# -*- coding: utf-8 -*-
import MeCab
import sys
import operator
import pyexcel as pe

test = open('/home/sean/Bureau/etudies programa/pythonjp/mishima', 'r').read()


t = MeCab.Tagger (" ".join(sys.argv))
#lista com as frases
sentences = test.split('。')

t.parse(test)

dicio = {}
m = t.parseToNode(test)


    
pyjp_current =  0 #contadordasfrases
def queisso(meca, dados):
    #separa as caracteristicas da palavra numa lista 
    #Original Form\t, Part of Speech, Part of Speech section 1, Part of Speech section 2, Part of Speech section 3, Conjugated form, Inflection, Reading, Pronounciation
    carac = meca.feature.split(',')
    #termo na forma de dicionário (conjugated form)
    termo = carac[6]
    
    #tirar ponto, vírgula, (binding particle 係助詞), (case making particle), ( n sei oq é 同じ 大きな *), (partícula conjuntiva 接続助詞), (paralel marker 　と　や　並立助詞), (fecha parenteses 括弧閉), (abre parenteses 括弧開)
    if carac[1] in ('句点', '読点', '係助詞', '格助詞', '連体化', '*', '接続助詞', '並立助詞', '括弧閉', '括弧開'):
        if carac[1] == '句点':
            global pyjp_current
            pyjp_current = pyjp_current + 1            
        else:
            return 1
    elif termo not in dados:
        dados[termo] = [1, sentences[pyjp_current].replace('\n', '').replace("\u3000", '')]
        #if carac[1] == '':
        #        print (termo)
    else:
        dados[termo][0]= dados[termo][0] + 1
        dados[termo].append(sentences[pyjp_current].replace('\n', '').replace("\u3000", ''))

def arrumaesalva(dic):
    tab = sorted(dic.items(), key=operator.itemgetter(1), reverse=True)
    print (type(tab[0]))    
    sheet = pe.Sheet(tab)
    sheet.save_as("/home/sean/Bureau/etudies programa/pythonjp/mishima.csv")    
  
while m:

    queisso(m, dicio)
    #print(m.id, " \t", m.feature)
    #print(dir(m))
    m = m.next
print(dicio)
arrumaesalva(dicio)