[31388f42] Implement end-to-end email ingest for Tradingtwins leads via MS Graph API

lead-engine/ingest.py

@@ -28,6 +28,40 @@ def parse_tradingtwins_email(body):
     data['raw_body'] = body
     return data
 
+def parse_roboplanet_form(body):
+    """
+    Parses the Roboplanet website contact form (HTML format).
+    Example: <b>Vorname:</b> BÄKO <br><b>Nachname:</b> eG <br><b>Email:</b> Alexander.Grau@baeko-hr.de ...
+    """
+    data = {}
+    
+    # Helper to strip HTML tags if needed, but we'll use regex on the content
+    patterns = {
+        'contact_first': r'Vorname:</b>\s*(.*?)\s*<br>',
+        'contact_last': r'Nachname:</b>\s*(.*?)\s*<br>',
+        'email': r'Email:</b>\s*(.*?)\s*<br>',
+        'phone': r'Telefon:</b>\s*(.*?)\s*<br>',
+        'company': r'Firma:</b>\s*(.*?)\s*<br>',
+        'zip': r'PLZ:</b>\s*(.*?)\s*<br>',
+        'message': r'Nachricht:</b>\s*(.*?)\s*(?:<br>|--|$)'
+    }
+    
+    for key, pattern in patterns.items():
+        # Use re.DOTALL for message if it spans lines, but usually it's one block
+        match = re.search(pattern, body, re.IGNORECASE | re.DOTALL)
+        if match:
+            # Clean HTML tags from the captured value if any
+            val = re.sub(r'<.*?>', '', match.group(1)).strip()
+            data[key] = val
+            
+    # Combine names
+    if 'contact_first' in data and 'contact_last' in data:
+        data['contact'] = f"{data['contact_first']} {data['contact_last']}"
+    
+    # For Roboplanet forms, we use the timestamp as ID or a hash if missing
+    data['raw_body'] = body
+    return data
+
 def ingest_mock_leads():
     # Mock data from the session context
     leads = [