postgres/typesense
1
0
Fork 0
mirror of https://github.com/typesense/typesense.git synced 2025-05-20 05:32:30 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Keep the hashset bounded by deleting the element to be replaced in heap from the set.

Browse Source
This commit is contained in:
Kishore Nallan 2017-01-09 19:17:53 +05:30
parent 3e8f9298a9
commit ce471c9bb1
2 changed files with 17 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
include/topster.h
View File

@ -33,18 +33,20 @@ struct Topster {
}
void add(const uint64_t &key, const uint64_t &match_score, const int64_t &primary_attr, const int64_t &secondary_attr){
if(dedup_keys.count(key) != 0) {
return ;
}
dedup_keys.insert(key);
if (size >= MAX_SIZE) {
if(!is_greater(data[0], match_score, primary_attr, secondary_attr)) {
// when incoming value is less than the smallest in the heap, ignore
return;
}
if(dedup_keys.count(key) != 0) {
// when the key already exists, ignore
return ;
}
dedup_keys.erase(data[0].key);
dedup_keys.insert(key);
data[0].key = key;
data[0].match_score = match_score;
data[0].primary_attr = primary_attr;
@ -67,6 +69,13 @@ struct Topster {
i = next;
}
} else {
if(dedup_keys.count(key) != 0) {
// when the key already exists, ignore
return ;
}
dedup_keys.insert(key);
data[size].key = key;
data[size].match_score = match_score;
data[size].primary_attr = primary_attr;

4
src/collection.cpp
View File

@ -537,9 +537,9 @@ void Collection::remove(std::string id) {
}
std::string Collection::get_seq_id_key(uint32_t seq_id) {
return collection_id + "_" + SEQ_ID_PREFIX + std::to_string(seq_id);
return std::to_string(collection_id) + "_" + SEQ_ID_PREFIX + std::to_string(seq_id);
}
std::string Collection::get_doc_id_key(std::string doc_id) {
return collection_id + "_" + DOC_ID_PREFIX + doc_id;
return std::to_string(collection_id) + "_" + DOC_ID_PREFIX + doc_id;
}