première mise à jour pour python 3

[iramuteq] / autres / usecorpusNG.py

diff --git a/autres/usecorpusNG.py b/autres/usecorpusNG.py
new file mode 100644 (file)
index 0000000..343144c
--- /dev/null
+++ b/autres/usecorpusNG.py
@@ -0,0 +1,28 @@
+from corpus import *
+from functions import DoConf
+
+corpus_encodage = 'cp1252'
+corpus_in = '/home/pierre/fac/identite/identite_sans_doublons_ok_corpus_6/Corpus/corpus.cira'
+
+corpus_parametres = DoConf('/home/pierre/.iramuteq/corpus.cfg').getoptions('corpus')
+corpus_parametres['filename'] = self.filename
+corpus_parametres['encoding'] = corpus_encodage
+corpus_parametres['syscoding'] = 'utf8'
+#corpus = BuildFromAlceste(self.filename, corpus_parametres, self.lexique, self.expressions).corpus
+                #with codecs.open(self.filename, 'r', self.corpus_encodage) as f:
+corpus = Corpus(self, parametres = {'filename': corpus_in, 'syscoding': 'utf8'}, read = corpus_in)
+
+corpus.conn_all()
+
+etoiles = ['*date_0211','*date_0311', '*date_0411', '*date_0511', '*date_0611', '*date_0711']
+corpus.read_corpus()
+corpus.make_lems()
+actives = corpus.make_actives(450)
+corpus.make_and_write_sparse_matrix_from_uces(actives, 'mm.mm')
+ucesize = corpus.getucesize()
+float(sum(ucesize)/len(ucesize)
+actives = corpus.make_actives(10)
+#ucesize = corpus.make_uceactsize(actives)
+#uc1, uc2 = corpus.make_uc(actives, 25, 27)
+corpus.make_and_write_sparse_matrix_from_uc(actives, 25, 27, 'uc1.mm', 'uc2.mm')
+tab = corpus.make_lexitable(100, etoiles)