From: Pierre Date: Mon, 10 Dec 2012 08:15:37 +0000 (+0100) Subject: corrections... X-Git-Url: http://iramuteq.org/git?p=iramuteq;a=commitdiff_plain;h=ab23968410d4e2eff482fd16a639801b457d5063 corrections... --- diff --git a/PrintRScript.py b/PrintRScript.py index 45022f3..f4e20a1 100644 --- a/PrintRScript.py +++ b/PrintRScript.py @@ -460,10 +460,12 @@ def print_simi3d(self): movie = "'" + ffr(os.path.dirname(self.DictPathOut['RData'])) + "'" else : movie = 'NULL' - if self.section == 'chd_dist_quest' : - header = 'TRUE' - else : - header = 'FALSE' + + #if self.corpus.parametres['type'] == 'corpus' : + # header = 'TRUE' + #else : + # header = 'FALSE' + header = 'FALSE' txt += """ dm<-read.csv2("%s",row.names=1,header = %s) load("%s") @@ -781,7 +783,8 @@ class PrintSimiScript(PrintRScript) : txt += """ et <- list() """ - for i,et in enumerate(self.tableau.etline) : + print self.parametres + for i,et in enumerate(self.parametres['stars']) : txt+= """ et[[%i]] <- c(%s) """ % (i+1, ','.join(et[1:])) diff --git a/ProfList.py b/ProfList.py index 73f8ff7..7a9c245 100644 --- a/ProfList.py +++ b/ProfList.py @@ -41,6 +41,7 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col self.Alceste = Alceste self.Source = gparent self.cl = cl + self.var_mod = {} line1 = profclasse.pop(0) classen = [line for line in profclasse if line[0] != '*' and line[0] != '*****'] @@ -367,11 +368,19 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col corpus = self.Source.corpus ClasseCHD(self.parent, corpus, self.cl) + def getselectedwords(self) : + words = [self.getColumnText(self.GetFirstSelected(), 6)] + last = self.GetFirstSelected() + while self.GetNextSelected(last) != -1: + last = self.GetNextSelected(last) + words.append(self.getColumnText(last, 6)) + return words + def quest_var_mod(self, evt) : if 'corpus' in dir(self.Source): corpus = self.Source.corpus if self.var_mod == {} : - self.var_mod = treat_var_mod([val for val in corpus.make_etoiles()]) + self.var_mod = self.Source.corpus.make_etoiles_dict() else : corpus = self.Source.tableau if self.var_mod == {} : @@ -383,10 +392,10 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col chistable.pop(0) vchistable = [line[1:] for line in chistable] fchistable = [line[0] for line in chistable] - word = self.getColumnText(self.GetFirstSelected(), 6) + word = self.getselectedwords()[0] if len(word.split('_')) > 1 : - var = word.split('_')[0] - words = [word for word in self.var_mod[var]] + var = word.split('_') + words = ['_'.join([var[0],word]) for word in self.var_mod[var[0]]] words.sort() tableout = [] kwords = [] @@ -650,14 +659,8 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col chistable.pop(0) vchistable = [line[1:] for line in chistable] fchistable = [line[0] for line in chistable] - words = [self.getColumnText(self.GetFirstSelected(), 6)] - tableout = [vchistable[fchistable.index(words[0])]] - last = self.GetFirstSelected() - while self.GetNextSelected(last) != -1: - last = self.GetNextSelected(last) - word = self.getColumnText(last, 6) - words.append(word) - tableout.append(vchistable[fchistable.index(word)]) + words = self.getselectedwords() + tableout = [vchistable[fchistable.index(word)] for word in words] tmpgraph = tempfile.mktemp(dir=self.Source.parent.TEMPDIR) nbcl = len(title) txt = barplot(tableout, words, title, self.Source.parent.RscriptsPath['Rgraph'], tmpgraph) @@ -685,14 +688,8 @@ class ProfListctrlPanel(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.Col chistable.pop(0) vchistable = [line[1:] for line in chistable] fchistable = [line[0] for line in chistable] - words = [self.getColumnText(self.GetFirstSelected(), 6)] - tableout = [vchistable[fchistable.index(words[0])]] - last = self.GetFirstSelected() - while self.GetNextSelected(last) != -1: - last = self.GetNextSelected(last) - word = self.getColumnText(last, 6) - words.append(word) - tableout.append(vchistable[fchistable.index(word)]) + words = self.getselectedwords() + tableout = [vchistable[fchistable.index(word)] for word in words] tmpgraph = tempfile.mktemp(dir=self.Source.parent.TEMPDIR) txttable = 'c(' + ','.join([','.join(line) for line in tableout]) + ')' rownames = 'c("' + '","'.join(words) + '")' diff --git a/Rscripts/Rgraph.R b/Rscripts/Rgraph.R index 85f1ff8..061c1dd 100644 --- a/Rscripts/Rgraph.R +++ b/Rscripts/Rgraph.R @@ -296,7 +296,7 @@ create_afc_table <- function(x) { res } -make_afc_graph <- function(toplot,classes,clnb, xlab, ylab, cex.txt = NULL, leg = FALSE, cmd = FALSE) { +make_afc_graph <- function(toplot,classes,clnb, xlab, ylab, cex.txt = NULL, leg = FALSE, cmd = FALSE, black = FALSE) { rain <- rainbow(clnb) compt <- 1 tochange <- NULL @@ -316,6 +316,9 @@ make_afc_graph <- function(toplot,classes,clnb, xlab, ylab, cex.txt = NULL, leg } } cl.color <- rain[classes] + if (black) { + cl.color <- 'black' + } plot(toplot[,1],toplot[,2], pch='', xlab = xlab, ylab = ylab) abline(h=0,v=0, lty = 'dashed') #print('ATTENTION Rgraph.R : utilisation de maptools !') @@ -509,8 +512,9 @@ make.simi.afc <- function(x,chitable,lim=0, alpha = 0.1, movie = NULL) { cc<-mc[cc] #mass<-(rowSums(x)/max(rowSums(x))) * 5 maxchi<-norm.vec(maxchi, 0.03, 0.3) - rglplot(g1,vertex.label = vire.nonascii(rownames(x)),vertex.label.color='black',vertex.color=cc,vertex.size = 0.1, layout=lo, rescale=FALSE) - rgl.spheres(lo, col = cc, radius = maxchi, alpha = alpha) + rglplot(g1,vertex.label = vire.nonascii(rownames(x)),vertex.label.color= cc,vertex.label.cex = maxchi, vertex.size = 0.1, layout=lo, rescale=FALSE) + text3d(lo[,1], lo[,2],lo[,3], rownames(x), cex=maxchi, col=cc) + #rgl.spheres(lo, col = cc, radius = maxchi, alpha = alpha) rgl.bg(color = c('white','black')) if (!is.null(movie)) { require(tcltk) diff --git a/chemins.py b/chemins.py index e448064..18ce8e8 100644 --- a/chemins.py +++ b/chemins.py @@ -146,6 +146,8 @@ def ConstructDicoPath(AppliPath): 'swedish_exp' : os.path.join(BasePath, 'expression_sw.txt'), 'portuguese' : os.path.join(BasePath, 'lexique_pt.txt'), 'portuguese_exp': os.path.join(BasePath, 'expression_pt.txt'), + 'greek' : os.path.join(BasePath, 'lexique_gr.txt'), + 'greek_exp' : os.path.join(BasePath, 'expression_gr.txt'), } return DictPath diff --git a/configuration/corpus.cfg b/configuration/corpus.cfg index 52491e6..9452f14 100644 --- a/configuration/corpus.cfg +++ b/configuration/corpus.cfg @@ -6,7 +6,7 @@ encoding = utf8 lang = french douce = 1 ucemethod = 1 -ucesize = 35 +ucesize = 40 keep_ponct = 0 tolist = 0 etoile = 1 diff --git a/corpusNG.py b/corpusNG.py index 6a027dc..7354463 100644 --- a/corpusNG.py +++ b/corpusNG.py @@ -512,18 +512,21 @@ class Corpus : et = etoile.split('_') if et[0] in det : try : - if et[1] in det[et[0]] : - det[et[0]][et[1]] += 1 + endet = '_'.join(et[1:]) + if endet in det[et[0]] : + det[et[0]][endet] += 1 else : - det[et[0]][et[1]] = 1 + det[et[0]][endet] = 1 except IndexError : det[et[0]] += 1 else : try : - det[et[0]] = {et[1] :1} + endet = '_'.join(et[1:]) + det[et[0]] = {endet :1} except IndexError : det[et[0]] = 1 print det + return det def make_and_write_profile_et(self, ucecl, fileout) : log.info('etoiles/classes') @@ -906,8 +909,9 @@ class BuildCorpus : self.cleans.append(self.dolower) if self.corpus.parametres.get('firstclean', 1) : self.cleans.append(self.firstclean) - self.rule = self.corpus.parametres.get('keep_caract', u"^a-zA-Z0-9àÀâÂäÄáÁéÉèÈêÊëËìÌîÎïÏòÒôÔöÖùÙûÛüÜçÇßœŒ’ñ.:,;!?*'_") - self.cleans.append(self.docharact) + if self.corpus.parametres['charact'] : + self.rule = self.corpus.parametres.get('keep_caract', u"^a-zA-Z0-9àÀâÂäÄáÁéÉèÈêÊëËìÌîÎïÏòÒôÔöÖùÙûÛüÜçÇßœŒ’ñ.:,;!?*'_") + self.cleans.append(self.docharact) if self.corpus.parametres.get('expressions', 1) : self.cleans.append(self.make_expression) if self.corpus.parametres.get('apos', 1) : @@ -1047,7 +1051,11 @@ class BuildFromAlceste(BuildCorpus) : ucetxt = txt.split('laphrasepoursplitter') else : txt = ' '.join(txt) + + print txt txt = self.make_cleans(txt) + + print txt ucetxt = self.make_uces(txt, self.corpus.parametres['douce']) if self.corpus.ucis[-1].paras == [] : idpara += 1 diff --git a/dialog.py b/dialog.py index 44b7dc8..1bdeedb 100755 --- a/dialog.py +++ b/dialog.py @@ -2550,6 +2550,21 @@ class PrefWordCloud ( wx.Dialog ): self.m_staticline6 = wx.StaticLine( self, wx.ID_ANY, wx.DefaultPosition, wx.DefaultSize, wx.LI_HORIZONTAL ) fgSizer1.Add( self.m_staticline6, 0, wx.EXPAND |wx.ALL, 5 ) + self.typeformes = wx.StaticText( self, wx.ID_ANY, u"formes retenues", wx.DefaultPosition, wx.DefaultSize, 0 ) + self.typeformes.Wrap( -1 ) + fgSizer1.Add( self.typeformes, 0, wx.ALIGN_CENTER_VERTICAL|wx.ALL, 5 ) + + typeformeschoiceChoices = [ u"actives", u"supplémentaires", u"actives et supplémentaires" ] + self.typeformeschoice = wx.Choice( self, wx.ID_ANY, wx.DefaultPosition, wx.DefaultSize, typeformeschoiceChoices, 0 ) + self.typeformeschoice.SetSelection( 0 ) + fgSizer1.Add( self.typeformeschoice, 0, wx.ALIGN_CENTER_VERTICAL|wx.ALL, 5 ) + + self.m_staticline11 = wx.StaticLine( self, wx.ID_ANY, wx.DefaultPosition, wx.DefaultSize, wx.LI_HORIZONTAL ) + fgSizer1.Add( self.m_staticline11, 0, wx.EXPAND |wx.ALL, 5 ) + + self.m_staticline12 = wx.StaticLine( self, wx.ID_ANY, wx.DefaultPosition, wx.DefaultSize, wx.LI_HORIZONTAL ) + fgSizer1.Add( self.m_staticline12, 0, wx.EXPAND |wx.ALL, 5 ) + self.m_staticText6 = wx.StaticText( self, wx.ID_ANY, u"Taille du texte", wx.DefaultPosition, wx.DefaultSize, 0 ) self.m_staticText6.Wrap( -1 ) fgSizer1.Add( self.m_staticText6, 0, wx.ALIGN_CENTER_VERTICAL|wx.ALIGN_LEFT|wx.ALL, 5 ) @@ -2767,8 +2782,8 @@ class CorpusPref ( wx.Dialog ): def __init__( self, parent, parametres ): wx.Dialog.