def test_NRPattern(self):
"""
"""
trie = DoubleArrayTrie()
NRPattern.sort()
trie.build(key=NRPattern)
self.assertTrue(trie.exact_match_search("BCD") != -1)
self.assertTrue(trie.exact_match_search("BBCD") != -1)
self.assertTrue(trie.exact_match_search("BG") != -1)
self.assertTrue(trie.exact_match_search("DG") != -1)
self.assertTrue(trie.exact_match_search("CD") == -1)
def test_combin_by_dict(self):
dat = DoubleArrayTrie()
dat.build([u"江", u"河", u"湖", "海"])
text = u"江河湖海"
word_net = WordNet(text)
gen_word_net(text, word_net, dat)
vertexs = [v[0] for v in word_net.vertexs]
self.assertEqual(len(word_net), 6, u"自定义字典分词")
combin_dat = DoubleArrayTrie()
combin_dat.build(key=[u"江河湖海"], v=[u"江河湖海 n 1"])
vertexs = combine_by_custom_dict(vertexs, combin_dat)
self.assertEqual(len(vertexs), 3, u"合并完成后应该只有前尾加中间词")
def test_combin_by_dict(self):
dat = DoubleArrayTrie()
dat.build([u"江", u"河", u"湖", "海"])
text = u"江河湖海"
word_net = WordNet(text)
gen_word_net(text, word_net, dat)
vertexs = [v[0] for v in word_net.vertexs]
self.assertEqual(len(word_net), 6, u"自定义字典分词")
combin_dat = DoubleArrayTrie()
combin_dat.build(key=[u"江河湖海"], v=[u"江河湖海 n 1"])
vertexs = combine_by_custom_dict(vertexs, combin_dat)
self.assertEqual(len(vertexs), 3, u"合并完成后应该只有前尾加中间词")
def __init__(self):
self.trie = DoubleArrayTrie()
NSPattern.sort()
self.trie.build(key=NSPattern)
def __init__(self):
self.trie = DoubleArrayTrie.load(config.PLACE_DICT_NAME, enum_cls=NS)
self.matrix = HMMMatrix.load(config.PLACE_TR_PATH, NS)
class NRPatternDict:
def __init__(self):
self.trie = DoubleArrayTrie()
NRPattern.sort()
self.trie.build(key=NRPattern)
def __init__(self):
self.trie = DoubleArrayTrie.load(config.TRADITIONAL_CHINESE_DICT_NAME,
lambda i: i[:i.find(u'=')],
lambda i: i.split('='))
self.trie.get_attr = lambda v: v
def __init__(self):
self.trie = DoubleArrayTrie.load(config.TRADITIONAL_CHINESE_DICT_NAME,
lambda i: i[i.find(u'=') + 1:],
lambda i: i.split('=')[::-1],
dict_bin_ext=config.DICT_BIN_REVERSE_EXT)
self.trie.get_attr = lambda v: v
