从入门到进阶：Elasticsearch高级查询技巧详解

时间：2023-02-18 20:57:28

Elasticsearch是一款功能强大的全文搜索引擎，它使用Lucene搜索库进行底层索引和搜索。Elasticsearch提供了许多高级查询技巧，可以帮助用户更准确、更高效地查询数据。本教程将介绍Elasticsearch的高级查询技巧，并提供一些示例代码来说明它们的使用。

一、布尔查询

Elasticsearch支持布尔查询，包括AND、OR和NOT运算符。这使得用户可以使用多个条件来限制查询结果。

例如，以下查询将返回所有匹配“foo”和“bar”的文档：

GET /_search
{
"query": {
"bool": {
"must": [
        { "match": { "content": "foo" }},
        { "match": { "content": "bar" }}
      ]
    }
  }
}

此外，可以使用“should”查询来匹配任意一个条件。以下查询将返回匹配“foo”或“bar”的所有文档：

GET /_search
{
"query": {
"bool": {
"should": [
        { "match": { "content": "foo" }},
        { "match": { "content": "bar" }}
      ]
    }
  }
}

二、范围查询

Elasticsearch支持范围查询，可以用于查询一个字段是否在指定范围内。范围查询有两种类型：数值范围和日期范围。

例如，以下查询将返回所有年龄在18到30岁之间的用户：

GET /_search
{
"query": {
"range": {
"age": {
"gte": 18,
"lte": 30
      }
    }
  }
}

以下查询将返回所有注册日期在2019年1月1日到2020年1月1日之间的用户：

GET /_search
{
"query": {
"range": {
"registered_at": {
"gte": "2019-01-01",
"lte": "2020-01-01"
      }
    }
  }
}

三、模糊查询

Elasticsearch支持模糊查询，可以用于查询包含拼写错误或近似匹配的文档。模糊查询使用模糊匹配算法（如编辑距离算法）来找到近似匹配的文档。

例如，以下查询将返回包含“fox”或“fix”的文档：

GET /_search
{
"query": {
"fuzzy": {
"content": {
"value": "fox",
"fuzziness": "2"
      }
    }
  }
}

“fuzziness”参数指定了允许的最大编辑距离。在上面的例子中，“fuzziness”为2，表示查询将匹配编辑距离为1或2的文档。

四、正则表达式查询

Elasticsearch支持正则表达式查询，可以用于查询符合指定模式的文本。正则表达式查询可以使用“regexp”查询类型。

例如，以下查询将返回包含“foo”或“bar”的文档：

GET /_search
{
"query": {
"regexp": {
"content": "foo|bar"
    }
  }
}

五、通配符查询

Elasticsearch支持通配符查询，可以用于查询包含通配符模式的文本。通配符查询可以使用“wildcard”查询类型。

例如，以下查询将返回包含以“foo”或“bar”开头的文档：

GET /_search
{
"query": {
"wildcard": {
"content": "foo* OR bar*"
    }
  }
}

六、短语查询

Elasticsearch支持短语查询，可以用于查询包含一个或多个短语的文档。短语查询可以使用“match_phrase”查询类型。

例如，以下查询将返回包含短语“quick brown fox”或“lazy dog”的文档：

GET /_search
{
"query": {
"match_phrase": {
"content": "quick brown fox lazy dog"
    }
  }
}

七、高亮显示

Elasticsearch支持高亮显示查询结果中的关键字，可以用于使查询结果更易于理解。可以使用“highlight”参数来启用高亮显示。

例如，以下查询将返回包含“foo”或“bar”的文档，并将查询结果中的关键字高亮显示：

GET /_search
{
"query": {
"bool": {
"should": [
        { "match": { "content": "foo" }},
        { "match": { "content": "bar" }}
      ]
    }
  },
"highlight": {
"fields": {
"content": {}
    }
  }
}

八、分页和排序

Elasticsearch支持分页和排序查询结果。可以使用“from”和“size”参数来指定返回结果的起始位置和数量。可以使用“sort”参数来指定排序方式。

例如，以下查询将返回从第10个文档开始的5个文档，并按照“age”字段进行升序排序：

GET /_search
{
"from": 10,
"size": 5,
"query": {
"match_all": {}
  },
"sort": [
    { "age": "asc" }
  ]
}

