fix: QC-Ratings aus 7 älteren Artikeln entfernt (CHECK 10)

scrape_dicebreaker.py

@@ -0,0 +1,37 @@
+import sys, re, gzip
+
+# Read binary data
+data = sys.stdin.buffer.read()
+
+# Check if gzip compressed
+if data[:2] == b'\x1f\x8b':
+    data = gzip.decompress(data)
+
+text = data.decode('utf-8', errors='replace')
+
+# Find article titles and links
+# Look for link patterns in the HTML
+links = re.findall(r'<a[^>]*href="([^"]+)"[^>]*>([^<]{10,200})</a>', text)
+for url, title in links[:30]:
+    clean = title.strip()
+    if any(kw in url.lower() for kw in ['news', 'game', 'feature', 'review', 'board']):
+        if not url.startswith('http'):
+            url = 'https://www.dicebreaker.com' + url
+        print(f'TITLE: {clean}')
+        print(f'URL: {url}')
+        print()
+
+# Also try h2/h3 with links
+titles = re.findall(r'<h[23][^>]*>(.*?)</h[23]>', text, re.DOTALL)
+for t in titles[:20]:
+    clean = re.sub(r'<[^>]+>', '', t).strip()
+    link = re.search(r'href="([^"]+)"', t)
+    if clean and len(clean) > 10 and any(kw in clean.lower() for kw in ['game','board','play','card','dice','tabletop','news','release','announc','new','review','launch']):
+        url = ''
+        if link:
+            u = link.group(1)
+            url = 'https://www.dicebreaker.com' + u if u.startswith('/') else u
+        print(f'H-TITLE: {clean}')
+        if url:
+            print(f'URL: {url}')
+        print()