Elasticsearch DSL 查询

太过爱你忘了你带给我的痛 2021-12-03 11:27 442阅读 0赞

**DSL** 是 Domain Specific Language（特定领域语言）的缩写。

Elasticsearch 提供了功能丰富且灵活的查询语言，即 **DSL 查询**。你可以用它构建出更加复杂、功能更加强大的查询语句。

先来一个简单的例子，比如，我们可以这样查询 first\_name 字段包含 John 的用户。

GET /alibaba/user/_search
    {
        "query" : {
            "match" : {
                "first_name" : "John"
            }
        }
    }

它会返回和上一节中的**查询字符串搜索**相同的结果。只不过这里我们不再使用查询字符串作为请求参数，而是使用 JSON 格式的请求体。

## bool 联合查询 ##

### 逻辑与（AND） ###

下面，我们来进行一个稍微复杂点的查询。比如，查询姓氏包含 Smith 且年龄大于 30 岁的用户。

GET /alibaba/user/_search
    {
        "query" : {
            "bool" : {
                "filter" : {
                    "range" : {
                        "info.age" : { "gt" : 30 }
                    }
                },
                "must" : [{
                    "match" : {
                        "last_name" : "smith"
                    }
                }]
            }
        }
    }

返回结果：

{
      "took": 4,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 1,
        "max_score": 0.2876821,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "3",
            "_score": 0.2876821,
            "_source": {
              "email": "john3@smith.com",
              "first_name": "John3",
              "last_name": "Smith",
              "full_name": "John3 Smith",
              "info": {
                "age": 35,
                "interests": [
                  "sports",
                  "musics"
                ],
                "address": "guangdong shenzhen longhua"
              },
              "created_at": "2019-06-08 09:40:10"
            }
          }
        ]
      }
    }

查询 last\_name 包含 Smith，且 full\_name 包含 John 的用户。

GET /alibaba/user/_search
    {
      "query": {
        "bool": {
          "must": [
            { "match": { "last_name": "Smith" } },
            { "match": { "full_name": "John" } }
          ]
        }
      }
    }

返回结果：

{
      "took": 4,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 1,
        "max_score": 0.5457982,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "1",
            "_score": 0.5457982,
            "_source": {
              "email": "john@smith.com",
              "first_name": "John",
              "last_name": "Smith",
              "full_name": "John Smith",
              "info": {
                "age": 25,
                "interests": [
                  "games",
                  "musics"
                ],
                "address": "guangdong shenzhen nanshan"
              },
              "created_at": "2019-06-06 08:30:10"
            }
          }
        ]
      }
    }

### 逻辑或（OR） ###

查询 info.age 大于 30 或 info.address 包含 baoan 的用户。

GET /alibaba/user/_search
    {
    	"query": {
    		"bool": {
    			"should": [{
    					"range": {
    						"info.age": {
    							"gt": 30
    						}
    					}
    				},
    				{
    					"match": {
    						"info.address": "baoan"
    					}
    				}
    			]
    		}
    	}
    }

返回结果：

{
      "took": 8,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 2,
        "max_score": 1,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "3",
            "_score": 1,
            "_source": {
              "email": "john3@smith.com",
              "first_name": "John3",
              "last_name": "Smith",
              "full_name": "John3 Smith",
              "info": {
                "age": 35,
                "interests": [
                  "sports",
                  "musics"
                ],
                "address": "guangdong shenzhen longhua"
              },
              "created_at": "2019-06-08 09:40:10"
            }
          },
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "2",
            "_score": 0.25316024,
            "_source": {
              "email": "john2@smith.com",
              "first_name": "John2",
              "last_name": "Smith",
              "full_name": "John2 Smith",
              "info": {
                "age": 28,
                "interests": [
                  "games",
                  "books"
                ],
                "address": "guangdong shenzhen baoan"
              },
              "created_at": "2019-06-07 08:30:10"
            }
          }
        ]
      }
    }

查询 last\_name 包含 Smith，或者 full\_name 包含 John 的用户。

GET /alibaba/user/_search
    {
      "query": {
        "bool": {
          "should": [
            { "match": { "last_name": "Smith" } },
            { "match": { "full_name": "John" } }
          ]
        }
      }
    }

返回结果：

{
      "took": 7,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 3,
        "max_score": 0.5457982,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "1",
            "_score": 0.5457982,
            "_source": {
              "email": "john@smith.com",
              "first_name": "John",
              "last_name": "Smith",
              "full_name": "John Smith",
              "info": {
                "age": 25,
                "interests": [
                  "games",
                  "musics"
                ],
                "address": "guangdong shenzhen nanshan"
              },
              "created_at": "2019-06-06 08:30:10"
            }
          },
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "2",
            "_score": 0.2876821,
            "_source": {
              "email": "john2@smith.com",
              "first_name": "John2",
              "last_name": "Smith",
              "full_name": "John2 Smith",
              "info": {
                "age": 28,
                "interests": [
                  "games",
                  "books"
                ],
                "address": "guangdong shenzhen baoan"
              },
              "created_at": "2019-06-07 08:30:10"
            }
          },
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "3",
            "_score": 0.2876821,
            "_source": {
              "email": "john3@smith.com",
              "first_name": "John3",
              "last_name": "Smith",
              "full_name": "John3 Smith",
              "info": {
                "age": 35,
                "interests": [
                  "sports",
                  "musics"
                ],
                "address": "guangdong shenzhen longhua"
              },
              "created_at": "2019-06-08 09:40:10"
            }
          }
        ]
      }
    }

### 总结说明 ###

*  bool：表示联合查询，用来合并多种查询条件。
 *  must：表示多个查询条件的肯定匹配，多个条件必须都满足，相当于 and。
 *  should：表示多个查询条件的肯定匹配，多个条件只要有一个满足，相当于 or。
 *  must\_not：表示多个查询条件的否定匹配，多个条件必须都满足，相当于 and。
 *  filter：用于快速过滤结果集。
 *  文档字段的内部字段，通过 **.** 号来访问，如 info.age。

## 全文搜索 ##

这里，我们搜索 info.address 包含 "shenzhen baoan" 的用户。

GET /alibaba/user/_search
    {
        "query" : {
            "match" : {
                "info.address" : "shenzhen baoan"
            }
        }
    }

返回结果：

{
      "took": 5,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 3,
        "max_score": 0.5063205,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "2",
            "_score": 0.5063205,
            "_source": {
              "email": "john2@smith.com",
              "first_name": "John2",
              "last_name": "Smith",
              "full_name": "John2 Smith",
              "info": {
                "age": 28,
                "interests": [
                  "games",
                  "books"
                ],
                "address": "guangdong shenzhen baoan"
              },
              "created_at": "2019-06-07 08:30:10"
            }
          },
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "1",
            "_score": 0.25316024,
            "_source": {
              "email": "john@smith.com",
              "first_name": "John",
              "last_name": "Smith",
              "full_name": "John Smith",
              "info": {
                "age": 25,
                "interests": [
                  "games",
                  "musics"
                ],
                "address": "guangdong shenzhen nanshan"
              },
              "created_at": "2019-06-06 08:30:10"
            }
          },
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "3",
            "_score": 0.25316024,
            "_source": {
              "email": "john3@smith.com",
              "first_name": "John3",
              "last_name": "Smith",
              "full_name": "John3 Smith",
              "info": {
                "age": 35,
                "interests": [
                  "sports",
                  "musics"
                ],
                "address": "guangdong shenzhen longhua"
              },
              "created_at": "2019-06-08 09:40:10"
            }
          }
        ]
      }
    }

可以发现，Elasticsearch 自动把要搜索匹配的字符串 "shenzhen baoan"，拆分成了两个词 "shenzhen" 和 "baoan"。然后再到 info.address 中进行全文搜索匹配。

搜索到的 3 条记录，默认是按照搜索排名分数降序排列的。

**搜索排名分数**（score），就是搜索结果相关性大小的评分，即文档与查询条件的匹配程度。相关性越大，搜索排名分数就越高，排名（排序）就越靠前。

这个概念在传统的关系型数据库中是无法想象的。

上面的查询语句，等价于下面的这种写法：

GET /alibaba/user/_search
    {
        "query" : {
            "bool":{
              "should": [
                {"match" : {"info.address" : "shenzhen"}},
                {"match" : {"info.address" : "baoan"}}
              ]
            }
        }
    }

## 短语搜索 ##

有时，我们确切地想要匹配多个单词或者短语（phrase）。这时，可以采用 match\_phrase 来实现。

GET /alibaba/user/_search
    {
        "query" : {
            "match_phrase" : {
                "info.address" : "shenzhen baoan"
            }
        }
    }

返回结果：

{
      "took": 21,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 1,
        "max_score": 0.5063205,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "2",
            "_score": 0.5063205,
            "_source": {
              "email": "john2@smith.com",
              "first_name": "John2",
              "last_name": "Smith",
              "full_name": "John2 Smith",
              "info": {
                "age": 28,
                "interests": [
                  "games",
                  "books"
                ],
                "address": "guangdong shenzhen baoan"
              },
              "created_at": "2019-06-07 08:30:10"
            }
          }
        ]
      }
    }

## 高亮显示 ##

有时，我们需要高亮显示匹配到的关键字。可以通过 highlight 来实现。

GET /alibaba/user/_search
    {
        "query" : {
            "match_phrase" : {
                "info.address" : "shenzhen baoan"
            }
        },
        "highlight": {
            "fields" : {
                "info.address" : {}
            }
        }
    }

返回结果：

{
      "took": 77,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 1,
        "max_score": 0.5063205,
        "hits": [
          {
            "_index": "alibaba",
            "_type": "user",
            "_id": "2",
            "_score": 0.5063205,
            "_source": {
              "email": "john2@smith.com",
              "first_name": "John2",
              "last_name": "Smith",
              "full_name": "John2 Smith",
              "info": {
                "age": 28,
                "interests": [
                  "games",
                  "books"
                ],
                "address": "guangdong shenzhen baoan"
              },
              "created_at": "2019-06-07 08:30:10"
            },
            "highlight": {
              "info.address": [
                "guangdong <em>shenzhen</em> <em>baoan</em>"
              ]
            }
          }
        ]
      }
    }

## 多字段搜索 ##

有时，我们想要让同一个查询字符串同时应用到多个字段进行搜索匹配。可以采用 multi\_match 来实现。

例如，查询 first\_name 或者 full\_name 包含 "smith" 的文档。

GET /alibaba/user/_search
    {
        "query" : {
            "multi_match": {
              "query": "smith",
              "fields": ["first_name", "full_name"]
            }
        }
    }