corpus

[iramuteq] / analysetxt.py

diff --git a/analysetxt.py b/analysetxt.py
index f5d9273..3edf0a9 100644 (file)
--- a/analysetxt.py
+++ b/analysetxt.py
@@ -1,48 +1,21 @@
 # -*- coding: utf-8 -*-
 #Author: Pierre Ratinaud
 # -*- coding: utf-8 -*-
 #Author: Pierre Ratinaud

-#from corpusNG import Corpus
+#lisence : GNU GPL
+#copyright : 2012-2013 (c) Pierre Ratinaud
+

 import logging
 from chemins import PathOut, ChdTxtPathOut
 from functions import exec_rcode, check_Rresult, DoConf, print_liste
 from time import time, sleep
 from uuid import uuid4
 import os
 import logging
 from chemins import PathOut, ChdTxtPathOut
 from functions import exec_rcode, check_Rresult, DoConf, print_liste
 from time import time, sleep
 from uuid import uuid4
 import os

-#ALCESTE

 from PrintRScript import RchdTxt, AlcesteTxtProf
 from OptionAlceste import OptionAlc 
 from layout import PrintRapport
 from openanalyse import OpenAnalyse
 from time import time
 from PrintRScript import RchdTxt, AlcesteTxtProf
 from OptionAlceste import OptionAlc 
 from layout import PrintRapport
 from openanalyse import OpenAnalyse
 from time import time

-######################################
-print '#######LOGGING TEST###########'
-log = logging.getLogger('iramuteq.analyse')
-#formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
-#ch = logging.StreamHandler()
-#ch.setFormatter(formatter)
-#log.addHandler(ch)
-#log.setLevel(logging.INFO)
-#######################################
-
-#def make_ucecl_from_R(filein) :
-#    with open(filein, 'rU') as f :
-#        c = f.readlines()
-#    c.pop(0)
-#    ucecl = []
-#    for line in c :
-#        line = line.replace('\n', '').replace('"', '').split(';')
-#        ucecl.append([int(line[0]) - 1, int(line[1])])
-#    classesl = [val[1] for val in ucecl]
-#    clnb = max(classesl)
-#    ucecl = sorted(ucecl, key=itemgetter(1))
-#    ucecl = [[uce[0] for uce in ucecl if uce[1] == i] for i in range(clnb+1)]
-#    return ucecl
-#
-#def make_lc(self, uces, classes, clnb) :    
-#    self.lc = [[] for classe in range(0,clnb)]
-#    for i in range(0,clnb):
-#        self.lc[i] = [uce for j, uce in enumerate(uces) if i+1 == classes[j]]
-#    self.lc0 = [uce for j, uce in enumerate(uces) if 0 == classes[j]]

 
 

+log = logging.getLogger('iramuteq.analyse')

 
 class AnalyseText :
     def __init__(self, ira, corpus, parametres = None, dlg = False) :
 
 class AnalyseText :
     def __init__(self, ira, corpus, parametres = None, dlg = False) :


@@ -52,7 +25,11 @@ class AnalyseText :
         self.dlg = dlg
         self.dialok = True
         self.parametres = parametres
         self.dlg = dlg
         self.dialok = True
         self.parametres = parametres

-        self.pathout = PathOut(corpus.parametres['originalpath'], analyse_type = parametres['type'], dirout = corpus.parametres['pathout'])
+        self.val = False
+        if not 'pathout' in self.parametres :
+            self.pathout = PathOut(corpus.parametres['originalpath'], analyse_type = parametres['type'], dirout = corpus.parametres['pathout'])
+        else :
+            self.pathout = PathOut(filename = corpus.parametres['originalpath'], dirout = self.parametres['pathout'], analyse_type = self.parametres['name'])

         self.parametres = self.make_config(parametres)
         log.info(self.pathout.dirout)
         if self.parametres is not None :
         self.parametres = self.make_config(parametres)
         log.info(self.pathout.dirout)
         if self.parametres is not None :


@@ -69,23 +46,28 @@ class AnalyseText :
             self.parametres['uuid'] = str(uuid4())
             self.parametres['name'] = os.path.split(self.parametres['pathout'])[1]
             self.parametres['type'] = parametres['type']
             self.parametres['uuid'] = str(uuid4())
             self.parametres['name'] = os.path.split(self.parametres['pathout'])[1]
             self.parametres['type'] = parametres['type']

