从入门到进阶：Elasticsearch高级查询技巧详解

淡淡的烟草味﹌ 2024-03-22 21:33 21阅读 0赞

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。