Update ai_mapping.py

ai_mapping.py

@@ -31,10 +31,11 @@ def extract_key_values_with_layoutlm(page_data: list, pdf_path: str) -> Dict[str
         key_values = {}
         # Enhanced regex patterns with flexibility
         dates = re.findall(r'(Agreement\s+(?:Start|End)\s+Date(?:s)?)\s*[:\s]*(\d{1,2}[/-]\d{1,2}[/-]\d{4})', text_data, re.IGNORECASE)
-        # Capture date near "Effective Date" or "Signed"
-        date_context = re.findall(r'(?:Effective\s+Date|Signed)\s*[:\s]*(\d{1,2}[/-]\d{1,2}[/-]\d{4})', text_data, re.IGNORECASE)
+        # Capture date near "Effective Date" or "executed as of"
+        date_context = re.findall(r'(?:Effective\s+Date|executed\s+as\s+of)\s*[:\s]*(\d{1,2}[/-]\d{1,2}[/-]\d{4})', text_data, re.IGNORECASE)
         amounts = re.findall(r'\$\d{1,3}(?:,\d{3})*(?:\.\d{2})?', text_data)
-        names = re.findall(r'(Agreement\s+Name)\s*[:\s]*([A-Za-z0-9\s-]+)', text_data, re.IGNORECASE)
+        # Heuristic for Agreement Name near Order Form or titles
+        name_context = re.findall(r'(?:Order\s+Form|Agreement)\s*[:\s]*([A-Za-z0-9\s-]+)(?=\s*(?:Product|Quantity|List|Net))', text_data, re.IGNORECASE)
         # Update key_values with matched fields
         for key, value in dates:
             key_values[key] = value
@@ -42,8 +43,8 @@ def extract_key_values_with_layoutlm(page_data: list, pdf_path: str) -> Dict[str
             key_values["Agreement Start Date"] = date_context[0]
         if amounts:
             key_values["Amount"] = amounts[0]
-        for key, value in names:
-            key_values[key] = value
+        if name_context:
+            key_values["Agreement Name"] = name_context[0].strip()
 
         # Attempt LayoutLMv3 processing
         doc = fitz.open(pdf_path)
@@ -121,13 +122,10 @@ def extract_clauses(page_data: list) -> Dict[str, str]:
     """
     clauses = {}
     text_data = "\n".join([page["text"] for page in page_data])  # Use newlines for better segmentation
-    clause_keywords = ["Clause", "Section", "Terms", "Condition", "Provision", "Exhibit"]
-    for keyword in clause_keywords:
-        matches = re.finditer(rf'(?:{keyword}\s+\d+\.?\s*|\b{keyword}\b)\s*(.+?)(?=(?:{keyword}\s+\d+\.?|\Z))', text_data, re.IGNORECASE | re.DOTALL)
-        for match in matches:
-            clause_text = match.group(1).strip()
-            if clause_text and len(clause_text.split()) > 5:  # Minimum length to ensure meaningful clause
-                clauses[f"{keyword} {len([k for k in clauses.keys() if k.startswith(keyword)]) + 1}"] = clause_text
+    # Target "NO WAIVER" specifically under "General Provisions"
+    no_waiver_match = re.search(r'(?:General\s+Provisions\s*[\s\S]*?NO\s+WAIVER\s*[:\s]*)([\s\S]*?)(?=(?:General\s+Provisions|\Z))', text_data, re.IGNORECASE)
+    if no_waiver_match and len(no_waiver_match.group(1).strip().split()) > 5:
+        clauses["NO WAIVER"] = no_waiver_match.group(1).strip()
     return clauses if clauses else {}
 
 def run_ai_mapping_with_layoutlm(key_values: Dict[str, str], object_field_names: List[str], pdf_path: str) -> Dict: