toolbox/to_markdown/pptx_to_markdown.py

#!/usr/bin/python3
# -*- coding: utf-8 -*-
import argparse
import os
from pathlib import Path
import shutil
import tempfile
import uuid

from project_settings import project_path
from toolbox.to_markdown.base_to_markdown import BaseToMarkdown
from toolbox.os.command import Command


def get_args():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--filename",
        default=(project_path / "data/files/ppt/AI能力分享.pptx").as_posix(),
        # default=(project_path / "data/files/ppt/钉钉基础使用手册.pptx").as_posix(),
        type=str
    )
    args = parser.parse_args()
    return args


@BaseToMarkdown.register("pptx2md")
class Pptx2md(BaseToMarkdown):
    """
    https://github.com/ptsefton/pptx_to_md
    """
    def __init__(self, filename: str):
        super().__init__(filename)

    def command(self, filename: str, output_file: str, image_dir: str = "media"):
        cmd = f'python -m pptx2md -o "{output_file}" -i "{image_dir}" "{filename}"'
        Command.popen(cmd)
        return cmd

    def save_to_zip(self, output_dir: str):
        basename = str(uuid.uuid4())

        temp_dir = Path(tempfile.gettempdir()) / basename
        temp_dir.mkdir(parents=True, exist_ok=False)

        md_file = temp_dir / f"{basename}.md"

        self.command(
            self.filename,
            output_file=md_file.as_posix(),
            image_dir=(temp_dir / "media").as_posix()
        )

        # zip
        output_zip_file = os.path.join(output_dir, f"{basename}.zip")
        self.zip_directory(temp_dir, output_zip_file)
        shutil.rmtree(temp_dir)
        return output_zip_file


def main():
    args = get_args()

    p2m = Pptx2md(args.filename)

    output_zip_file = p2m.save_to_zip(output_dir=".")
    print(output_zip_file)
    return


if __name__ == "__main__":
    main()