# -*- coding: utf-8 -*-
#Author: Pierre Ratinaud
#Copyright (c) 2008-2012 Pierre Ratinaud
-#Lisense: GNU/GPL
+#License: GNU/GPL
import wx
import re
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 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('\\\\') :
+ path = '\\' + path
+ return path
+
+class TGen :
+ def __init__(self, path = None, encoding = 'utf8'):
+ self.path = path
+ self.tgen = {}
+ self.encoding = encoding
+
+ def __getitem__(self, key):
+ return self.tgen[key]
+
+ def read(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()
+ with open(pathout, 'w') as f :
+ line = '\t'.join([u'tgens'] + etoiles) + '\n'
+ f.write(line.encode(self.encoding))
+ for t in tgens :
+ line = '\t'.join([t] + [`tgens[t][et]` for et in etoiles]) + '\n'
+ f.write(line.encode(self.encoding))
+ i = 0
+ totname = 'total'
+ while totname + `i` in tgens :
+ i += 1
+ totname = totname + `i`
+ line = '\t'.join([totname] + [`totocc[et]` for et in etoiles]) + '\n'
+ f.write(line.encode(self.encoding))
+
class History :
def __init__(self, filein, syscoding = 'utf8') :
self.filein = filein
self.syscoding = syscoding
- self.corpora = {}
+ self.corpus = {}
self.openedcorpus = {}
+ self.openedmatrix = {}
self.orph = []
self.analyses = {}
self.history = []
def read(self) :
d = shelve.open(self.filein)
self.history = d.get('history', [])
+ self.matrix = d.get('matrix', [])
self.ordercorpus = dict([[corpus['uuid'], i] for i, corpus in enumerate(self.history)])
- self.corpus = dict([[corpus['uuid'], corpus] for i, corpus in enumerate(self.history)])
+ self.corpus = dict([[corpus['uuid'], corpus] for corpus in self.history])
self.analyses = dict([[analyse['uuid'], analyse] for corpus in self.history for analyse in corpus.get('analyses', [])])
+ 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()
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']}
+ if tosave['uuid'] in self.corpus :
+ log.info('problem : this uuid is already in history : %s' % tosave['uuid'])
+ return
if analyse.get('corpus', False) :
if analyse['uuid'] in self.analyses :
return
tosave['corpus'] = analyse['corpus']
tosave['name'] = analyse['name']
acorpus_uuid = analyse['corpus']
- if acorpus_uuid in self.ordercorpus :
+ if acorpus_uuid in self.corpus :
if 'analyses' in self.history[self.ordercorpus[acorpus_uuid]] :
self.history[self.ordercorpus[acorpus_uuid]]['analyses'].append(tosave)
else :
self.orph.append(tosave)
else :
tosave['corpus_name'] = analyse['corpus_name']
+ #self.ordercorpus[tosave['uuid']] = len(history)
+ #self.corpus[tosave['uuid']] = analyse
self.history.append(tosave)
self.write()
self.read()
+ def addMatrix(self, analyse) :
+ tosave = analyse
+ #tosave['matrix_name'] = analyse['matrix_name']
+ tosave['analyses'] = []
+ self.matrix.append(tosave)
+ self.write()
+ self.read()
+
+ def addMatrixAnalyse(self, analyse) :
+ tosave = {'uuid' : analyse['uuid'], 'ira': analyse['ira'], 'type' : analyse['type'], 'matrix' : analyse['matrix']}
+ tosave['name'] = analyse['name']
+ if tosave['matrix'] in self.ordermatrix :
+ 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 :
+ tosave = {'uuid' : analyse['uuid'], 'ira': analyse['ira'], 'type' : analyse['type']}
+ corpus = analyse['corpus']
+ tosave['corpus'] = corpus
+ tosave['name'] = analyse['name']
+ if corpus in self.corpus :
+ if 'analyses' in self.history[self.ordercorpus[corpus]] :
+ self.history[self.ordercorpus[corpus]]['analyses'].append(tosave)
+ else :
+ self.history[self.ordercorpus[corpus]]['analyses'] = [tosave]
+ self.write()
+ self.read()
+
def delete(self, analyse, corpus = False) :
+ log.info('delete %s' % analyse.get('name', 'noname'))
if corpus :
self.history.pop(self.ordercorpus[analyse['uuid']])
