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
Fix stemming prefix highlight.
Kishore Nallan7e8aa01056e2
diff --git a/src/collection.cpp b/src/collection.cppindex ddaafe2e..735aeec8 100644--- a/src/collection.cpp+++ b/src/collection.cpp@@ -4151,6 +4151,11 @@ bool Collection::handle_highlight_text(std::string& text, bool normalise, const}}+ if(num_letters < prefix_token_num_chars) {+ // can happen in the case of stemming+ prefix_end = tok_start + num_letters;+ }+size_t char_diff = num_letters - prefix_letters;auto new_tok_end = (char_diff <= 2 && qtoken_it.value().num_typos != 0) ? tok_end : prefix_end;token_offsets.emplace(tok_start, new_tok_end);diff --git a/test/collection_specific_more_test.cpp b/test/collection_specific_more_test.cppindex a67b205a..8bb422c6 100644--- a/test/collection_specific_more_test.cpp+++ b/test/collection_specific_more_test.cpp@@ -2914,6 +2914,54 @@ TEST_F(CollectionSpecificMoreTest, StemmingEnglishWithCaps) {ASSERT_STREQ("1", res["hits"][1]["document"]["id"].get<std::string>().c_str());}+TEST_F(CollectionSpecificMoreTest, StemmingEnglishPrefixHighlight) {+ nlohmann::json schema = R"({+ "name": "coll1",+ "fields": [+ {+ "facet": false,+ "index": true,+ "infix": true,+ "locale": "",+ "name": "name",+ "optional": false,+ "sort": false,+ "stem": false,+ "store": true,+ "type": "string"+ },+ {+ "facet": true,+ "index": true,+ "infix": true,+ "locale": "",+ "name": "subClass",+ "optional": true,+ "sort": false,+ "stem": true,+ "store": true,+ "type": "string"+ }+ ]+ })"_json;++ Collection* coll1 = collectionManager.create_collection(schema).get();++ nlohmann::json doc;+ doc["id"] = "0";+ doc["name"] = "Generic Red Onions";+ doc["subClass"] = "ONIONS";++ ASSERT_TRUE(coll1->add(doc.dump()).ok());++ std::vector<sort_by> sort_fields = {};+ auto res = coll1->search("onions", {"subClass","name"}, "", {}, sort_fields, {2}, 10, 1, FREQUENCY, {true}, 0).get();++ ASSERT_EQ(1, res["hits"].size());+ ASSERT_EQ("Generic Red <mark>Onions</mark>", res["hits"][0]["highlight"]["name"]["snippet"].get<std::string>());+ ASSERT_EQ("<mark>ONIONS</mark>", res["hits"][0]["highlight"]["subClass"]["snippet"].get<std::string>());+}+TEST_F(CollectionSpecificMoreTest, StemmingEnglishHighlights) {nlohmann::json schema = R"({"name": "test",