Literate Markdown comments in C

litc.py

#!/usr/bin/env python3

import re
import subprocess
import sys

pattern = r'(?s)("[^"\\]*(?:\\.[^"\\]*)*")|/\*(.*?)\*/|([^"/]+)|(.)'
token = re.compile(pattern)

def put(text):
    print(text, end="")

def escape(text):
    text = text.replace("&", "&")
    text = text.replace("<", "&lt;")
    return text.replace(">", "&gt;")

def main():
    put('<meta charset="utf-8">\n')
    put('<title>%s</title>\n' % escape(sys.argv[1]))
    preformatted = False
    with open(sys.argv[1], encoding="utf-8") as f:
        text = f.read()
    for string, comment, code1, code2 in token.findall(text):
        code = string or code1 or code2
        if code:
            if preformatted is False:
                put("<pre>\n")
                preformatted = True
                code = code.lstrip()
            put(escape(code))
            continue
        comment = (comment.strip() + "\n").encode("utf-8")
        if preformatted is True:
            put("</pre>\n")
            preformatted = False
        try:
            command = ["cmark", "/dev/stdin"]
            html = subprocess.check_output(command, input=comment)
        except subprocess.CalledProcessError as err:
            put("<p>Markdown Error: %s</p>\n" % escape(str(err)))
        else:
            put(str(html, "utf-8"))
    if preformatted is True:
        put("</pre>\n")

if __name__ == "__main__":
    main()