九、聚合查询

Elasticsearch支持聚合查询，可以用于对文档进行统计和分组。聚合查询可以使用“aggs”参数来启用。

例如，以下查询将返回“content”字段中包含每个单词的文档数量：

GET /_search
{
"query": {
"match_all": {}
  },
"aggs": {
"word_count": {
"terms": {
"field": "content"
      }
    }
  }
}

以上就是Elasticsearch的一些高级查询技巧。下面将提供一些示例代码来说明它们的使用。

十、Java示例代码



示例代码如下：


import java.io.IOException;
import java.util.HashMap;
import java.util.Map;

import org.apache.http.HttpHost;
import org.elasticsearch.action.search.SearchRequest;
import org.elasticsearch.action.search.SearchResponse;
import org.elasticsearch.client.RequestOptions;
import org.elasticsearch.client.RestClient;
import org.elasticsearch.client.RestHighLevelClient;
import org.elasticsearch.common.text.Text;
import org.elasticsearch.index.query.MatchQueryBuilder;
import org.elasticsearch.index.query.QueryBuilders;
import org.elasticsearch.index.query.RangeQueryBuilder;
import org.elasticsearch.search.SearchHit;
import org.elasticsearch.search.SearchHits;
import org.elasticsearch.search.builder.SearchSourceBuilder;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightField;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder.Field;

public class ElasticsearchDemo {

public static void main(String[] args) throws IOException {
// 创建客户端
RestHighLevelClient client = new RestHighLevelClient(
RestClient.builder(new HttpHost("localhost", 9200, "http")));

// 创建索引和映射
createIndexAndMapping(client);

// 插入文档
insertDocument(client);

// 查询
MatchQueryBuilder matchQuery = QueryBuilders.matchQuery("content", "elasticsearch");
SearchRequest searchRequest = new SearchRequest("my_index");
SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
searchSourceBuilder.query(matchQuery);
searchRequest.source(searchSourceBuilder);
SearchResponse response = client.search(searchRequest, RequestOptions.DEFAULT);
printSearchResult(response);

// 带有高亮显示的查询
HighlightBuilder highlightBuilder = new HighlightBuilder();
highlightBuilder.field(new Field("content").preTags("<em>").postTags("</em>"));
searchSourceBuilder = new SearchSourceBuilder();
searchSourceBuilder.query(matchQuery);
searchSourceBuilder.highlighter(highlightBuilder);
searchRequest = new SearchRequest("my_index");
searchRequest.source(searchSourceBuilder);
response = client.search(searchRequest, RequestOptions.DEFAULT);
printSearchResultWithHighlight(response);

// 范围查询
RangeQueryBuilder rangeQuery = QueryBuilders.rangeQuery("publish_date")
                .from("2020-01-01")
                .to("2021-12-31");
searchSourceBuilder = new SearchSourceBuilder();
searchSourceBuilder.query(rangeQuery);
searchRequest = new SearchRequest("my_index");
searchRequest.source(searchSourceBuilder);
response = client.search(searchRequest, RequestOptions.DEFAULT);
printSearchResult(response);

// 排序
searchSourceBuilder = new SearchSourceBuilder();
searchSourceBuilder.query(matchQuery);
searchSourceBuilder.sort("publish_date");
searchRequest = new SearchRequest("my_index");
searchRequest.source(searchSourceBuilder);
response = client.search(searchRequest, RequestOptions.DEFAULT);
printSearchResult(response);

// 删除索引
deleteIndex(client);

// 关闭客户端
client.close();
    }

private static void createIndexAndMapping(RestHighLevelClient client) throws IOException {
// 创建索引
Map<String, Object> settings = new HashMap<>();
settings.put("number_of_shards", 1);
settings.put("number_of_replicas", 0);
Map<String, Object> mapping = new HashMap<>();
Map<String, Object> properties = new HashMap<>();
properties.put("title", Map.of("type", "text"));
properties.put("content", Map.of("type", "text"));
properties.put("publish_date", Map.of("type", "date"));
mapping.put("properties", properties);
client.indices().create(Map.of("index", "my_index", "settings", settings, "mapping", mapping),
RequestOptions.DEFAULT);
    }

