antipro pour matrix

diff --git a/ProfList.py b/ProfList.py
index 6d007a7..063a275 100644 (file)
--- a/ProfList.py
+++ b/ProfList.py
@@ -291,7 +291,7 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col
                 self.Bind(wx.EVT_MENU, self.onwordgraph, id = self.idwordgraph)
                 self.Bind(wx.EVT_MENU, self.onproxe, id = self.popup_proxe)
                 self.Bind(wx.EVT_MENU, self.onlexdendro, id = self.idlexdendro)
-                #self.Bind(wx.EVT_MENU, self.onexport, id = self.idexport)
+                self.Bind(wx.EVT_MENU, self.onexport, id = self.idexport)
              #  self.Bind(wx.EVT_MENU, self.on_export_classes, id = self.export_classes)
    #            self.Bind(wx.EVT_MENU, self.OnPopupThree, id=self.popupID3)
     
@@ -323,7 +323,7 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col
             menu.Append(self.popupIDgraph, u"Graphe de la classe")
             menu.Append(self.idseg, u"Segments répétés")
             menu.Append(self.iducecarac, u"UCE caractéristiques")
-            #menu.Append(self.idexport, 'Partitionner...')
+            menu.Append(self.idexport, 'Exporter...')
             #menu.Append(self.popupID2, u"Concordancier")
     #        menu.Append(self.popupID3, "recharger")
     
@@ -349,7 +349,13 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col
     def onexport(self, evt) :
         if 'corpus' in dir(self.Source):
             corpus = self.Source.corpus
-        ClasseCHD(self.parent, corpus, self.cl)
+            corpus.export_classe(self.Source.pathout['classe_%i_export.txt' % self.cl], self.cl)
+        dial = wx.MessageDialog(self, self.Source.pathout['classe_%i_export.txt' % self.cl], u"Export", wx.ID_OK|wx.ICON_INFORMATION)
+        dial.ShowModal()
+        dial.Destroy()
+        #if 'corpus' in dir(self.Source):
+        #    corpus = self.Source.corpus
+        #ClasseCHD(self.parent, corpus, self.cl)
 
     def getselectedwords(self) :
         words = [self.getColumnText(self.GetFirstSelected(), 6)]

diff --git a/corpus.py b/corpus.py
index 51061fb..76ebb66 100644 (file)
--- a/corpus.py
+++ b/corpus.py
@@ -433,6 +433,18 @@ class Corpus :
                 f.write(etline.encode(self.parametres['syscoding']) + '\n')
                 f.write(guce.encode(self.parametres['syscoding']) + '\n\n')
 
+    def export_classe(self, outf, classe, lem = False) :
+        sts = self.lc[classe] 
+        res = self.getconcorde(sts)
+        self.make_iduces()
+        with open(outf, 'w') as f :
+            for uce in res :
+                guce = uce[1]
+                f.write(' '.join(self.ucis[self.iduces[uce[0]].uci].etoiles).encode(self.parametres['syscoding']) + '\n')
+                if lem :
+                    guce = ' '.join([self.formes[forme].lem for forme in guce.split()])
+                f.write(guce.encode(self.parametres['syscoding']) + '\n\n')
+
     def make_and_write_sparse_matrix_from_uces(self, actives, outfile, listuce = False) :
         log.info('make_and_write_sparse_matrix_from_uces %s' % outfile)
         nbl = 0
@@ -707,27 +719,37 @@ class Corpus :
         self.lc0 = self.lc.pop(0)
         #return ucecl
     
