new tgen

[iramuteq] / textreinert.py

diff --git a/textreinert.py b/textreinert.py
index d32f24a..8bb9a6d 100644 (file)
--- a/textreinert.py
+++ b/textreinert.py
@@ -7,10 +7,10 @@ import os
 from time import time
 from analysetxt import AnalyseText
 from OptionAlceste import OptionAlc 
-from PrintRScript import RchdTxt, AlcesteTxtProf
+from PrintRScript import RchdTxt, ReinertTxtProf, TgenProfScript
 from layout import PrintRapport
-from chemins import ChdTxtPathOut
-from functions import DoConf, print_liste
+from chemins import ChdTxtPathOut, PathOut
+from functions import DoConf, print_liste, TGen
 
 
 class Reinert(AnalyseText) :
@@ -31,8 +31,9 @@ class Reinert(AnalyseText) :
             self.corpus.make_and_write_sparse_matrix_from_uci(self.actives, self.pathout['TableUc1'], self.pathout['listeuce1'])
             uci = True
         Rscript = self.printRscript()
-        self.doR(Rscript, dlg=self.dlg, message='CHD...')
-
+        result = self.doR(Rscript, dlg=self.dlg, message='CHD...')
+        if not result :
+            return 'NOK'
         self.corpus.make_ucecl_from_R(self.pathout['uce'])
         self.corpus.make_and_write_profile(self.actives, self.corpus.lc, self.pathout['Contout'], uci = uci)
         self.sup, lim = self.corpus.make_actives_nb(self.parametres['max_actives'], 2)
@@ -80,7 +81,7 @@ class Reinert(AnalyseText) :
         return self.pathout['Rchdtxt']
 
     def printRscript2(self) :
-        AlcesteTxtProf(self.pathout, self.parent.RscriptsPath, self.clnb, 0.9)
+        ReinertTxtProf(self.pathout, self.parent.RscriptsPath, self.clnb, 0.9)
         return self.pathout['RTxtProfGraph']
 
     def print_graph_files(self) :
@@ -97,4 +98,39 @@ class Reinert(AnalyseText) :
             chd_graph_list.append([os.path.basename(self.pathout['arbre2']), u'chd2'])       
         print_liste(self.pathout['liste_graph_afc'], afc_graph_list)
         print_liste(self.pathout['liste_graph_chd'], chd_graph_list)
-        PrintRapport(self, self.corpus, self.parametres)
\ No newline at end of file
+        PrintRapport(self, self.corpus, self.parametres)
+
+class TgenProf(AnalyseText):
+    def __init__(self, ira, corpus, parametres, cluster_size):
+        self.ira = ira
+        self.corpus = corpus
+        self.parametres = parametres
+        self.pathout = PathOut(dirout = self.parametres['pathout'])
+        self.cluster_size = [len(classe) for classe in corpus.lc]
+        self.doanalyse()
+        
+    def doanalyse(self):    
+        self.tgen = TGen(path = self.parametres['tgenpath'], encoding = self.ira.syscoding)
+        self.tgen.read(self.tgen.path)
+        #self.parametres['etoiles'].sort()
+        self.parametres['tgeneff'] = os.path.join(self.parametres['pathout'], 'tgeneff.csv') 
+        tgenst = self.corpus.make_tgen_profile(self.tgen.tgen, self.corpus.lc)
+        clnames = ['cluster_%03d' % i for i in range(1, len(self.cluster_size) + 1)]
+        et = dict(zip(clnames, self.cluster_size))
+        tgenst = dict([[line[0], dict(zip(clnames, line[1:]))] for line in tgenst]) 
+        self.tgen.writetable(self.parametres['tgeneff'], tgenst, et)
+        self.parametres['tgenspec'] = os.path.join(self.parametres['pathout'], 'tgenchi2.csv')
+        self.parametres['tgenlemeff'] = os.path.join(self.parametres['pathout'], 'tgenlemeff.csv')
+        self.parametres['tgenlemspec'] = os.path.join(self.parametres['pathout'], 'tgenlemchi2.csv')
+        tgenlemeff = dict([[lem, dict(zip(clnames, self.corpus.tgenlem[lem]))] for lem in self.corpus.tgenlem]) 
+        self.tgen.writetable(self.parametres['tgenlemeff'], tgenlemeff, et)
+        self.Rscript = TgenProfScript(self)
+        self.Rscript.make_script()
+        self.Rscript.write()
+        self.doR(self.Rscript.scriptout, dlg = False, message = 'R...')
+        
+        
+        
+        
+        
+        
\ No newline at end of file