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
Don't consider dynamic field name having .* as a field.
Kishore Nallana99dd45bdc26
diff --git a/src/collection.cpp b/src/collection.cppindex 67140cd6..ebc671a4 100644--- a/src/collection.cpp+++ b/src/collection.cpp@@ -3095,6 +3095,12 @@ Option<bool> Collection::detect_new_fields(nlohmann::json& document,break;}+ // to prevent confusion we also disallow dynamic field names that contain ".*"+ if((kv.key() != ".*" && kv.key().find(".*") != std::string::npos)) {+ skip_field = true;+ break;+ }+new_field = dynamic_field.second;new_field.name = fname;found_dynamic_field = true;diff --git a/src/index.cpp b/src/index.cppindex e686168c..7a84e07a 100644--- a/src/index.cpp+++ b/src/index.cpp@@ -689,10 +689,11 @@ void Index::index_field_in_memory(const field& afield, std::vector<index_record>fhashvalues.hashes[i] = field_index_it->second.facet_hashes[i];}- if(facet_index_v3[afield.name][seq_id % ARRAY_FACET_DIM] == nullptr) {+ auto& facet_dim_index = facet_index_v3[afield.name][seq_id % ARRAY_FACET_DIM];+ if(facet_dim_index == nullptr) {LOG(ERROR) << "Error, facet index not initialized for field " << afield.name;} else {- facet_index_v3[afield.name][seq_id % ARRAY_FACET_DIM]->emplace(seq_id, std::move(fhashvalues));+ facet_dim_index->emplace(seq_id, std::move(fhashvalues));}}diff --git a/test/collection_all_fields_test.cpp b/test/collection_all_fields_test.cppindex f3542b49..3d4c56cf 100644--- a/test/collection_all_fields_test.cpp+++ b/test/collection_all_fields_test.cpp@@ -1517,6 +1517,7 @@ TEST_F(CollectionAllFieldsTest, FieldNameMatchingRegexpShouldNotBeIndexed) {doc1["id"] = "0";doc1["title"] = "One Two Three";doc1["name.*"] = "Rowling";+ doc1["name.*barbaz"] = "JK";doc1[".*"] = "foo";std::vector<std::string> json_lines;@@ -1539,3 +1540,36 @@ TEST_F(CollectionAllFieldsTest, FieldNameMatchingRegexpShouldNotBeIndexed) {ASSERT_EQ(1, results["hits"].size());}+TEST_F(CollectionAllFieldsTest, FieldNameMatchingRegexpShouldNotBeIndexedInNonAutoSchema) {+ std::vector<field> fields = {field("title", field_types::STRING, false),+ field("name.*", field_types::STRING, true, true)};++ Collection* coll1 = collectionManager.create_collection("coll1", 1, fields, "", 0, field_types::AUTO).get();++ nlohmann::json doc1;+ doc1["id"] = "0";+ doc1["title"] = "One Two Three";+ doc1["name.*"] = "Rowling";+ doc1["name.*barbaz"] = "JK";+ doc1[".*"] = "foo";++ std::vector<std::string> json_lines;+ json_lines.push_back(doc1.dump());++ coll1->add_many(json_lines, doc1, UPSERT);+ json_lines[0] = doc1.dump();+ coll1->add_many(json_lines, doc1, UPSERT);++ ASSERT_EQ(1, coll1->_get_index()->_get_search_index().size());+ ASSERT_EQ(2, coll1->get_fields().size());++ auto results = coll1->search("one", {"title"},+ "", {}, {}, {2}, 10,+ 1, FREQUENCY, {true},+ 1, spp::sparse_hash_set<std::string>(),+ spp::sparse_hash_set<std::string>(), 10, "", 30, 4, "title", 5, {}, {}, {}, 0,+ "<mark>", "</mark>", {}, 1000, true).get();++ ASSERT_EQ(1, results["hits"].size());+}+