X-Git-Url: http://iramuteq.org/git?p=iramuteq;a=blobdiff_plain;f=corpusNG.py;fp=corpusNG.py;h=a950d0f2ba23f6c2a687d8e1bbbb9af4569f41a8;hp=380b9a2c8fc1a129ee74fadd4dea4c47382d1d7d;hb=81594f689f1e191599d96a2a503fbb5529df69d7;hpb=eb3837fa8ebc881ef1df81980ec576ca0451eb40;ds=sidebyside diff --git a/corpusNG.py b/corpusNG.py index 380b9a2..a950d0f 100644 --- a/corpusNG.py +++ b/corpusNG.py @@ -547,6 +547,21 @@ class Corpus : if len(l) > taille_limite : l = l[-taille_limite:] return l + + def find_segments_in_classe(self, list_uce, taille_segment, taille_limite): + d={} + for uce in self.getconcorde(list_uce) : + uce = uce[1].split() + d =self.count_from_list([' '.join(uce[i:i+taille_segment]) for i in range(len(uce)-(taille_segment - 1))], d) + l = [[d[val], val, taille_segment] for val in d if d[val] >= 3] + del(d) + l.sort() + if len(l) > taille_limite : + l = l[-taille_limite:] + return l + + + def make_ucecl_from_R(self, filein) : with open(filein, 'rU') as f : @@ -983,7 +998,7 @@ class BuildFromAlceste(BuildCorpus) : self.limitshow = 0 else : self.limitshow = self.last / 100000 - log.debug(`iduci`, `idpara`, `iduce`) + log.debug(' '.join([`iduci`,`idpara`,`iduce`])) if self.last > self.lim : self.backup_uce() self.last = 0