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
C++ algoliadatastoreelasticsearchenterprise-searchfacetingfull-text-searchfuzzy-searchgeosearchin-memoryinstantsearchmerchandisingpineconesearchsearch-enginesemantic-searchsimilarity-searchsite-searchsynonymstypo-tolerancevector-search
Enabling exhaustive search should automatically drop tokens.
Kishore Nallanf40637fe643d
diff --git a/src/index.cpp b/src/index.cppindex 94f084cd..f599eab6 100644--- a/src/index.cpp+++ b/src/index.cpp@@ -2708,7 +2708,7 @@ void Index::search(std::vector<query_tokens_t>& field_query_tokens, const std::v// gather up both original query and synonym queries and do drop tokens- if (all_result_ids_len < drop_tokens_threshold) {+ if (exhaustive_search || all_result_ids_len < drop_tokens_threshold) {for (size_t qi = 0; qi < all_queries.size(); qi++) {auto& orig_tokens = all_queries[qi];size_t num_tokens_dropped = 0;diff --git a/test/collection_specific_more_test.cpp b/test/collection_specific_more_test.cppindex 608c5fea..3527bbec 100644--- a/test/collection_specific_more_test.cpp+++ b/test/collection_specific_more_test.cpp@@ -1782,6 +1782,34 @@ TEST_F(CollectionSpecificMoreTest, SearchCutoffTest) {ASSERT_EQ(408, coll_op.code());}+TEST_F(CollectionSpecificMoreTest, ExhaustiveSearchWithoutExplicitDropTokens) {+ nlohmann::json schema = R"({+ "name": "coll1",+ "fields": [+ {"name": "title", "type": "string"}+ ]+ })"_json;++ Collection* coll1 = collectionManager.create_collection(schema).get();++ nlohmann::json doc;+ doc["title"] = "alpha beta gamma";+ ASSERT_TRUE(coll1->add(doc.dump()).ok());++ doc["title"] = "alpha";+ ASSERT_TRUE(coll1->add(doc.dump()).ok());++ bool exhaustive_search = true;+ size_t drop_tokens_threshold = 1;++ auto res = coll1->search("alpha beta", {"title"}, "", {}, {}, {0}, 3, 1, FREQUENCY, {false}, drop_tokens_threshold,+ spp::sparse_hash_set<std::string>(),+ spp::sparse_hash_set<std::string>(), 10, "", 30, 4, "title", 20, {}, {}, {}, 0,+ "<mark>", "</mark>", {}, 1000, true, false, true, "", exhaustive_search).get();++ ASSERT_EQ(2, res["hits"].size());+}+TEST_F(CollectionSpecificMoreTest, CrossFieldTypoAndPrefixWithWeights) {nlohmann::json schema = R"({"name": "coll1",