import json import os import secrets import hashlib import hmac import uuid import tempfile from io import BytesIO from queue import Queue from threading import Thread from dotenv import load_dotenv from flask import ( Flask, Response, render_template_string, request, send_file, stream_with_context, jsonify, ) from werkzeug.utils import secure_filename import razorpay from pypdf import PdfReader from openpyxl import load_workbook import pandas as pd from services import ( excel_to_csv_process, pdf_delete_pages_process, pdf_merge_process, pdf_split_process, pdf_to_word_process, summarize_process, pdf_ocr_process, bank_statement_process, ) from statement_parser import BANK_NAMES as STAT_BANK_NAMES from statement_parser import excel_bytes_from_df from statement_parser import parse_statement_pdf load_dotenv() OCR_PRICE_PER_PAGE_PAISE = 500 PDF_FREE_PAGES = 10 PDF_EXTRA_PAGE_PRICE_PAISE = 200 EXCEL_FREE_ROWS = 500 EXCEL_PRICE_PER_EXTRA_ROW_PAISE = 20 EXCEL_MAX_ROWS = 50000 PENDING_ORDERS = {} BYPASS_ACCESS_KEY = os.getenv("OCR_BYPASS_KEY", "215836").strip() PDF_SERVICES = { "pdf-ocr": "Searchable PDF OCR", "pdf-to-word": "PDF to Word", "pdf-split": "PDF Splitter", "pdf-merge": "PDF Merger", "pdf-delete-pages": "PDF Delete Pages", "summarize": "Document Summarizer", } EXCEL_SERVICE = "excel-to-csv" def razorpay_keys(): key_id = os.getenv("RAZORPAY_KEY_ID", "").strip() key_secret = os.getenv("RAZORPAY_KEY_SECRET", "").strip() if not key_id or not key_secret: raise RuntimeError("Razorpay keys are not configured. Set RAZORPAY_KEY_ID and RAZORPAY_KEY_SECRET.") return key_id, key_secret def verify_razorpay_signature(order_id, payment_id, signature): try: _, key_secret = razorpay_keys() except RuntimeError: return False message = f"{order_id}|{payment_id}".encode("utf-8") expected = hmac.new(key_secret.encode("utf-8"), message, hashlib.sha256).hexdigest() return hmac.compare_digest(expected, signature or "") def has_valid_bypass_key(): access_key = request.form.get("access_key", "").strip() return bool(BYPASS_ACCESS_KEY and hmac.compare_digest(access_key, BYPASS_ACCESS_KEY)) def count_pdf_pages(path, password=None): reader = PdfReader(path) if reader.is_encrypted: if password: try: reader.decrypt(password) except Exception: raise ValueError("Incorrect password for protected PDF.") else: raise ValueError("This PDF is password-protected. Please enter the PDF password.") return len(reader.pages) def count_excel_rows(input_path, extension): if extension in (".xlsx", ".xlsm"): workbook = load_workbook(input_path, read_only=True, data_only=True) try: sheet = workbook.active rows = sum(1 for _ in sheet.iter_rows(values_only=True)) finally: workbook.close() return rows return len(pd.read_excel(input_path, header=None)) def order_amount_for(service, path=None, pages=None, num_pdfs=None, excel_rows=None, bank_entries=None, password=None): if path and pages is None: try: pages = count_pdf_pages(path, password=password) except Exception: pages = 0 if service == "pdf-ocr": if pages is None or pages < 1: raise ValueError("This PDF does not contain any pages.") amount = pages * 200 # 2 Rs per page return amount, pages, f"{pages} page{'s' if pages != 1 else ''} x INR 2.00 = INR {amount / 100:.2f}" elif service == "pdf-split": if pages is None or pages < 1: raise ValueError("This PDF does not contain any pages.") if pages <= 100: amount = 0 summary = f"{pages} page{'s' if pages != 1 else ''}. Free tier applied (up to 100 pages free)." else: extra = pages - 100 if extra < 51: amount = 0 summary = f"{pages} page{'s' if pages != 1 else ''}. Free tier applied (first 100 pages free + {extra} extra page{'s' if extra != 1 else ''} free, as extra is less than 51)." else: amount = extra * 2 # 2 paise per page summary = f"First 100 pages free + {extra} extra page{'s' if extra != 1 else ''} x INR 0.02 = INR {amount / 100:.2f}" return amount, pages, summary elif service == "pdf-merge": if num_pdfs is None: raise ValueError("Number of PDFs is required for merge calculation.") if num_pdfs <= 10: amount = 0 summary = f"{num_pdfs} PDF{'s' if num_pdfs != 1 