2 # -*- coding: utf-8 -*-
3 #Author: Pierre Ratinaud
4 #Copyright (c) 2008-2012 Pierre Ratinaud
9 from ConfigParser import ConfigParser
10 from subprocess import Popen, call, PIPE
21 from shutil import copyfile
22 #from dialog import BugDialog
23 print 'TEST LOGGING funcion'
26 log = logging.getLogger('iramuteq')
29 indices_simi = [u'cooccurrence' ,'pourcentage de cooccurrence',u'Russel',u'Jaccard', 'Kulczynski1', 'Kulczynski2', 'Mountford', 'Fager', 'simple matching', 'Hamman', 'Faith', 'Tanimoto', 'Dice', 'Phi', 'Stiles', 'Michael', 'Mozley', 'Yule', 'Yule2', 'Ochiai', 'Simpson', 'Braun-Blanquet','Chi-squared', 'Phi-squared', 'Tschuprow', 'Cramer', 'Pearson', 'binomial']
33 def __init__(self, filein, syscoding = 'utf8') :
35 self.syscoding = syscoding
37 self.openedcorpus = {}
44 self.conf = DoConf(self.filein)
47 for i, section in enumerate(self.conf.conf.sections()) :
48 if self.conf.conf.has_option(section, 'corpus_name') :
49 self.corpora[section] = self.conf.getoptions(section)
50 self.order[len(self.order)] = section
52 self.analyses[section] = self.conf.getoptions(section)
53 self.ordera[len(self.ordera)] = section
55 for corpus in self.corpora :
56 self.history[corpus] = copy(self.corpora[corpus])
57 for analyse in self.analyses :
58 if self.analyses[analyse]['corpus'] in self.corpora :
59 if 'analyses' in self.history[self.analyses[analyse]['corpus']] :
60 self.history[self.analyses[analyse]['corpus']]['analyses'].append(self.analyses[analyse])
63 self.history[self.analyses[analyse]['corpus']]['analyses'] = [self.analyses[analyse]]
66 self.history[analyse] = self.analyses[analyse]
67 #for analyse in todel :
68 # del self.analyses[analyse]
71 sections = self.corpora.keys() + self.analyses.keys()
72 parametres = [self.corpora[key] for key in self.corpora.keys()] + [self.analyses[key] for key in self.analyses.keys()]
73 self.conf.makeoptions(sections, parametres)
75 def add(self, analyse) :
76 if 'corpus' in analyse :
77 if analyse['corpus'] in self.corpora :
78 if 'analyses' in self.history[analyse['corpus']] :
79 self.history[analyse['corpus']]['analyses'].append(analyse)
81 self.history[analyse['corpus']]['analyses'] = [analyse]
82 self.analyses[analyse['uuid']] = analyse
84 self.analyses[analyse['uuid']] = analyse
85 elif 'corpus_name' in analyse :
86 self.history[analyse['uuid']] = analyse
87 self.corpora[analyse['uuid']] = analyse
90 def addtab(self, analyse) :
91 self.opened[analyse['uuid']] = analyse
93 def rmtab(self, analyse) :
94 del self.opened[analyse['uuid']]
97 def __init__(self, configfile=None, diff = None, parametres = None) :
98 self.configfile = configfile
99 self.conf = ConfigParser()
100 if configfile is not None :
101 self.conf.readfp(codecs.open(configfile, 'r', 'utf8'))
103 if parametres is not None :
104 self.doparametres(parametres)
106 def doparametres(self, parametres) :
109 def getsections(self) :
110 return self.conf.sections()
112 def getoptions(self, section = None, diff = None):
115 section = self.conf.sections()[0]
116 for option in self.conf.options(section) :
117 if self.conf.get(section, option).isdigit() :
118 parametres[option] = int(self.conf.get(section, option))
120 parametres[option] = self.conf.get(section, option)
121 if 'type' not in parametres :
122 parametres['type'] = section
125 def makeoptions(self, sections, parametres, outfile = None) :
126 for i, section in enumerate(sections) :
127 if not self.conf.has_section(section) :
128 self.conf.add_section(section)
129 for option in parametres[i] :
130 if isinstance(parametres[i][option], int) :
131 self.conf.set(section, option, `parametres[i][option]`)
132 elif isinstance(parametres[i][option], basestring) :
133 self.conf.set(section, option, parametres[i][option].encode('utf8'))
135 outfile = self.configfile
136 with codecs.open(outfile, 'w', 'utf8') as f :
139 def totext(self, parametres) :
141 for val in parametres :
142 if isinstance(parametres[val], int) :
143 txt.append(' \t\t: '.join([val, `parametres[val]`]))
145 txt.append(' \t\t: '.join([val, parametres[val]]))
146 return '\n'.join(txt)
149 def write_tab(tab, fileout) :
150 writer = csv.writer(open(fileout, 'wb'), delimiter=';', quoting = csv.QUOTE_NONNUMERIC)
151 writer.writerows(tab)
153 class BugDialog(wx.Dialog):
154 def __init__(self, *args, **kwds):
155 # begin wxGlade: MyDialog.__init__
156 kwds["style"] = wx.DEFAULT_DIALOG_STYLE
157 kwds["size"] = wx.Size(500, 200)
158 wx.Dialog.__init__(self, *args, **kwds)
159 self.text_ctrl_1 = wx.TextCtrl(self, -1, "", style=wx.TE_MULTILINE)
160 self.button_1 = wx.Button(self, wx.ID_OK, "")
162 self.__set_properties()
166 def __set_properties(self):
167 # begin wxGlade: MyDialog.__set_properties
169 self.SetMinSize(wx.Size(500, 200))
170 self.text_ctrl_1.SetMinSize(wx.Size(500, 200))
174 def __do_layout(self):
175 # begin wxGlade: MyDialog.__do_layout
176 sizer_1 = wx.BoxSizer(wx.VERTICAL)
177 sizer_1.Add(self.text_ctrl_1, 1, wx.EXPAND, 0)
178 sizer_1.Add(self.button_1, 0, wx.ALIGN_CENTER_HORIZONTAL, 0)
179 self.SetSizer(sizer_1)
184 def CreateIraFile(DictPathOut, clusternb, corpname='corpus_name', section = 'analyse'):
185 AnalyseConf = ConfigParser()
186 AnalyseConf.read(DictPathOut['ira'])
187 AnalyseConf.add_section(section)
188 date = datetime.datetime.now().ctime()
189 AnalyseConf.set(section, 'date', str(date))
190 AnalyseConf.set(section, 'clusternb', clusternb)
191 AnalyseConf.set(section, 'corpus_name', corpname)
193 fileout = open(DictPathOut['ira'], 'w')
194 AnalyseConf.write(fileout)
197 def sortedby(list, direct, *indices):
200 sortedby: sort a list of lists (e.g. a table) by one or more indices
201 (columns of the table) and return the sorted list
204 for list = [[2,3],[1,2],[3,1]]:
205 sortedby(list,1) will return [[3, 1], [1, 2], [2, 3]],
206 sortedby(list,0) will return [[1, 2], [2, 3], [3, 1]]
209 nlist = map(lambda x, indices=indices:
210 map(lambda i, x=x: x[i], indices) + [x],
215 nlist.sort(reverse=True)
216 return map(lambda l: l[-1], nlist)
218 def add_type(line, dictlem):
219 if line[4] in dictlem:
220 line.append(dictlem[line[4]])
225 def treat_line_alceste(i, line) :
226 if line[0] == '*' or line[0] == '*****' :
231 elif float(line[5].replace(',', '.')) < 0.0001:
233 elif float(line[5].replace(',', '.')) > 0.05:
234 line[5] = 'NS (%s)' % str(float(line[5].replace(',', '.')))[0:7]
236 line[5] = str(float(line[5].replace(',', '.')))[0:7]
237 return [i, int(line[0]), int(line[1]), float(line[2]), float(line[3]), line[6], line[4], line[5]]
239 def ReadProfileAsDico(parent, File, Alceste=False, encoding = sys.getdefaultencoding()):
240 #print 'lecture des profils : ReadProfileAsDico'
242 # print 'lecture du dictionnaire de type'
244 # for line in parent.corpus.lem_type_list :
245 # dictlem[line[0]] = line[1]
247 print 'lecture des profiles'
248 #encoding = sys.getdefaultencoding()
250 FileReader = codecs.open(File, 'r', encoding)
251 Filecontent = FileReader.readlines()
255 rows = [row.replace('\n', '').replace("'", '').replace('\"', '').replace(',', '.').replace('\r','').split(';') for row in Filecontent]
257 ClusterNb = rows[0][2]
259 clusters = [row[2] for row in rows if row[0] == u'**']
260 valclusters = [row[1:4] for row in rows if row[0] == u'****']
261 lp = [i for i, line in enumerate(rows) if line[0] == u'****']
262 prof = [rows[lp[i] + 1:lp[i+1] - 1] for i in range(0, len(lp)-1)] + [rows[lp[-1] + 1:len(rows)]]
264 prof = [[add_type(row, dictlem) for row in pr] for pr in prof]
265 prof = [[treat_line_alceste(i,line) for i, line in enumerate(pr)] for pr in prof]
267 prof = [[line + [''] for line in pr] for pr in prof]
268 prof = [[treat_line_alceste(i,line) for i, line in enumerate(pr)] for pr in prof]
269 for i, cluster in enumerate(clusters):
270 DictProfile[cluster] = [valclusters[i]] + prof[i]
273 def GetTxtProfile(dictprofile) :
275 for classe in range(0, len(dictprofile)) :
276 prof = dictprofile[str(classe + 1)]
278 proflist.append('\n'.join([' '.join(['classe %i' % (classe + 1), '-', '%s uce sur %s - %s%%' % (clinfo[0], clinfo[1], clinfo[2])]), '\n'.join(['%5s|%5s|%6s|%6s|%8s|%8s|%20s\t%10s' % tuple([str(val) for val in line]) for line in prof if len(line)==8])]))
279 return '\n\n'.join(proflist)
281 def formatExceptionInfo(maxTBlevel=5):
282 cla, exc, trbk = sys.exc_info()
283 excName = cla.__name__
285 excArgs = exc.__dict__["args"]
287 excArgs = "<no args>"
288 excTb = traceback.format_tb(trbk, maxTBlevel)
289 return (excName, excArgs, excTb)
292 #fonction des etudiants de l'iut
293 def decoupercharact(chaine, longueur, longueurOptimale, separateurs = None) :
295 on part du dernier caractère, et on recule jusqu'au début de la chaîne.