- else :
+ if analyse['uuid'] in self.openedcorpus :
+ del self.openedcorpus[analyse['uuid']]
+ log.info('delete corpus : %s' % analyse['uuid'])
+ elif analyse['uuid'] in self.analyses :
todel = [i for i, ana in enumerate(self.corpus[analyse['corpus']]['analyses']) if ana['uuid'] == analyse['uuid']][0]
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']]
self.write()
self.read()
def rmtab(self, analyse) :
del self.opened[analyse['uuid']]
+
+ def clean(self) :
+ corpustodel = [corpus for corpus in self.history if not os.path.exists(corpus['ira'])]
+ print corpustodel
+ for corpus in corpustodel :
+ print 'cleaning :', corpus['corpus_name']
+ self.delete(corpus, corpus = True)
+ anatodel = [analyse for corpus in self.history for analyse in corpus.get('analyses', []) if not os.path.exists(analyse.get('ira', '/'))]
+ for analyse in anatodel :
+ print 'cleaning :', analyse['name']
+ self.delete(analyse)
def __str__(self) :
return str(self.history)
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'))
self.parametres = {}
if parametres is not None :
txt += '%s = %s\n' % (option, `parametres[i][option]`)
if outfile is None :
outfile = self.configfile
- with codecs.open(outfile, 'w', 'utf8') as f :
- f.write(txt)
+ outfile = normpath_win32(outfile)
+ with open(outfile, 'w') as f :
+ f.write(txt.encode('utf8'))
#self.conf.write(f)
def totext(self, parametres) :
kwds["style"] = wx.DEFAULT_DIALOG_STYLE
kwds["size"] = wx.Size(500, 200)
wx.Dialog.__init__(self, *args, **kwds)
+ self.SetTitle(kwds['title'])
self.text_ctrl_1 = wx.TextCtrl(self, -1, "", style=wx.TE_MULTILINE)
+ self.text_ctrl_1.SetBackgroundColour('#DDE8EB')
self.button_1 = wx.Button(self, wx.ID_OK, "")
self.__set_properties()
def __set_properties(self):
# begin wxGlade: MyDialog.__set_properties
- self.SetTitle("Bug")
self.SetMinSize(wx.Size(500, 200))
self.text_ctrl_1.SetMinSize(wx.Size(500, 200))
FileReader.close()
DictProfile = {}
count = 0
- rows = [row.replace('\n', '').replace("'", '').replace('\"', '').replace(',', '.').replace('\r','').split(';') for row in Filecontent]
+ #rows = [row.replace('\n', '').replace("'", '').replace('\"', '').replace(',', '.').replace('\r','').split(';') for row in Filecontent]
+ rows = [row.replace('\n', '').replace("'", '').replace('\"', '').replace('\r','').split(';') for row in Filecontent]
rows.pop(0)
ClusterNb = rows[0][2]
rows.pop(0)
def formatExceptionInfo(maxTBlevel=5):
cla, exc, trbk = sys.exc_info()
- excName = cla.__name__
+ try :
+ excName = cla.__name__
+ except :
+ excName = 'None'
try:
excArgs = exc.args[0]
except :
# si on a rien trouvé
return False, chaine.split(), ''
+
+exceptions = {'paragrapheOT' : u"Un problème de formatage (présence d'un marqueur de paragraphe (-*) en dehors d'un texte) est survenu à la ligne ",
+ '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 ",
+}
+
def BugReport(parent, error = None):
for ch in parent.GetChildren():
if "<class 'wx._windows.ProgressDialog'>" == str(type(ch)):
ch.Destroy()
excName, exc, excTb = formatExceptionInfo()
if excName == 'Exception' :
- txt = 'Message :\n'
+ print exc
+ if len(exc.split()) == 2 :
+ mss, linenb = exc.split()
+ if mss in exceptions :
+ txt = exceptions[mss] + linenb
+ else :
+ txt = exc
+ else :
+ if exc in exceptions :
+ txt = exceptions[exc]
+ else :
+ txt = exc
+ title = "Information"
else :
txt = u' !== BUG ==! \n'
txt += u'*************************************\n'
txt += '\n'.join(excTb).replace(' ', ' ')
txt += excName + '\n'
- txt += exc
+ txt += `exc`
+ title = "Bug"
- dial = BugDialog(parent)
+ dial = BugDialog(parent, **{'title' : title})
if 'Rerror' in dir(parent) :
txt += parent.Rerror
parent.Rerror = ''
def ReadDicoAsDico(dicopath):