-    def get_stat_by_cluster(self, outf) :
+    def get_stat_by_cluster(self, outf, lclasses = None) :
         log.info('get_stat_by_cluster')
+        if lclasses is None :
+            lclasses = self.lc
         t1 = time()
-        occurrences = dict([[i + 1, 0] for i in range(len(self.lc))])
-        formescl = dict([[i + 1, 0] for i in range(len(self.lc))])
-        hapaxcl = dict([[i + 1, 0] for i in range(len(self.lc))])
-        lenclasses = dict([[i+1,len(cl)] for i, cl in enumerate(self.lc)])
-        sets = [set(cl) for cl in self.lc]
+        occurrences = dict([[i + 1, 0] for i in range(len(lclasses))])
+        formescl = dict([[i + 1, 0] for i in range(len(lclasses))])
+        hapaxcl = dict([[i + 1, 0] for i in range(len(lclasses))])
+        lenclasses = dict([[i+1,len(cl)] for i, cl in enumerate(lclasses)])
+        sets = [set(cl) for cl in lclasses]
         for forme in self.formes :
             formeuceeff = self.getformeuceseff(forme)
-            for i, classe in enumerate(self.lc) :
+            for i, classe in enumerate(lclasses) :
                 concern = sets[i].intersection(formeuceeff.keys())
                 if len(concern) :
                     occurrences[i+1] += sum([formeuceeff[uce] for uce in concern])
                     formescl[i+1] += 1
                     if self.formes[forme].freq == 1 :
                         hapaxcl[i+1] += 1
-        toprint = '\n'.join([';'.join([`i`, `occurrences[i]`, `formescl[i]`, `hapaxcl[i]`, `lenclasses[i]`, `float(hapaxcl[i])/float(formescl[i])`]) for i in occurrences])
-        with open(outf, 'w') as f :
-            f.write(toprint)
         log.info('%f' % (time() - t1))        
+        if outf is not None :
+            toprint = '\n'.join([';'.join([`i`, `occurrences[i]`, `formescl[i]`, `hapaxcl[i]`, `lenclasses[i]`, `float(hapaxcl[i])/float(formescl[i])`]) for i in occurrences])
+            with open(outf, 'w') as f :
+                f.write(toprint)
+        else :
+            return [[`occurrences[i]`, `formescl[i]`, `hapaxcl[i]`, `lenclasses[i]`, `float(hapaxcl[i])/float(formescl[i])`] for i in occurrences]
+
+    def get_stat_by_et(self, outf, etoiles) :
+        lclasses = [self.getucesfrometoile(etoile) for etoile in etoiles]
+        stats = self.get_stat_by_cluster(None, lclasses)
+        stats = [[etoiles[i]] + val for i, val in enumerate(stats)]
 
     def gethapaxbyet(self, etoiles) :
         hapaxuces = [self.getlemuces(forme)[0] for forme in self.lems if self.lems[forme].freq == 1]

diff --git a/guifunct.py b/guifunct.py
index 8ee4879..b16c58c 100644 (file)
--- a/guifunct.py
+++ b/guifunct.py
@@ -14,7 +14,7 @@ from functions import treat_var_mod
 
 def OnOpen(self, type):
         if type == "Data":
-             wildcard = u"Fichiers supportés|*.ods;*.xls;*.csv;*.txt|Openoffice Clac|*.ods|Fichier excel|*.xls|Fichier csv|*.csv|Fichier texte|*.txt|Tous les fichiers|*"
+             wildcard = u"Fichiers supportés|*.ods;*.xls;*.csv;*.txt|Openoffice Calc|*.ods|Excel 97/2000/XP/2003|*.xls|Fichier csv|*.csv|Fichier texte|*.txt|Tous les fichiers|*"
         elif type == "Texte":
             wildcard = "Fichier texte|*.txt|Tous les fichiers|*"
         elif type == "Analyse":

diff --git a/tree.py b/tree.py
index c3f0ba6..6c28c09 100644 (file)
--- a/tree.py
+++ b/tree.py
@@ -352,7 +352,7 @@ class LeftTree(CT.CustomTreeCtrl):
             self.Bind(wx.EVT_MENU, self.OnSpec, spec)
             self.Bind(wx.EVT_MENU, self.OnSimiTxt, simi)
             self.Bind(wx.EVT_MENU, self.OnWordCloud, wdc)
-        elif pydata.get('type', False) == 'alceste'  and pydata['uuid'] in self.parent.history.opened :
+        elif pydata.get('type', False) == 'alceste' and pydata['uuid'] in self.parent.history.opened :
             openmenu = wx.Menu()
             antipro = openmenu.Append(wx.ID_ANY, u"antiprofils")
             menu.AppendMenu(wx.ID_ANY, u"Ouvrir...", openmenu)
@@ -379,6 +379,12 @@ class LeftTree(CT.CustomTreeCtrl):
             self.Bind(wx.EVT_MENU, self.OnExportDictionary, export_dictionary)
             self.Bind(wx.EVT_MENU, self.OnExportLems, export_lems)
             menu.AppendSeparator()
+        elif pydata.get('type', False) == 'gnepamatrix' and pydata['uuid'] in self.parent.history.opened :
+            openmenu = wx.Menu()
+            antipro = openmenu.Append(wx.ID_ANY, u"antiprofils")
+            menu.AppendMenu(wx.ID_ANY, u"Ouvrir...", openmenu)
+            self.Bind(wx.EVT_MENU, self.OpenAntipro, antipro)
+
 
         itemdelete = menu.Append(wx.ID_ANY, "Supprimer de l'historique")
         if item == self.GetRootItem():