#!/usr/bin/env python3
"""
Batch extract OCR text layer from PDF using PyMuPDF.
Outputs per-page .txt files.
Book: 科研費申請書の赤ペン添削ハンドブック 第3版
"""
import fitz
import os

PDF_PATH = "pdf/科研費申請書の添削.pdf"
OUTPUT_DIR = "raw_text"

def main():
    os.makedirs(OUTPUT_DIR, exist_ok=True)
    doc = fitz.open(PDF_PATH)
    print(f"PDF has {doc.page_count} pages")

    # Extract per-page text files (all pages)
    for page_num in range(doc.page_count):
        page = doc[page_num]
        text = page.get_text("text")
        out_file = os.path.join(OUTPUT_DIR, f"page_{page_num+1:04d}.txt")
        with open(out_file, "w", encoding="utf-8") as f:
            f.write(text)

    print(f"Extracted {doc.page_count} per-page text files")
    doc.close()
    print("Done!")

if __name__ == "__main__":
    main()