Query DSL¶
Query DSL (Query Domain Specified Language 特殊领域查询语言)，利用 REST API 传递 JSON 格式的请求数据与 ES 进行交互查询。这种方式的丰富查询方法让 ES 检索更强大、简洁。
Syntax¶
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 313
{
"took": 1,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 3,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "3",
"_score": 1.0,
"_source": {
"title": "IPhone 11",
"price": 7999.99,
"description": "balabala11",
"create_at": "2022-07-19"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "1",
"_score": 1.0,
"_source": {
"title": "Andriod1",
"price": 1999.00,
"description": "Androd yyds",
"create_at": "2022-06-06"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": {
"title": "Iphone1",
"price": 4999.00,
"description": "Iphone yyds",
"create_at": "2022-06-06"
}
}
]
}
}
常见检索¶
基本查询¶
查询所有 match_all¶
Note
返回索引中全部文档
关键词查询 term¶
Note
使用关键词查询
keyword、integer、double、date 这些都是不分词的。
text 类型默认使用 ES 标准分词器。标准分词器对中文是单字分词，对英文是单词分词。
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 240
{
"took": 2,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": {
"title": "Iphone1",
"price": 4999.00,
"description": "Iphone yyds",
"create_at": "2022-06-06"
}
}
]
}
}
范围 range¶
Note
用于指定查询范围内的文档
GET /<INDEX_NAME>/_search
{
"query": {
"range:" {
"<FIELD>": {
"<OPERATOR>": <VALUE>
"relation": "<INTERSECTS | CONTAINS | WITHIN>"
"boost": 1.0
}
}
}
}
Operator 操作符包括：
- gt: 大于
- gte: 大于等于
- lt: 小于
- lte: 小于等于
- format: VALUE为字符串，匹配 data 用
- time_zone: 指定时区的偏移量
relation 用于指定查询策略：
- INTERSECTS(默认)：将文档与查询的范围做交集运算然后匹配
- CONTAINS：完全包含查询范围才算命中
- WITHIN：在查询范围内才算命中
boost 权重：用于提升在多范围查询时，当前范围的权重，默认1.0
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 266
{
"took": 11,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 2,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "1",
"_score": 1.0,
"_source": {
"title": "Andriod1",
"price": 1999.00,
"description": "Androd yyds",
"create_at": "2022-06-06"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": {
"title": "Iphone1",
"price": 4999.00,
"description": "Iphone yyds",
"create_at": "2022-06-06"
}
}
]
}
}
前缀 prefix¶
Note
使用前缀进行匹配查询
GET {{host}}/product/_search HTTP/1.1
Content-Type: application/json
{
"query": {
"prefix": {
"title": {
"value": "Ip"
}
}
}
}
通配符 wildcard¶
Note
通配符查询，？ 匹配一个任意字符，* 匹配多个任意字符
GET {{host}}/product/_search HTTP/1.1
Content-Type: application/json
{
"query": {
"wildcard": {
"title": {
"value": "Ip*"
}
}
}
}
多ID查询 ids¶
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 290
{
"took": 27,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 3,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "1",
"_score": 1.0,
"_source": {
"title": "Andriod1",
"price": 1999.00,
"description": "Andriod1 yyds",
"create_at": "2022-06-06"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": {
"title": "Iphone1",
"price": 4999.00,
"description": "Iphone1 yyds",
"create_at": "2022-06-07"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "3",
"_score": 1.0,
"_source": {
"title": "Andriod2",
"price": 2999.00,
"description": "Andriod2 yyds",
"create_at": "2022-06-08"
}
}
]
}
}
模糊查询 fuzzy¶
Note
用来模糊查询包含有关键词的文档
Warning
fuzzy 模糊查询，最大模糊错误必须在 0-2 之间
- 搜索关键词长度为 2 不允许存在模糊错误
- 搜索关键词长度为 3-5 允许一次模糊错误
- 搜索关键词长度大于 5 允许最大2次模糊错误
关键词长度为6(> 5)允许最大2次模糊错误
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 302
{
"took": 14,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 3,
"relation": "eq"
},
"max_score": 0.9241961,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 0.9241961,
"_source": {
"id": 2,
"title": "Iphone1",
"price": 4999.00,
"description": "Iphone1 yyds",
"create_at": "2022-06-07"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "4",
"_score": 0.9241961,
"_source": {
"id": 4,
"title": "Iphone2",
"price": 3999.00,
"description": "Iphone2 yyds",
"create_at": "2022-06-09"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "5",
"_score": 0.9241961,
"_source": {
"id": 5,
"title": "Iphone3",
"price": 6978.00,
"description": "Iphone3 yyds",
"create_at": "2022-06-10"
}
}
]
}
}
复杂查询¶
布尔查询 bool¶
Note
bool 关键字用来组合多个条件实现复杂查询
must: 相当于&&should: 相当于||must_not: 相当于!
GET /<INDEX_NAME>/_search
{
"query": {
"bool": {
"< must | should | must_not >": [
{
"<term | range | fuzzy | ...>": {
"<FIELD>": <VALUE>
}
},
...other condition
]
}
}
}
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 246
{
"took": 15,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 2.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "5",
"_score": 2.0,
"_source": {
"id": 5,
"title": "Iphone3",
"price": 6978.00,
"description": "Iphone3 yyds",
"create_at": "2022-06-10"
}
}
]
}
}
多字段查询 multi_match¶
Note
multi_match 在多个字段中检索能匹配查询关键词的文档
GET /<INDEX_NAME>/_search
{
"query": {
"multi_match": {
"query": "<QUERY_STRING>"
"fields": ["<FIELD>", "<FIELD>", ...]
