correction labbé

diff --git a/Rscripts/Rgraph.R b/Rscripts/Rgraph.R
index af0b351..8407e31 100644 (file)
--- a/Rscripts/Rgraph.R
+++ b/Rscripts/Rgraph.R
@@ -14,7 +14,7 @@
 #   dev.off()
 #}
 
-PlotDendroCut <- function(chd,filename,reso,clusternb) {   
+PlotDendroCut <- function(chd,filename,reso,clusternb) {
    h.chd <- as.hclust(chd)
    memb <- cutree(h.chd, k = clusternb)
    cent <- NULL
@@ -951,7 +951,7 @@ plot.spec <- function(spec, nb.word = 20) {
            vcex <- norm.vec(word.size[,i], 2, 3)
                text(-0.9, -0.5, cn[i], cex = 1, adj=0, srt=90, col='black')
            for (j in 1:length(word.size[,i])) {
-               yval <- yval-(strheight(word.to.plot[j,i],cex=vcex[j])+0.02)
+               yval <- yval-(strheight(word.to.plot[j,i],cex=vcex[j])+0.01)
                text(-0.9, yval, word.to.plot[j,i], cex = vcex[j], col = col, adj=0)
            }
        }
@@ -996,6 +996,7 @@ plot.alceste.graph <- function(rdata,nd=3,layout='fruke', chilim = 2) {
 
 make.simi.afc <- function(x,chitable,lim=0, alpha = 0.1, movie = NULL) {
     library(igraph)
+    library(rgl)
     chimax<-as.matrix(apply(chitable,1,max))
     chimax<-as.matrix(chimax[,1][1:nrow(x)])
     chimax<-cbind(chimax,1:nrow(x))
@@ -1192,19 +1193,20 @@ graphml.to.file <- function(graph.path) {
 graph.to.file <- function(graph.simi, nodesfile = NULL, edgesfile = NULL, community = FALSE, color = NULL, sweight = NULL) {
        require(igraph)
        g <- graph.simi$graph
-    print(graph.simi$eff)
+    #print(graph.simi$eff)
     if (!is.null(graph.simi$eff)) {
            V(g)$weight <- graph.simi$eff
     } else {
         V(g)$weight <- graph.simi$label.cex
     }
-       layout <- layout.norm(graph.simi$layout,-5,5,-5,5,-5,5)
-       print(layout)
+       layout <- layout.norm(graph.simi$layout,-10,10,-10,10,-10,10)
+       #print(layout)
        V(g)$x <- layout[,1]
        V(g)$y <- layout[,2]
        if (ncol(layout) == 3) {
                V(g)$z <- layout[,3]
        }
+    E(g)$weight <- graph.simi$we.width
        if (community) {
                member <- graph.simi$communities$membership
                col <- rainbow(max(member))
@@ -1218,7 +1220,7 @@ graph.to.file <- function(graph.simi, nodesfile = NULL, edgesfile = NULL, commun
                v.colors <- col2rgb(color)
                V(g)$r <- v.colors[1,]
                V(g)$g <- v.colors[2,]
-               V(g)$b <- v.colors[3,]          
+               V(g)$b <- v.colors[3,]
        }
        if (!is.null(sweight)) {
                V(g)$sweight <- sweight
@@ -1238,6 +1240,7 @@ graph.to.file <- function(graph.simi, nodesfile = NULL, edgesfile = NULL, commun
 graph.to.file2 <- function(graph, layout, nodesfile = NULL, edgesfile = NULL, community = FALSE, color = NULL, sweight = NULL) {
        require(igraph)
        g <- graph
+    layout <- layout.norm(layout,-5,5,-5,5,-5,5)
        V(g)$x <- layout[,1]
        V(g)$y <- layout[,2]
        if (ncol(layout) == 3) {

