#!/usr/bin/python
# -*- coding: iso8859-1 -*- 

#### WAS: -*- coding: utf8 -*- 

#
# wordref_extract.py -- Simple program to extract some
#                       conjugation information out of pages like:
# http://www.wordreference.com/conj/ESverbs.asp?v=arg%FCir
#
# to be tested against the output of Finite State Tools.
#
# Usage:
# from wordref_extract import *
#
# extract('*.html') # (for example...)

# Written April 11 2008 by Antti Karttunen and placed in Public Domain.

# Following lines contains the stuff we are interested about:
#

#
# <table>
# 	<tr>
# 		<td  style="text-align:right"><strong>Infinitivo:<br>Gerundio:<br>Participio:</strong></td>
# 		<td valign="top">argüir<br><b>arguy</b><b>endo</b><br>
# 		argüido</td>
# 		<td valign="top"><a href="/definicion/argüir">Definición</a></td>
# 		<td valign="top"><a href="/es/en/translation.asp?spen=argüir">En Inglés</a><br>
# 		<a href="/esfr/argüir">En Francés</a><br>
# 		<a href="/espt/argüir">En Portugués</a></td>
# 		<td valign="top">&nbsp;</td>
# 		<td valign="top">&nbsp;</td>
# 	</tr>
# 	<tr>
# 		<th colspan="6">Indicativo</th>
# 	</tr>
# 	<tr>
# 		<td>&nbsp;</td>
# 		<td><u>presente</u></td>
# 		<td><u>imperfecto</u></td>
# 		<td><u title="pretérito indefinido">pretérito</u></td>
# 		<td><u>futuro</u></td>
# 		<td><u>condicional</u></td>
# 	</tr>
# 	<tr>
# 		<td style="white-space:nowrap;text-align:right">yo<br>tú<br>él, ella, <span title=usted>Ud.</span><br>nosotros<br>vosotros<br>ellos, ellas, <span title=ustedes>Uds.</span></td>
# 		<td><b>arguy</b>o<br><b>arguy</b>es<br><b>arguy</b>e<br>argüimos<br>argüís<br><b>arguy</b>en</td>
# 		<td>argüía<br>argüías<br>argüía<br>argüíamos<br>argüíais<br>argüían</td>
# 		<td>argüí<br>argüiste<br><b>arguy</b><b>ó</b><br>argüimos<br>argüisteis<br><b>arguy</b><b>eron</b></td>
# 		<td>argüiré<br>argüirás<br>argüirá<br>argüiremos<br>argüiréis<br>argüirán</td>
# 		<td>argüiría<br>argüirías<br>argüiría<br>argüiríamos<br>argüiríais<br>argüirían</td>
# 	</tr>
# 

#
# And print the extracted words out like this, in this order:
#
# amar+V+Ger               amando
# amar+V+Inf               amar
# amar+V+Pres+Osg+P1       amo
# amar+V+Pres+Osg+P2       amas
# amar+V+Pres+Osg+P3       ama
# amar+V+Pres+Plu+P1       amamos
# amar+V+Pres+Plu+P2       amáis
# amar+V+Pres+Plu+P3       aman
# amar+V+Pret+Osg+P1       amé
# amar+V+Pret+Osg+P2       amaste
# amar+V+Pret+Osg+P3       amó
# amar+V+Pret+Plu+P1       amamos
# amar+V+Pret+Plu+P2       amasteis
# amar+V+Pret+Plu+P3       amaron
# 
#

import os
import re


def listdir_by_pattern(path,filepattern):
  '''List all files matching with "filepattern" in the directory "path".'''
  filepattern = filepattern.replace('*','.*') + '$'
  filepat = re.compile(filepattern)

  olddir = os.getcwd()
  os.chdir(path)
  allfiles = os.listdir('.')
  files = filter(lambda(fn): filepat.match(fn),allfiles)
  os.chdir(olddir)

  return(files)



