86 lines
2.7 KiB
Python
86 lines
2.7 KiB
Python
import re
|
|
import argparse
|
|
import fitz
|
|
import pytesseract
|
|
import cv2
|
|
import numpy as np
|
|
|
|
def redact_text(text, validators, ignored):
|
|
redacted_text = text
|
|
|
|
# Apply validators to redact sensitive information
|
|
for validator_type, patterns in validators.items():
|
|
for pattern, replacement in patterns:
|
|
redacted_text = re.sub(pattern, replacement, redacted_text, flags=re.IGNORECASE)
|
|
|
|
# Apply ignored patterns to exclude certain patterns from redaction
|
|
for pattern in ignored:
|
|
redacted_text = re.sub(pattern, '', redacted_text, flags=re.IGNORECASE)
|
|
|
|
return redacted_text
|
|
|
|
|
|
def process_pdf(input_file, output_file, validators, ignored):
|
|
doc = fitz.open(input_file)
|
|
|
|
for page_index, page in enumerate(doc):
|
|
pixmap = page.get_pixmap()
|
|
|
|
# Convert the pixmap to a NumPy array
|
|
image_np = np.frombuffer(pixmap.samples, dtype=np.uint8).reshape(pixmap.h, pixmap.w, pixmap.n)
|
|
|
|
# Convert the image from BGR to RGB format
|
|
image_rgb = cv2.cvtColor(image_np, cv2.COLOR_BGR2RGB)
|
|
|
|
# Use Tesseract OCR to extract text from the image
|
|
extracted_text = pytesseract.image_to_string(image_rgb)
|
|
|
|
# Apply redaction on the extracted text
|
|
redacted_text = redact_text(extracted_text, validators, ignored)
|
|
|
|
# Print the extracted and redacted text for verification
|
|
print(f"Page {page_index + 1} - Extracted Text:")
|
|
print(extracted_text)
|
|
print(f"Page {page_index + 1} - Redacted Text:")
|
|
print(redacted_text)
|
|
print("-----")
|
|
|
|
doc.save(output_file)
|
|
doc.close()
|
|
|
|
print("Redacted PDF saved successfully!")
|
|
|
|
# Define the validators and ignored patterns
|
|
validators = {
|
|
|
|
"Hong Kong Identity Card Number": [
|
|
(r'\b[A-Z]\d{6}\([A-Z]\)', 'XXXXXXXXX'),
|
|
(r'\b[A-Z]\d{6}\(\d\)', 'XXXXXXXXX')
|
|
],
|
|
"Phone Numbers": [
|
|
(r'\+\d{1,3}[-\s]?\d{1,4}[-\s]?\d{1,4}[-\s]?\d{1,9}', '+XXX XXXX XXXX'),
|
|
(r'\d{1,4}[-\s]?\d{1,4}[-\s]?\d{1,9}', 'XXXX XXXX')
|
|
],
|
|
"Credit Card Numbers": [
|
|
(r'\b[0-9]{4}-[0-9]{4}-[0-9]{4}-[0-9]{4}\b', 'XXXX-XXXX-XXXX-XXXX')
|
|
],
|
|
"Email Addresses": [
|
|
(r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]{2,}\b', 'XXXXX@XXXXX.XXX')
|
|
],
|
|
|
|
}
|
|
|
|
ignored = [
|
|
r'0000[-\s]?0000[-\s]?0000[-\s]?0000',
|
|
r'1111[-\s]?1111[-\s]?1111[-\s]?1111'
|
|
]
|
|
|
|
# Parse command-line arguments
|
|
parser = argparse.ArgumentParser(description='PDF Redaction Script')
|
|
parser.add_argument('input_file', help='Path to the input PDF file')
|
|
parser.add_argument('output_file', help='Path to the output PDF file')
|
|
args = parser.parse_args()
|
|
|
|
# Call the process_pdf function with input and output file paths, validators, and ignored patterns
|
|
process_pdf(args.input_file, args.output_file, validators, ignored)
|