Skip to content

Open Source alternative to Algolia + Pinecone and an Easier-to-Use alternative to ElasticSearch ⚡ 🔍 ✨ Fast, typo tolerant, in-memory fuzzy Search Engine for building delightful search experiences

Aggregate match score when matched across array elements.

kishorencf4423367c856
diff --git a/src/index.cpp b/src/index.cpp
index de895219..07b23b7f 100644
--- a/src/index.cpp
+++ b/src/index.cpp
@@ -1524,10 +1524,24 @@ void Index::search(Option<uint32_t> & outcome,
art_fuzzy_search(search_index.at(field), (const unsigned char *) token.c_str(), token_len,
0, 0, 1, token_order, prefix_search, leaves);
- if(!leaves.empty() && leaves[0]->values->ids.contains(seq_id)) {
- words_present++;
+ if(leaves.empty()) {
+ continue;
+ }
+
+ uint32_t doc_index = leaves[0]->values->ids.indexOf(seq_id);
+
+ if (doc_index == leaves[0]->values->ids.getLength()) {
+ continue;
}
+ uint32_t start_offset = leaves[0]->values->offset_index.at(doc_index);
+ uint32_t end_offset = (doc_index == leaves[0]->values->ids.getLength() - 1) ?
+ leaves[0]->values->offsets.getLength() :
+ leaves[0]->values->offset_index.at(doc_index+1);
+
+ words_present += (end_offset - start_offset);
+ int k = 10;
+
/*if(!leaves.empty()) {
LOG(INFO) << "tok: " << leaves[0]->key;
}*/
@@ -1810,9 +1824,7 @@ void Index::score_results(const std::vector<sort_by> & sort_fields, const uint16
const Match & match = Match(seq_id, token_positions, false);
uint64_t this_match_score = match.get_match_score(total_cost, field_id);
- if(this_match_score > match_score) {
- match_score = this_match_score;
- }
+ match_score += this_match_score;
/*std::ostringstream os;
os << name << ", total_cost: " << (255 - total_cost)
diff --git a/test/collection_test.cpp b/test/collection_test.cpp
index fa783292..7c412c3a 100644
--- a/test/collection_test.cpp
+++ b/test/collection_test.cpp
@@ -2708,6 +2708,48 @@ TEST_F(CollectionTest, MultiFieldMatchRanking) {
collectionManager.drop_collection("coll1");
}
+TEST_F(CollectionTest, MultiFieldMatchRankingOnArray) {
+ Collection *coll1;
+
+ std::vector<field> fields = {field("name", field_types::STRING, false),
+ field("strong_skills", field_types::STRING_ARRAY, false),
+ field("skills", field_types::STRING_ARRAY, false),
+ field("points", field_types::INT32, false),};
+
+ coll1 = collectionManager.get_collection("coll1");
+ if(coll1 == nullptr) {
+ coll1 = collectionManager.create_collection("coll1", 1, fields, "points").get();
+ }
+
+ std::vector<std::vector<std::vector<std::string>>> records = {
+ {{"John Snow"}, {"Golang", "Vue", "React"}, {"Docker", "Goa", "Elixir"}},
+ {{"Jack Dan"}, {"Golang", "Phoenix", "React"}, {"Docker", "Vue", "Kubernetes"}},
+ };
+
+ for(size_t i=0; i<records.size(); i++) {
+ nlohmann::json doc;
+
+ doc["id"] = std::to_string(i);
+ doc["name"] = records[i][0][0];
+ doc["strong_skills"] = records[i][1];
+ doc["skills"] = records[i][2];
+ doc["points"] = i;
+
+ ASSERT_TRUE(coll1->add(doc.dump()).ok());
+ }
+
+ auto results = coll1->search("golang vue",
+ {"strong_skills", "skills"}, "", {}, {}, 0, 3, 1, FREQUENCY, true, 5).get();
+
+ ASSERT_EQ(2, results["found"].get<size_t>());
+ ASSERT_EQ(2, results["hits"].size());
+
+ ASSERT_STREQ("0", results["hits"][0]["document"]["id"].get<std::string>().c_str());
+ ASSERT_STREQ("1", results["hits"][1]["document"]["id"].get<std::string>().c_str());
+
+ collectionManager.drop_collection("coll1");
+}
+
TEST_F(CollectionTest, HighlightWithAccentedCharacters) {
Collection *coll1;