diff --git a/Rscripts/distance-labbe.R b/Rscripts/distance-labbe.R
index eccb3ae..747332b 100644 (file)
--- a/Rscripts/distance-labbe.R
+++ b/Rscripts/distance-labbe.R
@@ -29,8 +29,10 @@ compute.labbe <- function(x, y, tab) {
         U <- N1/N2
         mini.tab[,2] <- mini.tab[,2] * U
                col.plusgrand <- mini.tab[,2]
-               cs.plus.grand <- sum(col.plusgrand[col.plusgrand>1])
+               cs.plus.grand <- sum(col.plusgrand[col.plusgrand>=1])
     }
+    #print(U)
+    #print(cs.plus.grand)
     commun <- which((mini.tab[,1] > 0) & (mini.tab[,2] > 0))
     deA <- which((mini.tab[,plus.petit] > 0) & (mini.tab[,plus.grand] == 0))
     deB <- which((mini.tab[,plus.petit] == 0)  & (mini.tab[,plus.grand] >= 1))
@@ -39,9 +41,12 @@ compute.labbe <- function(x, y, tab) {
     dist.deA <- abs(mini.tab[deA, plus.petit] - mini.tab[deA, plus.grand])
     dist.deB <- abs(mini.tab[deB, plus.petit] - mini.tab[deB, plus.grand])
     dist.labbe <- sum(dist.commun) + sum(dist.deA) + sum(dist.deB)
+    #print(cs[plus.petit])
+    #print(dist.labbe)
 
     indice.labbe <- dist.labbe/(cs[plus.petit] + cs.plus.grand)
-    indice.labbe
+    res = list(indice.labbe = indice.labbe, commun=commun, deA=deA, deB=deB, dist.commun=dist.commun, dist.deA=dist.deA, dist.deB=dist.deB)
+       res
 }
 
 #calcul pour distance texte 1 et 2
@@ -54,9 +59,59 @@ dist.labbe <- function(tab) {
        for (i in 1:(ncol(tab)-1)) {
                for (j in (1+i):ncol(tab)) {
                        #lab <- compute.labbe(i,j,tab)
-                       mat[j,i] <- compute.labbe(i,j,tab)
+                       mat[j,i] <- compute.labbe(i,j,tab)$indice.labbe
                }
        }
     mat
 }
 
+dist.labbe2 <- function(tab) {
+    distance_from_idxs <- function (idxs) {
+        i1 <- idxs[1]
+        i2 <- idxs[2]
+        compute.labbe(i1, i2, tab)$indice.labbe
+    }
+
+    size <- ncol(tab)
+    d <- apply(utils::combn(size, 2), 2, distance_from_idxs)
+    attr(d, "Size") <- size
+    xnames <- colnames(tab)
+    if (!is.null(xnames)) {
+        attr(d, "Labels") <- xnames
+    }
+    attr(d, "Diag") <- FALSE
+    attr(d, "Upper") <- FALSE
+    class(d) <- "dist"
+    d
+       #mat <- matrix(NA, ncol=ncol(tab), nrow=ncol(tab))
+       #rownames(mat) <- colnames(tab)
+       #colnames(mat) <- colnames(tab)
+       #for (i in 1:(ncol(tab)-1)) {
+       #       for (j in (1+i):ncol(tab)) {
+                       #lab <- compute.labbe(i,j,tab)
+       #               mat[j,i] <- compute.labbe(i,j,tab)$indice.labbe
+       #       }
+       #}
+    #mat
+}
+
+dist.labbe3 <- function(tab) {
+    distance_from_idxs <- function (idxs) {
+        i1 <- tab[,idxs[1]]
+        i2 <- tab[,idxs[2]]
+        labbe(i1, i2)
+    }
+
+    size <- ncol(tab)
+    d <- apply(utils::combn(size, 2), 2, distance_from_idxs)
+    attr(d, "Size") <- size
+    xnames <- colnames(tab)
+    if (!is.null(xnames)) {
+        attr(d, "Labels") <- xnames
+    }
+    attr(d, "Diag") <- FALSE
+    attr(d, "Upper") <- FALSE
+    class(d) <- "dist"
+    d
+
+}

diff --git a/checkinstall.py b/checkinstall.py
index e7dca8a..17d32fd 100644 (file)
--- a/checkinstall.py
+++ b/checkinstall.py
@@ -123,7 +123,7 @@ def FindRPAthWin32():
         for progpath in progpaths :
             rpath = os.path.join(progpath, "R")
             if os.path.exists(rpath) :
-                for maj in range(2,4) :
+                for maj in range(2,6) :
                     for i in range(0,30):
                         for j in range(0,20):
                             for poss in ['', 'i386', 'x64'] :
@@ -138,6 +138,8 @@ def FindRPathNix():
         BestPath='/usr/bin/R'
     elif os.path.exists('/usr/local/bin/R'):
         BestPath='/usr/local/bin/R'
+    elif os.path.exists('/Library/Frameworks/R.framework/Resources/bin/R') :
+        BestPath='/Library/Frameworks/R.framework/Resources/bin/R'
     return BestPath
 
