SpringBoot Data整合ElasticSearch

pom依赖

<!-- spring data和es的start依赖，会引入关联的elasticsearch-rest-high-level-clinet等依赖 -->
<dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-data-elasticsearch</artifactId>
</dependency>

RestHighLevelClient方式整合

RestHighLevelClient是ElasticSearch为java语言提供的客户端。我们可以通过配置类的方式引入到spring boot项目中，代码如下：

@Configuration
public class RestClientConfig extends AbstractElasticsearchConfiguration {private static final int PORT = 9200;private static final String scheme = "http";@Value("${spring.elasticsearch.host}")private String hostAndPort;@Value("${spring.elasticsearch.username}")private String username;@Value("${spring.elasticsearch.password}")private String password;@Override@Beanpublic RestHighLevelClient elasticsearchClient() {final CredentialsProvider credentialsProvider = new BasicCredentialsProvider();credentialsProvider.setCredentials(AuthScope.ANY, new UsernamePasswordCredentials(username, password));// 通过builder创建rest client，配置http client的HttpClientConfigCallback。RestClientBuilder builder = RestClient.builder(new HttpHost(hostAndPort, PORT, scheme)).setHttpClientConfigCallback(httpClientBuilder -> httpClientBuilder.setDefaultCredentialsProvider(credentialsProvider));// RestHighLevelClient实例通过REST low-level client builder进行构造。return new RestHighLevelClient(builder);}
}

配置好RestHighLevelClient之后，就可以通过他操作es，下面列出了常见的几个方法的Api。关于更多的细节参考：elasticsearch.clients.rest

@Service
public class EsServiceDemo {@ResourceRestHighLevelClient highLevelClient;public void add(Log log) throws IOException {IndexRequest request = new IndexRequest(log.getIndex());request.source(JSONObject.toJSONString(log), XContentType.JSON);request.setRefreshPolicy(WriteRequest.RefreshPolicy.IMMEDIATE);highLevelClient.index(request, RequestOptions.DEFAULT);}public void update(Log log) throws IOException {GetRequest getRequest = new GetRequest(log.getIndex(), log.getId());boolean exists = highLevelClient.exists(getRequest, RequestOptions.DEFAULT);if (exists) {UpdateRequest updateRequest = new UpdateRequest(log.getIndex(), log.getId());updateRequest.doc(JSONObject.toJSONString(log), XContentType.JSON);highLevelClient.update(updateRequest, RequestOptions.DEFAULT);}}public void delete(String index, String id) throws IOException {DeleteRequest request = new DeleteRequest(index, id);highLevelClient.delete(request, RequestOptions.DEFAULT);}public Log searchById(String index, String id) throws IOException {GetRequest request = new GetRequest(index, id);GetResponse response = highLevelClient.get(request, RequestOptions.DEFAULT);String source = response.getSourceAsString();return JSONObject.parseObject(source, Log.class);}public List<Log> searchByKeyword(String index, String keyword) throws IOException {SearchRequest request = new SearchRequest(index);// 构建查询条件SearchSourceBuilder builder = new SearchSourceBuilder();MultiMatchQueryBuilder multiMatchQueryBuilder = QueryBuilders.multiMatchQuery(keyword, "clientIp", "message");builder.query(multiMatchQueryBuilder);HighlightBuilder highlightBuilder = new HighlightBuilder();highlightBuilder.field("message");highlightBuilder.preTags("<font color='red'>");highlightBuilder.postTags("</font>");builder.highlighter(highlightBuilder);// 将查询条件设置到请求中request.source(builder);SearchResponse response = highLevelClient.search(request, RequestOptions.DEFAULT);SearchHit[] hits = response.getHits().getHits();List<Log> logs = new LinkedList<>();for (SearchHit hit : hits) {String json = hit.getSourceAsString();Log log = JSONObject.parseObject(json, Log.class);// 设置高亮的文本到实体类中Map<String, HighlightField> highlightFields = hit.getHighlightFields();HighlightField message = highlightFields.get("message");if (null != message) {// 获取的高亮片段Text[] fragments = message.getFragments();StringBuilder sb = new StringBuilder();for (Text text : fragments) {sb.append(text);}log.setMessage(sb.toString());}logs.add(log);}return logs;}
}

ElasticsearchRepository方式整合

上面的方式我们在配置Bean中手动注册了客户端，spring boot也可以为我们自动装配客户端，当然这个需要我们将es的一些基本配置写在配置文件中。

spring.elasticsearch.uris=
spring.elasticsearch.username=
spring.elasticsearch.password=

