fix: Kingdom Rush #15540 online mit Bild + Workflow gehärtet

- verify_article.sh: +4 CSS-Compliance-Checks (vorbestell-box, bsn-facts, veraltete Boxen, Inline-Styles) - fix_article_format.py: Auto-Fixer für Inline-Style-Stripping - journalistic-article-writing SKILL.md: QC-Pipeline + Template mit vorbestell-box - artikel_kingdom_rush_deal_20260710.html: PATCH 200 auf Joomla + Bild
2026-07-10 23:41:04 +02:00
parent c8bddebdab
commit ef55212a7e
34 changed files with 5089 additions and 1103 deletions
@@ -1,67 +1,67 @@
 #!/usr/bin/env python3
-"""Fix article HTML format for Article-Server compatibility.
-Ensures: <p class="meta"> date, <meta name="description">, <meta name="keywords">
-Run: python3 fix_article_format.py /home/hermes/workspace/artikel.html
 """
-import sys, re, os
-from datetime import datetime
+fix_article_format.py — Stripped Inline-Styles von BSN-Standard-Klassen.
+Entfernt veraltete Box-Formate aus Artikel-HTML.
+Usage: python3 fix_article_format.py <artikel.html>
+"""
+import re, sys

-def fix_article(filepath):
-    with open(filepath, 'r') as f:
-        content = f.read()
-    
-    changed = False
-    
-    # 1. Fix <div class="meta"> → <p class="meta">
-    if re.search(r'<div class="meta">([^<]+)</div>', content):
-        content = re.sub(r'<div class="meta">([^<]+)</div>', r'<p class="meta">\1</p>', content)
-        changed = True
-        print(f"  ✓ <div class=\"meta\"> → <p class=\"meta\">")
-    
-    # 2. Add <p class="meta"> if missing (using dateline or default)
-    if '<p class="meta">' not in content:
-        date_str = datetime.now().strftime("%d. %B %Y")
-        # Try to extract date from dateline
-        dl = re.search(r'<p class="dateline">([^<]+)</p>', content)
-        if dl:
-            date_str = dl.group(1).split('·')[0].strip()
-        # Insert after <body> or after <h1>
-        meta_line = f'<p class="meta">{date_str} · Lesezeit: ca. 5 Minuten</p>\n'
-        if '<h1>' in content:
-            content = re.sub(r'(</h1>\n)', r'\1' + meta_line, content, count=1)
-        else:
-            content = re.sub(r'(<body>\n)', r'\1' + meta_line, content, count=1)
-        changed = True
-        print(f"  ✓ <p class=\"meta\"> hinzugefügt: {date_str}")
-    
-    # 3. Add <meta name="description"> if missing
-    if '<meta name="description" content="' not in content:
-        title_m = re.search(r'<title>(.*?)</title>', content)
-        if title_m:
-            title = title_m.group(1).strip()
-            desc = title[:160] + ('…' if len(title) > 160 else '')
-            desc_tag = f'<meta name="description" content="{desc}">\n'
-            content = re.sub(r'(<title>.*?</title>\n)', r'\1' + desc_tag, content, count=1)
-            changed = True
-            print(f"  ✓ <meta name=\"description\"> hinzugefügt")
-    
-    # 4. Add <meta name="keywords"> if missing  
-    if '<meta name="keywords" content="' not in content:
-        kw_tag = '<meta name="keywords" content="Brettspiel, News, brettspiel-news.de">\n'
-        content = re.sub(r'(<meta name="description".*?>\n)', r'\1' + kw_tag, content, count=1)
-        changed = True
-        print(f"  ✓ <meta name=\"keywords\"> hinzugefügt")
-    
-    if changed:
-        with open(filepath, 'w') as f:
-            f.write(content)
-        return True
-    return False
+if len(sys.argv) < 2:
+    print(f"Usage: {sys.argv[0]} <artikel.html>")
+    sys.exit(1)

-if __name__ == '__main__':
-    for f in sys.argv[1:]:
-        print(f"{os.path.basename(f)}:")
-        if fix_article(f):
-            print(f"  ✅ Gefixt")
-        else:
-            print(f"  ✓ Bereits korrekt")
+filepath = sys.argv[1]
+with open(filepath) as f:
+    html = f.read()
+
+original = html
+fixes = 0
+
+# 1. Strip style="" from vorbestell-box elements
+html, n = re.subn(
+    r'(<div[^>]*class="[^"]*vorbestell-box[^"]*")[^>]*>',
+    r'\1>',
+    html
+)
+fixes += n
+
+# 2. Strip style="" from bsn-facts elements
+html, n = re.subn(
+    r'(<div[^>]*class="[^"]*bsn-facts[^"]*")[^>]*>',
+    r'\1>',
+    html
+)
+fixes += n
+
+# 3. Remove <div class="affiliate-box">...</div> blocks entirely
+html, n = re.subn(
+    r'<div class="affiliate-box"[^>]*>.*?</div>\s*',
+    '',
+    html,
+    flags=re.DOTALL
+)
+fixes += n
+
+# 4. Strip style="background:..." / style="border:..." from <div>s
+#    (but ONLY if they also have a class= — pure style-only divs stay)
+html, n = re.subn(
+    r'(<div[^>]*class="[^"]*"[^>]*)\s+style="[^"]*(?:background|border)[^"]*"',
+    r'\1',
+    html
+)
+fixes += n
+
+# 5. Strip style="" from <p> inside old-format deal boxes
+html, n = re.subn(
+    r'<p style="[^"]*(?:margin|font-weight|font-size|color:#888)[^"]*"[^>]*>',
+    r'<p>',
+    html
+)
+fixes += n
+
+if html != original:
+    with open(filepath, 'w') as f:
+        f.write(html)
+    print(f"✅ {fixes} Fixes angewandt auf {filepath}")
+else:
+    print(f"ℹ️  Keine Fixes nötig für {filepath}")