#!/usr/bin/env python3
"""Step 2: run Tesseract OCR (German) over a preprocessed receipt image,
writing the raw recognized text to the output directory for inspection
before it is handed to the structuring step."""
import argparse
import os

import pytesseract
from PIL import Image


def ocr(path: str, out_dir: str) -> str:
    os.makedirs(out_dir, exist_ok=True)
    img = Image.open(path)
    text = pytesseract.image_to_string(img, lang="deu", config="--psm 3")

    out_path = os.path.join(out_dir, "ocr.txt")
    with open(out_path, "w", encoding="utf-8") as f:
        f.write(text)
    return out_path


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("-in", dest="in_path", required=True)
    parser.add_argument("-out", dest="out_dir", required=True)
    args = parser.parse_args()

    out_path = ocr(args.in_path, args.out_dir)
    print(out_path)


if __name__ == "__main__":
    main()