Profanity filter, please.

[kiosk.git] / seattletimes_rss_renderer.py

diff --git a/seattletimes_rss_renderer.py b/seattletimes_rss_renderer.py
index 906e00ef134a0e0959e0b8ee9b3fc2140d5809b1..653c74ac4c7e984f153cf189ca82b07623a1c4e0 100644 (file)
--- a/seattletimes_rss_renderer.py
+++ b/seattletimes_rss_renderer.py
@@ -1,90 +1,93 @@
-import datetime
+#!/usr/bin/env python3
+
+import logging
+from typing import Dict, List
+import xml
+
 import generic_news_rss_renderer as gnrss
-import sets
 
-class seattletimes_rss_renderer(gnrss.generic_news_rss_renderer):
-    interesting_categories = sets.ImmutableSet([
-        'Nation',
-        'World',
-        'Life',
-        'Technology'
-        'Local News',
-        'Food',
-        'Drink',
-        'Today File',
-        'Seahawks',
-        'Oddities',
-        'Packfic NW',
-        'Home',
-        'Garden',
-        'Travel',
-        'Outdoors',
-    ])
 
-    def __init__(self, name_to_timeout_dict, feed_site, feed_uris, page_title):
-        super(seattletimes_rss_renderer, self).__init__(
-            name_to_timeout_dict,
-            feed_site,
-            feed_uris,
-            page_title)
-        self.oldest = datetime.datetime.now() - datetime.timedelta(14)
-        self.debug_print("oldest story we'll keep: %s" % self.oldest)
+logger = logging.getLogger(__name__)
 
-    def debug_prefix(self):
-        return "seattletimes"
 
-    def get_headlines_page_prefix(self):
+class seattletimes_rss_renderer(gnrss.generic_news_rss_renderer):
+    interesting_categories = frozenset(
+        [
+            "Nation",
+            "World",
+            "Life",
+            "Technology",
+            "Local News",
+            "Food",
+            "Drink",
+            "Today File",
+            "Seahawks",
+            "Oddities",
+            "Packfic NW",
+            "Home",
+            "Garden",
+            "Travel",
+            "Outdoors",
+        ]
+    )
+
+    def __init__(
+        self,
+        name_to_timeout_dict: Dict[str, int],
+        feed_site: str,
+        feed_uris: List[str],
+        page_title: str,
+    ):
+        super().__init__(name_to_timeout_dict, feed_site, feed_uris, page_title)
+
+    def get_headlines_page_prefix(self) -> str:
         return "seattletimes-nonnews"
 
-    def get_details_page_prefix(self):
+    def get_details_page_prefix(self) -> str:
         return "seattletimes-details-nonnews"
 
-    def should_use_https(self):
+    def should_use_https(self) -> bool:
         return True
 
-    def item_is_interesting_for_headlines(self, title, description, item):
+    def item_is_interesting_for_headlines(
+        self, title: str, description: str, item: xml.etree.ElementTree.Element
+    ) -> bool:
         if item.tag != "item":
-            self.debug_print("Item.tag isn't item?!")
+            logger.debug(f'{title}: item.tag ({item}) isn\'t "item"?!')
+            return False
+        if self.is_item_older_than_n_days(item, 14):
+            logger.info(f"{title}: too old to be interesting.")
             return False
 
         details = {}
-        for detail in item.getchildren():
-            self.debug_print("detail %s => %s (%s)" % (detail.tag,
-                                                       detail.attrib,
-                                                       detail.text))
-            if detail.text != None:
+        for detail in list(item):
+            logger.debug(f"detail {detail.tag} => {detail.attrib} ({detail.text})")
+            if detail.text is not None:
                 details[detail.tag] = detail.text
         if "category" not in details:
-            self.debug_print("No category in details?!")
-            self.debug_print(details)
+            logger.debug(f"{title}: no category in details?")
             return False
-
         interesting = False
         for x in seattletimes_rss_renderer.interesting_categories:
             if x in details["category"]:
-                self.debug_print("%s looks like a good category." % x)
+                logger.debug(f"{x} looks like a good category.")
                 interesting = True
-        if not interesting:
-            return False
-
-        if 'enclosure' in details:
-            if 'pubDate' in details:
-                x = details['pubDate']
-                x = x.rsplit(' ', 1)[0]
-                # Fri, 13 Nov 2015 10:07:00
-                dt = datetime.datetime.strptime(x, '%a, %d %b %Y %H:%M:%S')
-                if dt < self.oldest:
-                    self.debug_print("%s is too old." % (
-                        details["pubDate"]))
-                    return False
-        return True
+                break
+        return interesting
 
-    def item_is_interesting_for_article(self, title, description, item):
+    def item_is_interesting_for_article(
+        self, title: str, description: str, item: xml.etree.ElementTree.Element
+    ) -> bool:
+        if self.is_item_older_than_n_days(item, 14):
+            logger.info(f"{title}: is too old to be interesting.")
+            return False
         return len(description) >= 65
 
-#x = seattletimes_rss_renderer({"Test", 123},
+
+# Test
+# x = seattletimes_rss_renderer({"Test", 123},
 #                              "www.seattletimes.com",
-#                              [ "/life/feed/" ],
+#                              [ "/outdoors/feed/", '/live/feed/' ],
 #                              "nonnews")
-#x.periodic_render("Fetch News")
-#x.periodic_render("Shuffle News")
+# x.periodic_render("Fetch News")
+# x.periodic_render("Shuffle News")