convert_hatena_blog_html_to_md.py

from pathlib import Path
from typing import Optional
import re
import html
import urllib.request

import markdownify


class Converter(markdownify.MarkdownConverter):
    def convert_img(self, el, text, convert_as_inline):
        src = el.attrs["src"]
        print(src)
        filename = src.rsplit("/")[-1]
        dst = Path("./public/i") / filename
        with urllib.request.urlopen(src) as webf:
            with dst.open("wb") as fp:
                fp.write(webf.read())
        el.attrs["src"] = "/i/" + filename
        return super().convert_img(el, text, convert_as_inline)


def filepath_to_html(filepath: Path) -> Optional[str]:
    with open(filepath, "r", encoding="utf-8") as fp:
        contents = fp.read()

    frontmatter, contents = contents.split("\n\n", maxsplit=1)

    if not contents.startswith("<"):
        return None

    converter = Converter(autolinks=False, heading_style="ATX", strip=["iframe"])
    return frontmatter + "\n\n" + converter.convert(contents)


# [C++](http://d.hatena.ne.jp/keyword/C%2B%2B)
def cleanup(s: str) -> str:
    return re.sub(r"\[([^\]]+)\]\(http://d.hatena.ne.jp/keyword/[^\)]*\)", r"\1", s)


def main() -> None:
    basedir = Path("./posts/")

    for filepath in basedir.glob("*.md"):
        s = filepath_to_html(filepath)
        if not s:
            continue
        s = html.unescape(cleanup(s))
        filepath.open("w", encoding="utf-8").write(s)


if __name__ == "__main__":
    main()