private static void insertDocument(RestHighLevelClient client) throws IOException {
// 插入文档
Map<String, Object> document = new HashMap<>();
document.put("title", "Elasticsearch Guide");
document.put("content", "This is a guide to Elasticsearch.");
document.put("publish_date", "2021-03-01");
client.index(Map.of("index", "my_index", "id", "1", "body", document), RequestOptions.DEFAULT);
    }

private static void deleteIndex(RestHighLevelClient client) throws IOException {
// 删除索引
client.indices().delete(Map.of("index", "my_index"), RequestOptions.DEFAULT);
    }

private static void printSearchResult(SearchResponse response) {
// 打印查询结果
SearchHits hits = response.getHits();
System.out.println("Total hits: " + hits.getTotalHits().value);
System.out.println("Hits:");
for (SearchHit hit : hits) {
System.out.println("Id: " + hit.getId());
System.out.println("Score: " + hit.getScore());
System.out.println("Title: " + hit.getSourceAsMap().get("title"));
System.out.println("Content: " + hit.getSourceAsMap().get("content"));
System.out.println("Publish date: " + hit.getSourceAsMap().get("publish_date"));
        }
    }

private static void printSearchResultWithHighlight(SearchResponse response) {
// 打印带有高亮显示的查询结果
SearchHits hits = response.getHits();
System.out.println("Total hits: " + hits.getTotalHits().value);
System.out.println("Hits:");
for (SearchHit hit : hits) {
System.out.println("Id: " + hit.getId());
System.out.println("Score: " + hit.getScore());
System.out.println("Title: " + hit.getSourceAsMap().get("title"));
HighlightField highlightField = hit.getHighlightFields().get("content");
if (highlightField != null) {
Text[] fragments = highlightField.fragments();
String content = "";
for (Text fragment : fragments) {
content += fragment;
                }
System.out.println("Content: " + content);
            } else {
System.out.println("Content: " + hit.getSourceAsMap().get("content"));
            }
System.out.println("Publish date: " + hit.getSourceAsMap().get("publish_date"));
        }
    }
}


 


这里我们使用了Elasticsearch高级REST客户端API来实现示例代码，相较于低级API，使用高级API的好处在于更易用，而且使用方式更加接近面向对象编程，提高了开发效率。


十一、使用Spring Boot框架


​首先，我们需要添加相关依赖。在​​pom.xml​​文件中添加以下依赖：


<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-data-elasticsearch</artifactId>
</dependency>
<dependency>
<groupId>org.elasticsearch.client</groupId>
<artifactId>elasticsearch-rest-high-level-client</artifactId>
<version>7.15.2</version>
</dependency>


其中，​​spring-boot-starter-data-elasticsearch​​​依赖为Spring Boot提供的与Elasticsearch集成的基础依赖，​​elasticsearch-rest-high-level-client​​为Elasticsearch高级REST客户端API的依赖。


接下来，我们创建一个Spring Boot主类，并在其中添加如下代码：


import org.elasticsearch.action.delete.DeleteRequest;
import org.elasticsearch.action.index.IndexRequest;
import org.elasticsearch.action.search.SearchRequest;
import org.elasticsearch.action.search.SearchResponse;
import org.elasticsearch.client.RequestOptions;
import org.elasticsearch.client.RestHighLevelClient;
import org.elasticsearch.common.settings.Settings;
import org.elasticsearch.common.unit.TimeValue;
import org.elasticsearch.index.query.BoolQueryBuilder;
import org.elasticsearch.index.query.QueryBuilders;
import org.elasticsearch.search.SearchHit;
import org.elasticsearch.search.builder.SearchSourceBuilder;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightField;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder.Field;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder.HighlightQuery;
import org.elasticsearch.search.sort.SortOrder;
import org.springframework.boot.CommandLineRunner;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.context.annotation.Bean;
import org.springframework.data.elasticsearch.client.RestClients;

import java.io.IOException;
import java.util.HashMap;
import java.util.Map;

@SpringBootApplication
public class ElasticsearchDemoApplication implements CommandLineRunner {

public static void main(String[] args) {
SpringApplication.run(ElasticsearchDemoApplication.class, args);
    }

