utils.py

import fitz  # PyMuPDF
import pytesseract
from PIL import Image
import io

def extract_text_from_pdf(pdf_stream: io.BytesIO) -> str:
    """
    Extracts text from a PDF file using PyMuPDF.
    Falls back to OCR if no text found on a page.
    """
    try:
        doc = fitz.open(stream=pdf_stream.read(), filetype="pdf")
        full_text = []

        for page_num, page in enumerate(doc):
            text = page.get_text().strip()
            if text:
                full_text.append(f"--- Page {page_num + 1} ---\n{text}")
            else:
                pix = page.get_pixmap(dpi=300)
                img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                ocr_text = pytesseract.image_to_string(img).strip()
                full_text.append(f"--- Page {page_num + 1} (OCR) ---\n{ocr_text}")

        return "\n\n".join(full_text)

    except Exception as e:
        return f"Error occurred while processing the PDF: {str(e)}"