 def RLibsAreInstalled(self) :
@@ -148,7 +150,8 @@ def RLibsAreInstalled(self) :
         return True
 
 def CheckRPackages(self):
-    listdep = ['ca', 'rgl', 'gee', 'ape', 'igraph','proxy', 'wordcloud', 'irlba', 'textometry']
+    listdep = ['ca', 'rgl', 'gee', 'ape', 'igraph','proxy', 'wordcloud',
+               'irlba', 'textometry', 'intergraph', 'sna', 'network']
     nolib = []
     i=0
     dlg = wx.ProgressDialog("Test des librairies de R", "test en cours...", maximum = len(listdep), parent=self, style=wx.PD_APP_MODAL | wx.PD_AUTO_HIDE | wx.PD_ELAPSED_TIME | wx.PD_CAN_ABORT)
@@ -174,12 +177,12 @@ def CheckRPackages(self):
 %s
 
 Sans ces bibliothèques, IRamuteq ne fonctionnera pas.
-    
+
 - Vous pouvez installer ces bibliothèques manuellement :
         Cliquez sur Annuler
         Lancez R
         Tapez install.packages('nom de la bibiothèque')
-        
+
 - ou laisser IRamuteq les installer automatiquement en cliquant sur VALIDER .
         Les bibliothèques seront téléchargées depuis le site miroir de R %s.
         """ % (txt, self.pref.get('iramuteq','rmirror'))

diff --git a/functions.py b/functions.py
index 8c0c66c..6a3d968 100755 (executable)
--- a/functions.py
+++ b/functions.py
@@ -114,6 +114,11 @@ 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)
@@ -805,7 +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 :
@@ -813,8 +818,11 @@ 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 = ['<font color=red><b>%s</b></font>' % mot for mot in listmot]
     dmots = dict(zip(listmot, mothtml))
     for uce in uces :
@@ -877,13 +885,12 @@ def gettranslation(words, lf, lt) :
     .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"
-    print len(words)
     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').replace(u"'", u'_').replace(u' | ', u'|').replace(u' ', u'_').replace(u'-',u'_').replace(u'\n','') for line in data[0]]
+    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=[]
@@ -901,7 +908,7 @@ def treatempty(val) :
     else :
         return val
 
-def translateprofile(corpus, dictprofile, lf='it', lt='fr', maxword = 50) :
+def translateprofile(corpus, dictprofile, lf='it', lt='fr', maxword = 20) :
     nprof = {}
     lems = {}
     for i in range(len(dictprofile)) :
@@ -943,7 +950,7 @@ def translateprofile(corpus, dictprofile, lf='it', lt='fr', maxword = 50) :
                 nlensup = maxword
             else :
                 nlensup = lensup
-            supori = [line[6] for line in prof[(1+lenact):(lenact+nlensup)]]
+            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)

diff --git a/iramuteq.py b/iramuteq.py
index 501b0f6..8531535 100755 (executable)
--- a/iramuteq.py
+++ b/iramuteq.py
@@ -1,8 +1,9 @@
 #!/bin/env python
 # -*- coding: utf-8 -*-
 #Author: Pierre Ratinaud
-#Copyright (c) 2008-2016, Pierre Ratinaud
+#Copyright (c) 2008-2020, Pierre Ratinaud
 #License: GNU GPL
+#test
 
 from optparse import OptionParser
 
@@ -13,6 +14,7 @@ parser.add_option("-f", "--file", dest="filename",
 
 import sys
 reload(sys)
+#sys.path.insert(0, '/usr/lib/python2.7/dist-packages/wxPython-4.0.1-py2.7-linux-amd64.egg')
 import locale
 import tempfile
 import codecs
@@ -28,6 +30,7 @@ import wx.lib.agw.aui as aui
 import wx.html
 import wx.grid
 import wx.lib.hyperlink as hl
+#import wx.lib.agw.hyperlink as hl
 #------------------------------------
 from functions import BugReport, PlaySound, History, progressbar
 from checkversion import NewVersion
@@ -51,7 +54,7 @@ from textaslexico import Lexico
 from textlabbe import DistLabbe
 from textsimi import SimiTxt, SimiFromCluster
 from textwordcloud import WordCloud, ClusterCloud
-from textreinert import Reinert
+from textreinert import Reinert#, ReDoReinert
 #from textcheckcorpus import checkcorpus
 from openanalyse import OpenAnalyse
 from corpus import Builder, SubBuilder, MergeClusters
@@ -1031,6 +1034,9 @@ Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, États-Unis."""
         #FIXME
         AnalyseMerge(self, {'type': 'merge', 'fileout' : '/tmp/test.txt'}, dlg = 5)
 
