[2ff88f42] multiplikation vorbereitet

multiplikation vorbereitet

company-explorer/backend/scripts/sync_notion_to_ce_enhanced.py

@@ -7,7 +7,7 @@ import logging
 # /app/backend/scripts/sync.py -> /app
 sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), "../..")))
 
-from backend.database import SessionLocal, Industry, RoboticsCategory, init_db
+from backend.database import SessionLocal, Industry, RoboticsCategory, Persona, init_db
 from dotenv import load_dotenv
 
 # Try loading from .env in root if exists
@@ -76,6 +76,21 @@ def extract_number(prop):
     if not prop or "number" not in prop: return None
     return prop["number"]
 
+def extract_rich_text_to_list(prop):
+    if not prop or "rich_text" not in prop: return []
+    full_text = "".join([t.get("plain_text", "") for t in prop.get("rich_text", [])])
+    lines = full_text.split('\n')
+    cleaned_lines = []
+    for line in lines:
+        line = line.strip()
+        if not line: continue
+        if line.startswith("- "):
+            line = line[2:]
+        elif line.startswith("• "):
+            line = line[2:]
+        cleaned_lines.append(line)
+    return cleaned_lines
+
 def sync():
     logger.info("--- Starting Enhanced Sync ---")
     
@@ -83,6 +98,48 @@ def sync():
     init_db()
     session = SessionLocal()
 
+    # --- 4. Sync Personas (NEW) ---
+    # Sector & Persona Master ID
+    PERSONAS_DB_ID = "2e288f42-8544-8113-b878-ec99c8a02a6b"
+    VALID_ARCHETYPES = {
+        "Wirtschaftlicher Entscheider",
+        "Operativer Entscheider",
+        "Infrastruktur-Verantwortlicher",
+        "Innovations-Treiber"
+    }
+    
+    if PERSONAS_DB_ID:
+        logger.info(f"Syncing Personas from {PERSONAS_DB_ID}...")
+        pages = query_all(PERSONAS_DB_ID)
+        p_count = 0
+        
+        # We assume Personas are cumulative, so we don't delete all first (safer for IDs)
+        # But we could if we wanted a clean slate. Upsert is better.
+        
+        for page in pages:
+            props = page["properties"]
+            name = extract_title(props.get("Name"))
+            
+            if name not in VALID_ARCHETYPES:
+                continue
+                
+            import json
+            pains_list = extract_rich_text_to_list(props.get("Pains"))
+            gains_list = extract_rich_text_to_list(props.get("Gains"))
+            
+            persona = session.query(Persona).filter(Persona.name == name).first()
+            if not persona:
+                persona = Persona(name=name)
+                session.add(persona)
+            
+            persona.pains = json.dumps(pains_list, ensure_ascii=False)
+            persona.gains = json.dumps(gains_list, ensure_ascii=False)
+            
+            p_count += 1
+            
+        session.commit()
+        logger.info(f"✅ Synced {p_count} Personas.")
+
     # 2. Sync Categories (Products)
     cat_db_id = find_db_id("Product Categories") or find_db_id("Products")
     if cat_db_id: