# -*- coding: utf-8 -*-
#Author: Pierre Ratinaud
-#Copyright (c) 2008-2011 Pierre Ratinaud
+#Copyright (c) 2008-2013 Pierre Ratinaud
#Lisense: GNU/GPL
from chemins import ffr, simipath
#from corpus import Corpus
import os
from analysetxt import AnalyseText
-from ConfigParser import RawConfigParser
-from guifunct import getPage, getCorpus
+#from ConfigParser import RawConfigParser
+#from guifunct import getPage, getCorpus
from dialog import StatDialog
from guifunct import SelectColumn, PrepSimi
from functions import indices_simi, progressbar, treat_var_mod, read_list_file, print_liste
-from tableau import Tableau
-from tabsimi import DoSimi
+#from tableau import Tableau
+#from tabsimi import DoSimi
from PrintRScript import PrintSimiScript
import wx
from copy import copy
import logging
-logger = logging.getLogger('iramuteq.textsimi')
+log = logging.getLogger('iramuteq.textsimi')
class SimiTxt(AnalyseText):
def doanalyse(self) :
#FIXME
self.actives = self.corpus.make_actives_limit(3)
dictcol = dict([[i, [act, self.corpus.getlemeff(act)]] for i, act in enumerate(self.actives)])
- SelectColumn(self.ira, dictcol, self.actives, self.pathout['selected.csv'], dlg = self.dlg)
- self.makefiles()
+ continu = False
if self.dlg :
- prep = PrepSimi(self.ira, self.parametres, indices_simi)
- self.parametres = prep.parametres
- script = PrintSimiScript(self)
- script.make_script()
- self.doR(script.scriptout)
- if self.parametres['type_graph'] == 1:
- if os.path.exists(self.pathout['liste_graph']):
- graph_simi = read_list_file(self.pathout['liste_graph'])
- graph_simi.append([os.path.basename(script.filename), script.txtgraph])
- else :
- graph_simi = [[os.path.basename(script.filename), script.txtgraph]]
- print_liste(self.pathout['liste_graph'], graph_simi)
+ #cont = SelectColumn(self.ira, dictcol, self.actives, self.pathout['selected.csv'], dlg = self.dlg)
+ #if cont.ok :
+ self.listet = self.corpus.make_etoiles()
+ self.listet.sort()
+ self.stars = copy(self.listet)
+ self.parametres['stars'] = copy(self.listet)
+ self.parametres['sfromchi'] = False
+ prep = PrepSimi(self.ira, self, self.parametres, self.pathout['selected.csv'], self.actives, indices_simi, wordlist=dictcol)
+ if prep.val == wx.ID_OK :
+ continu = True
+ self.parametres = prep.parametres
+ if continu :
+ self.makefiles()
+ script = PrintSimiScript(self)
+ script.make_script()
+ if not self.doR(script.scriptout, dlg = self.dlg, message = 'R...') :
+ log.info('Problem')
+ return False
+ if self.parametres['type_graph'] == 1:
+ if os.path.exists(self.pathout['liste_graph']):
+ graph_simi = read_list_file(self.pathout['liste_graph'])
+ graph_simi.append([os.path.basename(script.filename), script.txtgraph])
+ else :
+ graph_simi = [[os.path.basename(script.filename), script.txtgraph]]
+ print_liste(self.pathout['liste_graph'], graph_simi)
+ else :
+ return False
def preferences(self) :
dial = StatDialog(self, self.parent)
#self.actives, lim = self.corpus.make_actives_nb(self.parametres.get('max_actives',1500), 1)
self.parametres['eff_min_forme'] = lim
self.parametres['nbactives'] = len(self.actives)
- self.parametres['fromprof'] = True
+ self.parametres['fromprof'] = False
self.corpus.make_and_write_sparse_matrix_from_uces(self.actives, self.pathout['mat01.csv'], self.pathout['listeuce1.csv'])
with open(self.pathout['actives.csv'], 'w') as f :
f.write('\n'.join(self.actives).encode(self.ira.syscoding))
- self.listet = self.corpus.make_etoiles()
- self.listet.sort()
- self.parametres['stars'] = copy(self.listet)
- self.stars = copy(self.listet)
- self.parametres['sfromchi'] = False
+class SimiFromCluster(SimiTxt) :
+ def __init__(self, ira, corpus, actives, numcluster, parametres = None, dlg = False) :
+ self.actives = actives
+ self.numcluster = numcluster
+ parametres['name'] = 'simi_classe_%i' % (numcluster + 1)
+ SimiTxt.__init__(self, ira, corpus, parametres, dlg)
+
+ def preferences(self) :
+ return self.parametres
+
+ def doanalyse(self) :
+ self.parametres['type'] = 'clustersimitxt'
+ self.pathout.basefiles(simipath)
+ self.indices = indices_simi
+ self.makesimiparam()
+ if 'bystar' in self.parametres :
+ del self.parametres['bystar']
+ dictcol = dict([[i, [act, self.corpus.getlemclustereff(act, self.numcluster)]] for i, act in enumerate(self.actives)])
+ continu = True
+ if self.dlg :
+ #self.listet = self.corpus.make_etoiles()
+ #self.listet.sort()
+ self.stars = []#copy(self.listet)
+ self.parametres['stars'] = False#copy(self.listet)
+ self.parametres['sfromchi'] = True
+ prep = PrepSimi(self.ira, self, self.parametres, self.pathout['selected.csv'], self.actives, indices_simi, wordlist=dictcol)
+ if prep.val == wx.ID_OK :
+ continu = True
+ self.parametres = prep.parametres
+ else :
+ continu = False
+ if continu :
+ self.makefiles()
+ script = PrintSimiScript(self)
+ script.make_script()
+ if not self.doR(script.scriptout, dlg = self.dlg, message = 'R ...') :
+ return False
+ if self.parametres['type_graph'] == 1:
+ if os.path.exists(self.pathout['liste_graph']):
+ graph_simi = read_list_file(self.pathout['liste_graph'])
+ graph_simi.append([os.path.basename(script.filename), script.txtgraph])
+ else :
+ graph_simi = [[os.path.basename(script.filename), script.txtgraph]]
+ print_liste(self.pathout['liste_graph'], graph_simi)
+ else :
+ return False
+
+ def makefiles(self) :
+ self.parametres['eff_min_forme'] = 3
+ self.parametres['nbactives'] = len(self.actives)
+ self.parametres['fromprof'] = True
+ self.corpus.make_and_write_sparse_matrix_from_classe(self.actives, self.corpus.lc[self.numcluster], self.pathout['mat01.csv'])
+ with open(self.pathout['actives.csv'], 'w') as f :
+ f.write('\n'.join(self.actives).encode(self.ira.syscoding))
# self.tableau = Tableau(self.parent, '')
# self.tableau.listactives = self.actives
projects / iramuteq / blobdiff