Update app.py

app.py

@@ -90,7 +90,7 @@ def extract_items(text):
         # Find the end of the main table
         table_end = len(lines)
         for i in range(table_start, len(lines)):
-            if "Item(s) Total" in lines[i] or "Total Value" in lines[i]:
+            if "Item(s) Total" in lines[i] or "Total Value" in lines[i] or "Sr.No Particulars" in lines[i]:
                 table_end = i
                 break
         
@@ -104,11 +104,12 @@ def extract_items(text):
             table_row_pattern = r"\|?\s*([A-Za-z\s\d-]+(?:\s[A-Za-z\s\d-]+)*?)\s*\|?\s*(\d+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?"
         else:
             # Pattern for Invoice_6164752968.pdf: "1 x Chicken Frankie | 60 | 6 | 54 | 2.5% | 1.35 | 2.5% | 1.35 | 56.7"
-            table_row_pattern = r"\|?\s*(\d+\s*x\s*[A-Za-z\s\d-]+(?:\s[A-Za-z\s\d-]+)*?)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*[\d.%]+\s*\|?\s*[\d.]+(?:\s*\|?\s*[\d.%]+\s*\|?\s*[\d.]+)?\s*\|?\s*([\d.]+)\s*\|?"
+            # Adjusted to be more flexible for potential spacing or formatting issues
+            table_row_pattern = r"\|?\s*(\d+\s*x\s*[A-Za-z\s\d-]+(?:\s[A-Za-z\s\d-]+)*?)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*\|?\s*([\d.]+)\s*(?:\|?\s*[\d.%]+\s*\|?\s*[\d.]+){2}\s*\|?\s*([\d.]+)\s*\|?"
 
         for line in table_lines:
             line = line.strip()
-            if not line:
+            if not line or "HSN Code" in line or "Total" in line:
                 continue
             # Skip alignment rows (e.g., "|---|---|")
             if re.match(r"\|?\s*[-:]+(\s*\|\s*[-:]+)*\s*\|?", line):
@@ -147,7 +148,7 @@ def extract_items(text):
     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]:
+            if "Total" in lines[i] and "Sr.No" not in lines[i]:
                 platform_fee_end = i + 1
                 break
         platform_fee_lines = lines[platform_fee_start:platform_fee_end]
@@ -198,7 +199,6 @@ def extract_entities(text):
         invoice_numbers.append(invoice_number)
         print(f"Matched Invoice Number: {invoice_number}")  # Debug
     if invoice_numbers:
-        # Prioritize the invoice number associated with "Restaurant Service" (HSN Code: 996331)
         for i, num in enumerate(invoice_numbers):
             start_idx = text.find(num)
             context = text[max(0, start_idx-100):start_idx+100]
@@ -206,7 +206,7 @@ def extract_entities(text):
                 primary_invoice_number = num
                 break
         if primary_invoice_number == "Unknown":
-            primary_invoice_number = invoice_numbers[0]  # Fallback to the first invoice number
+            primary_invoice_number = invoice_numbers[0]
     print(f"Primary Invoice Number: {primary_invoice_number}")  # Debug
 
     # Vendor Name
@@ -268,10 +268,7 @@ def extract_entities(text):
             print(f"Matched Amount: {amount}")  # Debug
         except ValueError:
             continue
-    # Sum only the final totals (e.g., after taxes)
     if total_amounts:
-        # In this invoice, "Total Value" appears twice: ₹184.5 (net value before taxes) and ₹193.726 (after taxes)
-        # We want the final total after taxes for the main items, plus the platform fee
         main_total = max([amt for amt in total_amounts if amt > 100], default=0.0)  # ₹193.726
         platform_fee = min([amt for amt in total_amounts if amt < 10], default=0.0)  # ₹3.54
         total_amount = main_total + platform_fee