with codecs.open(dicopath, 'r', 'UTF8') as f:
content = f.readlines()
- dico = {}
- for line in content :
- if line[0] != u'':
- line = line.replace(u'\n', '').replace('"', '').split('\t')
- dico[line[0]] = line[1:]
- return dico
+ lines = [line.rstrip('\n\r').replace(u'\n', '').replace('"', '').split('\t') for line in content if line != u'']
+ return dict([[line[0], line[1:]] for line in lines])
def ReadLexique(parent, lang = 'french', filein = None):
if lang != 'other' :
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()):
+def ReadList(filein, encoding = sys.getdefaultencoding(), sep = ';'):
#file = open(filein)
- file = codecs.open(filein, 'r', encoding)
- content = file.readlines()
- file.close()
+ with codecs.open(filein, 'r', encoding) as f :
+ content = f.read()
+ content = [line.replace('\n', '').replace('\r','').replace('\"', '').replace(',', '.').split(sep) for line in content.splitlines()]
+ #file = codecs.open(filein, 'r', encoding)
+ #content = file.readlines()
+ #file.close()
first = content.pop(0)
- first = first.replace('\n', '').replace('\r','').replace('\"', '').split(';')
+ #first = first.replace('\n', '').replace('\r','').replace('\"', '').split(sep)
dict = {}
i = 0
for line in content:
- line = line.replace('\n', '').replace('\r','').replace('\"', '').replace(',', '.')
- line = line.split(';')
+ #line = line.replace('\n', '').replace('\r','').replace('\"', '').replace(',', '.')
+ #line = line.split(';')
nline = [line[0]]
for val in line[1:]:
if val == u'NA' :
if sys.platform == 'darwin' :
try :
macversion = platform.mac_ver()[0].split('.')
- print macversion
if int(macversion[1]) < 5 :
needX11 = True
else :
needX11 = False
rpath = rpath.replace('\\','\\\\')
+ env = os.environ.copy()
+ if sys.platform == 'darwin' and 'LC_ALL' not in env:
+ env['LC_ALL'] = 'en_US.UTF-8'
if not graph :
if wait :
if sys.platform == 'win32':
error = call(["%s" % rpath, "--vanilla","--slave","-f", "%s" % rcode])
else :
- error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
+ error = call([rpath, '--slave', "--vanilla", "-f %s" % rcode, "--encoding=UTF-8"], env = env)
return error
else :
if sys.platform == 'win32':
pid = Popen(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
else :
- pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
+ pid = Popen([rpath, '--slave', "--vanilla", "-f %s" % rcode, "--encoding=UTF-8"], stderr = PIPE, env = env)
return pid
else :
if wait :
error = call(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
elif sys.platform == 'darwin' and needX11:
os.environ['DISPLAY'] = ':0.0'
- error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
+ error = call([rpath, '--vanilla','--slave',"-f %s" % rcode, "--encoding=UTF-8"], env = env)
else :
- error = call([rpath, '--vanilla','--slave',"-f %s" % rcode])
+ error = call([rpath, '--vanilla','--slave',"-f %s" % rcode, "--encoding=UTF-8"], env = env)
return error
else :
if sys.platform == 'win32':
pid = Popen(["%s" % rpath, '--vanilla','--slave','-f', "%s" % rcode])
elif sys.platform == 'darwin' and needX11:
os.environ['DISPLAY'] = ':0.0'
- pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
+ pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode, "--encoding=UTF-8"], stderr = PIPE, env = env)
else :
- pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode], stderr = PIPE)
+ pid = Popen([rpath, '--vanilla','--slave',"-f %s" % rcode, "--encoding=UTF-8"], stderr = PIPE, env = env)
return pid
def check_Rresult(parent, pid) :
if error[1] is None :
error[1] = 'None'
parent.Rerror = '\n'.join([str(pid.returncode), '\n'.join(error)])
- #try :
- raise Exception('\n'.join([u'Erreur R', '\n'.join(error[1:])]))
+ try :
+ raise Exception('\n'.join([u'Erreur R', '\n'.join(error[1:])]))