296 Si on trouve un '$', c'est fini.
297 Sinon, on cherche le meilleur candidat. C'est-Ã -dire le rapport poids/distance le plus important.
299 separateurs = [[u'.', 60.0], [u'?', 60.0], [u'!', 60.0], [u'£', 60], [u':', 50.0], [u';', 40.0], [u',', 10.0], [u' ', 0.1]]
300 trouve = False # si on a trouvé un bon séparateur
301 iDecoupe = 0 # indice du caractere ou il faut decouper
303 # on découpe la chaine pour avoir au maximum 240 caractères
304 longueur = min(longueur, len(chaine) - 1)
305 chaineTravail = chaine[:longueur + 1]
307 meilleur = ['', 0, 0] # type, poids et position du meilleur separateur
309 # on vérifie si on ne trouve pas un '$'
310 indice = chaineTravail.find(u'$')
315 # si on ne trouve rien, on cherche le meilleur séparateur
318 caractere = chaineTravail[nbCar]
319 distance = abs(longueurOptimale - nbCar) + 1
320 meilleureDistance = abs(longueurOptimale - meilleur[2]) + 1
322 # on vérifie si le caractére courant est une marque de ponctuation
323 for s in separateurs:
324 if caractere == s[0]:
325 # si c'est une ponctuation
327 if s[1] / distance > float(meilleur[1]) / meilleureDistance:
335 # et on termine la recherche
338 # on passe au caractère précédant
343 fin = chaine[iDecoupe + 1:]
344 retour = chaineTravail[:iDecoupe]
345 return len(retour) > 0, retour.split(), fin
346 # si on a rien trouvé
347 return False, chaine.split(), ''
349 def BugReport(parent):
350 for ch in parent.GetChildren():
351 if "<class 'wx._windows.ProgressDialog'>" == str(type(ch)):
353 dial = BugDialog(parent)
354 txt = u' !== BUG ==! \n'
355 txt += u'*************************************\n'
356 for line in formatExceptionInfo():
357 if type(line) == type([]):
359 txt += don.replace(' ', ' ')
362 if 'Rerror' in dir(parent) :
365 print formatExceptionInfo()
367 dial.text_ctrl_1.write(txt)
368 dial.CenterOnParent()
370 raise NameError('Bug')
372 def PlaySound(parent):
373 if parent.pref.getboolean('iramuteq', 'sound') :
375 if "gtk2" in wx.PlatformInfo:
376 error = Popen(['aplay','-q',os.path.join(parent.AppliPath,'son_fin.wav')])
378 sound = wx.Sound(os.path.join(parent.AppliPath, 'son_fin.wav'))
379 sound.Play(wx.SOUND_SYNC)
383 def ReadDicoAsDico(dicopath):
384 with codecs.open(dicopath, 'r', 'UTF8') as f:
385 content = f.readlines()
387 for line in content :
389 line = line.replace(u'\n', '').replace('"', '').split('\t')
390 dico[line[0]] = line[1:]
393 def ReadLexique(parent, lang = 'french'):
394 parent.lexique = ReadDicoAsDico(parent.DictPath.get(lang, 'french'))
396 def ReadList(filein, encoding = sys.getdefaultencoding()):
398 file = codecs.open(filein, 'r', encoding)
399 content = file.readlines()
401 first = content.pop(0)
402 first = first.replace('\n', '').replace('\r','').replace('\"', '').split(';')
406 line = line.replace('\n', '').replace('\r','').replace('\"', '').replace(',', '.')
407 line = line.split(';')
416 don = float('%.5f' % float(val))
422 def exec_rcode(rpath, rcode, wait = True, graph = False):
425 if sys.platform == 'darwin' :
427 macversion = platform.mac_ver()[0].split('.')
