Update app.py

app.py

@@ -87,6 +87,36 @@ def extract_items(text):
     
     if table_start == -1:
         print("Table header not found.")
+        # Look for platform fee as a separate table
+        platform_fee_start = -1
+        for i, line in enumerate(lines):
+            if "Sr.No Particulars" in line:
+                platform_fee_start = i + 1
+                break
+        if platform_fee_start != -1:
+            platform_fee_end = len(lines)
+            for i in range(platform_fee_start, len(lines)):
+                if "Total" in lines[i] and not "Sr.No" in lines[i]:
+                    platform_fee_end = i + 1
+                    break
+            platform_fee_lines = lines[platform_fee_start:platform_fee_end]
+            print("Platform fee lines:", platform_fee_lines)  # Debug
+            platform_fee_pattern = r"\|?\s*\d+\s*\|?\s*([A-Za-z\s]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?"
+            for line in platform_fee_lines:
+                line = line.strip()
+                if not line or "Total" in line:
+                    continue
+                match = re.match(platform_fee_pattern, line)
+                if match:
+                    description = match.group(1).strip()
+                    total_price = float(match.group(5))
+                    items.append({
+                        "description": description,
+                        "quantity": 1,  # Platform fee is a single item
+                        "unit_price": float(match.group(2)),  # Taxable amount
+                        "total_price": total_price
+                    })
+                    print(f"Extracted Platform Fee: {description}, Total Price: {total_price}")  # Debug
         return items
 
     # Find the end of the table (before "Total Amount", "Total Value", or end of text)
@@ -139,11 +169,43 @@ def extract_items(text):
         else:
             print(f"Failed to match row: {line}")
 
+    # Look for platform fee as a separate table
+    platform_fee_start = -1
+    for i, line in enumerate(lines):
+        if "Sr.No Particulars" in line:
+            platform_fee_start = i + 1
+            break
+    if platform_fee_start != -1:
+        platform_fee_end = len(lines)
+        for i in range(platform_fee_start, len(lines)):
+            if "Total" in lines[i] and not "Sr.No" in lines[i]:
+                platform_fee_end = i + 1
+                break
+        platform_fee_lines = lines[platform_fee_start:platform_fee_end]
+        print("Platform fee lines:", platform_fee_lines)  # Debug
+        platform_fee_pattern = r"\|?\s*\d+\s*\|?\s*([A-Za-z\s]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?"
+        for line in platform_fee_lines:
+            line = line.strip()
+            if not line or "Total" in line:
+                continue
+            match = re.match(platform_fee_pattern, line)
+            if match:
+                description = match.group(1).strip()
+                total_price = float(match.group(5))
+                items.append({
+                    "description": description,
+                    "quantity": 1,  # Platform fee is a single item
+                    "unit_price": float(match.group(2)),  # Taxable amount
+                    "total_price": total_price
+                })
+                print(f"Extracted Platform Fee: {description}, Total Price: {total_price}")  # Debug
+
     return items
 
 def extract_entities(text):
     """Extract structured invoice details including recipient name using flexible regex patterns."""
     invoice_numbers = []
+    primary_invoice_number = "Unknown"
     vendor_name = "Unknown"
     invoice_date = datetime.now().date()
     total_amount = 0.0
@@ -157,15 +219,27 @@ def extract_entities(text):
     invoice_num_pattern = r"(?:Invoice\s*(?:Number|No\.?|#)|Order\s*(?:Number|No\.?))\s*[:\-\s#]*([\w-]+)|(?:INV-|ORD-|Z\d{2}APOT\d{9})([\w-]+)"
     vendor_pattern = r"(?:Vendor\s*(?:Name|Company)?|Supplier|Company\s*Name|From|Sold\s*By|Restaurant\s*Name)\s*[:\-\s]*([A-Za-z\s&\.\-]+)(?=\s*(?:Address|Invoice\s*(?:No|Number)|Date|Phone|Email|\n|$))"
     invoice_date_pattern = r"(?:Invoice\s*Date|Date|Issue\s*Date)\s*[:\-\s]*(\d{4}-\d{2}-\d{2}|\d{2}/\d{2}/\d{4}|\d{2}-\d{2}-\d{4}|[A-Za-z]+\s*\d{1,2},\s*\d{4})"
-    total_amount_pattern = r"(?:Total\s*(?:Amount|Due|Value))?[^:\n]*[:\-\s]*[₹$£€]?\s*([\d,]+\.?\d*)\s*(?:USD|GBP|EUR|INR)?"
-    recipient_pattern = r"(?:Customer\s*Name|Recipient|Bill\s*To)\s*[:\-\s]*([A-Za-z\s]+)(?=\s*(?:Address|Phone|Email|\n|$))"
+    total_amount_pattern = r"(?:Total\s*(?:Amount|Due|Value))\s*[:\-\s]*[₹$£€]?\s*([\d,]+\.?\d*)\s*(?:USD|GBP|EUR|INR)?"
+    recipient_pattern = r"(?:Customer\s*Name|Recipient|Bill\s*To)\s*[:\-\s]*([A-Za-z]+)\s*(?=\s*(?:Address|Phone|Email|\n|$))"
 
