]eed', title) is not None
+            or re.search('[Cc]annabis', title) is not None
+            or re.search('[Cc]annabis', description) is not None
+            or 'THC' in title
+            or re.search('[Ll]ose [Ww]eight', title) is not None
+            or re.search('[Ll]ose [Ww]eight', description) is not None
         )
 
     def item_is_interesting_for_headlines(
         self, title: str, description: str, item: xml.etree.ElementTree.Element
     ) -> bool:
+        unfiltered_description = item.findtext("description")
         if self.is_item_older_than_n_days(item, 10):
-            self.debug_print("%s: is too old!" % title)
+            logger.info(f'{title}: is too old!')
+            return False
+        if bellevue_reporter_rss_renderer.looks_like_spam(title, unfiltered_description):
+            logger.debug(f'{title}: looks like spam')
             return False
         if bellevue_reporter_rss_renderer.looks_like_football(title, description):
-            self.debug_print("%s: looks like it's about football." % title)
+            logger.debug(f'{title}: looks like it\'s about football.')
+            return False
+        if bellevue_reporter_rss_renderer.looks_like_review(title, description):
+            logger.debug(f'{title}: looks like a review.')
             return False
         return True
 
     def item_is_interesting_for_article(
         self, title: str, description: str, item: xml.etree.ElementTree.Element
     ) -> bool:
+        unfiltered_description = item.findtext("description")
         if self.is_item_older_than_n_days(item, 10):
-            self.debug_print("%s: is too old!" % title)
+            logger.debug(f'{title}: is too old!')
+            return False
+        if bellevue_reporter_rss_renderer.looks_like_spam(title, unfiltered_description):
+            logger.debug(f'{title}: looks like spam')
             return False
         if bellevue_reporter_rss_renderer.looks_like_football(title, description):
-            self.debug_print("%s: looks like it's about football." % title)
+            logger.debug(f'{title}: looks like it\'s about football.')
+            return False
+        if bellevue_reporter_rss_renderer.looks_like_review(title, description):
+            logger.debug(f'{title}: looks like a review.')
             return False
         return True