def handleWord(self, word):
#logger.debug(util.funcName('begin'))
wordText1 = ""
wordText2 = ""
punct = None
is_first_text = True
for childNode in word.childNodes:
if not childNode.attributes:
continue
elif childNode.getAttribute("type") == "txt":
text = xmlutil.getElemText(childNode)
if is_first_text:
wordText1 = text
is_first_text = False
else:
wordText2 = text
elif childNode.getAttribute("type") == "punct":
punct = xmlutil.getElemText(childNode)
break
if punct:
if self.ex.wordList:
self.ex.addPunctuation(punct)
else:
self.ex.appendWord(punct, punct)
#logger.debug(util.funcName('return', args=punct))
return
morphemes = word.getElementsByTagName("morph")
if len(morphemes):
self.handleWordMorphemes(morphemes)
else:
self.ex.appendMorphObj(singleMorphemeWord(word))
self.ex.appendWord(wordText1, wordText2)
def singleMorphemeWord(word):
"""For words consisting of a single morpheme, get word-level
attributes instead of morpheme-level.
"""
morph = lingex_structs.LingGramMorph()
items = word.getElementsByTagName("item")
for item in items:
if item.attributes is None:
continue
itemType = item.getAttribute("type")
if itemType == "gls":
if morph.gloss and not morph.text1:
morph.text1 = morph.gloss
morph.gloss = xmlutil.getElemText(item)
elif itemType == "msa":
morph.pos = xmlutil.getElemText(item)
return morph
def handleSentence(self, sentence):
logger.debug(util.funcName('begin'))
if self.use_segnum:
for childNode in sentence.childNodes:
if not childNode.attributes:
continue
if childNode.getAttribute("type") == "segnum":
self.ex.refText = xmlutil.getElemText(childNode).strip()
break
words = sentence.getElementsByTagName("word")
for childNode in sentence.childNodes:
if childNode.attributes is None:
continue
if childNode.getAttribute("type") == "gls":
self.ex.freeTrans = xmlutil.getElemText(childNode)
for word in words:
self.handleWord(word)
if self.prefix:
self.ex.refText = self.prefix + self.ex.refText
def handleWordMorphemes(self, morphemes):
#logger.debug(util.funcName('begin'))
mergedMorphemes = MergedMorphemes()
for morpheme in morphemes:
items = morpheme.getElementsByTagName("item")
morph = lingex_structs.LingGramMorph()
is_first_text = True
for item in items:
if item.attributes is None:
continue
itemType = item.getAttribute("type")
if itemType == "txt":
text = xmlutil.getElemText(item)
if is_first_text:
morph.text1 = text
is_first_text = False
else:
morph.text2 = text
elif itemType == "cf":
# lex entry, typically same as morph text
pass
elif itemType == "gls":
morph.gloss = xmlutil.getElemText(item)
elif itemType == "msa":
morph.pos = xmlutil.getElemText(item)
if self.config.separateMorphColumns:
## store each morpheme separately
#logger.debug(morph.text)
self.ex.appendMorphObj(morph)
else:
#logger.debug(morph.text)
mergedMorphemes.add(morph)
if not self.config.separateMorphColumns:
self.ex.appendMorphObj(
mergedMorphemes.getMorph(self.config.get_showMorphemeBreaks()))