scraper/code/_3picture.py

from __future__ import annotations

import os
from concurrent.futures import ThreadPoolExecutor

import pillow_heif
from pdf2image import convert_from_path

INPUT_DIR = "downloads"
OUTPUT_DIR = "images"
PDF_DPI = 200
HEIF_QUALITY = 90
MAX_WORKERS = 10

pillow_heif.register_heif_opener()


def pdf_to_heif(pdf_path: str, output_folder: str) -> None:
    os.makedirs(output_folder, exist_ok=True)
    print(f"Converting PDF: {pdf_path}")

    pages = convert_from_path(pdf_path, dpi=PDF_DPI)
    pdf_name = os.path.splitext(os.path.basename(pdf_path))[0]

    for page_index, page in enumerate(pages, start=1):
        filename = f"{pdf_name}_page_{page_index}.heif"
        filepath = os.path.join(output_folder, filename)
        page.save(filepath, "HEIF", quality=HEIF_QUALITY)
        print(f"Saved: {filename}")

    print(f"Converted {len(pages)} pages for {pdf_name}.")


def convert_single_pdf(filename: str) -> int:
    returni = 0
    # prevents pdf from converting
    if not filename.lower().endswith(".pdf"):
        return returni

    pdf_stem = os.path.splitext(filename)[0]
    output_path = os.path.join(OUTPUT_DIR, pdf_stem)
    pdf_path = os.path.join(INPUT_DIR, filename)

    if os.path.isdir(output_path) and os.listdir(output_path):
        print(f"Skipping {filename}: output already exists.")
        return returni

    try:
        pdf_to_heif(pdf_path, output_path)
        returni = 1
    except Exception as exc:
        print(f"Failed conversion for {filename}: {exc}")

    return returni


def run() -> None:
    os.makedirs(OUTPUT_DIR, exist_ok=True)
    os.makedirs(INPUT_DIR, exist_ok=True)

    # Don't exclude already converted?
    files = sorted(os.listdir(INPUT_DIR))
    print(f"Found {len(files)} files in {INPUT_DIR}.")

    with ThreadPoolExecutor(max_workers=MAX_WORKERS) as executor:
        executor.map(convert_single_pdf, files)

    print("Conversion process completed.")


if __name__ == "__main__":
    run()