else ''}. Free tier applied (up to 10 PDFs free)." else: extra = num_pdfs - 10 amount = extra * 2 # 2 paise per PDF summary = f"10 PDFs free + {extra} extra PDF{'s' if extra != 1 else ''} x INR 0.02 = INR {amount / 100:.2f}" return amount, num_pdfs, summary elif service == "pdf-to-word": if pages is None or pages < 1: raise ValueError("This PDF does not contain any pages.") if pages <= 100: amount = 0 summary = f"{pages} page{'s' if pages != 1 else ''}. Free tier applied (up to 100 pages free)." else: extra = pages - 100 amount = extra * 2 # 2 paise per page summary = f"100 pages free + {extra} extra page{'s' if extra != 1 else ''} x INR 0.02 = INR {amount / 100:.2f}" return amount, pages, summary elif service == "pdf-delete-pages": return 0, pages or 0, "Always free." elif service == "summarize": # Keep summarize free or default return 0, pages or 0, "Always free." elif service == "excel-to-csv": if excel_rows is None: raise ValueError("Excel row count is required.") if excel_rows > EXCEL_MAX_ROWS: raise ValueError("Excel files above 50,000 rows are not supported.") if excel_rows <= 10000: amount = 0 summary = f"{excel_rows} rows. Free tier applied (up to 10,000 entries free)." else: extra = excel_rows - 10000 groups = (extra + 9) // 10 amount = groups * 2 # 2 paise for each 10 entry summary = f"10,000 entries free + {extra} extra entries ({groups} groups of 10) x INR 0.02 = INR {amount / 100:.2f}" return amount, excel_rows, summary elif service.startswith("bank-"): if bank_entries is None: raise ValueError("Bank entry count is required.") amount = bank_entries * 8 # 8 paise per entry summary = f"{bank_entries} entries x INR 0.08 = INR {amount / 100:.2f}" return amount, bank_entries, summary raise ValueError("Unknown paid service.") def create_order(service, amount, pages_or_rows): if amount == 0: return "", None key_id, key_secret = razorpay_keys() client = razorpay.Client(auth=(key_id, key_secret)) order = client.order.create( { "amount": amount, "currency": "INR", "receipt": f"scanly_{uuid.uuid4().hex[:25]}", "notes": { "service": service, "pages_or_rows": str(pages_or_rows), "amount": str(amount), }, } ) PENDING_ORDERS[order["id"]] = { "amount": amount, "pages_or_rows": pages_or_rows, "service": service, "currency": "INR", } return key_id, order app = Flask(__name__) ROOT = os.path.dirname(__file__) INDEX_HTML = os.path.join(ROOT, "templatemo_622_clearwave", "index.html") # ── In-memory task store ────────────────────────────────────────────────── # task_id → {"queue": Queue, "result": bytes|None, "error": str|None, "fname": str, "mime": str} _pending: dict[str, dict] = {} def _run_task(task_id: str, fn, *args, **kwargs): entry = _pending[task_id] q: Queue = entry["queue"] def _progress(pct, msg): q.put({"type": "progress", "percent": pct, "message": msg}) try: result = fn(*args, progress_callback=_progress, **kwargs) entry["result"] = result if isinstance(result, bytes) else result.read() q.put({"type": "done"}) except Exception as exc: entry["error"] = str(exc) q.put({"type": "error", "message": str(exc)}) # ── Index ────────────────────────────────────────────────────────────────── @app.route("/") def index(): if not os.path.exists(INDEX_HTML): return "SCANLY home page not found.", 404 with open(INDEX_HTML, "r", encoding="utf-8") as fh: return fh.read() @app.route("/assets/") def assets(filename): from flask import send_from_directory return send_from_directory(ROOT, filename) # ── SSE progress endpoint ───────────────────────────────────────────────── @app.route("/progress/") def progress_stream(task_id): entry = _pending.get(task_id) if not entry: return Response( "data: " + json.dumps({"type": "done"}) + "\n\n", mimetype="text/event-stream", ) q = entry["queue"] def generate(): while True: msg = q.get() yield "data: " + json.dumps(msg) + "\n\n" if msg["type"] in ("done", "error"): break return Response(stream_with_context(generate()), mimetype="text/event-stream") # ── Multi-purpose service page template ─────────────────────────────────── SERVICE_SHELL = r""" {{ title }} | SCANLY
← Home 💰 Pricing
SCANLY

