source: pycodebattler/trunk/pycodebattler/score.py @ 79

# -*- coding: utf-8 -*-
#
#  Copyright (C) 2010 by Atzm WATANABE <atzm@atzm.org>
#
#  This program is free software; you can redistribute it and/or modify it
#  under the terms of the GNU General Public License (version 2) as
#  published by the Free Software Foundation.  It is distributed in the
#  hope that it will be useful, but WITHOUT ANY WARRANTY; without even the
#  implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
#  PURPOSE.  See the GNU General Public License for more details.
#
# $Id$
#

import os
import sys
import zlib
import token
import parser

import cc
import pep8


def _compseq(seq, level):
    c = zlib.compressobj(level)
    return ''.join([c.compress(chunk) for chunk in seq] + [c.flush()])


def _maxstrlen(seq, iferr=0):
    try:
        return max(len(l) for l in seq)
    except:
        return iferr


class Scorer:
    def __init__(self, lines, compress_level=9):
        self._code = ''.join(lines)
        self._ccode = _compseq(lines, compress_level)
        self._lines = lines
        self._nlines = [l for l in lines
                        if l.strip() and not l.strip().startswith('#')]

        self._astpl = parser.suite(self.code()).totuple()
        self._token_cache = {}

        self.complexity()
        self.style_errors()

    def tupletree(self):
        return self._astpl

    def code(self):
        return self._code

    def lines(self):
        return self._lines

    def uncommented_lines(self):
        return self._nlines

    def bytes(self):
        return len(self.code()) or 1

    def uncommented_bytes(self):
        return sum(len(l) for l in self.uncommented_lines()) or 1

    def rows(self):
        return len(self.lines()) or 1

    def uncommented_rows(self):
        return len(self.uncommented_lines()) or 1

    def compressibility(self):
        return float(len(self._ccode)) / self.bytes()

    def token_bytes(self, types):
        def _bytes(stp, bytes=0):
            if not stp:
                return bytes
            if stp[0] in types:
                bytes += len(stp[1])
            else:
                for p in stp[1:]:
                    bytes = _bytes(p, bytes)
            return bytes

        types = tuple(types)

        try:
            ret = self._token_cache[types]
        except KeyError:
            ret = self._token_cache[types] = _bytes(self.tupletree())

        return ret

    def complexity(self, item='all', iferr=(256, 1)):
        try:
            stat = cc.measure_complexity(self._code, item)
            ccpp = cc.PrettyPrinter(None, True)
            cplx = zip(*ccpp.flatten_stats(stat))[-1]
            self._ccsum = sum(cplx) or 1
            self._cclen = len(cplx) or 1
        except:
            self._ccsum, self._cclen = iferr
        return self._ccsum, self._cclen

    def style_errors(self, name='', iferr=256):
        pep8.process_options([name])
        p8ck = pep8.Checker(None)
        p8ck.lines = self._lines
        p8ck.filename = name

        try:
            tmpfp = sys.stdout
            sys.stdout = open(os.devnull, 'w')
            self._estyle = p8ck.check_all() or 1
        except:
            self._estyle = iferr
        finally:
            sys.stdout = tmpfp
        return self._estyle

    def complex_score(self, coefficient=48):
        maxlen = _maxstrlen(self.lines(), 256)
        score = self._cclen * coefficient
        score /= (maxlen / (coefficient * 2)) or 1
        score /= (self._ccsum / ((coefficient / 3) or 1)) or 1
        score *= self.compressibility()
        return int(score) or 1

    def style_score(self, coefficient=64):
        data_bytes = self.token_bytes([token.NUMBER, token.STRING,
                                       token.NAME, token.COLON])
        score = self.uncommented_bytes() - data_bytes
        if score < 1:
            score = 1
        score /= (self.uncommented_rows() / coefficient) or 1
        score /= self._estyle
        score *= (coefficient / 12) or 1
        score *= self.compressibility()
        return int(score) or 1

    def lines_score(self, coefficient=2.5):
        maxlen = _maxstrlen(self.lines(), 256)
        data_bytes = self.token_bytes([token.NUMBER, token.STRING,
                                       token.NAME, token.COLON])
        score = self.uncommented_bytes() - data_bytes
        if score < 1:
            score = 1
        score /= (maxlen / self.uncommented_rows()) or 1
        score /= self.uncommented_rows()
        score *= coefficient
        score *= self.compressibility()
        return int(score) or 1

    def const_score(self, coefficient=2):
        const_bytes = self.token_bytes([token.NUMBER, token.STRING]) or 1
        score = self.uncommented_bytes() / const_bytes
        score *= coefficient
        score *= self.compressibility()
        return int(score) or 1

    def colon_score(self, coefficient=1.5):
        colon_bytes = self.token_bytes([token.COLON]) or 1
        score = self.uncommented_bytes() / colon_bytes
        score *= coefficient
        score *= self.compressibility()
        return int(score) or 1

    def name_score(self, coefficient=48):
        name_bytes = self.token_bytes([token.NAME]) or 1
        score = self.uncommented_bytes()
        score /= (name_bytes / coefficient) or 1
        score *= self.compressibility()
        return int(score) or 1