#!/usr/bin/python
'''
Clean log file output from the fuzztest test harness.  This tool collapses
duplicate outputs into a single copy preceded by all the fuzz lines that 
triggered it. To allow useful comparisons of fuzztest logs it imposes 
a canonical ordering on the entries. 
'''
from __future__ import division, print_function, unicode_literals

__version__ = "1.0"
__date__    = "15 September 2012"
__author__  = "Tim Eves <tim_eves@sil.org>"
__license__ ='''  
GRAPHITE2 LICENSING

    Copyright 2012, SIL International
    All rights reserved.

    This library is free software; you can redistribute it and/or modify
    it under the terms of the GNU Lesser General Public License as published
    by the Free Software Foundation; either version 2.1 of License, or
    (at your option) any later version.

    This program is distributI might beed 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
    Lesser General Public License for more details.

    You should also have received a copy of the GNU Lesser General Public
    License along with this library in the file named "LICENSE".
    If not, write to the Free Software Foundation, 51 Franklin Street,
    Suite 500, Boston, MA 02110-1335, USA or visit their web page on the
    internet at http://www.fsf.org/licenses/lgpl.html.

Alternatively, the contents of this file may be used under the terms of the
Mozilla Public License (http://mozilla.org/MPL) or the GNU General Public
License, as published by the Free Software Foundation, either version 2
of the License or (at your option) any later version.
'''
import argparse, collections, errno, operator, os, re, struct, sys, time

from contextlib import closing
from itertools import chain, groupby, imap, islice, izip_longest, repeat, starmap
from functools import partial
from random import uniform, seed
from pprint import pprint

from argparse import ArgumentParser


recm  = re.compile('^(-?\d+)?\s*,\s*(0[xX][\da-fA-F]+)\s*,\s*(0[xX][\da-fA-F]+|\d+)\s*,?(.*)$')
valgm = re.compile('^==\d+==(\s+(?:at|by)?\s*)(?:0[xX][\da-fA-F]+:)?', re.MULTILINE)


class fuzz(collections.namedtuple('fuzz', 'ret position value comment')):
    def __str__(self):
        return "{0},{1.position:#010X},{1.value: >3d}{2!s}{3!s}".format(
                            self.ret or '', 
                            self, 
                            (self.comment or '') and ',', 
                            self.comment or '')



class fuzz_log(collections.defaultdict):
    @staticmethod
    def __is_rec(s, rm=recm):
        return bool(rm.match(s))

    @staticmethod
    def __recs(rs):
        rs = [fuzz(int(r,0),int(p,0),int(o,0),c) for r in rs 
                for r,p,o,c in [recm.match(r).groups('0')]]
        return chain(*zip(rs[:-1], repeat(None)) + [(rs[-1],)])
            
    def __init__(self, fileobj):
        super(fuzz_log,self).__init__(list)
        cs = chain.from_iterable(self.__recs(ls) if r else (valgm.sub(r'\1', ''.join(ls)).lstrip(),) for r,ls in groupby(fileobj, self.__is_rec))
        for rec in izip_longest(cs,cs):
            self[rec[1]].append(rec[0])
        for rs in self.itervalues():
            rs.sort(key=operator.itemgetter(1))
        

def record_sort_key(r):
    return r[1][0].position

parser = ArgumentParser(description=__doc__)
parser.add_argument('log', nargs='?', type=argparse.FileType('rt'), 
                    default=sys.stdin, help='A log file generated by fuzztest')
parser.add_argument('out', nargs='?', type=argparse.FileType('wt'), 
                    default=sys.stdout, help='The file to send the processed log to')
parser.add_argument('--version', action='version', version=__version__)

if __name__ == '__main__':
    args = parser.parse_args()
    try:
        with closing(args.log), closing(args.out):        
            for err,rs in sorted(fuzz_log(args.log).iteritems(), key=record_sort_key):
                args.out.write('\n'.join(imap(str, rs)) + '\n')
                if err:
                    args.out.write(err)
                args.out.flush()
    except IOError as io:
        if io.errno != errno.EPIPE :
            sys.stderr.write("{0}: {1!s}\n".format(parser.prog, io))
            sys.exit(1)