convert_dico.py

   1 #!/usr/bin/python
   2 # -*- coding:utf8 -*-
   3
   4
   5 import re
   6
   7 f=open("dico_for_python.txt")
   8 t=f.read()
   9 f.close()
  10
  11 l=re.findall('"(.*)","(.*)"',t)
  12
  13 def maju(texte):
  14     texte=unicode(texte,"utf8").lower()
  15     remplacements = {
  16                 u"á":u"a",
  17                 u"à":u"a",
  18                 u"â":u"a",
  19                 u"ä":u"a",
  20                 u"é":u"e",
  21                 u"è":u"e",
  22                 u"ê":u"e",
  23                 u"ë":u"e",
  24                 u"í":u"i",
  25                 u"ì":u"i",
  26                 u"î":u"i",
  27                 u"ï":u"i",
  28                 u"ó":u"o",
  29                 u"ò":u"o",
  30                 u"ô":u"o",
  31                 u"ö":u"o",
  32                 u"ú":u"u",
  33                 u"ù":u"u",
  34                 u"û":u"u",
  35                 u"ü":u"u",
  36                 u"ý":u"y",
  37                 u"ỳ":u"y",
  38                 u"ŷ":u"y",
  39                 u"ÿ":u"y",
  40                 u"œ":u"oe",
  41                 u"æ":u"ae"
  42     }
  43     for avant,apres in remplacements.items():
  44         texte=texte.replace(avant,apres)
  45     return texte.upper()
  46
  47 l=[[maju(i[0]),i[0]+" : "+i[1].replace("\\n","; ")] for i in l
  48     if len(i[0])>=8
  49        and not " " in i[0]
  50        and not i[0][-1]=='-'] # les mots finissant par un tiret foiraient
  51                               # en plus ils sont pas intéressants, donc on vire
  52
  53 f1=open("mots.txt.out","w")
  54 f2=open("definitions.txt.out","w")
  55 f1.writelines([(i[0]+"\n").encode("utf8") for i in l])
  56 f2.writelines([i[1].decode("utf8").replace(u"note",u"nøte").replace(u"Note",u"Nøte").encode("utf8")+"\n" for i in l])
  57 f1.close()
  58 f2.close()