Update bhel.py

bhel.py

@@ -1,83 +1,110 @@
 import pdfplumber
 import pandas as pd
 import re
+import gradio as gr
 
-# Function: Extract Text from PDF
-def extract_text_from_pdf(pdf_file):
-    with pdfplumber.open(pdf_file.name) as pdf:
-        text = ""
+def extract_data(pdf_file):
+    """
+    Extract data from the uploaded PDF for dynamic ranges (e.g., 10 to n).
+    """
+    data = []
+    columns = ["SI No", "Material Description", "Unit", "Quantity", "Dely Qty", "Dely Date", "Unit Rate", "Value"]
+
+    start_si = 10  # Start from SI No 10
+    end_si = None  # Dynamically detect the end SI No
+
+    with pdfplumber.open(pdf_file) as pdf:
         for page in pdf.pages:
-            text += page.extract_text()
-    return text
+            full_text = page.extract_text()  # Get the text content for the page
+            lines = full_text.splitlines() if full_text else []
+
+            for line in lines:
+                try:
+                    # Parse the first column for SI No
+                    si_no_match = re.match(r"^\s*(\d+)\s", line)
+                    if not si_no_match:
+                        continue
+
+                    si_no = int(si_no_match.group(1))
+                    # Dynamically set the end SI No if higher SI Nos are found
+                    if end_si is None or si_no > end_si:
+                        end_si = si_no
+
+                    if si_no < start_si:
+                        continue  # Skip rows below the start SI No
+
+                    # Extract Material Description and details dynamically
+                    material_desc = extract_material_description(full_text, si_no)
+
+                    # Extract remaining fields
+                    parts = line.split()
+                    unit = parts[3]
+                    quantity = int(parts[4])
+                    dely_qty = int(parts[5])
+                    dely_date = parts[6]
+                    unit_rate = float(parts[7])
+                    value = float(parts[8])
+
+                    # Append row data
+                    data.append([si_no, material_desc, unit, quantity, dely_qty, dely_date, unit_rate, value])
+
+                except (ValueError, IndexError):
+                    # Skip invalid rows or rows with missing data
+                    continue
 
-# Function: Parse PO Items
-def parse_po_items(text):
+    # Convert data to DataFrame and save as Excel
+    df = pd.DataFrame(data, columns=columns)
+    excel_path = "/tmp/Extracted_PO_Data_Dynamic.xlsx"
+    df.to_excel(excel_path, index=False)
+    return excel_path
+
+def extract_material_description(full_text, si_no):
     """
-    Parses purchase order items from the extracted text.
-    Handles split descriptions across lines and filters unwanted text.
+    Extract Material Description, including Material Number, HSN Code, and IGST, using unique patterns.
     """
-    lines = text.splitlines()
-    data = []
-    current_item = {}
-    description_accumulator = []
-
-    for line in lines:
-        # Match the start of an item row
-        item_match = re.match(r"^(?P<Item>\d+)\s+(?P<Description>.+)", line)
-        if item_match:
-            # Save the previous item and start a new one
-            if current_item:
-                current_item["Description"] = " ".join(description_accumulator).strip()
-                data.append(current_item)
-                description_accumulator = []
-
-            current_item = {
-                "Item": item_match.group("Item"),
-                "Description": "",
-                "Qty": "",
-                "Unit": "",
-                "Unit Price": "",
-                "Total Price": "",
-            }
-            description_accumulator.append(item_match.group("Description"))
-        elif current_item:
-            # Handle additional description lines or split descriptions
-            description_accumulator.append(line.strip())
-
-        # Match Qty, Unit, Unit Price, and Total Price
-        qty_match = re.search(r"(?P<Qty>\d+)\s+(Nos\.|Set)", line)
-        if qty_match:
-            current_item["Qty"] = qty_match.group("Qty")
-            current_item["Unit"] = qty_match.group(2)
-
-        price_match = re.search(r"(?P<UnitPrice>[\d.]+)\s+(?P<TotalPrice>[\d.]+)$", line)
-        if price_match:
-            current_item["Unit Price"] = price_match.group("UnitPrice")
-            current_item["Total Price"] = price_match.group("TotalPrice")
-
-    # Save the last item
-    if current_item:
-        current_item["Description"] = " ".join(description_accumulator).strip()
-        data.append(current_item)
-
-    if not data:
-        return None, "No items found. Please check the PDF file format."
-    df = pd.DataFrame(data)
-    return df, "Data extracted successfully."
-
-# Function: Save to Excel
-def save_to_excel(df, output_path="bhel_extracted_data.xlsx"):
-    df.to_excel(output_path, index=False)
-    return output_path
-
-# Main function to process PDF
-def process_pdf(file):
-    try:
-        text = extract_text_from_pdf(file)
-        df, status = parse_po_items(text)
-        if df is not None:
-            output_path = save_to_excel(df)
-            return output_path, status
-        return None, status
-    except Exception as e:
-        return None, f"Error during processing: {str(e)}"
+    material_desc = ""
+
+    # Match the specific SI No row to extract details
+    si_no_pattern = rf"{si_no}\s+(BPS\s+\d+).*?Material\s+Number:\s+(\d+)"
+    match = re.search(si_no_pattern, full_text, re.DOTALL)
+    if match:
+        bps_code = match.group(1)
+        material_number = match.group(2)
+        material_desc += f"{bps_code}\nMaterial Number: {material_number}\n"
+
+    # Extract HSN Code
+    hsn_code_match = re.search(r"HSN\s+Code:\s*(\d+)", full_text)
+    if hsn_code_match:
+        hsn_code = hsn_code_match.group(1)
+        material_desc += f"HSN Code: {hsn_code}\n"
+    else:
+        material_desc += "HSN Code: Not Found\n"
+
+    # Extract IGST
+    igst_match = re.search(r"IGST\s*:\s*(\d+)\s*%", full_text)
+    if igst_match:
+        igst = igst_match.group(1)
+        material_desc += f"IGST: {igst} %"
+    else:
+        material_desc += "IGST: Not Found"
+
+    return material_desc.strip()
+
+# Gradio Interface
+def gradio_interface(pdf_file):
+    """
+    Interface function for Gradio to process the PDF and return the Excel file.
+    """
+    return extract_data(pdf_file.name)
+
+# Define Gradio interface
+interface = gr.Interface(
+    fn=gradio_interface,
+    inputs=gr.File(label="Upload PDF"),
+    outputs=gr.File(label="Download Extracted Excel"),
+    title="Dynamic BHEL PO Data Extractor",
+    description="Upload a PDF to extract accurate Material Numbers and related data dynamically into an Excel file."
+)
+
+if __name__ == "__main__":
+    interface.launch()