}
}
}
Warning
字段类型如果分词，则将查询关键词分词之后去指定的那些字段检索
字段类型如果不分词，则将查询关键词分词作为整体去指定的那些字段检索
Example
GET {{host}}/product/_search HTTP/1.1
Content-Type: application/json
{
"query": {
"multi_match": {
"query": "五分钟",
"fields": [ "title", "description" ]
}
}
}
ES 会将 “五分钟” 作为整体去 title 中检索，检索不到会分词成为 “五”、“分”、“钟”，然后去 description 中检索。最后检索到 id 为 4 的文档。
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 296
{
"took": 2,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 3.0841155,
"hit
{
"_index": "product",
"_type": "_doc",
"_id": "4",
"_score": 3.0841155,
"_source": {
"id": 4,
"title": "Oppo",
"price": 3999.00,
"description": "充电两小时, 通话五分钟",
"create_at": "2022-06-09"
}
}
]
}
}
默认字段分词查询 query_string¶
GET /<INDEX_NAME>/_search
{
"query": {
"query_string": {
"default_field": "<FIELD>"
"query": ""
}
}
}
过滤查询¶
ES 的查询操作分为两种：
- 匹配查询 (query)
- 过滤查询 (filter)
匹配查询 query 就是上面的关键词、范围、前缀、通配符、多ID、模糊查询等，这种查询是直接送达 ES 服务的。 ES 服务会计算每个返回文档的得分，然后根据得分排序。
过滤查询则会先筛选出符合的文档，并不计算得分，然后送达 ES 服务。而且过滤查询还可以缓存起来。所以从性能角度考虑，过滤比匹配更快。
Tip
过滤查询适合在大范围筛选数据，而匹配查询适合精确匹配数据。一般应用时，应先使用过滤操作过滤数据，然后使用匹配查询匹配数据。
过滤查询必须在布尔查询中。
过滤有 term、terms、exists、range、ids 几种。
GET /<INDEX_NAME>/_search
{
"query": {
"bool": {
"must": [ {...} ],
"filter": [
{
"<term | terms | exists | range | ids>": { ... }
}
]
}
}
}
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 252
{
"took": 2,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 1.0700173,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0700173,
"_source": {
"id": 2,
"title": "Iphone",
"price": 4999.00,
"description": "Iphone just yyds",
"create_at": "2022-06-07"
}
}
]
}
}
其他特性¶
高亮查询 highlight¶
Warning
ES7 中只有能分词的才能高亮
GET /<INDEX_NAME>/_search
{
"query:" { ... },
"highlight": {
"fields": {
"<FIELD | *>": {},
"pre_tags": ["<PRE_TAGS>"],
"post_tags": ["<POST_TAGS>"],
"requiree_field_match": "<true | false>" // 其他字段也一起高亮，false开启，true 只高亮对应字段
}
}
}
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 278
{
"took": 18,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 1.129573,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.129573,
"_source": {
"id": 2,
"title": "Iphone",
"price": 4999.00,
"description": "Iphone just yyds",
"create_at": "2022-06-07"
},
"highlight": { // 多出这一段即为高亮检索词的结果
"description": [
"<em>Iphone</em> just yyds"
]
}
}
]
}
}
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 278
{
"took": 18,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 1.129573,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.129573,
"_source": {
"id": 2,
"title": "Iphone",
"price": 4999.00,
"description": "Iphone just yyds",
"create_at": "2022-06-07"
},
"highlight": { // 多出这一段即为高亮检索词的结果
"description": [
"<span style='color:red;'>Iphone</span> just yyds"
]
}
}
]
}
}
分页 from, size¶
Note
from 用来指定起始返回位置，默认0；size 用来指定返回条数，默认10。
size 可以单独使用，两者结合可以实现分页效果。
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 381
{
"took": 1,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 5,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "3",
"_score": 1.0,
"_source": {
"id": 3,
"title": "Xiaomi",
"price": 2999.00,
"description": "Xiaomi 为发烧而生",
"create_at": "2022-06-08"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "4",
"_score": 1.0,
"_source": {
"id": 4,
"title": "Oppo",
"price": 3999.00,
"description": "充电两小时, 通话五分钟",
"create_at": "2022-06-09"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "5",
"_score": 1.0,
"_source": {
"id": 5,
"title": "Vivo",
"price": 6978.00,
"description": "拍出你的美",
"create_at": "2022-06-10"
}
}
]
}
}
排序 sort¶
默认降序 desc
GET /<INDEX_NAME>/_search
{
"query": { ... },
"sort": [
{ "<FIELD>": "<desc | asc>" },
{
"<FIELD>": {
"order": "<desc | asc>",
"format": "strict_date_optional_time_nanos"
}
},
{
"<FIELD>": {
"order": "<desc | asc>",
"mode": "<min | max | sum | avg | median>"
}
}
]
}
返回指定字段 _source¶
Example
HTTP/1.1 200 OK
content-type: application/json; charset=UTF-8
content-encoding: gzip
content-length: 213
{
"took": 4,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 5,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "product",
"_type": "_doc",
"_id": "1",
"_score": 1.0,
"_source": { // 只返回指定字段
"id": 1,
"title": "Andriod"
}
},
{
"_index": "product",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": { // 只返回指定字段
"id": 2,
"title": "Iphone"
}
}
]
}
}