{{ title }}

{{ subtitle }}

🔒 We don't save your data. All processing is done in-memory and immediately deleted.
{% if hint %}
💡 {{ hint }}
{% endif %}
{% if select_input %} {% endif %} {% if extra_input %} {% endif %}

Starting…

""" PRICING_TEMPLATE = r""" Pricing Plans | SCANLY
← Back to Dashboard
SCANLY

Simple & Transparent Pricing

No subscriptions, no hidden fees. Generous free tiers on almost all services, and pay-as-you-go rates when you scale.

🎁 What's Free for You

Free Tier

Delete Pages

₹0.00

Permanently remove selected pages (e.g. `1,3-5`) from your PDF documents.

  • 100% Free always
  • Unlimited files and pages
  • In-memory local secure processing
100 Pages Free

PDF Splitter

₹0.00/ up to 100 pages

Extract single pages or split a PDF into range groups inside a downloadable ZIP.

  • First 100 pages are completely free
  • Next 50 pages are also free
  • Only pays if extra pages exceed 50
100 Pages Free

PDF to Word

₹0.00/ up to 100 pages

Convert PDF text layout into an editable Microsoft Word document.

  • First 100 pages are completely free
  • Extracts all text structure
  • Download as editable .docx
10 PDFs Free

PDF Merger

₹0.00/ up to 10 PDFs

Combine multiple PDF files into a single master document.

  • First 10 PDFs merged are free
  • Keep your pages in ordered structure
10,000 Rows Free

Excel → CSV

₹0.00/ up to 10k entries

Instantly convert large Excel spreadsheets into clean flat CSV files.

  • First 10,000 entries (rows) are free
  • Highly optimized conversion speed
Always Free

Summarizer

₹0.00

Generate clean bullet-point document summaries using AI.

  • AI-powered summaries
  • Outputs formatted DOCX files

⚡ Pay As You Go (Beyond Free limits)

Paid Service

Searchable PDF OCR

₹2.00/ page

Run advanced ocrmypdf engines to overlay selectable text layers on scanned PDFs.

  • Accurate deskew and rotation
  • Runs local OCR models
  • Make any image PDF selectable
Paid Service

Bank Statement Parser

₹0.08/ entry

Extract structured transactions from Axis, SBI, ICICI, Kotak, HDFC, and other PDF statements into Excel.

  • Select specific pages to parse
  • Pay only for extracted entries
  • Highly accurate rule-based extraction
Paid Service

Split PDF Extra

₹0.02/ extra page

When splitting extremely large PDFs containing more than 150 pages.

  • Applies only from page 151+
  • ₹0.02 per additional page
Paid Service

Merge PDF Extra

₹0.02/ extra PDF

When merging large batches containing more than 10 documents.

  • Applies only from 11th PDF+
  • ₹0.02 per additional file
Paid Service

PDF to Word Extra

₹0.02/ extra page

When converting large PDFs containing more than 100 pages.

  • Applies only from page 101+
  • ₹0.02 per additional page
Paid Service

Excel → CSV Extra

₹0.02/ 10 extra entries

When converting spreadsheets containing more than 10,000 entries (rows).