+ except :
+ BugReport(parent)
return False
- #except :
- # BugReport(parent)
else :
- return None
+ return True
else :
if pid != 0 :
- #try :
- raise Exception(u'Erreur R')
+ try :
+ raise Exception(u'Erreur R')
+ except :
+ BugReport(parent)
return False
- #except :
- # BugReport(parent)
else :
- return None
+ return True
def print_liste(filename,liste):
with open(filename,'w') as f :
ncontent=[line.replace('\n','').split(';') for line in content if line.strip() != '']
return ncontent
-class MessageImage(wx.Frame):
- def __init__(self, parent,title, size):
- wx.Frame.__init__ ( self, parent, id = wx.ID_ANY, title = title, pos = wx.DefaultPosition, size = size, style = wx.DEFAULT_FRAME_STYLE )
- self.SetSizeHintsSz( wx.DefaultSize, wx.DefaultSize )
- self.imageFile = False
- self.imagename = u"chi_classe.png"
- self.HtmlPage = wx.html.HtmlWindow(self, -1)
- self.HtmlPage.SetMinSize(size)
- if "gtk2" in wx.PlatformInfo:
- self.HtmlPage.SetStandardFonts()
- self.HtmlPage.SetFonts('Courier', 'Courier')
-
- self.button_1 = wx.Button(self, wx.ID_CANCEL)
- self.button_2 = wx.Button(self, wx.ID_SAVE)
- self.Bind(wx.EVT_BUTTON, self.OnCloseMe, self.button_1)
- self.Bind(wx.EVT_BUTTON, self.OnSaveImage, self.button_2)
- self.do_layout()
-
- def do_layout(self):
- self.sizer_1 = wx.BoxSizer(wx.VERTICAL)
- self.sizer_2 = wx.BoxSizer(wx.HORIZONTAL)
- self.sizer_1.Add(self.HtmlPage, 2, wx.EXPAND, 0)
-
- self.m_sdbSizer1 = wx.StdDialogButtonSizer()
- self.m_sdbSizer1.AddButton( self.button_2 )
- self.m_sdbSizer1.AddButton( self.button_1 )
- self.m_sdbSizer1.Realize()
- self.sizer_1.Add(self.m_sdbSizer1, 0, wx.EXPAND, 5)
- self.SetSizer(self.sizer_1)
- self.Layout()
- self.sizer_1.Fit( self )
- def addsaveimage(self, imageFile) :
- self.imageFile = imageFile
-
- def OnCloseMe(self, event):
- self.Destroy()
-
- def OnSaveImage(self, event) :
- dlg = wx.FileDialog(
- self, message="Enregistrer sous...", defaultDir=os.getcwd(),
- defaultFile= self.imagename, wildcard="png|*.png", style=wx.SAVE | wx.OVERWRITE_PROMPT
- )
- dlg.SetFilterIndex(2)
- dlg.CenterOnParent()
- if dlg.ShowModal() == wx.ID_OK:
- path = dlg.GetPath()
- copyfile(self.imageFile, path)
def progressbar(self, maxi) :
parent = self.parent
else :
parent = self
+ try :
+ print '###horrible hack progressbar'
+ maxi = int(maxi)
+ except :
+ maxi = 1
return wx.ProgressDialog("Traitements",
"Veuillez patienter...",
maximum=maxi,
def treat_var_mod(variables) :
var_mod = {}
- for variable in variables :
- if u'_' in variable :
- forme = variable.split(u'_')
- var = forme[0]
- mod = forme[1]
- if not var in var_mod :
- var_mod[var] = [variable]
- else :
- if not mod in var_mod[var] :
- var_mod[var].append(variable)
+ variables = list(set(variables))
+ varmod = [variable.split('_') for variable in variables]
+ vars = list(set([var[0] for var in varmod if len(var) >=2]))
+ 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'_')
+# var = forme[0]
+# mod = forme[1]
+# if not var in var_mod :
+# var_mod[var] = [variable]
+# else :
+# if not mod in var_mod[var] :
+# var_mod[var].append(variable)
return var_mod
+
+def doconcorde(corpus, uces, mots, uci = False) :
+ if not uci :
+ ucestxt1 = [row for row in corpus.getconcorde(uces)]
+ else :
+ ucestxt1 = [row for row in corpus.getuciconcorde(uces)]
+ 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]
+ mothtml = ['<font color=red><b>%s</b></font>' % 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('<p><b>' + ' '.join(corpus.ucis[corpus.getucefromid(uce).uci].etoiles) + '</b></p>')
+ else :
+ ucis_txt.append('<p><b>' + ' '.join(corpus.ucis[uce].etoiles) + '</b></p>')
+ 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