def _addClusterText(self, utterance, languageId):
"""Add 'utterance' as a text cluster.
param utterance: an utf-8 encoded string
"""
utterance = utterance.strip()
if len(utterance) > 0:
c = TextCluster(self, utterance)
c.setLanguage(languageId)
self.addDocumentLine(c)
param 'sentencesList': a list of text sentences
param 'languageId' : the language id for the sentences list
param 'bEmpty' : empty current document is set otherwise
add to existing clusters
"""
if bEmpty:
self.reset()
#Add sentences as clusters
for line in sentencesList:
#Further sentence split to avoid long paragraphes
for utterance in re.split(ur"\t|;|:|!|\?", line, flags=re.UNICODE):
utterance = utterance.strip()
if len(utterance) > 0:
c = TextCluster(self, utterance)
c.setLanguage(languageId)
self.addDocumentLine(c)
def _getLanguage2ClustersDict(self):
"""Map languages with a list of clusters.
return a dictionary with one entry per
language.
"""
languageDict = {}
for textCluster in self.listContent:
clusterLanguageId = textCluster.getLanguageId()
#First cluster
if clusterLanguageId not in languageDict:
languageDict[clusterLanguageId] = []