上面用到的是es为我们提供的原生客户端，其实还是有一些麻烦的，和业务无关的代码确实有点多了。那spring data致力于为所有的数据库提供一整套统一的增删改查接口，当然也支持es。上面我们的依赖就会引入spring-data-elasticsearch的相关依赖。这里有一个很重要的接口，就是ElasticsearchRepository这个接口，他继承自CrudRepository，默认提供了增删改查的基本方法，并基于es扩展了一些方法。我们只需要新加一个接口，指定泛型，他就会为我们自动提供增删改查的能力。

public interface ProductRepository extends ElasticsearchRepository<Product, Long> {}

上面的接口我们指定泛型为Product，就是说要对Product做一些CRUD操作，但是他ES在查询的时候怎么知道你这个数据是在那个索引呢，es中的Field和你的Bean中的属性怎么对应呢。这里我们就需要在这个Product对象中进行一些配置了。

@Data
@NoArgsConstructor
@AllArgsConstructor
@Document(indexName = "products")  // 表示是ES中的一个文档，对应的索引是products
public class Product {@Idprivate Long id;@Field(type = FieldType.Text)  // Field注解将bean中的属性和es中的field进行映射private String title;@Field(type = FieldType.Keyword)private String category;@Field(type = FieldType.Double)private Double price;@Field(type = FieldType.Text)private String brand;@Field(type = FieldType.Float)private Float discount;
}

在上面的准备工作做好以后，es就会为我们自动装配es客户端，然后我们只要注入我们定义的这个接口ProductRepository，就可以使用他的基础的CRUD方法，下面是通过单元测试的方法进行验证的，亲测可行。

@SpringBootTest
@RunWith(SpringRunner.class)
public class ProductTest {@Autowiredprivate ProductRepository productRepository;@Testpublic void testAdd() {Product product = new Product();product.setId(1L);product.setTitle("Hongxing ERKE (ERKE) official flagship sportswear men's winter winter warm casual windproof sportswear men's coat black L(weight 120-140 kg)");product.setBrand("ERKE");product.setPrice(100.0);product.setDiscount(0.7F);product.setCategory("clothes");Product save = productRepository.save(product);Assert.assertEquals(1L, save.getId().longValue());}@Testpublic void testBatchAdd() {List<Product> products = new LinkedList<>();products.add(new Product(2L, "qiaodan trousers", "clothes", 97.5, "qiaodan", 0.9F));products.add(new Product(3L, "qiaodan Casual shoes", "shoes", 87.5, "qiaodan", 0.85F));products.add(new Product(4L, "nike fleece", "clothes", 197.5, "nike", 0.8F));products.add(new Product(5L, "nike trousers", "shoes", 97.5, "nike", 0.8F));Iterable<Product> result = productRepository.saveAll(products);}@Testpublic void testFindById() {Optional<Product> optional = productRepository.findById(1L);if (optional.isPresent()) {Product product = optional.get();Assert.assertEquals(1L, product.getId().longValue());}}
}

这还不是最牛逼的，最牛逼的是，他除了上面的功能之外，还支持自定义方法，并且将自定义的方法自动翻译成es对应的查询语言，比如我们定义这样一个方法。那么不需要多说什么，es就会知道这个方法的意思是查询价格介于lowPrice和highPrice之间的商品。

public interface ProductRepository extends ElasticsearchRepository<Product, Long> {List<Product> findByPriceBetween(double lowPrice, double highPrice);
}

@Test
public void testFindByPriceBetween() {List<Product> result = productRepository.findByPriceBetween(95.0, 150.0);Assert.assertEquals(3, result.size());
}

这里举一个简单的例子，说明如何使用，关于更多更详细的介绍参考：Elasticsearch Repositories

ElasticsearchRestTemplate方式整合

除了上面的方式，sping data还支持通过ElasticsearchRestTemplate的方式去访问es，这种方式我们可以灵活的构建原生的es查询语句，功能更加丰富。这个ElasticsearchRestTemplate对象也是spring boot为我们自动装配的。

@SpringBootTest
@RunWith(SpringRunner.class)
public class ProductTest {@Autowiredprivate ElasticsearchRestTemplate elasticsearchRestTemplate;@Testpublic void testSearch() {NativeSearchQueryBuilder queryBuilder = new NativeSearchQueryBuilder();// 通过QueryBuilders构建一个dis_max查询，这里可以构建所有的es的查询DisMaxQueryBuilder disMaxQueryBuilder = QueryBuilders.disMaxQuery();disMaxQueryBuilder.add(QueryBuilders.matchQuery("title", "qiaodan"));disMaxQueryBuilder.add(QueryBuilders.matchQuery("category", "qiaodan"));disMaxQueryBuilder.tieBreaker(0.7F);// 通过search方法发起请求，都有一样的套路SearchHits<Product> searchHits = elasticsearchRestTemplate.search(queryBuilder.withQuery(disMaxQueryBuilder).build(), Product.class);for (SearchHit<Product> searchHit : searchHits) {Product content = searchHit.getContent();System.out.println(JSONObject.toJSON(content));}}
}