+            self.parametres['encoding'] = self.ira.syscoding

             self.t1 = time()
             self.t1 = time()

-            #if self.corpus.lems is None :

             self.corpus.make_lems(lem = self.parametres['lem'])
             corpus.parse_active(gramact, gramsup)
             self.corpus.make_lems(lem = self.parametres['lem'])
             corpus.parse_active(gramact, gramsup)

-            self.doanalyse()
-            self.time = time() - self.t1
-            minutes, seconds = divmod(self.time, 60)
-            hours, minutes = divmod(minutes, 60)            
-            self.parametres['time'] = '%.0fh %.0fm %.0fs' % (hours, minutes, seconds)
-            self.parametres['ira'] = self.pathout['ira']
-            DoConf().makeoptions([self.parametres['type']], [self.parametres], self.pathout['ira'])
-            self.ira.history.add(self.parametres)
-            if dlg :
-                dlg.Destroy()
-                OpenAnalyse(self.parent, self.parametres['ira'])
-                self.ira.tree.AddAnalyse(self.parametres)
-                self.val = 5100
+            result_analyse = self.doanalyse()
+            if result_analyse is None :
+                 self.time = time() - self.t1
+                 minutes, seconds = divmod(self.time, 60)
+                 hours, minutes = divmod(minutes, 60)            
+                 self.parametres['time'] = '%.0fh %.0fm %.0fs' % (hours, minutes, seconds)
+                 self.parametres['ira'] = self.pathout['Analyse.ira']
+                 DoConf().makeoptions([self.parametres['type']], [self.parametres], self.pathout['Analyse.ira'])
+                 self.ira.history.add(self.parametres)
+                 if dlg :
+                     dlg.Destroy()
+                     OpenAnalyse(self.parent, self.parametres['ira'])
+                     self.ira.tree.AddAnalyse(self.parametres)
+                     self.val = 5100
+            else :
+                 self.val = False
+                 if dlg :
+                     dlg.Destroy()

         else :
             if dlg :
                 dlg.Destroy()
         else :
             if dlg :
                 dlg.Destroy()


@@ -96,8 +78,8 @@ class AnalyseText :
 
     def make_config(self, config) :
         if config is not None :
 
     def make_config(self, config) :
         if config is not None :

-            if isinstance(config, basestring) : 
-                return self.readconfig(config)
+            if not self.dlg : 
+                return config

             else :
                 return self.preferences()
 
             else :
                 return self.preferences()
 


@@ -107,9 +89,6 @@ class AnalyseText :
     def preferences(self) :
         return {}
 
     def preferences(self) :
         return {}
 

-    def doR(self):
-        pass
-

     def printRscript(self) :
         pass
 
     def printRscript(self) :
         pass
 


@@ -122,13 +101,12 @@ class AnalyseText :
                 sleep(0.2)
             else :
                 sleep(0.2)
                 sleep(0.2)
             else :
                 sleep(0.2)

-        check_Rresult(self.ira, pid)
+        return check_Rresult(self.ira, pid)

 
 
 
 class Alceste(AnalyseText) :
     def doanalyse(self) :
 
 
 
 class Alceste(AnalyseText) :
     def doanalyse(self) :

-        #self.pathout = PathOut(self.corpus.parametres['filename'], 'alceste')

         self.parametres['type'] = 'alceste'
         self.pathout.basefiles(ChdTxtPathOut)
         self.actives, lim = self.corpus.make_actives_nb(self.parametres['max_actives'], 1)
         self.parametres['type'] = 'alceste'
         self.pathout.basefiles(ChdTxtPathOut)
         self.actives, lim = self.corpus.make_actives_nb(self.parametres['max_actives'], 1)


@@ -143,9 +121,7 @@ class Alceste(AnalyseText) :
         elif self.parametres['classif_mode'] == 2 :
             self.corpus.make_and_write_sparse_matrix_from_uci(self.actives, self.pathout['TableUc1'], self.pathout['listeuce1'])
         Rscript = self.printRscript()
         elif self.parametres['classif_mode'] == 2 :
             self.corpus.make_and_write_sparse_matrix_from_uci(self.actives, self.pathout['TableUc1'], self.pathout['listeuce1'])
         Rscript = self.printRscript()

-        self.doR(Rscript)
-        #self.lc = make_ucecl_from_R(self.pathout['uce'])
-        #self.lc0 = self.lc.pop(0)
+        self.doR(Rscript, dlg = self.dlg, message = 'CHD...')

