import os

import re

import streamlit as st

import pytesseract

from PIL import Image

from PyPDF2 import PdfReader

import fitz # PyMuPDF

# Setup upload folder

UPLOAD_FOLDER = 'uploads/'

os.makedirs(UPLOAD_FOLDER, exist_ok=True)

# Pytesseract setup

# Check if Tesseract is available

import shutil

TESSERACT_AVAILABLE = shutil.which('tesseract') is not None

if not TESSERACT_AVAILABLE:

# Try common installation paths

common_paths = [

r'C:\Program Files\Tesseract-OCR\tesseract.exe',

r'C:\Program Files (x86)\Tesseract-OCR\tesseract.exe'

]

for path in common_paths:

if os.path.exists(path):

pytesseract.pytesseract.tesseract_cmd = path

TESSERACT_AVAILABLE = True

break

# PII Regex Patterns

aadhaar_pattern = re.compile(r'\b\d{4}\s?\d{4}\s?\d{4}\b')

pan_pattern = re.compile(r'\b[A-Z]{5}[0-9]{4}[A-Z]{1}\b')

dl_pattern = re.compile(r'\b[0-9]{2}[A-Z]{2}[0-9]{2,7}\b')

voter_id_pattern = re.compile(r'\b[A-Z]{3}[0-9]{7}\b')

# Function to extract text from files

def extract_text(file_path):

text = ''

if file_path.lower().endswith('.pdf'):

reader = PdfReader(file_path)

text = "".join([page.extract_text() for page in reader.pages if page.extract_text()])

elif file_path.lower().endswith(('.png', '.jpg', '.jpeg')):

if TESSERACT_AVAILABLE:

text = pytesseract.image_to_string(Image.open(file_path), config='--psm 6')

else:

text = "OCR not available - Tesseract not installed"

elif file_path.lower().endswith('.txt'):

with open(file_path, 'r', encoding='utf-8') as file:

text = file.read()

return text

# Function to detect and mask PII

def detect_and_mask_pii(text):

masked_pii = {

'aadhaar': [re.sub(r'\S', 'X', match) for match in aadhaar_pattern.findall(text)],

'pan': [re.sub(r'\S', 'X', match) for match in pan_pattern.findall(text)],

'driving_license': [re.sub(r'\S', 'X', match) for match in dl_pattern.findall(text)],

'voter_id': [re.sub(r'\S', 'X', match) for match in voter_id_pattern.findall(text)]

}

pii_found = {

'aadhaar': aadhaar_pattern.findall(text),

'pan': pan_pattern.findall(text),

'driving_license': dl_pattern.findall(text),

'voter_id': voter_id_pattern.findall(text)

}

return pii_found, masked_pii

# Function to redact PII in PDF

def redact_pii_in_pdf(pdf_path, pii_data):

doc = fitz.open(pdf_path)

redacted = False

for page in doc:

for pii_list in pii_data.values():

for pii in pii_list:

text_instances = page.search_for(pii)

if text_instances:

redacted = True

for inst in text_instances:

page.add_redact_annot(inst, fill=(0, 0, 0))

if redacted:

page.apply_redactions()

modified_pdf_path = os.path.join(UPLOAD_FOLDER, 'redacted_' + os.path.basename(pdf_path))

doc.save(modified_pdf_path, deflate=True)

doc.close()

return modified_pdf_path if redacted else None

# Streamlit UI

st.title("PII Detection & Redaction App")

if not TESSERACT_AVAILABLE:

st.warning("⚠️ Tesseract OCR is not installed. Image files (PNG, JPG, JPEG) cannot be processed. Only PDF and TXT files will work.")

st.info("To process image files, please install Tesseract OCR from: https://github.com/tesseract-ocr/tesseract")

file = st.file_uploader("Upload a document (PDF, PNG, JPG, TXT)", type=["pdf", "png", "jpg", "jpeg", "txt"])

if file:

file_path = os.path.join(UPLOAD_FOLDER, file.name)

with open(file_path, "wb") as f:

f.write(file.getbuffer())

st.success("File uploaded successfully!")

text = extract_text(file_path)

pii_results, masked_pii = detect_and_mask_pii(text)

if any(pii_results.values()):

st.subheader("Detected PII")

st.json(pii_results)

st.subheader("Masked PII")

st.json(masked_pii)

if file.name.endswith('.pdf'):

modified_pdf_path = redact_pii_in_pdf(file_path, pii_results)

if modified_pdf_path:

with open(modified_pdf_path, "rb") as f:

st.download_button("Download Redacted PDF", f, file_name="redacted_pii.pdf")

else:

st.warning("No PII found for redaction in the PDF.")

else:

st.warning("No PII detected in the document.")