__init__ ( self, parent, id = wx.ID_ANY, title = u"Préférences", pos = wx.DefaultPosition, size = wx.DefaultSize, style = wx.DEFAULT_DIALOG_STYLE ) self.parent = parent - langues_n = [u'français', u'english', u'german (expérimentale)', u'italian (expérimentale)', u'swedish (exp.)', u'portuguese (exp.)'] - self.langues = [u'french', u'english', u'german', 'italian', 'swedish', u'portuguese'] + langues_n = [u'français', u'english', u'german (expérimentale)', u'italian (expérimentale)', u'swedish (exp.)', u'portuguese (exp.)', 'greek', 'other...'] + self.langues = [u'french', u'english', u'german', 'italian', 'swedish', u'portuguese', 'greek', 'other'] self.encodages = [enc[0].lower() for enc in encodages] ucimark = [u'****', u'0000'] @@ -2930,7 +2945,7 @@ class CorpusPref ( wx.Dialog ): self.check_ponct = wx.CheckBox( self.m_panel2, wx.ID_ANY, wx.EmptyString, wx.DefaultPosition, wx.DefaultSize, 0 ) fgSizer3.Add( self.check_ponct, 0, wx.ALL, 5 ) - self.m_staticText16 = wx.StaticText( self.m_panel2, wx.ID_ANY, u"Pas d'espace entre deux formes (pour le chinois par exemple)", wx.DefaultPosition, wx.DefaultSize, 0 ) + self.m_staticText16 = wx.StaticText( self.m_panel2, wx.ID_ANY, u"Pas d'espace entre deux formes", wx.DefaultPosition, wx.DefaultSize, 0 ) self.m_staticText16.Wrap( -1 ) fgSizer3.Add( self.m_staticText16, 0, wx.ALIGN_CENTER_VERTICAL|wx.ALIGN_LEFT|wx.ALL, 5 ) diff --git a/guifunct.py b/guifunct.py index f7e6249..981f58b 100644 --- a/guifunct.py +++ b/guifunct.py @@ -9,6 +9,7 @@ import sys from copy import copy import dialog from listlex import * +from functions import treat_var_mod def OnOpen(self, type): @@ -156,6 +157,21 @@ class PrepSimi : self.dial.CenterOnParent() self.val = self.dial.ShowModal() if self.val == wx.ID_OK : + if self.dial.check_bystar.GetValue() : + variables = treat_var_mod(self.parametres['stars']) + vardial = dialog.OptLexi(parent) + vardial.listet = self.parametres['stars'] + vardial.variables = [v for v in variables] + for et in vardial.variables : + vardial.list_box_1.Append(et) + nval = vardial.ShowModal() + if nval == wx.ID_OK : + if vardial.choice.GetSelection() == 1 : + listet = [vardial.listet[i] for i in dial.list_box_1.GetSelections()] + else : + listet = variables[vardial.variables[dial.list_box_1.GetSelections()[0]]] + dial.Destroy() + self.tableau.etline = self.Source.corpus.make_etline(listet) self.make_param() def make_param(self) : diff --git a/iramuteq.py b/iramuteq.py index 3ebb9dd..699bde5 100644 --- a/iramuteq.py +++ b/iramuteq.py @@ -13,7 +13,6 @@ parser.add_option("-f", "--file", dest="filename", #print args #print options - import sys reload(sys) import locale @@ -104,8 +103,9 @@ if sys.platform == 'darwin' : else : sys.setdefaultencoding(locale.getpreferredencoding()) #chemin de l'application - +#print 'ATTENTION APPLIPATH' AppliPath = os.path.abspath(os.path.dirname(os.path.realpath(sys.argv[0]))) +#AppliPath = '/home/pierre/workspace/iramuteqNG' #chemin des images ImagePath = os.path.join(AppliPath, 'images') #configuration generale @@ -1123,7 +1123,6 @@ class MyApp(wx.App): Create and show the splash screen. It will then create and show the main frame when it is time to do so. """ - wx.SystemOptions.SetOptionInt("mac.window-plain-transition", 