ocr_utils.py

import re
import pdf2image
import pytesseract
from pathlib import Path
import os

# pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"

def extract_pdf_text(pdf_file):
    if pdf_file is None:    
        return None
    
    try:
        if isinstance(pdf_file, (str, Path)):
            pdf_bytes = Path(pdf_file).read_bytes()
        elif hasattr(pdf_file, "read"):
            pdf_bytes = pdf_file.read()
        else:
            return None

        images = pdf2image.convert_from_bytes(pdf_bytes)
        all_text = [pytesseract.image_to_string(img) for img in images]
        combined_text = "\n".join(all_text)
        cleaned = re.sub(r'\s+', ' ', combined_text)
        cleaned = re.sub(r'[^\w\s.,&%]', '', cleaned)
        return cleaned.strip()
    
    except Exception as e:
         print(f"❌ Error during OCR: {e}")
         return None