X-Git-Url: http://iramuteq.org/git?p=iramuteq;a=blobdiff_plain;f=functions.py;h=6a3d9689a4298c5cfbf9bf41d78c9006a80aa886;hp=9b1c8bf0c43e1286d4b27f62e77c9f1dd42947cf;hb=ae32485b960c98387ae7987751198335ad549ab3;hpb=53df4af897991ea260f0b57e4b0f7826f6409860 diff --git a/functions.py b/functions.py old mode 100644 new mode 100755 index 9b1c8bf..6a3d968 --- a/functions.py +++ b/functions.py @@ -21,6 +21,7 @@ import datetime from copy import copy from shutil import copyfile import shelve +import json #from dialog import BugDialog import logging @@ -30,12 +31,21 @@ log = logging.getLogger('iramuteq') 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'] + +def open_folder(folder): + if sys.platform == "win32": + os.startfile(folder) + else: + opener ="open" if sys.platform == "darwin" else "xdg-open" + #call([opener, folder]) + call([u"%s %s &" % (opener, folder)], shell=True) + def normpath_win32(path) : if not sys.platform == 'win32' : return path while '\\\\' in path : path = path.replace('\\\\', '\\') - if sys.platform == 'win32' and path.startswith('\\') and not path.startswith('\\\\') : + if path.startswith('\\') and not path.startswith('\\\\') : path = '\\' + path return path @@ -44,24 +54,26 @@ class TGen : self.path = path self.tgen = {} self.encoding = encoding - + def __getitem__(self, key): return self.tgen[key] - - def read(self, path): + + def read(self, path = None): + if path is None : + path = self.path with codecs.open(path, 'r', self.encoding) as f : tgen = f.read() tgen = [line.split('\t') for line in tgen.splitlines()] tgen = dict([[line[0], line[1:]] for line in tgen]) self.tgen = tgen self.path = path - + def write(self, path = None): if path is None : path = self.path with open(path, 'w') as f : f.write('\n'.join(['\t'.join([val] + self.tgen[val]) for val in self.tgen]).encode(self.encoding)) - + def writetable(self, pathout, tgens, totocc): etoiles = totocc.keys() etoiles.sort() @@ -102,13 +114,18 @@ class History : self.matrixanalyse = dict([[mat['uuid'], mat] for mat in self.matrix]) self.ordermatrix = dict([[matrix['uuid'], i] for i, matrix in enumerate(self.matrix)]) d.close() + d = {} + d['history'] = self.history + d['matrix'] = self.matrix +# with open('/home/pierre/hystory.json', 'w') as f : +# f.write(json.dumps(d, indent=4, default=str)) def write(self) : d = shelve.open(self.filein) d['history'] = self.history d['matrix'] = self.matrix d.close() - + def add(self, analyse) : log.info('add to history %s' % analyse.get('corpus_name', 'pas un corpus')) tosave = {'uuid' : analyse['uuid'], 'ira': analyse['ira'], 'type' : analyse['type']} @@ -151,7 +168,7 @@ class History : self.matrix[self.ordermatrix[tosave['matrix']]]['analyses'].append(tosave) self.write() self.read() - + def addmultiple(self, analyses) : log.info('add multiple') for analyse in analyses : @@ -179,6 +196,11 @@ class History : self.history[self.ordercorpus[analyse['corpus']]]['analyses'].pop(todel) elif analyse['uuid'] in self.matrixanalyse : self.matrix = [mat for mat in self.matrix if mat['uuid'] != analyse['uuid']] + elif analyse.get('matrix', False) in self.matrixanalyse : + analyses = self.matrix[self.ordermatrix[analyse['matrix']]]['analyses'] + topop = [i for i, val in enumerate(analyses) if analyse['uuid'] == val['uuid']][0] + analyses.pop(topop) + self.matrix[self.ordermatrix[analyse['matrix']]]['analyses'] = analyses self.write() self.read() @@ -188,6 +210,19 @@ class History : def rmtab(self, analyse) : del self.opened[analyse['uuid']] + def update(self, analyse) : + if 'matrix_name' in analyse : + self.matrixanalyse[analyse['uuid']].update(analyse) + elif 'corpus_name' in analyse : + self.corpus[analyse['uuid']].update(analyse) + elif 'corpus' in analyse : + self.analyses[analyse['uuid']].update(analyse) + else : + toupdate = [an for an in self.matrixanalyse[analyse['matrix']]['analyses'] if an['uuid'] == analyse['uuid']] + toupdate[0].update(analyse) + self.write() + self.read() + def clean(self) : corpustodel = [corpus for corpus in self.history if not os.path.exists(corpus['ira'])] print corpustodel @@ -198,7 +233,64 @@ class History : for analyse in anatodel : print 'cleaning :', analyse['name'] self.delete(analyse) - + + def dostat(self): + todel = {} + tokens = 0 + corpusnb = {} + subnb = 0 + analysenb = 0 + hours = 0 + minutes = 0 + secondes = 0 + ha = 0 + ma = 0 + sa = 0 + for corpus in self.history : + analysenb += len(corpus.get('analyses', [])) + analyses = corpus.get('analyses', []) + for analyse in analyses : + if os.path.exists(analyse['ira']) : + ana = DoConf(analyse['ira']).getoptions() + if 'time' in ana : + time = ana['time'].split() + ha += int(time[0].replace('h','')) * 3600 + ma += int(time[1].replace('m','')) * 60 + sa += int(time[2].replace('s','')) + if os.path.exists(corpus['ira']) : + param = DoConf(corpus['ira']).getoptions() + time = param.get('time','0h 0m 0s') + time = time.split() + hours += int(time[0].replace('h','')) * 3600 + minutes += int(time[1].replace('m','')) * 60 + secondes += int(time[2].replace('s','')) + if param.get('originalpath', False) : + if param['originalpath'] in corpusnb : + corpusnb[param['originalpath']] += 1 + tokens += int(param['occurrences']) + else : + corpusnb[param['originalpath']] = 1 + #print param + else : + subnb += 1 + else : + if corpus['ira'] in todel : + todel['ira'] += 1 + else : + todel['ira'] = 1 + print u'Nbr total de corpus : %s' % len(self.history) + corpus_nb = len(corpusnb) + len(todel) + print u'Nbr de corpus différents : %s' % corpus_nb + lentodel = len(todel) + print u'Nbr de corpus à supprimer : %s' % lentodel + print u'Nbr de sous corpus : %s' % subnb + print u"Nbr total d'occurrences : %s" % tokens + print u'Moyenne occurrences par corpus : %f' % (tokens/corpus_nb) + print '---------------------' + print u"Nbr total d'analyses : %s" % analysenb + print u'Temps total indexation : %f h' % ((hours+minutes+secondes) / 3600) + print u'Temps total analyses : %f h' % ((ha+ma+sa) / 3600) + def __str__(self) : return str(self.history) @@ -206,7 +298,7 @@ class DoConf : def __init__(self, configfile=None, diff = None, parametres = None) : self.configfile = configfile self.conf = ConfigParser() - + if configfile is not None : configfile = normpath_win32(configfile) self.conf.readfp(codecs.open(configfile, 'r', 'utf8')) @@ -240,7 +332,7 @@ class DoConf : if 'type' not in parametres : parametres['type'] = section return parametres - + def makeoptions(self, sections, parametres, outfile = None) : txt = '' for i, section in enumerate(sections) : @@ -291,7 +383,7 @@ def write_tab(tab, fileout) : class BugDialog(wx.Dialog): def __init__(self, *args, **kwds): # begin wxGlade: MyDialog.__init__ - kwds["style"] = wx.DEFAULT_DIALOG_STYLE + kwds["style"] = wx.DEFAULT_DIALOG_STYLE | wx.STAY_ON_TOP kwds["size"] = wx.Size(500, 200) wx.Dialog.__init__(self, *args, **kwds) self.SetTitle(kwds['title']) @@ -307,7 +399,7 @@ class BugDialog(wx.Dialog): # begin wxGlade: MyDialog.__set_properties self.SetMinSize(wx.Size(500, 200)) self.text_ctrl_1.SetMinSize(wx.Size(500, 200)) - + # end wxGlade def __do_layout(self): @@ -345,7 +437,7 @@ def sortedby(list, direct, *indices): sortedby(list,0) will return [[1, 2], [2, 3], [3, 1]] """ - nlist = map(lambda x, indices=indices: + nlist = map(lambda x, indices=indices: map(lambda i, x=x: x[i], indices) + [x], list) if direct == 1: @@ -376,15 +468,8 @@ def treat_line_alceste(i, line) : return [i, int(line[0]), int(line[1]), float(line[2]), float(line[3]), line[6], line[4], line[5]] def ReadProfileAsDico(File, Alceste=False, encoding = sys.getdefaultencoding()): - #print 'lecture des profils : ReadProfileAsDico' - #if Alceste : - # print 'lecture du dictionnaire de type' - # dictlem = {} - # for line in parent.corpus.lem_type_list : - # dictlem[line[0]] = line[1] dictlem = {} print 'lecture des profiles' - #encoding = sys.getdefaultencoding() FileReader = codecs.open(File, 'r', encoding) Filecontent = FileReader.readlines() FileReader.close() @@ -398,10 +483,10 @@ def ReadProfileAsDico(File, Alceste=False, encoding = sys.getdefaultencoding()): clusters = [row[2] for row in rows if row[0] == u'**'] valclusters = [row[1:4] for row in rows if row[0] == u'****'] lp = [i for i, line in enumerate(rows) if line[0] == u'****'] - prof = [rows[lp[i] + 1:lp[i+1] - 1] for i in range(0, len(lp)-1)] + [rows[lp[-1] + 1:len(rows)]] + prof = [rows[lp[i] + 1:lp[i+1] - 1] for i in range(0, len(lp)-1)] + [rows[lp[-1] + 1:len(rows)]] if Alceste : prof = [[add_type(row, dictlem) for row in pr] for pr in prof] - prof = [[treat_line_alceste(i,line) for i, line in enumerate(pr)] for pr in prof] + prof = [[treat_line_alceste(i,line) for i, line in enumerate(pr)] for pr in prof] else : prof = [[line + [''] for line in pr] for pr in prof] prof = [[treat_line_alceste(i,line) for i, line in enumerate(pr)] for pr in prof] @@ -418,17 +503,17 @@ def GetTxtProfile(dictprofile, cluster_size) : return '\n\n'.join(proflist) def formatExceptionInfo(maxTBlevel=5): - cla, exc, trbk = sys.exc_info() - try : - excName = cla.__name__ - except : - excName = 'None' - try: - excArgs = exc.args[0] - except : - excArgs = "" - excTb = traceback.format_tb(trbk, maxTBlevel) - return (excName, excArgs, excTb) + cla, exc, trbk = sys.exc_info() + try : + excName = cla.__name__ + except : + excName = 'None' + try: + excArgs = exc.args[0] + except : + excArgs = "" + excTb = traceback.format_tb(trbk, maxTBlevel) + return (excName, excArgs, excTb) #fonction des etudiants de l'iut @@ -441,13 +526,13 @@ def decoupercharact(chaine, longueur, longueurOptimale, separateurs = None) : separateurs = [[u'.', 60.0], [u'?', 60.0], [u'!', 60.0], [u'£$£', 60], [u':', 50.0], [u';', 40.0], [u',', 10.0], [u' ', 0.1]] trouve = False # si on a trouvé un bon séparateur iDecoupe = 0 # indice du caractere ou il faut decouper - + # on découpe la chaine pour avoir au maximum 240 caractères longueur = min(longueur, len(chaine) - 1) chaineTravail = chaine[:longueur + 1] nbCar = longueur meilleur = ['', 0, 0] # type, poids et position du meilleur separateur - + # on vérifie si on ne trouve pas un '$' indice = chaineTravail.find(u'$') if indice > -1: @@ -464,8 +549,8 @@ def decoupercharact(chaine, longueur, longueurOptimale, separateurs = None) : # on vérifie si le caractére courant est une marque de ponctuation for s in separateurs: if caractere == s[0]: - # si c'est une ponctuation - + # si c'est une ponctuation + if s[1] / distance > float(meilleur[1]) / meilleureDistance: # print nbCar, s[0] meilleur[0] = s[0] @@ -473,13 +558,13 @@ def decoupercharact(chaine, longueur, longueurOptimale, separateurs = None) : meilleur[2] = nbCar trouve = True iDecoupe = nbCar - + # et