// This creates the final keyword set that composes each of the excerpts. Only // the top N keywords are stored which are ranked on their keyword weight. // @param hit_list_breadth - this is the number of hit list divisions // @param max_keyword_size - this is the maximum number of keywords // - allowed in the initial excerp set void CompileExcerptKeywordSet(int client_id, int client_num, int hit_list_breadth, int max_keyword_size) { CNodeStat::SetClientID(client_id); CNodeStat::SetClientNum(client_num); CNodeStat::SetHashDivNum(hit_list_breadth); m_tok_queue.Initialize(max_keyword_size, CompareTokens); m_sorted_tok_set.Initialize(max_keyword_size); Initialize(); m_hit_enc_file.OpenReadFile(CUtility::ExtendString ("LocalData/hit_encoding", GetClientID())); m_excerpt_occur_map_file.OpenReadFile(CUtility::ExtendString ("LocalData/excerpt_occurr", GetClientID())); for(int i=0; i<m_assoc_id_file.OverflowSize(); i++) { m_assoc_id_file[i].OpenReadFile(CUtility::ExtendString ("LocalData/mapped_assoc_file", i, ".client", GetClientID())); } cout<<endl<<"Writing Excerpt Documents"<<endl; WriteDocuments(); }