Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| from PIL import Image | |
| import pytesseract | |
| import pandas as pd | |
| import re | |
| import cv2 | |
| import numpy as np | |
| from PyPDF2 import PdfReader | |
| import tempfile | |
| import os | |
| st.set_page_config(page_title="CA Invoice Automation", layout="wide") | |
| st.title("📑 CA Invoice Automation App") | |
| uploaded_file = st.file_uploader( | |
| "Upload Invoice (Image or PDF)", | |
| type=["jpg", "jpeg", "png", "pdf"] | |
| ) | |
| # -------------------------------- | |
| # Classification | |
| # -------------------------------- | |
| def classify_invoice(text): | |
| if "tax invoice" in text.lower(): | |
| return "GST_PURCHASE" | |
| elif "credit note" in text.lower(): | |
| return "CREDIT_NOTE" | |
| else: | |
| return "UNKNOWN" | |
| # -------------------------------- | |
| # Extraction | |
| # -------------------------------- | |
| def extract_data(text): | |
| gstin_pattern = r"\d{2}[A-Z]{5}\d{4}[A-Z]\dZ\d" | |
| date_pattern = r"\d{2}[-/]\d{2}[-/]\d{4}" | |
| amount_pattern = r"\d{1,3}(,\d{3})*(\.\d{2})?" | |
| gstin = re.search(gstin_pattern, text) | |
| date = re.search(date_pattern, text) | |
| amount = re.search(amount_pattern, text) | |
| return { | |
| "Date": date.group() if date else None, | |
| "GSTIN": gstin.group() if gstin else None, | |
| "Total Amount": amount.group() if amount else None | |
| } | |
| # -------------------------------- | |
| # OCR for Images | |
| # -------------------------------- | |
| def ocr_image(pil_image): | |
| image_np = np.array(pil_image) | |
| gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY) | |
| return pytesseract.image_to_string(gray) | |
| # -------------------------------- | |
| # MAIN | |
| # -------------------------------- | |
| if uploaded_file: | |
| full_text = "" | |
| try: | |
| # ================================ | |
| # -------- PDF HANDLING ---------- | |
| # ================================ | |
| if uploaded_file.type == "application/pdf": | |
| st.info("Extracting text directly from PDF...") | |
| with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp: | |
| tmp.write(uploaded_file.getvalue()) | |
| tmp_path = tmp.name | |
| reader = PdfReader(tmp_path) | |
| for page in reader.pages[:5]: # Limit to first 5 pages | |
| extracted_text = page.extract_text() | |
| if extracted_text: | |
| full_text += extracted_text + "\n" | |
| os.remove(tmp_path) | |
| if not full_text.strip(): | |
| st.warning("No embedded text found in PDF (might be scanned).") | |
| # ================================ | |
| # -------- IMAGE HANDLING -------- | |
| # ================================ | |
| else: | |
| image = Image.open(uploaded_file) | |
| st.image(image, caption="Uploaded Invoice", use_column_width=True) | |
| full_text = ocr_image(image) | |
| # -------------------------------- | |
| # Show OCR / Extracted Text | |
| # -------------------------------- | |
| st.subheader("📝 Extracted Text") | |
| st.text(full_text) | |
| # -------------------------------- | |
| # Classification | |
| # -------------------------------- | |
| invoice_type = classify_invoice(full_text) | |
| st.success(f"Invoice Type: {invoice_type}") | |
| # -------------------------------- | |
| # Extraction | |
| # -------------------------------- | |
| data = extract_data(full_text) | |
| formatted_data = { | |
| "Date": data["Date"], | |
| "Voucher Type": "Purchase" if invoice_type=="GST_PURCHASE" else "Unknown", | |
| "Vendor": "To be detected", | |
| "GSTIN": data["GSTIN"], | |
| "Invoice No": "To be extracted", | |
| "Taxable": None, | |
| "CGST": None, | |
| "SGST": None, | |
| "Total": data["Total Amount"], | |
| "Debit Ledger": "Purchase A/c", | |
| "Credit Ledger": "Vendor A/c", | |
| "Narration": f"Being purchase invoice dated {data['Date']}", | |
| } | |
| df = pd.DataFrame([formatted_data]) | |
| st.subheader("📊 Structured Voucher Data") | |
| st.dataframe(df, use_container_width=True) | |
| st.download_button( | |
| "⬇ Download CSV", | |
| df.to_csv(index=False), | |
| "voucher_output.csv", | |
| "text/csv" | |
| ) | |
| except Exception as e: | |
| st.error("Processing failed.") | |
| st.write(str(e)) | |