         self.corpus.make_ucecl_from_R(self.pathout['uce'])
         self.corpus.make_and_write_profile(self.actives, self.corpus.lc, self.pathout['Contout'])
         self.sup, lim = self.corpus.make_actives_nb(self.parametres['max_actives'], 2)
         self.corpus.make_ucecl_from_R(self.pathout['uce'])
         self.corpus.make_and_write_profile(self.actives, self.corpus.lc, self.pathout['Contout'])
         self.sup, lim = self.corpus.make_actives_nb(self.parametres['max_actives'], 2)


@@ -154,7 +130,7 @@ class Alceste(AnalyseText) :
         self.clnb = len(self.corpus.lc)
         self.parametres['clnb'] = self.clnb
         Rscript = self.printRscript2()
         self.clnb = len(self.corpus.lc)
         self.parametres['clnb'] = self.clnb
         Rscript = self.printRscript2()

-        self.doR(Rscript)
+        self.doR(Rscript, dlg = self.dlg, message = 'profils et A.F.C. ...')

         self.time = time() - self.t1
         minutes, seconds = divmod(self.time, 60)
         hours, minutes = divmod(minutes, 60)            
         self.time = time() - self.t1
         minutes, seconds = divmod(self.time, 60)
         hours, minutes = divmod(minutes, 60)            


@@ -201,14 +177,11 @@ class Alceste(AnalyseText) :
         return self.pathout['RTxtProfGraph']
 
     def print_graph_files(self) :
         return self.pathout['RTxtProfGraph']
 
     def print_graph_files(self) :

-        afc_graph_list = [[os.path.basename(self.pathout['AFC2DL_OUT']), u'Variables actives - coordonnées - facteurs 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DSL_OUT']), u'variables supplémentaires - coordonnées - facteurs 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DEL_OUT']), u'Variables illustratives - Coordonnées - facteur 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DCL_OUT']), u'Classes - Coordonnées - facteur 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DCoul']), u'Variables actives - Corrélation - facteur 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DCoulSup']), u'Variables supplémentaires - Corrélation - facteur 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DCoulEt']), u'Variables illustratives - Corrélations - facteur 1 / 2'],
-                      [os.path.basename(self.pathout['AFC2DCoulCl']), u'Classes - Corrélations - facteurs 1 / 2'],]
+        mess_afc = u"La position des points n'est peut être pas exacte"
+        afc_graph_list = [[os.path.basename(self.pathout['AFC2DL_OUT']), u'Variables actives - coordonnées - 30 points par classes - facteurs 1 / 2 - %s' % mess_afc],
+                      [os.path.basename(self.pathout['AFC2DSL_OUT']), u'variables supplémentaires - coordonnées - 30 points par classes - facteurs 1 / 2 - %s' % mess_afc],
+                      [os.path.basename(self.pathout['AFC2DEL_OUT']), u'Variables illustratives - Coordonnées - 30 points par classes - facteur 1 / 2 - %s' % mess_afc],
+                      [os.path.basename(self.pathout['AFC2DCL_OUT']), u'Classes - Coordonnées - facteur 1 / 2']]

         chd_graph_list = [[os.path.basename(self.pathout['dendro1']), u'dendrogramme à partir de chd1']]
         if self.parametres['classif_mode'] == 0 :
             chd_graph_list.append([os.path.basename(self.pathout['dendro2']), u'dendrogramme à partir de chd2'])
         chd_graph_list = [[os.path.basename(self.pathout['dendro1']), u'dendrogramme à partir de chd1']]
         if self.parametres['classif_mode'] == 0 :
             chd_graph_list.append([os.path.basename(self.pathout['dendro2']), u'dendrogramme à partir de chd2'])


@@ -251,8 +224,3 @@ keys = {'art_def' : 2,
 
 gramact = [k for k in keys if keys[k] == 1]
 gramsup = [k for k in keys if keys[k] == 2]
 
 gramact = [k for k in keys if keys[k] == 1]
 gramsup = [k for k in keys if keys[k] == 2]

-
-#corpus = Corpus('', {'filename': '/home/pierre/workspace/iramuteq/dev/testcorpus.txt','formesdb':'formes.db', 'ucesdb': 'uces.db', 'corpusdb' : 'corpus.db', 'syscoding' : 'utf-8'})
-#corpus.read_corpus()
-#corpus.parse_active(gramact, gramsup)
-#Alceste(corpus).doanalyse()