X-Git-Url: http://iramuteq.org/git?p=iramuteq;a=blobdiff_plain;f=corpus.py;h=ee80d9e8fc60f2d7b05e26dd807e3417a15ed3ea;hp=2e85ed58c413b031f7c4a3bb142de0da4a134755;hb=191e64482209e897e7bf1853646c0d2ca84db1f2;hpb=87842df83ba95117fcda5575bc60067a6d3654b0

diff --git a/corpus.py b/corpus.py
index 2e85ed5..ee80d9e 100644
--- a/corpus.py
+++ b/corpus.py
@@ -77,10 +77,10 @@ class Corpus :
                 gramtype = self.parent.lexique[word][1]
                 lem = self.parent.lexique[word][0]
             elif word.isdigit() :
-                gramtype = 'num'
+                gramtype = u'num'
                 lem = word
             else :
-                gramtype = 'nr'
+                gramtype = u'nr'
                 lem = word
             self.formes[word] =  Word(word, gramtype, len(self.formes), lem)
             self.idformesuces[self.formes[word].ident] = {self.ucis[-1].uces[-1].ident : 1}
@@ -434,7 +434,7 @@ class Corpus :
                 f.write(guce.encode(self.parametres['syscoding']) + '\n\n')
 
     def export_classe(self, outf, classe, lem = False) :
-        sts = self.lc[classe] 
+        sts = self.lc[classe - 1] 
         res = self.getconcorde(sts)
         self.make_iduces()
         with open(outf, 'w') as f :
@@ -1081,7 +1081,7 @@ class BuildCorpus :
     def firstclean(self, txt) :
         txt = txt.replace(u'’',"'")
         txt = txt.replace(u'œ', u'oe')
-        return txt.replace('...',u' £$£ ').replace('?',' ? ').replace('.',' . ').replace('!', ' ! ').replace(',',' , ').replace(';', ' ; ').replace(':',' : ').replace(u'…', ' £$£ ')
+        return txt.replace('...',u' £$£ ').replace('?',' ? ').replace('.',' . ').replace('!', ' ! ').replace(',',' , ').replace(';', ' ; ').replace(':',' : ').replace(u'…', u' £$£ ')
 
     def make_cleans(self, txt) :
         for clean in self.cleans :