+    def OnEliminate(self, corpus, parametres) :
+        print parametres
+
     def OnMergeClusters(self, evt) :
         builder = MergeClusters(self, {})
         if builder.res == wx.ID_OK :

diff --git a/layout.py b/layout.py
index 4ea02da..80ca8b5 100644 (file)
--- a/layout.py
+++ b/layout.py
@@ -767,14 +767,20 @@ class dolexlayout :
         self.TabStat = aui.AuiNotebook(ira.nb, -1, wx.DefaultPosition)
         self.TabStat.parametres = parametres
         self.ListPan = ListForSpec(ira, self, self.DictSpec, self.etoiles)
+        self.ListPan.pathout = self.pathout
         if os.path.exists(self.pathout['banalites.csv']) :
             self.listban = ListForSpec(ira, self, self.dictban, ['eff'] + self.etoiles, usefirst = True)
         #self.ListPan2 = ListForSpec(sash.rightwin1, self, self.DictSpec, first)
         self.ListPant = ListForSpec(ira, self, self.DictType, self.etoiles)
+        self.ListPant.pathout = self.pathout
         self.ListPanEff = ListForSpec(ira, self, self.DictEff, self.etoiles)
+        self.ListPanEff.pathout = self.pathout
         self.ListPanEffType = ListForSpec(ira, self, self.DictEffType, self.etoiles)
+        self.ListPanEffType.pathout = self.pathout
         self.ListPanEffRelForme = ListForSpec(ira, self, self.DictEffRelForme, self.etoiles)
+        self.ListPanEffRelForme.pathout = self.pathout
         self.ListPanEffRelType = ListForSpec(ira, self, self.DictEffRelType, self.etoiles)
+        self.ListPanEffRelType.pathout = self.pathout
 
         self.TabStat.AddPage(self.ListPan, _(u'Forms').decode('utf8'))
         if os.path.exists(self.pathout['banalites.csv']) :
@@ -799,6 +805,7 @@ class dolexlayout :
 
         self.TabStat.corpus = self.corpus
         self.TabStat.etoiles = self.etoiles
+        self.TabStat.pathout = self.pathout
         if os.path.exists(os.path.join(self.parametres['pathout'], 'tgenspec.csv')) :
             self.parametres['tgenspec'] = os.path.join(self.parametres['pathout'], 'tgenspec.csv')
             TgenLayout(self.TabStat)
@@ -822,6 +829,8 @@ class StatLayout:
 
         #self.TabStatTot = wx.TextCtrl(self.TabStat, -1, style=wx.NO_BORDER | wx.TE_MULTILINE | wx.TE_RICH2)
         list_graph = [['zipf.png', 'zipf']]
+        if os.path.exists(self.pathout['segments_size.png']) :
+            list_graph.append(['segments_size.png', _(u'Taille des segments')])
         self.TabStatTot = GraphPanel(ira.nb, self.pathout, list_graph, self.result['glob'])
         self.TabStat.AddPage(self.TabStatTot, _(u'Abstract').decode('utf8'))
 
@@ -856,6 +865,7 @@ class GraphPanelDendro(wx.Panel):
         self.graphnb = 1
         self.dictpathout = dico
         self.dirout = os.path.dirname(self.dictpathout['ira'])
+        self.pathout = self.dictpathout
         self.list_graph = list_graph
         self.parent = self.GetParent()#parent
         self.SetFont(wx.Font(10, wx.DEFAULT, wx.NORMAL, wx.NORMAL, 0, "Arial"))

diff --git a/textstat.py b/textstat.py
index 7fdf5bd..f0a92fd 100755 (executable)
--- a/textstat.py
+++ b/textstat.py
@@ -84,7 +84,7 @@ class Stat(AnalyseText) :
         open_file_graph("%s", width = 400, height = 400)
         barplot(table(stsize[,1]))
         dev.off()
-        """ % (self.pathout['stsize.csv'], self.pathout['segments_size.png'])
+        """ % (ffr(self.pathout['stsize.csv']), ffr(self.pathout['segments_size.png']))
         tmpscript = tempfile.mktemp(dir=self.parent.TEMPDIR)
         with open(tmpscript, 'w') as f :
             f.write(txt)