#import cStringIO
import tempfile
import codecs
-from functions import exec_rcode, MessageImage, progressbar, treat_var_mod
+from functions import exec_rcode, MessageImage, progressbar, treat_var_mod, doconcorde
from PrintRScript import barplot
from textclassechd import ClasseCHD
from shutil import copyfile
#events
#self.Bind(wx.EVT_LIST_ITEM_SELECTED, self.OnItemSelected)
- #self.Bind(wx.EVT_LIST_ITEM_ACTIVATED, self.OnItemActivated)
+ self.Bind(wx.EVT_LIST_ITEM_ACTIVATED, self.OnPopupTwo, self)
#self.Bind(wx.EVT_LIST_ITEM_DESELECTED, self.OnItemDeselected)
self.Bind(wx.EVT_LIST_COL_CLICK, self.OnColClick)
#for searching
search_id = wx.NewId()
searchall_id = wx.NewId()
+ concord_id = wx.NewId()
self.parent.Bind(wx.EVT_MENU, self.onsearch, id = search_id)
self.parent.Bind(wx.EVT_MENU, self.onsearchall, id = searchall_id)
self.accel_tbl = wx.AcceleratorTable([(wx.ACCEL_CTRL, ord('F'), search_id),
parametres['clusterprof'] = prof
parametres['type'] = 'clustercloud'
parametres['prof'] = self.Source.pathout['actprof_classe_%i.csv' % self.cl]
- print parametres['prof']
+ del parametres['uuid']
#if not os.path.exists(self.Source.pathout['actprof_classe_%i.csv' % self.lc]) :
# with open(self.Source.pathout['actprof_classe_%i.csv' % self.lc], 'w') as f :
# f.write('\n'.join(prof).encode(self.parent.syscoding))
def onexport(self, evt) :
if 'corpus' in dir(self.Source):
corpus = self.Source.corpus
- corpus.export_classe(self.Source.pathout['classe_%i_export.txt' % self.cl], self.cl)
+ if self.Source.parametres['classif_mode'] != 2 :
+ uci = False
+ else :
+ uci = True
+ corpus.export_classe(self.Source.pathout['classe_%i_export.txt' % self.cl], self.cl, uci = uci)
dial = wx.MessageDialog(self, self.Source.pathout['classe_%i_export.txt' % self.cl], u"Export", wx.OK|wx.ICON_INFORMATION)
dial.ShowModal()
dial.Destroy()
def quest_simi(self, evt) :
tableau = self.Source.tableau
+ if self.tmpchi is None :
+ self.tmpchi = tempfile.mktemp(dir=self.Source.parent.TEMPDIR)
+ with open(self.tmpchi, 'w') as f:
+ f.write('\n'.join([str(val) for val in self.lchi]))
tab = tableau.make_table_from_classe(self.cl, self.la)
pathout = ConstructPathOut(self.Source.pathout.dirout, 'simi_classe_%i' %self.cl)
self.filename = os.path.join(pathout,'mat01.csv')
'film': False,
'com' : 0,
'communities' : 0,
- 'halo' : 0
+ 'halo' : 0,
+ 'tmpchi': self.tmpchi
}
act = {}
tableau.chi = {}
dlg = progressbar(self, 2)
corpus = self.Source.corpus
uces = corpus.lc[self.cl-1]
+ if self.Source.parametres['classif_mode'] != 2 :
+ uci = False
+ else :
+ uci = True
l = []
dlg.Update(1, u'Segments...')
for i in range(2,10) :
- li = corpus.find_segments_in_classe(uces, i, 1000)
+ li = corpus.find_segments_in_classe(uces, i, 1000, uci = uci)
if li == [] :
break
else :
dlg = progressbar(self,maxi = 4)
corpus = self.Source.corpus
uces = corpus.lc[self.cl-1]
- tab = corpus.make_table_with_classe(uces, self.la)
+ if self.Source.parametres['classif_mode'] != 2 :
+ uci = False
+ else :
+ uci = True
+ tab = corpus.make_table_with_classe(uces, self.la, uci = uci)
tab.pop(0)
dlg.Update(2, u'score...')
if atype == 0 :
ntab2 = ntab2[:limite]
nuces = [val[1] for val in ntab2]
dlg.Update(3, u'concordancier...')
