Spring Boot 2.0.5 Integration with Elasticsearch: Installation, Configuration, Entity Mapping, CRUD, and Full‑Text Search

This article provides a step‑by‑step tutorial for integrating Spring Boot 2.0.5 (springBootVersion = '2.0.5.RELEASE') with Elasticsearch 5.6.11.

Installation : Download Elasticsearch and Kibana from the official Elastic website.

Dependencies (add to build.gradle or pom.xml):

springBootVersion = '2.0.5.RELEASE'
compile group: 'org.springframework.boot', name: 'spring-boot-starter-data-elasticsearch'
// Ensure the transport client version matches the Elasticsearch version
compile 'org.elasticsearch.client:transport:5.6.11'

Configuration can be done either in application.yml:

spring:
  data:
    elasticsearch:
      cluster-name: springboot-elasticsearch
      cluster-nodes: 127.0.0.1:9300
      repositories:
        enabled: true

or programmatically with a Kotlin configuration class:

@Configuration
@EnableElasticsearchRepositories(basePackages = ["com.futao.springmvcdemo.dao"])
open class ElasticSearchConfiguration {
    @Bean
    fun client(): TransportClient {
        val node = InetSocketTransportAddress(InetAddress.getByName("127.0.0.1"), 9300)
        val settings = Settings.builder()
            .put("cluster.name", "springboot-elasticsearch")
            .build()
        return PreBuiltTransportClient(settings).addTransportAddress(node)
    }
}

Entity Mapping uses the @Document annotation to map a Kotlin/Java class to an Elasticsearch index and type:

@Document(indexName = "futao", type = "article")
class Article : BaseEntity() {
    private String title;
    private String description;
    private String content;
    // getters and setters omitted for brevity
}

DAO Layer defines an interface extending ElasticsearchRepository:

interface ArticleSearchDao : ElasticsearchRepository<Article, String>

Service Layer saves articles and performs full‑text search using QueryBuilders:

@Service
open class ArticleServiceImpl : ArticleService {
    @Resource
    lateinit var elasticsearch: ArticleSearchDao
    @Resource
    lateinit var client: Client

    override fun list(): List<Article> {
        val list = articleDao.list()
        client.saveAll(list)
        return list
    }

    override fun search(key: String): ArrayList<Article> {
        val hits = client.prepareSearch("futao")
            .setTypes("article")
            .setQuery(QueryBuilders.boolQuery()
                .should(QueryBuilders.matchQuery(Article::getContent.getFieldName(), key))
                .should(QueryBuilders.matchQuery(Article::getTitle.getFieldName(), key))
                .should(QueryBuilders.matchQuery(Article::getDescription.getFieldName(), key)))
            .execute().actionGet().hits
        val result = arrayListOf<Article>()
        hits.forEach { result.add(JSONObject.parseObject(it.sourceAsString, Article::class.java)) }
        return result
    }
}

Controller Layer exposes REST endpoints for adding, listing, and searching articles:

@RestController
@RequestMapping(path = "article", produces = MediaType.APPLICATION_JSON_UTF8_VALUE)
class ArticleController {
    @Resource
    lateinit var articleService: ArticleService

    @PostMapping("add")
    fun add(@RequestParam("title") title: String,
            @RequestParam("desc") desc: String,
            @RequestParam("content") content: String): SingleValueResult {
        articleService.add(title, desc, content)
        return SingleValueResult("success")
    }

    @GetMapping("list")
    fun list(): List<Article> = articleService.list()

    @GetMapping("search")
    fun search(@RequestParam("key") key: String): List<Article> = articleService.search(key)
}

Startup Fix for Netty Conflict : Before running the application, set the system property to disable Netty's processor detection:

System.setProperty("es.set.netty.runtime.available.processors", "false")

Optionally enable FastJSON auto‑type support for Redis deserialization: ParserConfig.getGlobalInstance().setAutoTypeSupport(true) Testing : After starting the Spring Boot application, you can verify the health endpoint, use the /article/list endpoint to insert data into Elasticsearch, view the data in Kibana, and perform full‑text searches via /article/search?key=.... The stored data resides under Elasticsearch’s data directory, which can be cleared to delete all indices.