bool init(const string& dictPath, const string& hmmFilePath, const string& charStatus, const string& startProb, const string& emitProb, const string& endProb, const string& transProb)
{
assert(!_getInitFlag());
_dictTrie.init(dictPath);
assert(_dictTrie);
return _setInitFlag(_segment.init(dictPath, hmmFilePath));
};
bool init(const string& dictPath, const string& userDictPath = "")
{
if(_getInitFlag())
{
LogError("already inited before now.");
return false;
}
_dictTrie.init(dictPath, userDictPath);
assert(_dictTrie);
LogInfo("MPSegment init(%s) ok", dictPath.c_str());
return _setInitFlag(true);
}
TEST(DictTrieTest, Test1) {
string s1, s2;
DictTrie trie;
trie.init(DICT_FILE);
ASSERT_LT(trie.getMinWeight() + 15.6479, 0.001);
string word("来到");
Unicode uni;
ASSERT_TRUE(TransCode::decode(word, uni));
DictUnit nodeInfo;
nodeInfo.word = uni;
nodeInfo.tag = "v";
nodeInfo.weight = -8.87033;
s1 << nodeInfo;
s2 << (*trie.find(uni.begin(), uni.end()));
EXPECT_EQ("[\"26469\", \"21040\"] v -8.870", s2);
word = "清华大学";
LocalVector<pair<size_t, const DictUnit*> > res;
//vector<pair<size_t, const DictUnit* > resMap;
LocalVector<pair<size_t, const DictUnit*> > res2;
const char * words[] = {"清", "清华", "清华大学"};
for(size_t i = 0; i < sizeof(words)/sizeof(words[0]); i++) {
ASSERT_TRUE(TransCode::decode(words[i], uni));
res.push_back(make_pair(uni.size() - 1, trie.find(uni.begin(), uni.end())));
//resMap[uni.size() - 1] = trie.find(uni.begin(), uni.end());
}
//DictUnit
//res.push_back(make_pair(0, ))
vector<pair<size_t, const DictUnit*> > vec;
ASSERT_TRUE(TransCode::decode(word, uni));
ASSERT_TRUE(trie.find(uni.begin(), uni.end(), res2, 0));
s1 << res;
s2 << res;
ASSERT_EQ(s1, s2);
}
bool init(const string& dictPath, const string& userDictPath = "")
{
LIMONP_CHECK(_dictTrie.init(dictPath, userDictPath));
LogInfo("MPSegment init(%s) ok", dictPath.c_str());
return true;
}