Flake8 output was giving me a headache, so i wrote this quick parser. It will convert a `--file` of flake8 text to a "checklist" format , json or pretty-printed json.

parse_flake8.py

import argparse
import os.path
import pprint

"""
flake8 parser.
this is a quick and dirty tool.

Flake8 output was giving me a headache, so i wrote this quick parser. It will convert a `--file` of flake8 text to a "checklist" format , json or pretty-printed json.

This will strip a given prefix from the data, and return it in less of a 'WALL OF TEXT'.

I didn't build in support for piping in data directly from STDIN, but the (simple) routines can be imported into a project.


json-pretty

    {'foo.py': ['133:76: W291 trailing whitespace'],
     'lib/data/bar.py': ['52:1: W293 blank line contains whitespace'],
     'lib/data/foo.py': ['101:1: W293 blank line contains whitespace'],
     'lib/utils/bar.py': ['605:1: W293 blank line contains whitespace'],
     'lib/utils/date.py': ["328:5: F811 redefinition of unused 'currentYear' from line 306",
                           "331:5: F811 redefinition of unused 'currentMonth' from line 310",
                           "334:5: F811 redefinition of unused 'currentDay' from line 314"],
     'lib/utils/foo.py': ['595:1: W293 blank line contains whitespace']}


checklist

    foo.py
        133:76: W291 trailing whitespace
    lib/data/bar.py
        52:1: W293 blank line contains whitespace
    lib/data/foo.py
        101:1: W293 blank line contains whitespace
    lib/utils/bar.py
        605:1: W293 blank line contains whitespace
    lib/utils/date.py
        328:5: F811 redefinition of unused 'currentYear' from line 306
        331:5: F811 redefinition of unused 'currentMonth' from line 310
        334:5: F811 redefinition of unused 'currentDay' from line 314
    lib/utils/foo.py
        595:1: W293 blank line contains whitespace

"""

_sample_data = """/Users/me/path/to/project/files/lib/data/bar.py:52:1: W293 blank line contains whitespace
/Users/me/path/to/project/files/lib/data/foo.py:101:1: W293 blank line contains whitespace
/Users/me/path/to/project/files/lib/utils/date.py:328:5: F811 redefinition of unused 'currentYear' from line 306
/Users/me/path/to/project/files/lib/utils/date.py:331:5: F811 redefinition of unused 'currentMonth' from line 310
/Users/me/path/to/project/files/lib/utils/date.py:334:5: F811 redefinition of unused 'currentDay' from line 314
/Users/me/path/to/project/files/lib/utils/foo.py:595:1: W293 blank line contains whitespace
/Users/me/path/to/project/files/lib/utils/bar.py:605:1: W293 blank line contains whitespace
/Users/me/path/to/project/files/foo.py:133:76: W291 trailing whitespace"""

def flake8_to_json(data, prefix=None):
    output = {}
    prefix_length = None if prefix is None else len(prefix)
    data = [l.strip() for l in data.split('\n')]
    for line in data:
        _newline = False
        (_path, _row, _col, _code_text) = line.split(':')
        if prefix is not None:
            if _path[:prefix_length] == prefix:
                _path = _path[prefix_length:]
        if _path not in output:
            output[_path] = []
        output[_path].append(':'.join([_row, _col, _code_text]))
    return output
    

def to_checklist(output):
    ks = output.keys()
    ks.sort()
    as_checklist = []
    for k in ks:
        as_checklist.append(k)
        as_checklist.extend(['\t%s' % i for i in output[k]])
    return '\n'.join(as_checklist)    


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    
    parser.add_argument("--file",
                        help="Which file?",
                        )
    parser.add_argument("--mode",
                        help="how should the output be formatted? defaults to `checklist`, but `json` and `json-pretty` are valid",
                        choices=['json', 'json-pretty', 'checklist'],
                        default='checklist',
                        )
    parser.add_argument("--prefix",
                        type=str,
                        help="Use a prefix?",
                        )
    parser.add_argument("--demo",
                        help="Run in demo mode?",
                        action='store_true',
                        )

    options = parser.parse_args()

    if options.demo:
        data = _sample_data
        output = flake8_to_json(data, prefix='/Users/me/path/to/project/files/')
        as_checklist = to_checklist(output)
        print "#" * 20
        print "flake8_to_json -----"
        pprint.pprint(output)

        print "#" * 20
        print "as_checklist -----"
        print as_checklist

        exit()

    if not options.file:
        print "Outside of `--demo` mode, you must supply a `--file`"
        exit()
    
    if not os.path.exists(options.file):
        raise ValueError("`%s` is not a valid file." % options.file)
        
    data = open(options.file, 'r').read()
    output = flake8_to_json(data, prefix=options.prefix)

    if options.mode == 'json-pretty':
        pprint.pprint(output)
    elif options.mode == 'json':
        print output
    elif options.mode == 'checklist':
        output = to_checklist(output)
        print output
    
    
"""
http://pep8.readthedocs.org/en/latest/intro.html#example-usage-and-output    

    $ pep8 testsuite/E40.py --format=default
    testsuite/E40.py:2:10: E401 multiple imports on one line

    $ pep8 testsuite/E40.py --format=pylint
    testsuite/E40.py:2: [E401] multiple imports on one line

    $ pep8 testsuite/E40.py --format='%(path)s\n\y%(row)d|%(col)d| %(code)s %(text)s'
    testsuite/E40.py|2|10| E401 multiple imports on one line

    Variables in the custom format option

    Variable	Significance
    path	File name
    row	Row number
    col	Column number
    code	Error code
    text	Error text    

"""