- #ucestxt = [corpus.ucis_paras_uces[val[1][0]][val[1][1]][val[1][2]] for val in ntab2]
- ucestxt1 = [row for row in corpus.getconcorde(nuces)]
- ucestxt = []
- ucis_txt = []
- for uce in ucestxt1 :
- ucetxt = ' '+uce[1]+' '
- ucis_txt.append(' '.join(corpus.ucis[corpus.getucefromid(uce[0]).uci].etoiles) + '<br>')
- for lem in self.la :
- listmot = corpus.getlems()[lem].formes
- for id in listmot :
- forme = corpus.getforme(id).forme
- ucetxt = ucetxt.replace(' '+forme+' ', '<font color=red> ' + forme + ' </font>')
- ucestxt.append(ucetxt)
- #ucestxt = [corpus.make_concord(self.la, ' '.join(uce), 'red') for uce in ucestxt]
+ ucis_txt, ucestxt = doconcorde(corpus, nuces, self.la, uci = uci)
dlg.Update(4, u'texte...')
- #ucis_txt = [' '.join(corpus.ucis[val[1][0]][0]) for val in ntab2]
win = message(self, u"Segments de texte caractéristiques - Classe %i" % self.cl, (750, 600))
- win.html = '<html>\n' + '<br><br>'.join(['<br>'.join([ucis_txt[i], 'score : ' + str(ntab2[i][0]), ucestxt[i]]) for i in range(0,len(ucestxt))]) + '\n</html>'
+ win.html = '<html>\n' + '<br>'.join(['<br>'.join([ucis_txt[i], '<table bgcolor = #1BF0F7 border=0><tr><td><b>score : %.2f</b></td></tr></table>' % ntab2[i][0], ucestxt[i]]) for i in range(0,len(ucestxt))]) + '\n</html>'
win.HtmlPage.SetPage(win.html)
dlg.Destroy()
win.Show(True)
win.HtmlPage.SetPage(txt)
win.Show(True)
-
def make_concord(self, uces, title, color = 'red') :
corpus = self.Source.corpus
ListWord = [self.getColumnText(self.GetFirstSelected(), 6)]
while self.GetNextSelected(last) != -1:
last = self.GetNextSelected(last)
ListWord.append(self.getColumnText(last, 6))
- listmot = [forme for item in ListWord for forme in corpus.getlems()[item].formes]
- win = message(self, title, size=(750, 600))
- toshow = ['<html>\n<H1>Concordancier</H1>\n']
- toshow.append('<h3><font color=%s>' % color + ' '.join(ListWord) + '</font></h3><br>')
- duce = {}
ucef = []
- for word in ListWord :
- ucef += list(set(corpus.getlemuces(word)).intersection(uces))
- ucef = list(set(ucef))
- ucef.sort()
- res = corpus.getconcorde(ucef)
- txt = '<br>'.join(toshow) +'<br><br>'
- for uce in res :
- ucetxt = ' '+uce[1]+' '
- txt += ' '.join(corpus.ucis[corpus.getucefromid(uce[0]).uci].etoiles) + '<br>'
- for forme in listmot:
- forme = corpus.getforme(forme).forme
- ucetxt = ucetxt.replace(' '+forme+' ', '<font color=red> ' + forme + ' </font>')
- txt += ucetxt + '<br><br>'
- win.HtmlPage.SetPage(txt)
+ if self.Source.parametres['classif_mode'] != 2 :
+ for word in ListWord :
+ uci = False
+ ucef += list(set(corpus.getlemuces(word)).intersection(uces))
+ else :
+ for word in ListWord :
+ ucef += list(set(corpus.getlemucis(word)).intersection(uces))
+ uci = True
+ ucis_txt, ucestxt = doconcorde(corpus, ucef, ListWord, uci = uci)
+ win = message(self, title, size=(750, 600))
+ win.html = ('<html>\n<h1>%s</h1>' % ' '.join(ListWord)) + '<br>'.join(['<br>'.join([ucis_txt[i], ucestxt[i]]) for i in range(0,len(ucestxt))]) + '\n</html>'
+ win.HtmlPage.SetPage(win.html)
return win
def OnPopupTwo(self, event):
rep = []
#FIXME : donner aussi eff reel a la place de nb uce
for forme in lems[word].formes :
- ucef = list(set(corpus.getworduces(forme)).intersection(uces))
+ if self.Source.parametres['classif_mode'] != 2 :
+ ucef = list(set(corpus.getworduces(forme)).intersection(uces))
+ else :
+ ucef = list(set(corpus.getworducis(forme)).intersection(uces))
#ucef = [uce for uce in corpus.formes[forme][1] if uce in uces]
if ucef != [] :
nb = len(ucef)