Introduction into Spring Data Elasticsearch

A presentation at Elastic User Group by Alexander Reelsen

Spring Data Elasticsearch - 2021 edition Alexander Reelsen alex@elastic.co | @spinscale

Agenda Spring Data Introduction Spring Data Elasticsearch Spring Data Elasticsearch Reactive Complex Queries What’s next? Q&A

Spring Data Spring Data’s mission is to provide a familiar and consistent, Spring-based programming model for data access while still retaining the special traits of the underlying data store.

Spring Data Implementations: JPA, JDBC, LDAP, MongoDB, Redis, Cassandra Community: Aerospike, ArangoDB, Couchbase, Google Spanner, DynamoDb, Hazelcast, Neo4j, Solr, Yugabyte, Elasticsearch

Spring Data ElasticsearchTemplate helper class Object Mapping, Annotation based metadata ( @Id , @Field ) Repositories

Entity @Document(indexName = “products”) public class Product { @Id private String id; @Field(type = FieldType.Text) private String name; @Field(type = FieldType.Text) private String description; @Field(type = FieldType.Keyword) private List<String> tags; @Field(type = FieldType.Keyword, name = “image_url”) private String imageUrl; @Field(type= FieldType.Date, format = DateFormat.date_time) private ZonedDateTime created; … }

Repositories method names reflect finders select fields, ranges, specify sorting, i.e. findByNameAndPrice , findByPriceGreaterThan Note: Inefficient queries like findByNameEndingWith / findByNameStartingWith / findByNameContaining

Repositories public interface UserProfileRepository extends ElasticsearchRepository<UserProfile, String> { }

Reactive Repostiories public interface ProductRepository extends ReactiveSortingRepository<Product, String> { { @Query(“”” “bool” : { “must” : [ { “multi_match” : { “query”: “?0”, “fields”: [ “description”, “name”, “tags” ] } } ], “should” : [ { “match” : { “name” : “?0” } } ] } } “”“) Flux<Product> findProducts(String q); }

Reactive Repositories @RestController @RequestMapping(“/products”) public class ProductController { @GetMapping(value = “/product/{id}”) public Mono<Product> retrieveProduct(@PathVariable String id) { // this does not return a 404 when the Mono is empty… weird default mode IMO return repository.findById(id); } @DeleteMapping(“/product/{id}”) public Mono<Void> deleteProduct(@PathVariable String id) { return repository.deleteById(id); } } @GetMapping(value = “/search”) public Flux<Product> search(@RequestParam String q) { return repository.findProducts(q); }

ElasticsearchTemplate final BoolQueryBuilder qb = QueryBuilders.boolQuery() .must(QueryBuilders.termQuery(“url”, contribution.getUrl())) .must(QueryBuilders.termQuery(“submitted_by.email”, login(principal))) .must(QueryBuilders.termQuery(“category”, contribution.getCategory())); final SearchHit<Contribution> hit = elasticsearchRestTemplate.searchOne(new NativeSearchQuery(qb), Contribution.class);

Complex queries QueryBuilder qb = QueryBuilders.boolQuery() .mustNot(QueryBuilders.termQuery(“category”, Contribution.Category.CONTRIBUTION_RATING)) .filter(QueryBuilders.termQuery(“region”, region)) .filter(QueryBuilders.termQuery(“state”, Contribution.State.APPROVED)) .filter(QueryBuilders.termQuery(“cycle”, Cycle.current())); final NativeSearchQuery query = new NativeSearchQuery(qb); query.setPageable(Pageable.unpaged()); // ensure uniqueness by email, then retrieve latest fullname query.addAggregation(AggregationBuilders.terms(“by_user”).field(“submitted_by.email”).size(100) .subAggregation(AggregationBuilders.topHits(“by_name”).size(1).sort(SortBuilders.fieldSort(“created_at”) .order(SortOrder.DESC)).fetchSource(“submitted_by.full_name”, “”))); final SearchHits<Contribution> hits = elasticsearchTemplate.search(query, Contribution.class);

Client configuration @Configuration public class ReactiveRestClientConfig extends AbstractReactiveElasticsearchConfiguration { @Override @Bean public ReactiveElasticsearchClient reactiveElasticsearchClient() { final String elasticsearchUrl = System.getenv(“ELASTICSEARCH_URL”); if (elasticsearchUrl != null) { final URI uri = URI.create(elasticsearchUrl); final int port = uri.getPort(); final String host = uri.getHost(); final InetSocketAddress addr = new InetSocketAddress(host, port); final ClientConfiguration.MaybeSecureClientConfigurationBuilder builder = ClientConfiguration.builder().connectedTo(addr); if (uri.getScheme().equals(“https”)) { builder.usingSsl(); } if (uri.getUserInfo() != null) { final String[] data = uri.getUserInfo().split(“:”, 2); builder.withBasicAuth(data[0], data[1]); } } } return ReactiveRestClients.create(builder.build()); } else { // fallback to localhost:9200 as default, with SSL final InetSocketAddress addr = new InetSocketAddress(“localhost”, 9200); final ClientConfiguration.MaybeSecureClientConfigurationBuilder builder = ClientConfiguration.builder().connectedTo(addr); return ReactiveRestClients.create(builder.build()); }

What’s next New Elasticsearch Client without Elasticsearch dependency Client gets generated from specification Spring Data Elasticsearch will move to the client at some point, still in development New client has the concept of a Transport, which could be reused when talking to Elastic solutions

Resources https://github.com/spinscale/spring-boot-reactive-observability-demo https://github.com/spinscale/link-rating https://docs.spring.io/springdata/elasticsearch/docs/current/reference/html/ https://github.com/elastic/ecs-logging-java https://spinscale.de/posts/2020-04-15-introduction-into-the-elasticsearchjava-rest-client.html https://spinscale.de/posts/2020-08-06-introduction-into-spring-dataelasticsearch.html

Books

Thanks for listening Q&A Alexander Reelsen alex@elastic.co | @spinscale

Alexander Reelsen
@spinscale

1 / 17

A quick introduction into Spring Data Elasticsearch, covering a few code samples to make it easier to follow.