1、索引相关

查看集群状态

http://127.0.0.1:9200/_cluster/health

创建索引

curl -XPUT 10.9.39.37:9200/test_cycle-order_20227_1

查看所有索引

http://127.0.0.1:9200/_cat/indices

查看索引信息

http://127.0.0.1:9200/index_name
{
    "user-info": {
        "aliases": {},
        "mappings": {
            "_doc": {
                "properties": {
                    "address": {
                        "type": "text",
                        "fields": {
                            "keyword": {
                                "type": "keyword",
                                "ignore_above": 256
                            }
                        }
                    },
                    "age": {
                        "type": "long"
                    },
                    "name": {
                        "type": "text",
                        "fields": {
                            "keyword": {
                                "type": "keyword",
                                "ignore_above": 256
                            }
                        }
                    },
                    "sex": {
                        "type": "long"
                    }
                }
            }
        },
        "settings": {
            "index": {
                "creation_date": "1670915768780",
                "number_of_shards": "5",
                "number_of_replicas": "1",
                "uuid": "An2CYpuxSq-6sh0vWRjS2g",
                "version": {
                    "created": "6040299"
                },
                "provided_name": "user-info"
            }
        }
    }
}

删除索引

curl -XDELETE 127.0.0.1:9200/index_name

2、高级查询 DSL

2.1 查询所有 match_all

使用match_all，默认只会返回10条数据

curl -X GET 'http://10.9.39.37:9200/user-info/_doc/_search' \
-H 'Accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "query":{
        "match_all":{
        }
    }
}'
{
    //查询花费的总时间
    "took": 1,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        //符合条件的总文档数
        "total": 1,
        "max_score": 1.0,
        //结果集，默认10
        "hits": [
            {
                "_index": "user-info",
                "_type": "_doc",
                "_id": "1",
                "_score": 1.0,
                "_source": {
                    "name": "张三",
                    "age": 18,
                    "address": "北京市朝阳区",
                    "sex": 1
                }
            }
        ]
    }
}

返回指定条数

{
    "query": {
        "match_all": {}
    },
    "size":50
}

size 不能无限大，如果过大会出现异常

{
    "error": {
        "root_cause": [
            {
                "type": "query_phase_execution_exception",
                "reason": "Result window is too large, from + size must be less than or equal to: [10000] but was [99999]. See the scroll api for a more efficient way to request large data sets. This limit can be set by changing the [index.max_result_window] index level setting."
            }
        ],
        "type": "search_phase_execution_exception",
        "reason": "all shards failed",
        "phase": "query",
        "grouped": true,
        "failed_shards": [
            {
                "shard": 0,
                "index": "user-info",
                "node": "uyvZR82uTVGEdPxthCCYHg",
                "reason": {
                    "type": "query_phase_execution_exception",
                    "reason": "Result window is too large, from + size must be less than or equal to: [10000] but was [99999]. See the scroll api for a more efficient way to request large data sets. This limit can be set by changing the [index.max_result_window] index level setting."
                }
            }
        ]
    },
    "status": 500
}

1、查询结果的窗口太大，from + size的结果必须小于或等于10000，而当前查询结果的窗
口为20000。
2、可以采用scroll api更高效的请求大量数据集。

3、查询结果的窗口的限制可以通过参数index.max_result_window进行设置。

注意：参数index.max_result_window主要用来限制单次查询满足查询条件的结果窗口的 大小，窗口大小由from + size共同决定。不能简单理解成查询返回给调用方的数据量。这 样做主要是为了限制内存的消耗。 比如：from为1000000，size为10，逻辑意义是从满足条件的数据中取1000000到 （1000000 + 10）的记录。这时ES一定要先将（1000000 + 10）的记录（即 result_window）加载到内存中，再进行分页取值的操作。尽管最后我们只取了10条数据返 回给客户端，但ES进程执行查询操作的过程中确需要将（1000000 + 10）的记录都加载到 内存中，可想而知对内存的消耗有多大。这也是ES中不推荐采用（from + size）方式进行 深度分页的原因。
同理，from为0，size为1000000时，ES进程执行查询操作的过程中确需要将1000000 条 记录都加载到内存中再返回给调用方，也会对ES内存造成很大压力。

2.2 分页查询 from

{
    "query": {
        "match_all": {}
    },
    "from": 0,
    "size": 9999
}

2.3 深分页查询 scroll

最佳实践还是根据异常提示中的采用scroll api更高效的请求大量数据集。

  1. 查询命令中新增scroll=3000ms,说明采用游标查询，保持游标查询窗口3000毫秒
  2. 这里由于测试数据量不够，所以size值设置为
  3. 实际使用中为了减少游标查询的次数，可以将size值适当增大，比如设置为1000
curl -X GET 'http://10.9.39.37:9200/user-info/_doc/_search?scroll=3000ms' \
-H 'Accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "query": {
        "match_all": {}
    },
    "size": 2
}'
{
    "_scroll_id": "DnF1ZXJ5VGhlbkZldGNoBQAAAAACCBzjFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aKqBZzVkIwYk5kMlNVbWwybnJDaXZwb1VnAAAAAAqiPgUWaUxYa0tPcHZUaHE0bm16TUpZVkpOZwAAAAACCBziFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aKpxZzVkIwYk5kMlNVbWwybnJDaXZwb1Vn",
    "took": 1,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": 7,
        "max_score": 1.0,
        "hits": [
            {
                "_index": "user-info",
                "_type": "_doc",
                "_id": "5",
                "_score": 1.0,
                "_source": {
                    "name": "1211111",
                    "age": 18,
                    "address": "北京市朝阳区",
                    "sex": 1
                }
            },
            {
                "_index": "user-info",
                "_type": "_doc",
                "_id": "8",
                "_score": 1.0,
                "_source": {
                    "name": "娃娃",
                    "age": 18,
                    "address": "北京市朝阳区",
                    "sex": 1
                }
            }
        ]
    }
}

除了返回前2条记录，还返回了一个游标ID值_scroll_id。
采用游标ID查询

curl -X GET 'http://10.9.39.37:9200/_search/scroll' \
-H 'Accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "scroll": "1m",
    "scroll_id": "DnF1ZXJ5VGhlbkZldGNoBQAAAAACCDZHFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aaaxZzVkIwYk5kMlNVbWwybnJDaXZwb1VnAAAAAAqiVaMWaUxYa0tPcHZUaHE0bm16TUpZVkpOZwAAAAACCDZIFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aabBZzVkIwYk5kMlNVbWwybnJDaXZwb1Vn"
}'
{
    "_scroll_id": "DnF1ZXJ5VGhlbkZldGNoBQAAAAACCDZHFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aaaxZzVkIwYk5kMlNVbWwybnJDaXZwb1VnAAAAAAqiVaMWaUxYa0tPcHZUaHE0bm16TUpZVkpOZwAAAAACCDZIFnV5dlpSODJ1VFZHRWRQeHRoQ0NZSGcAAAAAB9aabBZzVkIwYk5kMlNVbWwybnJDaXZwb1Vn",
    "took": 1,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": 8,
        "max_score": 1.0,
        "hits": []
    }
}

多次根据scroll_id游标查询，直到没有数据返回则结束查询。采用游标查询索引全量数据， 更安全高效，限制了单次对内存的消耗。

2.3 指定字段排序 sort

注意：会让得分失效

curl -X GET 'http://10.9.39.37:9200/user-info/_doc/_search' \
-H 'Accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "query": {
        "match_all": {}
    },
    "size": 5,
    "sort":[
        {
            "age":"desc"
        }
       
    ]
}'

排序分页

{
    "query": {
        "match_all": {}
    },
    "size": 5,
    "from": 5,
    "sort": [
        {
            "age": "desc"
        }
    ]
}

2.4 返回指定字段_source

{
    "query": {
        "match_all": {}
    },
    "size": 5,
    "from": 5,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source":["age","name"]
}

2.5 关键词 match

match在匹配时会对所查找的关键词进行分词，然后按分词匹配查找

{
    "query": {
        "match": {
            "age":18
        }
    },
    "size": 5,
    "from": 5,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source":["age","name"]
}

2.5 多字段查询 multi_match

{
    "query": {
        "multi_match": {
            "query": "张三",
            "fields": [
                "name",
                "address"
            ]
        }
    },
    "size": 5,
    "from": 5,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source": [
        "age",
        "name"
    ]
}

2.6 query_string

允许我们在单个查询字符串中指定AND | OR | NOT条件，同时也和 multi_match query 一样，支持多字段搜索。和match类似，但是match需要指定字段名，query_string是在所 有字段中搜索，范围更广泛

{
    "query": {
        "query_string": {
            "query": "张三 or 娃娃22211"
        }
    },
    "size": 100,
    "from": 0,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source": [
        "age",
        "name"
    ]
}

指定单个字段查询

{
    "query": {
        "query_string": {
            "default_field": "address",
            "query": "张三 or 娃娃22211"
        }
    },
    "size": 100,
    "from": 0,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source": [
        "age",
        "name"
    ]
}

指定多个字段查询

{
    "query": {
        "query_string": {
            "fields": [
                "name",
                "address"
            ],
            "query": "张三 or 娃娃22211"
        }
    },
    "size": 100,
    "from": 0,
    "sort": [
        {
            "age": "desc"
        }
    ],
    "_source": [
        "age",
        "name"
    ]
}

2.6 关键词查询 term

Term用来使用关键词查询(精确匹配),还可以用来查询没有被进行分词的数据类型。Term是 表达语意的最小单位，搜索和利用统计语言模型进行自然语言处理都需要处理Term。 match在匹配时会对所查找的关键词进行分词，然后按分词匹配查找，而term会直接对关 键词进行查找。一般模糊查找的时候，多用match，而精确查找时可以使用term

{
    "query": {
        "term": {
            "name": {
                "value": "张三"
            }
        }
    },
    "size": 100,
    "from": 0,
    "sort": [
        {
            "age": "desc"
        }
    ]
}

采用term精确查询, 查询字段映射类型为keyword

{
    "query": {
        "term": {
            "address.keyword": {
                "value": "北京市朝阳区"
            }
        }
    },
    "size": 100,
    "from": 0,
    "sort": [
        {
            "age": "desc"
        }
    ]
}