  • Applies only from row 10,001+
  • ₹0.02 per group of 10 extra rows
""" @app.route("/pricing") def pricing_page(): return render_template_string(PRICING_TEMPLATE) # ── Service definitions ─────────────────────────────────────────────────── SERVICE_ROUTES = { "pdf-ocr": { "title": "Searchable PDF OCR", "subtitle": "Convert scanned PDF documents into searchable PDFs with selectable text overlay.", "field_label": "Choose a scanned PDF", "accept": ".pdf", "btn_label": "Run OCR", "hint": "This service runs ocrmypdf locally to create a searchable text layer. Rate: INR 5.00 per PDF page.", "select_input": { "name": "mode", "label": "OCR Mode", "options": [ ("fast", "Fast OCR - skip pages that already have text"), ("accurate", "High accuracy - slower full OCR") ] } }, "pdf-split": { "title": "PDF Splitter", "subtitle": "Split a PDF into individual pages or groups of pages.", "field_label": "Choose a PDF", "accept": ".pdf", "btn_label": "Split PDF", "extra_input": {"name": "split_size", "type": "number", "placeholder": "Group size (e.g. 5, default: 1)", "label": "Pages per split group"}, "hint": "Default splits into individual pages. Enter a group size (e.g., 5) to split into groups of pages.", }, "pdf-merge": { "title": "PDF Merger", "subtitle": "Upload multiple PDFs and merge them into a single document.", "field_label": "Choose PDF files", "accept": ".pdf", "btn_label": "Merge PDFs", "multiple": True, "hint": "Select two or more PDFs. Order them correctly before uploading.", }, "pdf-to-word": { "title": "PDF to Word", "subtitle": "Convert a PDF document into an editable Word (.docx) file.", "field_label": "Choose a PDF", "accept": ".pdf", "btn_label": "Convert to Word", "hint": "Extracts selectable text from PDF pages and writes as a nicely formatted Word document.", }, "pdf-delete-pages": { "title": "PDF Delete Pages", "subtitle": "Remove specific pages from a PDF and download the cleaned version.", "field_label": "Choose a PDF", "accept": ".pdf", "btn_label": "Delete Pages", "extra_input": {"name": "pages", "type": "text", "placeholder": "Pages to delete, e.g. 1,3-5,7", "label": "Page numbers to remove"}, "hint": "Specify pages like '1,3-5,7' to remove. The PDF must have at least one page remaining.", }, "excel-to-csv": { "title": "Excel → CSV Converter", "subtitle": "Convert XLSX or XLS files to clean CSV format.", "field_label": "Choose a spreadsheet", "accept": ".xlsx,.xls,.csv", "btn_label": "Convert to CSV", "hint": "Supports .xlsx, .xls. Already-CSV files pass through.", }, "summarize": { "title": "Document Summarizer", "subtitle": "Upload a PDF, DOCX, or TXT file and get a bullet-point summary as a DOCX.", "field_label": "Choose a document", "accept": ".pdf,.docx,.txt", "btn_label": "Summarize", "hint": "Output is a clean bullet-point DOCX file.", }, } for slug, info in SERVICE_ROUTES.items(): action_url = f"/api/{slug}" def _make_route(_slug, _info, _action): @app.route(f"/service/{_slug}", methods=["GET"], endpoint=f"service_{_slug}") def service_page(): return render_template_string(SERVICE_SHELL, **_info, action=_action, slug=_slug) _make_route(slug, info, action_url) # ── API endpoints ───────────────────────────────────────────────────────── def verify_payment_before_processing(service, request_files_field="file"): if has_valid_bypass_key(): return None amount = 0 pages_or_rows = 0 if service in PDF_SERVICES: if service == "pdf-merge": files = request.files.getlist(request_files_field) num_pdfs = len([f for f in files if f and f.filename]) amount, pages_or_rows, _ = order_amount_for(service, num_pdfs=num_pdfs) else: f = request.files.get(request_files_field) if not f or not f.filename: return "No file uploaded", 400 temp = tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) temp.write(f.read()) temp.close() f.seek(0) try: pages = count_pdf_pages(temp.name) amount, pages_or_rows, _ = order_amount_for(service, pages=pages) finally: try: os.remove(temp.name) except OSError: pass elif service == EXCEL_SERVICE: f = request.files.get(request_files_field) if not f or not f.filename: return "No file uploaded", 400 ext = os.path.splitext(secure_filename(f.filename))[1].lower() temp = tempfile.NamedTemporaryFile(suffix=ext, delete=False) temp.write(f.read()) temp.close() f.seek(0) try: rows = count_excel_rows(temp.name, ext) amount, pages_or_rows, _ = order_amount_for(service, excel_rows=rows) finally: try: os.remove(temp.name) except OSError: pass elif service.startswith("bank-"): bank_slug = service.split("bank-")[1] uploads = request.files.getlist(request_files_field) valid_files = [up for up in uploads if up and up.filename and up.filename.lower().endswith(".pdf") and not os.path.basename(up.filename).startswith(".")] if not valid_files: return "No PDF file uploaded", 400 total_entries = 0 page_numbers = request.form.get("pages", "").strip() pdf_password = request.form.get("pdf_password", "").strip() for up in valid_files: pdf_bytes = up.read() up.seek(0) df = parse_statement_pdf(pdf_bytes, bank_slug=bank_slug, page_numbers=page_numbers, password=pdf_password) total_entries += len(df) amount, pages_or_rows, _ = order_amount_for(service, bank_entries=total_entries) if amount == 0: return None order_id = request.form.get("razorpay_order_id", "") payment_id = request.form.get("razorpay_payment_id", "") signature = request.form.get("razorpay_signature", "") order = PENDING_ORDERS.get(order_id) if not order_id or not payment_id or not signature: return "Complete the Razorpay payment before processing.", 402 if not order: return "Payment order expired or invalid. Please pay again.", 402 if order["service"] != service or order["amount"] != amount or order["pages_or_rows"] != pages_or_rows: return "Uploaded file page/row count does not match the payment. Please calculate payment again.", 400 if not verify_razorpay_signature(order_id, payment_id, signature): return "Payment signature verification failed.", 402 PENDING_ORDERS.pop(order_id, None) return None @app.route("/payment/order", methods=["POST"]) def create_payment_order(): service = request.form.get("service", "").strip() try: if has_valid_bypass_key(): return jsonify({ "key": "", "order_id": "", "amount": 0, "currency": "INR", "pages_or_rows": 0, "display_amount": "0.00", "summary": "Bypass key validated.", "description": "Bypass access key entered.", "free": True }) if service in PDF_SERVICES: if service == "pdf-merge": files = request.files.getlist("file") if not files or len(files) < 2: return jsonify({"error": "Upload at least 2 PDF files"}), 400 num_pdfs = len([f for f in files if f and f.filename]) amount, pages_or_rows, summary = order_amount_for(service, num_pdfs=num_pdfs) else: uploaded = request.files.get("file") if not uploaded or uploaded.filename == "": return jsonify({"error": "Please upload a PDF file"}), 400 filename = secure_filename(uploaded.filename) ext = os.path.splitext(filename)[1].lower() if ext != ".pdf": return jsonify({"error": "Only PDFs are supported for this service"}), 400 temp = tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) temp.close() uploaded.save(temp.name) try: amount, pages_or_rows, summary = order_amount_for(service, path=temp.name) finally: try: os.remove(temp.name) except OSError: pass description = PDF_SERVICES[service] elif service == EXCEL_SERVICE: uploaded = request.files.get("file") if not uploaded or uploaded.filename == "": return jsonify({"error": "Please upload an Excel file"}), 400 filename = secure_filename(uploaded.filename) ext = os.path.splitext(filename)[1].lower() if ext not in (".xlsx", ".xls", ".xlsm"): return jsonify({"error": "Please upload a valid Excel file"}), 400 temp = tempfile.NamedTemporaryFile(suffix=ext, delete=False) temp.close() uploaded.save(temp.name) try: rows = count_excel_rows(temp.name, ext) amount, pages_or_rows, summary = order_amount_for(service, excel_rows=rows) finally: try: os.remove(temp.name) except OSError: pass description = "Excel to CSV" elif service.startswith("bank-"): bank_slug = service.split("bank-")[1] if bank_slug not in STAT_BANK_NAMES: return jsonify({"error": "Unknown bank service"}), 400 uploads = request.files.getlist("file") valid_files = [up for up in uploads if up and up.filename and up.filename.lower().endswith(".pdf") and not os.path.basename(up.filename).startswith(".")] if not valid_files: return jsonify({"error": "Please upload one or more PDF statements"}), 400 total_entries = 0 page_numbers = request.form.get("pages", "").strip() pdf_password = request.form.get("pdf_password", "").strip() for up in valid_files: pdf_bytes = up.read() up.seek(0) df = parse_statement_pdf(pdf_bytes, bank_slug=bank_slug, page_numbers=page_numbers, password=pdf_password) total_entries += len(df) amount, pages_or_rows, summary = order_amount_for(service, bank_entries=total_entries) if len(valid_files) == 1: description = f"{STAT_BANK_NAMES[bank_slug]} Statement Parsing" else: description = f"{STAT_BANK_NAMES[bank_slug]} Batch Parsing ({len(valid_files)} statements)" else: return jsonify({"error": "Unknown service."