Create filter_news.py
Browse files- filter_news.py +47 -0
filter_news.py
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# filter_news.py (erweitert)
|
| 2 |
+
import re
|
| 3 |
+
from config import CATEGORY_FILTERS
|
| 4 |
+
import logging
|
| 5 |
+
|
| 6 |
+
logger = logging.getLogger(__name__)
|
| 7 |
+
|
| 8 |
+
class NewsFilter:
|
| 9 |
+
def __init__(self):
|
| 10 |
+
self.keyword_filters = CATEGORY_FILTERS
|
| 11 |
+
self.blacklist = self.load_wordlist("blacklist.txt")
|
| 12 |
+
self.whitelist = self.load_wordlist("whitelist.txt")
|
| 13 |
+
|
| 14 |
+
def load_wordlist(self, filename):
|
| 15 |
+
try:
|
| 16 |
+
with open(f"config/{filename}") as f:
|
| 17 |
+
return [line.strip().lower() for line in f if line.strip()]
|
| 18 |
+
except FileNotFoundError:
|
| 19 |
+
return []
|
| 20 |
+
|
| 21 |
+
def filter_articles(self, articles):
|
| 22 |
+
filtered = []
|
| 23 |
+
for article in articles:
|
| 24 |
+
if self.is_blacklisted(article) and not self.is_whitelisted(article):
|
| 25 |
+
continue
|
| 26 |
+
filtered.append(article)
|
| 27 |
+
return self.prioritize_articles(filtered)
|
| 28 |
+
|
| 29 |
+
def is_blacklisted(self, article):
|
| 30 |
+
text = f"{article['title']} {article['description']}".lower()
|
| 31 |
+
return any(re.search(rf'\b{word}\b', text) for word in self.blacklist)
|
| 32 |
+
|
| 33 |
+
def is_whitelisted(self, article):
|
| 34 |
+
text = f"{article['title']} {article['description']}".lower()
|
| 35 |
+
return any(re.search(rf'\b{word}\b', text) for word in self.whitelist)
|
| 36 |
+
|
| 37 |
+
def prioritize_articles(self, articles):
|
| 38 |
+
# Priorisierung nach Kategorie und Keywords
|
| 39 |
+
def sort_key(article):
|
| 40 |
+
score = 0
|
| 41 |
+
category_keywords = self.keyword_filters.get(article["category"], [])
|
| 42 |
+
text = f"{article['title']} {article['description']}".lower()
|
| 43 |
+
for keyword in category_keywords:
|
| 44 |
+
if re.search(rf'\b{keyword.lower()}\b', text):
|
| 45 |
+
score += 1
|
| 46 |
+
return -score # Höhere Priorität zuerst
|
| 47 |
+
return sorted(articles, key=sort_key)
|