on termine la recherche break # on passe au caractère précédant nbCar = nbCar - 1 - + # si on a trouvé if trouve: fin = chaine[iDecoupe + 1:] @@ -493,12 +578,13 @@ exceptions = {'paragrapheOT' : u"Un problème de formatage (présence d'un marqu 'EmptyText' : u"Texte vide (probablement un problème de formatage du corpus). Le problème est apparu à la ligne ", 'CorpusEncoding' : u"Problème d'encodage.", 'TextBeforeTextMark' : u"Problème de formatage : du texte avant le premier marqueur de texte (****). Le problème est survenu à la ligne ", + 'MissingAnalyse' : u'Aucun fichier à cet emplacement :\n', } def BugReport(parent, error = None): for ch in parent.GetChildren(): if "" == str(type(ch)): - ch.Destroy() + ch.Destroy() excName, exc, excTb = formatExceptionInfo() if excName == 'Exception' : print exc @@ -519,7 +605,7 @@ def BugReport(parent, error = None): txt += u'*************************************\n' txt += '\n'.join(excTb).replace(' ', ' ') txt += excName + '\n' - txt += exc + txt += `exc` title = "Bug" dial = BugDialog(parent, **{'title' : title}) @@ -531,13 +617,13 @@ def BugReport(parent, error = None): dial.CenterOnParent() dial.ShowModal() dial.Destroy() - + def PlaySound(parent): if parent.pref.getboolean('iramuteq', 'sound') : try: if "gtk2" in wx.PlatformInfo: error = Popen(['aplay','-q',os.path.join(parent.AppliPath,'son_fin.wav')]) - else : + else : sound = wx.Sound(os.path.join(parent.AppliPath, 'son_fin.wav')) sound.Play(wx.SOUND_SYNC) except : @@ -556,7 +642,10 @@ def ReadLexique(parent, lang = 'french', filein = None): else : parent.lexique = ReadDicoAsDico(filein) else : - parent.lexique = {} + if filein is None : + parent.lexique = {} + else : + parent.lexique = ReadDicoAsDico(filein) def ReadList(filein, encoding = sys.getdefaultencoding(), sep = ';'): #file = open(filein) @@ -577,7 +666,7 @@ def ReadList(filein, encoding = sys.getdefaultencoding(), sep = ';'): for val in line[1:]: if val == u'NA' : don = '' - else: + else: try: don = int(val) except: @@ -668,37 +757,37 @@ def check_Rresult(parent, pid) : else : return True + +def launchcommand(mycommand): + Popen(mycommand) + def print_liste(filename,liste): with open(filename,'w') as f : for graph in liste : - f.write(';'.join(graph)+'\n') + f.write(';'.join(graph).encode(sys.getdefaultencoding(), errors='replace')+'\n') def read_list_file(filename, encoding = sys.getdefaultencoding()): with codecs.open(filename,'rU', encoding) as f : content=f.readlines() ncontent=[line.replace('\n','').split(';') for line in content if line.strip() != ''] return ncontent - - - def progressbar(self, maxi) : - if 'parent' in dir(self) : - parent = self.parent - else : - parent = self + ira = wx.GetApp().GetTopWindow() + parent = ira try : - print '###horrible hack progressbar' maxi = int(maxi) except : maxi = 1 - return wx.ProgressDialog("Traitements", + prog = wx.ProgressDialog("Traitements", "Veuillez patienter...", maximum=maxi, parent=parent, style=wx.PD_APP_MODAL | wx.PD_AUTO_HIDE | wx.PD_ELAPSED_TIME | wx.PD_CAN_ABORT ) - + prog.SetSize((400,150)) + #prog.SetIcon(ira._icon) + return prog def treat_var_mod(variables) : var_mod = {} @@ -708,7 +797,7 @@ def treat_var_mod(variables) : for var in vars : mods = ['_'.join(v) for v in varmod if v[0] == var] var_mod[var] = mods - + # for variable in variables : # if u'_' in variable : # forme = variable.split(u'_') @@ -721,7 +810,7 @@ def treat_var_mod(variables) : # var_mod[var].append(variable) return var_mod -def doconcorde(corpus, uces, mots, uci = False) : +def doconcorde(corpus, uces, mots, uci = False, et = False) : if not uci : ucestxt1 = [row for row in corpus.getconcorde(uces)] else : @@ -729,22 +818,229 @@ def doconcorde(corpus, uces, mots, uci = False) : ucestxt1 = dict(ucestxt1) ucestxt = [] ucis_txt = [] - listmot = [corpus.getlems()[lem].formes for lem in mots] - listmot = [corpus.getforme(fid).forme for lem in listmot for fid in lem] + if not et : + listmot = [corpus.getlems()[lem].formes for lem in mots] + listmot = [corpus.getforme(fid).forme for lem in listmot for fid in lem] + else : + listmot = mots mothtml = ['%s' % mot for mot in listmot] dmots = dict(zip(listmot, mothtml)) for uce in uces : ucetxt = ucestxt1[uce].split() ucetxt = ' '.join([dmots.get(mot, mot) for mot in ucetxt]) if not uci : - ucis_txt.append('

' + ' '.join(corpus.ucis[corpus.getucefromid(uce).uci].etoiles) + '

') + uciid = corpus.getucefromid(uce).uci + ucis_txt.append('

' + ' '.join(corpus.ucis[corpus.getucefromid(uce).uci].etoiles) + ' *%i_%i

' % (uciid, uce, uciid, uce)) else : ucis_txt.append('

' + ' '.join(corpus.ucis[uce].etoiles) + '

') ucestxt.append(ucetxt) return ucis_txt, ucestxt - + def getallstcarac(corpus, analyse) : pathout = PathOut(analyse['ira']) profils = ReadProfileAsDico(pathout['PROFILE_OUT'], Alceste, self.encoding) print profils + +def read_chd(filein, fileout): + with open(filein, 'r') as f : + content = f.read() + #content = [line[3:].replace('"',"").replace(' ','') for line in content.splitlines()] + content = [line.split('\t') for line in content.splitlines()] + chd = {'name':1, 'children':[]} + mere={} + for i, line in enumerate(content) : + if i == 0 : + chd['children'] = [{'name': line[1],'size' : content[i+1][0]}, {'name':line[2], 'size': content[i+1][1]}] + mere[line[1]] = chd['children'][0] + mere[line[2]] = chd['children'][1] + elif not i % 2 : + if 'children' in mere[line[0]]: + mere[line[0]]['children'].append({'name': line[1],'size' : content[i+1][0]}) + mere[line[1]] = mere[line[0]]['children'][-1] + mere[line[0]]['children'].append({'name': line[2],'size' : content[i+1][1]}) + mere[line[2]] = mere[line[0]]['children'][-1] + else : + mere[line[0]]['children'] = [{'name': line[1],'size' : content[i+1][0]}, {'name':line[2], 'size': content[i+1][1]}] + mere[line[1]] = mere[line[0]]['children'][-2] + mere[line[2]] = mere[line[0]]['children'][-1] + with open(fileout, 'w') as f : + f.write(json.dumps(chd)) + + +translation_languages = {"Afrikaans":"af", "Albanian":"sq", "Amharic":"am", "Arabic":"ar", "Armenian":"hy", "Azeerbaijani":"az", "Basque":"eu", "Belarusian":"be", "Bengali":"bn", "Bosnian":"bs", "Bulgarian":"bg", "Catalan":"ca", "Cebuano":"ceb", "Chichewa":"ny", "Chinese (Simplified)":"zh-CN", "Chinese (Traditional)":"zh-TW", "Corsican":"co", "Croatian":"hr", "Czech":"cs", "Danish":"da", "Dutch":"nl", "English":"en", "Esperanto":"eo", "Estonian":"et", "Filipino":"tl", "Finnish":"fi", "French":"fr", "Frisian":"fy", "Galician":"gl", "Georgian":"ka", "German":"de", "Greek":"el", "Gujarati":"gu", "Haitian Creole":"ht", "Hausa":"ha", "Hawaiian":"haw", "Hebrew":"iw", "Hindi":"hi", "Hmong":"hmn ", "Hungarian":"hu", "Icelandic":"is", "Igbo":"ig", "Indonesian":"id", "Irish":"ga", "Italian":"it", "Japanese":"ja", "Javanese":"jw", "Kannada":"kn", "Kazakh":"kk", "Khmer":"km", "Korean":"ko", "Kurdish":"ku", "Kyrgyz":"ky", "Lao":"lo", "Latin":"la", "Latvian":"lv", "Lithuanian":"lt", "Luxembourgish":"lb", "Macedonian":"mk", "Malagasy":"mg", "Malay":"ms", "Malayalam":"ml", "Maltese":"mt", "Maori":"mi", "Marathi":"mr", "Mongolian":"mn", "Burmese":"my", "Nepali":"ne", "Norwegian":"no", "Pashto":"ps", "Persian":"fa", "Polish":"pl", "Portuguese":"pt", "Punjabi":"ma", "Romanian":"ro", "Russian":"ru", "Samoan":"sm", "Scots