}), 400 key_id, order = create_order(service, amount, pages_or_rows) order_id = order["id"] if order else "" return jsonify({ "key": key_id, "order_id": order_id, "amount": amount, "currency": "INR", "pages_or_rows": pages_or_rows, "display_amount": f"{amount / 100:.2f}", "summary": summary, "description": description, "free": amount == 0, }) except ValueError as exc: return jsonify({"error": str(exc)}), 400 except Exception as exc: return jsonify({"error": f"Could not create payment order: {exc}"}), 500 def _start_task(task_id: str, fn, *, fname: str, mime: str, **kw): entry = { "queue": Queue(), "result": None, "error": None, "fname": fname, "mime": mime, } _pending[task_id] = entry t = Thread(target=_run_task, args=(task_id, fn), kwargs=kw, daemon=True) t.start() @app.route("/api/pdf-ocr", methods=["POST"]) def api_pdf_ocr(): payment_err = verify_payment_before_processing("pdf-ocr") if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 data = f.read() filename = secure_filename(f.filename) or "scan.pdf" mode = request.form.get("mode", "fast") task_id = secrets.token_hex(8) base = os.path.splitext(filename)[0] def process(**kw): return pdf_ocr_process(data, filename, mode, **kw) _start_task(task_id, process, fname=f"{base}_ocr.pdf", mime="application/pdf") return _accepted(task_id) @app.route("/api/pdf-split", methods=["POST"]) def api_pdf_split(): payment_err = verify_payment_before_processing("pdf-split") if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 split_size_str = request.form.get("split_size", "1").strip() try: split_size = int(split_size_str) if split_size_str else 1 if split_size < 1: split_size = 1 except ValueError: split_size = 1 data = f.read() task_id = secrets.token_hex(8) base = os.path.splitext(secure_filename(f.filename))[0] _start_task(task_id, pdf_split_process, fname=f"{base}_pages.zip", mime="application/zip", file_bytes=data, split_size=split_size) return _accepted(task_id) @app.route("/api/pdf-merge", methods=["POST"]) def api_pdf_merge(): payment_err = verify_payment_before_processing("pdf-merge") if payment_err: return payment_err files = request.files.getlist("file") if len(files) < 2: return "Upload at least 2 PDF files", 400 task_id = secrets.token_hex(8) files_data = [(secure_filename(f.filename) or "doc.pdf", f.read()) for f in files] _start_task(task_id, pdf_merge_process, fname="merged.pdf", mime="application/pdf", files_data=files_data) return _accepted(task_id) @app.route("/api/excel-to-csv", methods=["POST"]) def api_excel_to_csv(): payment_err = verify_payment_before_processing(EXCEL_SERVICE) if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 data = f.read() filename = secure_filename(f.filename) or "sheet.xlsx" task_id = secrets.token_hex(8) base = os.path.splitext(filename)[0] def process(**kw): return excel_to_csv_process(data, filename, **kw) _start_task(task_id, process, fname=f"{base}.csv", mime="text/csv") return _accepted(task_id) @app.route("/api/summarize", methods=["POST"]) def api_summarize(): payment_err = verify_payment_before_processing("summarize") if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 data = f.read() filename = secure_filename(f.filename) or "doc.pdf" task_id = secrets.token_hex(8) base = os.path.splitext(filename)[0] def process(**kw): return summarize_process(data, filename, **kw) _start_task( task_id, process, fname=f"{base}_summary.docx", mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document", ) return _accepted(task_id) @app.route("/api/pdf-to-word", methods=["POST"]) def api_pdf_to_word(): payment_err = verify_payment_before_processing("pdf-to-word") if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 data = f.read() filename = secure_filename(f.filename) or "doc.pdf" task_id = secrets.token_hex(8) base = os.path.splitext(filename)[0] def process(**kw): return pdf_to_word_process(data, filename, **kw) _start_task( task_id, process, fname=f"{base}.docx", mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document", ) return _accepted(task_id) @app.route("/api/pdf-delete-pages", methods=["POST"]) def api_pdf_delete_pages(): payment_err = verify_payment_before_processing("pdf-delete-pages") if payment_err: return payment_err f = request.files.get("file") if not f or not f.filename: return "No file uploaded", 400 pages_to_delete = request.form.get("pages", "").strip() if not pages_to_delete: return "Please specify pages to delete (e.g., '1,3-5,7')", 400 data = f.read() filename = secure_filename(f.filename) or "doc.pdf" task_id = secrets.token_hex(8) base = os.path.splitext(filename)[0] def process(**kw): return pdf_delete_pages_process(data, pages_to_delete, **kw) _start_task( task_id, process, fname=f"{base}_cleaned.pdf", mime="application/pdf", ) return _accepted(task_id) def _accepted(task_id: str): resp = Response(status=202) resp.headers["X-Task-Id"] = task_id return resp @app.route("/download/") def download_result(task_id): entry = _pending.pop(task_id, None) if entry is None: return "Task not found", 404 if entry["error"]: return entry["error"], 400 if entry["result"] is None: return "Result not ready", 202 return send_file( BytesIO(entry["result"]), mimetype=entry["mime"], as_attachment=True, download_name=entry["fname"], ) # ── Bank statements ─────────────────────────────────────────────────────── BANK_STATEMENT_TEMPLATE = """ Bank Statement Parser | SCANLY
SCANLY
Home Pricing Services
Bank Parsers