1) self.SetAppName("Iramuteq") splash = MySplashScreen() diff --git a/layout.py b/layout.py index 642121c..0211668 100644 --- a/layout.py +++ b/layout.py @@ -952,7 +952,7 @@ class SimiLayout(DefaultTextLayout) : if self.actives is None : with codecs.open(self.pathout['actives.csv'], 'r', self.parametres['encoding']) as f : self.actives = f.read() - self.actives = [act for act in self.actives.splitlines()] + self.actives = self.actives.splitlines()#[act for act in self.actives.splitlines()] 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'], selected = selected) prep = PrepSimi(self.ira, self.parametres, indices_simi) diff --git a/listlex.py b/listlex.py index 43f4564..4b417e3 100644 --- a/listlex.py +++ b/listlex.py @@ -168,12 +168,19 @@ class ListForSpec(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.ColumnSor self.PopupMenu(menu) menu.Destroy() + def getselectedwords(self) : + words = [self.getColumnText(self.GetFirstSelected(), 6)] + last = self.GetFirstSelected() + while self.GetNextSelected(last) != -1: + last = self.GetNextSelected(last) + words.append(self.getColumnText(last, 1)) + return words def OnPopupOne(self, event): activenotebook = self.parent.nb.GetSelection() page = self.parent.nb.GetPage(activenotebook) corpus = page.corpus - word = self.getColumnText(self.GetFirstSelected(), 0) + word = self.getselectedwords()[0] lems = corpus.getlems() rep = [] for forme in lems[word].formes : diff --git a/search_list.py b/search_list.py index eb80f07..65b83c4 100644 --- a/search_list.py +++ b/search_list.py @@ -120,6 +120,11 @@ class SearchList(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.ColumnSort s = self.itemDataMap[index][col] return s + def OnGetItemData(self, item) : + index = self.itemIndexMap[item] + s = self.itemDataMap[index] + return s + def OnItemSelected(self, event): self.currentItem = event.m_itemIndex event.Skip() @@ -180,6 +185,13 @@ class SearchList(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.ColumnSort self.PopupMenu(menu) menu.Destroy() + def getselectedwords(self) : + words = [self.getColumnText(self.GetFirstSelected(), 6)] + last = self.GetFirstSelected() + while self.GetNextSelected(last) != -1: + last = self.GetNextSelected(last) + words.append(self.getColumnText(last, 6)) + return words def OnPopupOne(self, event): activenotebook = self.parent.nb.GetSelection() @@ -248,11 +260,11 @@ class SearchList(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.ColumnSort win.Show(True) def OnPopupThree(self, event) : - datas = [self.dlist[self.GetItemData(self.GetFirstSelected())]] + datas = [self.OnGetItemData(self.GetFirstSelected())] last = self.GetFirstSelected() while self.GetNextSelected(last) != -1: last = self.GetNextSelected(last) - data = self.dlist[self.GetItemData(last)] + data = self.OnGetItemData(last) datas += [data] colnames = self.first[2:] rownames = [val[1] for val in datas] @@ -271,11 +283,11 @@ class SearchList(wx.ListCtrl, listmix.ListCtrlAutoWidthMixin, listmix.ColumnSort def ongraphdendro(self, evt) : corpus = self.parent.corpus - datas = [self.dlist[self.GetItemData(self.GetFirstSelected())]] + datas = [self.OnGetItemData(self.GetFirstSelected())] last = self.GetFirstSelected() while