Convert all MDX to MD in the doc folder of a HF repo

convert_to_md.py

import argparse
import re
import shutil
from pathlib import Path
from tqdm import tqdm


DISCLAIMER = """⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer."""


def treat_doc_file(doc_file):
    doc_file = str(doc_file)
    if not doc_file.endswith(".mdx"):
        return
    new_path = doc_file.replace(".mdx", ".md")
    shutil.move(doc_file, new_path)
    
    with open(new_path, "r", encoding="utf-8") as f:
        content = f.read()
    if re.search("^\s*<!--", content) is not None:
        # Add disclaimer to first comment
        lines = content.split("\n")
        idx = 0
        while not lines[idx].rstrip().endswith("-->"):
            idx += 1
        
        lines = lines[:idx] + ["", DISCLAIMER] + lines[idx:]
        content = "\n".join(lines)
    else:
        content = f"<!--{DISCLAIMER}\n-->\n\n" + content
        
    with open(new_path, "w", encoding="utf-8") as f:
        f.write(content)


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--doc_folder", type=str, help="Where the doc folder is.")
    args = parser.parse_args()

    all_doc_files = list(Path(args.doc_folder).glob("**/*.mdx"))
    for doc_file in tqdm(all_doc_files):
        treat_doc_file(doc_file)