#-*- coding:utf-8 -*-

import os,re,codecs
finput=codecs.open('./dicovalence.txt', encoding='utf-8')
input_lines = finput.readlines()
finput.close()
#input_lines = input_lines[:3000]
def extraire_verbes_avec_etre (lentries):
    pronominals =[]
    verbs_with_etre =[]
    length = len(lentries)
    n=0
    while n < length:
        line = lentries[n]
        if line == u'\n':
            n +=1
        else:
            if line[:3] == 'VAL':
                i=1
                finished = False
                while not finished:
                    if n+i < length:
                        #print n+i
                        fline = lentries[n+i]
                        #print "champ", fline
                        if fline == u'\n':
                            finished = True
                            n +=i+1
                        else:
                            l=fline.split()
                            f=l[0][:-1]
                            if f == u'AUX' and l[1] == u'être':
                                finished = True
                                n += i+1
                                verb = line[4:].split(':')[0][1:]
                                if verb[:4] == u'être':
                                    n += i+1
                                elif verb[0:2] == u's\'':
                                    pronominals.append(verb[2:])
                                elif verb[0:3] == u'se ':
                                     pronominals.append(verb[3:])
                                else:
                                    verbs_with_etre.append(verb)
                            else:
                                i +=1
                    else:
                        finished = True
                        n +=1
            else:
                n +=1
    return (pronominals, verbs_with_etre)

def extraire_verbes_passivables (lentries):
    verbes_passivables =[]
    length = len(lentries)
    n=0
    while n < length:
        line = lentries[n]
        if line == u'\n':
            n +=1
        else:
            if line[:3] == 'VAL':
                i=1
                finished = False
                while not finished:
                    if n+i < length:
                        #print n+i
                        fline = lentries[n+i]
                        #print "champ", fline
                        if fline == u'\n':
                            finished = True
                            n +=i+1
                        else:
                            l=fline.split()
                            f=l[0][:-1]
                            if f == u'RP':
                                finished = True
                                n += i+1
                                verb = line[4:].split(':')[0][1:]
                                if verb[:4] == u'être':
                                    n += i+1
                                else:
                                    verbes_passivables.append(verb)
                            else:
                                i +=1
                    else:
                        finished = True
                        n +=1
            else:
                n +=1
    return verbes_passivables

def supprimer_doublons(input_list):
    output_list =[]
    item_prec = ''
    for item in input_list:
        if item <> item_prec:
            output_list.append(item)
            item_prec= item
    return output_list

def afficher(l) :
    for e in l:
        print e
        
def stocker(output, file_path):
    foutput=codecs.open(file_path, mode='w', encoding='utf-8')
    for entry in output:
        foutput.write(entry+"\n")
    foutput.close()


(pronominals_brut, verbs_with_etre_brut) = extraire_verbes_avec_etre(input_lines)
verbes_passivables_brut = extraire_verbes_passivables(input_lines)
pronominals = supprimer_doublons(pronominals_brut)
verbs_with_etre = supprimer_doublons(verbs_with_etre_brut)
verbes_passivables = supprimer_doublons(verbes_passivables_brut)
#afficher(extraire(input_lines, u'AUX', u'être'))
stocker(pronominals, '../rewriting_rules/lexicons/pronominal_verbs.lp')
stocker(verbs_with_etre, '../rewriting_rules/lexicons/verbs_with_etre.lp')
stocker(verbes_passivables, '../rewriting_rules/lexicons/passivable_verbs.lp')