self.GetNextSelected(last) != -1: last = self.GetNextSelected(last) - data = self.dlist[self.GetItemData(last)] + data = self.OnGetItemData(last) datas += [data] colnames = self.first[2:] rownames = [val[1] for val in datas] diff --git a/textaslexico.py b/textaslexico.py index d78c8fd..4982cbd 100644 --- a/textaslexico.py +++ b/textaslexico.py @@ -278,15 +278,15 @@ class Lexico(AnalyseText) : mineff = self.parametres['mineff'] #dlg = progressbar(self, maxi = 3) tabout = self.corpus.make_lexitable(mineff, self.listet) - log.warning('Fmax a 200') - Fmax = [line for line in tabout[1:] if sum(line[1:]) > 199] - 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) - tabout = [line for line in tabout if line[0] not in formesmax] - log.warning('ATTENTION : hapax par etoile') - tabout.append(['hapax'] + self.corpus.gethapaxbyet(self.listet)) + #log.warning('Fmax a 200') + #Fmax = [line for line in tabout[1:] if sum(line[1:]) > 199] + #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) + #tabout = [line for line in tabout if line[0] not in formesmax] + #log.warning('ATTENTION : hapax par etoile') + #tabout.append(['hapax'] + self.corpus.gethapaxbyet(self.listet)) write_tab(tabout, self.dictpathout['tableafcm']) #log.warning('ATTENTION : gethapaxuces') diff --git a/textsimi.py b/textsimi.py index c8d1d62..8b296bb 100644 --- a/textsimi.py +++ b/textsimi.py @@ -110,6 +110,7 @@ class SimiTxt(AnalyseText): self.listet = self.corpus.make_etoiles() self.listet.sort() self.parametres['stars'] = copy(self.listet) + self.stars = copy(self.listet) self.parametres['sfromchi'] = False diff --git a/textwordcloud.py b/textwordcloud.py index 01ccf9a..199d4a6 100644 --- a/textwordcloud.py +++ b/textwordcloud.py @@ -27,15 +27,16 @@ class WordCloud(AnalyseText): self.parametres['type'] = 'wordcloud' #FIXME limit = 3 - mode = 'both' - if mode == 'both' : + self.make_option() + if self.parametres['mode'] == 2 : self.actives = self.corpus.make_actives_limit(limit, 1) self.actives += self.corpus.make_actives_limit(limit, 2) - elif mode == 'act' : + elif self.parametres['mode'] == 0 : self.actives = self.corpus.make_actives_limit(limit, 1) + elif self.parametres['mode'] == 1 : + self.actives = self.corpus.make_actives_limit(limit, 2) 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']) - self.make_option() + SelectColumn(self.ira, dictcol, self.actives, self.pathout['selected.csv'], dlg = True) self.make_wordcloud() script = WordCloudRScript(self) script.make_script() @@ -48,12 +49,12 @@ class WordCloud(AnalyseText): if res == wx.ID_OK : self.parametres['width'] = dial.spin_L.GetValue() self.parametres['height'] = dial.spin_H.GetValue() - #maxword = dial.spin_maxword.GetValue() self.parametres['maxword'] = dial.spin_maxword.GetValue() self.parametres['mincex'] = float(dial.spin_mincex.GetValue())/float(10) self.parametres['maxcex'] = float(dial.spin_maxcex.GetValue())/float(10) self.parametres['col_text'] = dial.color_text.GetColour() self.parametres['col_bg'] = dial.color_bg.GetColour() + self.parametres['mode'] = dial.typeformeschoice.GetSelection() outgraph = os.path.join(os.path.dirname(self.pathout['zipf.png']), 'nuage_') nb = 1 while os.path.exists(outgraph + str(nb) + '.png') :