def wordref_extract_aux(filename):
  '''Opens file "filename" for reading and scans some information from it.'''
  infp = open(filename,'r')

  person_et_number = ['+Osg+P1','+Osg+P2','+Osg+P3',
                      '+Plu+P1','+Plu+P2','+Plu+P3']

  next_infinitivo_et_gerundio = False
  indicative_mood = False
  indicativo_presente_despues_n_lineas = 0
  indicativo_preterito_despues_n_lineas = 0

  anyspaceandtagspat = re.compile(r'\s*<[^>]*>\s*')
  boldtagspat = re.compile(r'</?b>')
  trigger_pat_inf_et_ger = re.compile(r'.*Infinitivo:.*Gerundio:.*')

  trigger_pat_indicativo = re.compile(r'.*Indicativo.*')
  trigger_pat_compuestos = re.compile(r'.*Tiempos compuestos.*')
  trigger_pat_subjuntivo = re.compile(r'.*Subjuntivo.*')
  trigger_pat_imperativo = re.compile(r'.*Imperativo.*')

  trigger_pat_presente = re.compile(r'.*presente.*')
  trigger_pat_imperfecto = re.compile(r'.*imperfecto.*')
  trigger_pat_preterito = re.compile(r'.*pretérito.*')


  for line in infp:
#   print "Line before sub-operation:\n" + line
    line = boldtagspat.sub('', line) # Eradicate all <b> and </b> tags at once.
#   print "Line after sub-operation:\n" + line

    if(trigger_pat_inf_et_ger.match(line)):
      next_infinitivo_et_gerundio = True
#     print "The next is it, line is now:\n" + line
    elif(next_infinitivo_et_gerundio):
      next_infinitivo_et_gerundio = False;
      words = anyspaceandtagspat.split(line)
      words = [x for x in words if x != ''] # Delete blankos.

      infinitivo = words[0];
      gerundio = words[1];

      print '%-24s %s' % ((infinitivo + "+V+Ger"),gerundio)
      print '%-24s %s' % ((infinitivo + "+V+Inf"),infinitivo)

    elif(trigger_pat_indicativo.match(line)):
      indicative_mood = True
    elif(trigger_pat_compuestos.match(line)
          or
         trigger_pat_subjuntivo.match(line)
          or
         trigger_pat_imperativo.match(line)):
      indicative_mood = False

    elif(indicative_mood):
      if(trigger_pat_presente.match(line)):
        indicativo_presente_despues_n_lineas = 9;
      if(trigger_pat_preterito.match(line)):
        indicativo_preterito_despues_n_lineas = 9;

      if(0 != indicativo_presente_despues_n_lineas):
        indicativo_presente_despues_n_lineas -= 1
        if(0 == indicativo_presente_despues_n_lineas):
          words = anyspaceandtagspat.split(line)
          words = [x for x in words if x != ''] # Delete blankos.
          print_finite_forms(infinitivo,'+Pres',person_et_number,words)

      if(0 != indicativo_preterito_despues_n_lineas):
        indicativo_preterito_despues_n_lineas -= 1
        if(0 == indicativo_preterito_despues_n_lineas):
          words = anyspaceandtagspat.split(line)
          words = [x for x in words if x != ''] # Delete blankos.
          print_finite_forms(infinitivo,'+Pret',person_et_number,words)


    else:
# Do nothing so far.
      next_infinitivo_et_gerundio = False;


def print_finite_forms(infinitivo,mood,person_et_number,forms):
  '''Print some of the finite conjugations of verb "infinitivo" in mood "mood".'''
  for i in range(len(forms)):
    word = forms[i]
    pn = person_et_number[i];
    print '%-24s %s' % ((infinitivo + "+V" + mood + pn),word)


def extract(filepat):
  '''Opens files specified with "filepat" for reading and extract some information from them.'''
  files = listdir_by_pattern('.',filepat)

  for filename in files:
    wordref_extract_aux(filename)


# This module ends here.