Gaelic":"gd", "Serbian":"sr", "Sesotho":"st", "Shona":"sn", "Sindhi":"sd", "Sinhala":"si", "Slovak":"sk", "Slovenian":"sl", "Somali":"so", "Spanish":"es", "Sundanese":"su", "Swahili":"sw", "Swedish":"sv", "Tajik":"tg", "Tamil":"ta", "Telugu":"te", "Thai":"th", "Turkish":"tr", "Ukrainian":"uk", "Urdu":"ur", "Uzbek":"uz", "Vietnamese":"vi", "Welsh":"cy", "Xhosa":"xh", "Yiddish":"yi", "Yoruba":"yo", "Zulu":"zu", } + + +def gettranslation(words, lf, lt) : + import urllib2 + import json + agent = {'User-Agent': + "Mozilla/4.0 (\ + compatible;\ + MSIE 6.0;\ + Windows NT 5.1;\ + SV1;\ + .NET CLR 1.1.4322;\ + .NET CLR 2.0.50727;\ + .NET CLR 3.0.04506.30\ + )"} + base_link = "https://translate.googleapis.com/translate_a/single?client=gtx&sl=%s&tl=%s&dt=t&q=%s" + totrans = urllib2.quote('\n'.join(words).encode('utf8')) + link = base_link % (lf, lt, totrans) + request = urllib2.Request(link, headers=agent) + raw_data = urllib2.urlopen(request).read() + data = json.loads(raw_data) + return [line[0].decode('utf8', errors='replace').replace(u"'", u'_').replace(u' | ', u'|').replace(u' ', u'_').replace(u'-',u'_').replace(u'\n','') for line in data[0]] + +def makenprof(prof, trans, deb=0) : + nprof=[] + if deb == 0 : + nprof.append(prof[0]) + for i, val in enumerate(trans) : + line = prof[deb+i+1][:] + line[6] = val + nprof.append(line) + return nprof + +def treatempty(val) : + if val.strip() == '' : + return '_' + else : + return val + +def translateprofile(corpus, dictprofile, lf='it', lt='fr', maxword = 20) : + nprof = {} + lems = {} + for i in range(len(dictprofile)) : + prof = dictprofile[`i+1`] + try : + lenact = prof.index([u'*****', u'*', u'*', u'*', u'*', u'*', '', '']) + lensup = -1 + except ValueError: + try : + lenact = prof.index([u'*', u'*', u'*', u'*', u'*', u'*', '', '']) + lensup = 0 + except ValueError: + lenact = len(prof) + lensup = 0 + try : + lensup += prof.index([u'*', u'*', u'*', u'*', u'*', u'*', '', '']) + lensup = lensup - lenact + except ValueError: + lensup += len(prof) - lenact + if lenact != 0 : + if lenact > maxword : + nlenact = maxword + else : + nlenact = lenact + actori = [line[6] for line in prof[1:nlenact]] + act = [val.replace(u'_', u' ') for val in actori] + act = gettranslation(act, lf, lt) + for j, val in enumerate(actori) : + if act[j] not in lems : + lems[act[j]] = val + else : + while act[j] in lems : + act[j] = act[j] + u"+" + lems[act[j]] = val + nprof[`i+1`] = makenprof(prof, act) + + if lensup != 0 : + if lensup > maxword : + nlensup = maxword + else : + nlensup = lensup + supori = [line[6] for line in prof[(1+lenact):(lenact+nlensup+1)]] + sup = [val.replace(u'_', u' ') for val in supori] + sup = [treatempty(val) for val in sup] + sup = gettranslation(sup, lf, lt) + for j, val in enumerate(supori) : + if sup[j] not in lems : + lems[sup[j]] = val + else : + while sup[j] in lems : + sup[j] = sup[j] + u"+" + lems[sup[j]] = val + nprof[`i+1`].append([u'*****', u'*', u'*', u'*', u'*', u'*', '', '']) + nprof[`i+1`] += makenprof(prof, sup, deb=lenact) + + try : + lenet = prof.index([u'*', u'*', u'*', u'*', u'*', u'*', '', '']) + nprof[`i+1`].append([u'*', u'*', u'*', u'*', u'*', u'*', '', '']) + nprof[`i+1`] += prof[(lenet+1):] + except : + pass + return nprof, lems + +def write_translation_profile(prof, lems, language, dictpathout) : + if os.path.exists(dictpathout['translations.txt']) : + with