@Bean
public RestHighLevelClient client() {
return RestClients.create(RestClients.createLocalHost()).rest();
    }

@Override
public void run(String... args) throws Exception {
RestHighLevelClient client = client();
try {
createIndex(client);
insertDocument(client);
searchDocument(client);
deleteIndex(client);
        } catch (IOException e) {
e.printStackTrace();
        } finally {
client.close();
        }
    }

private static void createIndex(RestHighLevelClient client) throws IOException {
// 创建索引
Settings.Builder settings = Settings.builder()
                .put("index.number_of_shards", 1)
                .put("index.number_of_replicas", 0);
Map<String, Object> mapping = new HashMap<>();
Map<String, Object> properties = new HashMap<>();
properties.put("title", Map.of("type", "text"));
properties.put("content", Map.of("type", "text"));
properties.put("publish_date", Map.of("type", "date"));
mapping.put("properties", properties);
client.indices().create(Map.of("index", "my_index", "settings", settings, "mapping", mapping),
RequestOptions.DEFAULT);
    }

private static void insertDocument(RestHighLevelClient client) throws IOException {
// 插入文档
Map<String, Object> document = new HashMap<>();
document.put("title", "Elasticsearch Guide");
document.put("content", "This is a guide to
IndexRequest request = new IndexRequest("my_index")
                .id("1")
                .source(document);
client.index(request, RequestOptions.DEFAULT);
    }

private static void searchDocument(RestHighLevelClient client) throws IOException {
// 搜索文档
SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery()
                .must(QueryBuilders.matchQuery("title", "Elasticsearch"))
                .should(QueryBuilders.matchQuery("content", "guide"));
sourceBuilder.query(boolQueryBuilder)
                .sort("publish_date", SortOrder.DESC)
                .from(0)
                .size(10)
                .timeout(TimeValue.timeValueSeconds(1))
                .fetchSource(new String[]{"title", "publish_date"}, new String[]{"content"});
HighlightBuilder highlightBuilder = new HighlightBuilder()
                .field(new Field("title"))
                .highlightQuery(new HighlightQuery().matchQuery(new HashMap<String, Object>() {{
put("title", new HashMap<>());
                }}));
sourceBuilder.highlighter(highlightBuilder);
SearchRequest request = new SearchRequest("my_index").source(sourceBuilder);
SearchResponse response = client.search(request, RequestOptions.DEFAULT);
System.out.println("Total hits: " + response.getHits().getTotalHits().value);
for (SearchHit hit : response.getHits().getHits()) {
System.out.println("Title: " + hit.getSourceAsMap().get("title"));
System.out.println("Publish date: " + hit.getSourceAsMap().get("publish_date"));
System.out.println("Content: " + hit.getHighlightFields().get("title").fragments()[0].string());
System.out.println("--------------------------");
        }
    }

private static void deleteIndex(RestHighLevelClient client) throws IOException {
// 删除索引
DeleteRequest request = new DeleteRequest("my_index");
client.indices().delete(request, RequestOptions.DEFAULT);
    }

}


我们在主类​​ElasticsearchDemoApplication​​​中实现了​​CommandLineRunner​​​接口，以便在应用启动时执行相关方法。在​​run​​方法中，我们调用了创建索引、插入文档、搜索文档和删除索引的方法。这些方法的具体实现与示例代码中的实现相同。


接下来，我们可以运行应用程序并查看结果。在终端中输入以下命令：


mvn spring-boot:run


通过这个Spring Boot实现，我们可以更方便地与Elasticsearch进行交互，而不必手动设置连接和释放资源等操作。此外，Spring Boot还提供了许多其他特性，例如自动配置和依赖注入等。这使得我们可以更加专注于业务逻辑，而不必过多关注与Elasticsearch的交互。


总结


Elasticsearch是一个功能强大的搜索引擎，拥有许多高级查询技巧。在实际使用中，可以根据具体需求选择合适的查询方式，并使用查询语句中的高级功能，来实现更复杂的查询操作。本教程介绍了Elasticsearch的基本查询方式和高级查询技巧，并提供了相应的代码示例，希望能帮助读者更好地掌握Elasticsearch的查询功能。


标签：
相关文章