Count lines in python functions

pyfunclen.py

#!/usr/bin/env python
"""
Count lines in python functions
"""
import re


def funcs_lengths(lines):
    """
    Loop through lines and yield (function name, line length)

    Nested funcs are ignored.

    Does not count decorator lines.

    Because I try to ignore lines comprising only triple-quoted strings,
    and I'm not actually parsing them, I can be fooled by unexpected
    usage of triple quotes. Typical usage should be fine.

    Example invocation:

    find . -type f -name '*.py' \
        | egrep -v 'virtualenv|node_modules|/tests/' \
        | xargs pyfunclen \
        | egrep -v '\b(routes|upgrade|downgrade|api_routes_v2)\b' \
        | sort -n | vim -

    Example of finding average function length:

    find . -type f -name '*.py' \
        | egrep -v 'virtualenv|node_modules|/tests/' \
        | xargs pyfunclen \
        | egrep -v '\b(routes|upgrade|downgrade|api_routes_v2)\b' \
        | awk '{n++;sum+=$1} END {print n ? sum/n : 0}'
    """
    start_re = re.compile(r'(?P<leadingws>\s*)(async)?\s*def\s+(?P<funcname>[^\s(]+)')
    doc_1_line_re = re.compile(r'^""".*"""$')
    close_bracket_re = re.compile(r"""^[)\]}"']$""")
    length = 0
    current_func = None
    current_ws = None
    in_comment = False
    for line in lines:
        stripped = line.strip()
        if stripped == '' and not lines.isfirstline():
            continue
        if stripped.startswith('#'):
            continue
        if doc_1_line_re.match(stripped):
            continue
        if in_comment:
            if '"""' in stripped:
                in_comment = False
            if stripped == '"""':
                in_comment = False
            continue
        elif '"""' in stripped:
            in_comment = True
            continue
        if close_bracket_re.match(stripped):
            continue
        if current_func:
            # look for end of func -- a line with the same leading WS
            m = re.match('^{}\S'.format(current_ws), line)
            if m or lines.isfirstline():
                yield length, current_func
                current_func = None
            else:
                length += 1
        if not current_func:
            m = start_re.match(line)
            if m:
                current_func = m.group('funcname')
                current_ws = m.group('leadingws')
                length = 1


if __name__ == '__main__':
    def main():
        import fileinput
        for length, funcname in funcs_lengths(fileinput.input()):
            print length, funcname
    main()