-    # Invoice Numbers (capture multiple if present)
-    for match in re.finditer(invoice_num_pattern, text, re.IGNORECASE):
+    # Invoice Numbers (capture all, then prioritize)
+    invoice_num_matches = list(re.finditer(invoice_num_pattern, text, re.IGNORECASE))
+    for match in invoice_num_matches:
         invoice_number = match.group(1) if match.group(1) else match.group(2)
         invoice_numbers.append(invoice_number)
         print(f"Matched Invoice Number: {invoice_number}")  # Debug
-    invoice_numbers = invoice_numbers if invoice_numbers else ["Unknown"]
+    if invoice_numbers:
+        # Prioritize the invoice number associated with "Restaurant Service" (HSN Code: 996331)
+        for i, num in enumerate(invoice_numbers):
+            # Find the context of this invoice number in the text
+            start_idx = text.find(num)
+            context = text[max(0, start_idx-100):start_idx+100]
+            if "996331" in context:  # HSN Code for Restaurant Service
+                primary_invoice_number = num
+                break
+        if primary_invoice_number == "Unknown":
+            primary_invoice_number = invoice_numbers[0]  # Fallback to the first invoice number
+    print(f"Primary Invoice Number: {primary_invoice_number}")  # Debug
 
     # Vendor Name
     vendor_match = re.search(vendor_pattern, text, re.IGNORECASE)
@@ -193,7 +267,7 @@ def extract_entities(text):
     # Invoice Date (prioritize "Invoice Date")
     invoice_date_match = None
     for line in text.split('\n'):
-        if "Invoice Date" in line:
+        if "Invoice Date" in line and not "Order Date" in line:
             match = re.search(invoice_date_pattern, line, re.IGNORECASE)
             if match:
                 invoice_date_match = match
@@ -216,14 +290,16 @@ def extract_entities(text):
         except ValueError as e:
             print(f"Failed to parse Invoice Date '{date_str}': {str(e)}")  # Debug
 
-    # Total Amount (sum all "Total Value" entries)
+    # Total Amount (sum all "Total Value" entries, fix parsing)
     total_amount_matches = re.finditer(total_amount_pattern, text, re.IGNORECASE)
     total_amounts = []
     for match in total_amount_matches:
         amount_str = match.group(1).replace(",", "")
         try:
             amount = float(amount_str)
-            total_amounts.append(amount)
+            # Ignore amounts that are unrealistically large (likely parsing errors)
+            if amount < 1000000:  # Arbitrary threshold to exclude erroneous large numbers
+                total_amounts.append(amount)
             print(f"Matched Amount: {amount}")  # Debug
         except ValueError:
             continue
@@ -236,9 +312,9 @@ def extract_entities(text):
         recipient_name = recipient_match.group(1).strip()
         print(f"Matched Recipient Name: {recipient_name}")  # Debug
 
-    return invoice_numbers, vendor_name, invoice_date, total_amount, recipient_name
+    return primary_invoice_number, vendor_name, invoice_date, total_amount, recipient_name
 
-def fetch_vendor_history(vendor_name, invoice_numbers, time_window_days=30):
+def fetch_vendor_history(vendor_name, invoice_number, time_window_days=30):
     """Fetch historical invoices for the vendor from Salesforce."""
     if sf is None:
         return pd.DataFrame()
@@ -265,15 +341,14 @@ def fetch_vendor_history(vendor_name, invoice_numbers, time_window_days=30):
         print(f"Failed to fetch vendor history: {str(e)}")
         return pd.DataFrame()
 
-def check_data_consistency(invoice_numbers, vendor_name, invoice_date, history_df):
+def check_data_consistency(invoice_number, vendor_name, invoice_date, history_df):
     """Check for data consistency issues like duplicates."""
     consistency_issues = []
 
     if not history_df.empty:
-        for invoice_number in invoice_numbers:
-            duplicate_invoices = history_df[history_df['Invoice_Number__c'] == invoice_number]
-            if not duplicate_invoices.empty:
-                consistency_issues.append(f"Duplicate invoice number '{invoice_number}' found for vendor '{vendor_name}'.")
+        duplicate_invoices = history_df[history_df['Invoice_Number__c'] == invoice_number]
+        if not duplicate_invoices.empty:
+            consistency_issues.append(f"Duplicate invoice number '{invoice_number}' found for vendor '{vendor_name}'.")
 
     return consistency_issues
 
@@ -375,16 +450,16 @@ def process_invoice(pdf_file):
     if "Error" in text:
         return f"**Error**: {text}"
 
-    invoice_numbers, vendor_name, invoice_date, total_amount, recipient_name = extract_entities(text)
+    invoice_number, vendor_name, invoice_date, total_amount, recipient_name = extract_entities(text)
     items = extract_items(text)
     text_length = len(text)
 
-    history_df = fetch_vendor_history(vendor_name, invoice_numbers)
-    consistency_issues = check_data_consistency(invoice_numbers, vendor_name, invoice_date, history_df)
+    history_df = fetch_vendor_history(vendor_name, invoice_number)
+    consistency_issues = check_data_consistency(invoice_number, vendor_name, invoice_date, history_df)
 
     data = {
         "invoice_id": str(uuid.uuid4()),
-        "invoice_number": "; ".join(invoice_numbers),
+        "invoice_number": invoice_number,
         "vendor_name": vendor_name,
         "amount": total_amount,
         "invoice_date": invoice_date,
@@ -424,8 +499,7 @@ def process_invoice(pdf_file):
 
     output = [
         "## Fraud Detection Summary",
-        f"- **Invoice Number**: {'; '.join(invoice_numbers)}",
-        f"- **Recipient Name**: {recipient_name}",
+        f"- **Invoice Number**: {invoice_number}",
         f"- **Vendor Name**: {vendor_name}",
         f"- **Invoice Date**: {invoice_date}",
         f"- **Invoice Amount**: ₹{total_amount:,.2f}",  # Assuming INR for this PDF
@@ -456,7 +530,7 @@ def process_invoice(pdf_file):
     if sf is not None:
         try:
             record_data = {
-                "Invoice_Number__c": "; ".join(invoice_numbers),
+                "Invoice_Number__c": invoice_number,
                 "Vendor_Name__c": vendor_name,
                 "Invoice_Amount__c": total_amount,
                 "Invoice_Date__c": str(invoice_date),