429 if int(macversion[1]) < 5 :
436 rpath = rpath.replace('\\','\\\\')
439 if sys.platform == 'win32':
440 error = call(["%s" % rpath, "--vanilla","--slave","-f", "%s" % rcode])
442 error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
445 if sys.platform == 'win32':
446 pid = Popen(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
448 pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
452 if sys.platform == 'win32':
453 error = call(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
454 elif sys.platform == 'darwin' and needX11:
455 os.environ['DISPLAY'] = ':0.0'
456 error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
458 error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
461 if sys.platform == 'win32':
462 pid = Popen(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
463 elif sys.platform == 'darwin' and needX11:
464 os.environ['DISPLAY'] = ':0.0'
465 pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
467 pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
470 def check_Rresult(parent, pid) :
471 if isinstance(pid, Popen) :
472 if pid.returncode != 0 :
473 error = pid.communicate()
474 error = [str(error[0]), error[1]]
475 if error[1] is None :
477 parent.Rerror = '\n'.join([str(pid.returncode), '\n'.join(error)])
479 raise Exception('\n'.join(u'Erreur R', '\n'.join(error[1:])))
485 raise Exception(u'Erreur R')
489 def print_liste(filename,liste):
490 with open(filename,'w') as f :
492 f.write(';'.join(graph)+'\n')
494 def read_list_file(filename, encoding = sys.getdefaultencoding()):
495 with codecs.open(filename,'rU', encoding) as f :
496 content=f.readlines()
497 ncontent=[line.replace('\n','').split(';') for line in content if line.strip() != '']
500 class MessageImage(wx.Frame):
501 def __init__(self, *args, **kwds):
502 # begin wxGlade: MyFrame.__init__
503 kwds["style"] = wx.DEFAULT_FRAME_STYLE
504 wx.Frame.__init__(self, *args, **kwds)
505 #self.text_ctrl_1 = wx.TextCtrl(self, -1, "", style=wx.TE_MULTILINE)
506 self.imageFile = False
507 self.imagename = u"chi_classe.png"
508 self.HtmlPage = wx.html.HtmlWindow(self, -1)
509 if "gtk2" in wx.PlatformInfo:
510 self.HtmlPage.SetStandardFonts()
511 self.HtmlPage.SetFonts('Courier', 'Courier')
513 self.button_1 = wx.Button(self, -1, u"Fermer")
514 self.Bind(wx.EVT_BUTTON, self.OnCloseMe, self.button_1)
515 self.Bind(wx.EVT_CLOSE, self.OnCloseWindow)
519 def __do_layout(self):
520 # begin wxGlade: MyFrame.__do_layout
521 sizer_1 = wx.BoxSizer(wx.VERTICAL)
522 self.sizer_2 = wx.BoxSizer(wx.VERTICAL)
523 self.sizer_2.Add(self.HtmlPage, 1, wx.EXPAND | wx.ADJUST_MINSIZE, 0)
524 self.sizer_2.Add(self.button_1, 0, wx.ALIGN_CENTER_HORIZONTAL | wx.ADJUST_MINSIZE, 0)
525 sizer_1.Add(self.sizer_2, 1, wx.EXPAND, 0)
526 self.SetAutoLayout(True)
527 self.SetSizer(sizer_1)
530 def addsaveimage(self, imageFile) :
531 self.imageFile = imageFile
532 self.button_2 = wx.Button(self, -1, u"Enregistrer l'image...")
533 self.Bind(wx.EVT_BUTTON, self.OnSaveImage, self.button_2)
534 self.sizer_2.Add(self.button_2, 0, wx.ALIGN_CENTER_HORIZONTAL | wx.ADJUST_MINSIZE, 0)
537 def OnCloseMe(self, event):
540 def OnCloseWindow(self, event):
543 def OnSaveImage(self, event) :
545 self, message="Enregistrer sous...", defaultDir=os.getcwd(),
546 defaultFile= self.imagename, wildcard="png|*.png", style=wx.SAVE | wx.OVERWRITE_PROMPT
548 dlg.SetFilterIndex(2)
550 if dlg.ShowModal() == wx.ID_OK:
552 copyfile(self.imageFile, path)
555 def progressbar(self, maxi) :
556 if 'parent' in dir(self) :
560 return wx.ProgressDialog("Traitements",
561 "Veuillez patienter...",
564 style=wx.PD_APP_MODAL | wx.PD_AUTO_HIDE | wx.PD_ELAPSED_TIME | wx.PD_CAN_ABORT
568 def treat_var_mod(variables) :
570 for variable in variables :
571 if u'_' in variable :
572 forme = variable.split(u'_')
575 if not var in var_mod :
576 var_mod[var] = [variable]
578 if not mod in var_mod[var] :
579 var_mod[var].append(variable)