Select the bank and upload the statement PDF.

Each sub-card runs its own parser. Supported options include Axis, SBI, ICICI, Kotak, HDFC, IDFC, IndusInd, Union Bank, PNB, IDBI, Credit Card Parser, and Other Bank (AI Parser).

🔒 We don't save your data. All processing is done in-memory and immediately deleted.

Starting…

{% for slug, name in banks.items() %}
{% if slug == 'other' %}🤖 AI{% elif slug == 'creditcard' %}💳 CC{% elif slug == 'amex' %}💳 AMEX{% else %}{{ slug[:3].upper() }}{% endif %}

{{ name }}

{% if slug == 'other' %}Upload any bank PDF statement. Uses AI API parsing directly with zero hardcoded heuristics.{% elif slug == 'amex' %}Upload American Express (Amex) Credit Card PDF statement. Extracts Spends (Debits), Payments/Credits, Membership Fees, and Installments with running balances.{% elif slug == 'creditcard' %}Upload any Credit Card PDF statement (SBI Card, HDFC, ICICI, Axis, Amex, RBL). Parses Spends (Dr) and Refunds/Payments (Cr) into Excel.{% else %}Upload this bank's PDF statement and download parsed transactions as Excel.{% endif %}

{% endfor %}
""" @app.route("/bank-statements", methods=["GET"]) def bank_statements(): return render_template_string(BANK_STATEMENT_TEMPLATE, banks=STAT_BANK_NAMES, message=None) @app.route("/api/clear-cache", methods=["GET", "POST"]) def api_clear_cache(): from statement_parser import clear_all_caches clear_all_caches() return jsonify({"status": "success", "message": "All statement parser caches cleared!"}) @app.route("/api/bank-statements/", methods=["POST"]) def api_bank_statement_parser(bank_slug): if bank_slug not in STAT_BANK_NAMES: return "Unknown bank", 404 payment_err = verify_payment_before_processing(f"bank-{bank_slug}") if payment_err: return payment_err uploads = request.files.getlist("file") valid_files = [up for up in uploads if up and up.filename and up.filename.lower().endswith(".pdf") and not os.path.basename(up.filename).startswith(".")] if not valid_files: return "No PDF files uploaded", 400 page_numbers = request.form.get("pages", "").strip() pdf_password = request.form.get("pdf_password", "").strip() task_id = secrets.token_hex(8) if len(valid_files) == 1: upload = valid_files[0] data = upload.read() base_name = os.path.splitext(secure_filename(os.path.basename(upload.filename)))[0] or bank_slug def process(**kw): return bank_statement_process(data, bank_slug, page_numbers, password=pdf_password, **kw) out_fname = f"{base_name}_{bank_slug}_parsed.xlsx" else: valid_files.sort(key=lambda f: os.path.basename(f.filename).lower()) file_inputs = [(os.path.basename(f.filename), f.read()) for f in valid_files] def process(**kw): return bank_statement_process(file_inputs, bank_slug, page_numbers, password=pdf_password, **kw) out_fname = f"{bank_slug}_batch_{len(valid_files)}_statements_parsed.xlsx" _start_task( task_id, process, fname=out_fname, mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet", ) return _accepted(task_id) if __name__ == "__main__": host = os.getenv("HOST", "0.0.0.0") port = int(os.getenv("PORT", 7860)) app.run(host=host, port=port, debug=False)