#! /usr/bin/env python

# Copyright (c) 2014, U Chun Lao All rights reserved.

# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met:

# 1. Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.

# 2. Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.

# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.



# extract the best lattice path for each sentence based on the score

import sys, numpy, argparse

parser = argparse.ArgumentParser()

parser.add_argument('-latt', '-l', default=False, required=True, help='lattice paths file')
parser.add_argument('-group', '-g', default=False, required=True, \
	help='file that divides the lattice file into sentence groups')
parser.add_argument('-score', '-s', default=False, required=True, \
	help='file that contains scores for each lattice path in LATT')
parser.add_argument('-output', '-o', default=False, \
	help='file that the best lattice paths will be written to')

args=parser.parse_args(sys.argv[1:])

# lattice paths file
lattF = args.latt

# file that divides the lattice file into sentence groups
groupF = args.group

# file that the best lattice paths will be written to
bestF = output

# file that contains scores for each lattice path in lattF
scoreF = score


# read lattices, groupings and scores

lattCount = [int(l.split('\t')[2]) for l in open(groupF, 'r')]

print 'read in %d groups' % len(lattCount)

scores = []

for l in open(scoreF, 'r'):
	if '(AD3)' in l:
		scores.append(float(l.split()[-1]))

print 'read in %d scores' % len(scores)

latt = []
buff = []
gbuff = []
total = 0
scid = 0

for l in open(lattF, 'r'):
	if len(l.strip()) > 0:
		buff.append(l.replace('\n', ''))
	else:
		gbuff.append((scores[scid], buff))
		buff = []
		scid += 1
		total += 1
		if len(gbuff) >= lattCount[len(latt)]:
			latt.append(gbuff)
			gbuff = []

print 'read in %d sentences in %d groups' % (total, len(latt))


# process scores
bestPath = []
goldBest = 0

for l in latt:
	best = l[0]
	gold = True
	for cand in l[1:]:
		if cand[0] > best[0]:
			best = cand
			gold = False
	bestPath.append(best)
	if gold:
		goldBest += 1

print '%d gold standards are rated as the best' % (goldBest)

if bestF:
	with open(bestF, 'w') as f:
		for b in bestPath:
			for l in b[1]:
				f.write('%s\n' % l)
			f.write('\n')