encoding

[iramuteq] / textaslexico.py

diff --git a/textaslexico.py b/textaslexico.py
index de0f181..35007b2 100644 (file)
--- a/textaslexico.py
+++ b/textaslexico.py
@@ -1,21 +1,22 @@
 # -*- coding: utf-8 -*-
 #Author: Pierre Ratinaud
 #Copyright (c) 2008-2011 Pierre Ratinaud
-#Lisense: GNU/GPL
+#License: GNU/GPL
 
-from chemins import ConstructPathOut, StatTxtPathOut
+from chemins import ConstructPathOut, StatTxtPathOut, PathOut
 #from corpus import Corpus
 from analysetxt import AnalyseText
 import wx
 import os
-import sys
-from listlex import *
-from functions import exec_rcode, progressbar, check_Rresult, CreateIraFile, print_liste, treat_var_mod, write_tab, DoConf
-from dialog import OptLexi, StatDialog #LexDialog
-from openanalyse import OpenAnalyse
+#import sys
+#from listlex import *
+from functions import exec_rcode, progressbar, check_Rresult, CreateIraFile, print_liste, treat_var_mod, write_tab, DoConf, TGen
+from dialog import OptLexi#, StatDialog 
+#from openanalyse import OpenAnalyse
 import tempfile
-from ConfigParser import RawConfigParser
-from guifunct import getPage, getCorpus
+#from ConfigParser import RawConfigParser
+#from guifunct import getPage, getCorpus
+from PrintRScript import TgenSpecScript
 from time import sleep
 import logging
 
@@ -46,6 +47,9 @@ class Lexico(AnalyseText) :
         txt += """
         dmt<-read.csv2("%s",row.names=1)
         """ % self.dictpathout['tabletypem']
+        txt += """
+        indice <- "%s"
+        """ % self.parametres['indice']
         if self.parametres['indice'] == 'hypergeo' :
             txt += """
             outf <- make.spec.hypergeo(dmf)
@@ -56,7 +60,19 @@ class Lexico(AnalyseText) :
             outf<-AsLexico2(dmf)
             outt<-AsLexico2(dmt)
             """
-
+        txt += """
+        if (indice == 'hypergeo') {
+            banseuil <- 2
+        } else if (indice == 'chi2') {
+            banseuil <- 3
+        }
+        banal <- apply(abs(outf[[1]]), 1, max)
+        banal <- which(banal < banseuil)
+        banalfreq <- rowSums(dmf[banal,])
+        banalspec <- specf<-outf[[1]][banal,]
+        banal <- cbind(banalfreq, banalspec)
+        write.csv2(banal,file="%s")
+        """ % self.pathout['banalites.csv']
         txt += """
         specf<-outf[[1]]
         spect<-outt[[1]]
@@ -151,6 +167,7 @@ class Lexico(AnalyseText) :
                 self.parametres['indice'] = 'hypergeo'
             else :
                 self.parametres['indice'] = 'chi2'
+            self.parametres['typeformes'] = dial.typeformes.GetSelection()
             self.parametres['clnb'] = len(ListEt)
             dial.Destroy()
             return self.parametres
@@ -161,10 +178,10 @@ class Lexico(AnalyseText) :
     def make_lexico(self) :
         mineff = self.parametres['mineff']
         #dlg = progressbar(self, maxi = 3)
-        tabout = self.corpus.make_lexitable(mineff, self.listet)
+        tabout = self.corpus.make_lexitable(mineff, self.listet, gram = self.parametres['typeformes'])
         #log.warning('Fmax a 200')
         #Fmax = [line for line in tabout[1:] if sum(line[1:]) > 199]
-        #formesmax = [line[0] for line in Fmax]
+        #formesmax = [line[0] for line in Fmax
         #Fmax = [line[1:] for line in Fmax]
         #summax = [sum(col) for col in zip(*Fmax)]
         #tabout.append(['Fmax'] + summax)
@@ -189,4 +206,26 @@ class Lexico(AnalyseText) :
                             [os.path.basename(self.dictpathout['afct_col']), u'colonnes']]
         print_liste(self.dictpathout['liste_graph_afcf'],afcf_graph_list)
         print_liste(self.dictpathout['liste_graph_afct'],afct_graph_list)
-        DoConf().makeoptions(['spec'],[self.parametres], self.dictpathout['ira'])
+        #DoConf().makeoptions(['spec'],[self.parametres], self.dictpathout['ira'])
+
+class TgenSpec(AnalyseText):
+    def __init__(self, ira, corpus, parametres):
+        self.ira = ira
+        self.corpus = corpus
+        self.parametres = parametres
+        self.pathout = PathOut(dirout = self.parametres['pathout'])
+        self.doanalyse()
+        
+    def doanalyse(self):
+        self.tgen = TGen(path = self.parametres['tgenpath'], encoding = self.ira.syscoding)
+        self.tgen.read(self.tgen.path)
+        tgenocc, totocc = self.corpus.make_tgen_table(self.tgen, self.parametres['etoiles'])
+        self.parametres['tgeneff'] = os.path.join(self.parametres['pathout'], 'tgeneff.csv')
+        self.tgen.writetable(self.parametres['tgeneff'], tgenocc, totocc)
+        self.parametres['tgenspec'] = os.path.join(self.parametres['pathout'], 'tgenspec.csv')
+        self.Rscript = TgenSpecScript(self)
+        self.Rscript.make_script()
+        self.Rscript.write()
+        self.doR(self.Rscript.scriptout, dlg = False, message = 'R...')
+
+    
\ No newline at end of file