void getNB(HashGraph* G, int v, int distance, hash_set<int>& result, vector<bool>& mark)
{
assert(distance==1 || distance==2);
EdgeMap* p_neighbors=G->getNeighbors(v);
EdgeMap::iterator pnb;
for (pnb=p_neighbors->begin(); pnb!=p_neighbors->end(); pnb++)
{
UINT w=pnb->first;
if(!mark[w])
result.insert(w);
}
if(distance==1)
return;
hash_set<int>::iterator p;
vector<int> temp;
for(p=result.begin(); p!=result.end(); p++)
temp.push_back(*p);
int imnb_size=result.size();
//result.clear();
for(int i=0; i<imnb_size; i++)
{
p_neighbors=G->getNeighbors(temp[i]);
for (pnb=p_neighbors->begin(); pnb!=p_neighbors->end(); pnb++)
{
UINT w=pnb->first;
if(!mark[w])
result.insert(w); //automatically handel duplication.
}
}
for(int i=0; i<imnb_size; i++)
result.erase(temp[i]);
}
int trim_entropy_filter(vector<string>* keep_words, hash_set<string>& cad_words_set, WordInfoMap& wordinfo_map)
{
keep_words->reserve(cad_words_set.size());
for (hash_set<string>::iterator it = cad_words_set.begin(); it != cad_words_set.end(); ++it) {
WordInfoMap::iterator it_map = wordinfo_map.find(*it);
if (it_map == wordinfo_map.end()) {
fprintf(stderr, "WARNING, word[%s] in cad_word, not in word_info", it->c_str());
continue;
}
if (it_map->first.size() <= WORD_LEN - 4 && it_map->second.calc_is_keep()) {
keep_words->push_back(*it);
}
}
return 0;
}