codecs.open(dictpathout['translations.txt'], 'r', 'utf8') as f : + translist = f.read() + translist = [line.split('\t') for line in translist.splitlines()] + else : + translist = [] + toprint = [] + toprint.append(['','','','','','']) + toprint.append([u'***', u'nb classes', `len(prof)`, u'***', '', '']) + for i in range(len(prof)) : + toprint.append([u'**', u'classe', `i+1`, u'**', '', '']) + toprint.append([u'****'] + prof[`i+1`][0] + [u'****']) + rest = [[`line[1]`, `line[2]`, `line[3]`, `line[4]`, line[6], line[7].replace('< 0,0001', '0.00009').replace('NS (','').replace(')','')] for line in prof[`i+1`][1:]] + for i, line in enumerate(prof[`i+1`][1:]) : + if line[0] == u'*' : + rest[i] = [u'*', u'*', u'*', u'*', u'*', u'*'] + elif line[0] == u'*****' : + rest[i] = [u'*****',u'*',u'*', u'*', u'*', u'*'] + toprint += rest + with open(dictpathout['translation_profile_%s.csv' % language], 'w') as f : + f.write('\n'.join([';'.join(line) for line in toprint]).encode('utf8')) + with open(dictpathout['translation_words_%s.csv' % language], 'w') as f : + f.write('\n'.join(['\t'.join([val, lems[val]]) for val in lems]).encode('utf8')) + if 'translation_profile_%s.csv' % language not in [val[0] for val in translist] : + translist.append(['translation_profile_%s.csv' % language, 'translation_words_%s.csv' % language]) + with open(dictpathout['translations.txt'], 'w') as f : + f.write('\n'.join(['\t'.join(line) for line in translist]).encode('utf8')) + +def makesentidict(infile, language) : + #'/home/pierre/workspace/iramuteq/dev/langues/NRC/NRC-Emotion-Lexicon.csv' + with codecs.open(infile,'r', 'utf8') as f : + content = f.read() + content = [line.split('\t') for line in content.splitlines()] + titles = content.pop(0) + senti = ['Positive', 'Negative', 'Anger', 'Anticipation', 'Disgust', 'Fear', 'Joy', 'Sadness', 'Surprise', 'Trust'] + sentid = {} + for sent in senti : + sentid[sent] = titles.index(sent) + frtitle = [val for val in titles if '(fr)' in val] + frid = titles.index(frtitle[0]) + sentidict = [[line[frid].lower(), [line[sentid[sent]] for sent in senti]] for line in content] + pos = ['positive'] + [line[0] for line in sentidict if line[1][0] == '1'] + neg = ['negative'] + [line[0] for line in sentidict if line[1][1] == '1'] + anger = ['anger'] + [line[0] for line in sentidict if line[1][2] == '1'] + anticipation = ['anticipation'] + [line[0] for line in sentidict if line[1][3] == '1'] + disgust = ['disgust'] + [line[0] for line in sentidict if line[1][4] == '1'] + fear = ['fear'] + [line[0] for line in sentidict if line[1][5] == '1'] + joy = ['joy'] + [line[0] for line in sentidict if line[1][6] == '1'] + sadness = ['sadness'] + [line[0] for line in sentidict if line[1][7] == '1'] + surprise = ['surprise'] + [line[0] for line in sentidict if line[1][8] == '1'] + trust = ['trust'] + [line[0] for line in sentidict if line[1][9] == '1'] + with open('/tmp/tgenemo.csv', 'w') as f : + for val in [pos, neg, anger, anticipation, disgust, fear, joy, sadness, surprise, trust] : + f.write('\t'.join(val).encode('utf8') + '\n') + +def countsentfromprof(prof, encoding, sentidict) : + with codecs.open(prof, 'r', encoding) as f : + content = f.read() + content = [line.split(';') for line in content.splitlines()] + print content + content = [[line[0], [int(val) for val in line[1:]]] for line in content] + print content + content = dict(content) + print content + +def iratolexico(infile, outfile, encoding) : + with codecs.open(infile, 'r', encoding) as f : + for line in f : + if line.startswith(u'**** ') : + line = line.split() +