#!/usr/bin/env python
# -*- coding: utf-8 -*-
import sys
from genscheme import pattern, freq
from operator import itemgetter
from pprint import pprint

"""
Examples for introduction to transliteration code.
"""

# open the source file

def build_lexicon(a2b):
  lexicon = open(a2b).read().decode('utf-8').splitlines()
  return [line.strip() for line in lexicon]


def collect_matches(lexicon):
  matches = []
  for line in lexicon:
    if len(line.split('\t')) == 2:
      a, b = line.split('\t')
      a, b = a.lower().strip(), b.lower().strip()
      if pattern(a) == pattern(b) and a != b: 
        matches.append((a,b))
  return matches

def collect_pairs(matches):
  pairs = []
  for a,b in matches:
    for x,y in zip(a,b):
      pairs.append((x,y))
  return pairs

def display_matches(matches):
  for i, (a,b) in enumerate(sorted(matches)): 
    #if i % 5 == 0: print i, a,b
    print a,'\n', b, '\n'

def count_pairs(pairs):
  return freq(pairs).items()

def display_pairs(pair_freq):
  byfreq = sorted(count_pairs(pairs), key=itemgetter(1), reverse=True) 
  for (x,y), fq in byfreq: 
    if fq> 1: 
      print x, y, '\t', fq

if __name__ == "__main__":
  a2b = 'corpora/' + sys.argv[1] + '.txt'
  lexicon = build_lexicon(a2b)
  matches = collect_matches(lexicon)
  display_matches(matches)
  pairs = collect_pairs(matches)
  #display_pairs(count_pairs(pairs))