Parse the Pylint message list

parse-pylint-list-msgs.py

import pathlib
import re

SUMMARY_LINE_RE = re.compile(r':(?P<name>\S+) \((?P<code>.+?)\):( \*(?P<summary>.+)\*)?')


def parse_summary_line(line):
    line_match = SUMMARY_LINE_RE.match(line)
    if not line_match:
        raise ValueError(f"ill-formed line: {line!r}")
    parts = line_match.groupdict()
    return parts['name'], parts['code'], parts.get('summary')


def filter_summary_lines(lines):
    for line in lines:
        if line.startswith(':'):
            yield line


def main(msgs_list):
    error_codes = []
    for line in filter_summary_lines(msgs_list.splitlines()):
        _, code, _ = parse_summary_line(line)
        if code.startswith('E'):
            error_codes.append(code)
    print(','.join(sorted(error_codes)))


if __name__ == '__main__':
    import sys
    if len(sys.argv) == 1:  # Pipe
        main(sys.stdin.read())
    elif len(sys.argv) == 2:  # File path
        with open(sys.argv[1], encoding='utf-8') as file:
            main(file.read())
    else:  # Beats me
        raise RuntimeError('too many CLI arguments')