es怎么查询 - es哪些查询

小科普 170

基本查询

基本语法：

GET /索引库名/_search
{"query":{"查询类型":{"查询条件":"查询条件值"}
}
}

这里的 query 代表一个查询对象，里面可以有不同的查询属性。

查询类型：match_all、match、term、range 等等。

查询条件：查询条件会根据类型的不同，写法也有差异。

查询所有（match_all)

示例：

GET /renda/_search
{"query":{"match_all": {}
}
}

query：代表查询对象。

match_all：代表查询所有。

结果：

{"took":1,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":3,"max_score":1,"hits": [
{"_index":"renda","_type":"goods","_id":"2","_score":1,"_source": {"title":"白米手机","images":"","price":2699}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":1,"_source": {"title":"小米手机","images":"","price":2699}
},
{"_index":"renda","_type":"goods","_id":"3","_score":1,"_source": {"title":"超大米手机","images":"","price":3299,"stock":200,"saleable":true,"subTitle":"大米"}
}
]
}
}

结果解析：

took- 检索所耗费的时间，单位是毫秒。timed_out- 是否超时。​_shards- 分片信息。​hits- 命中结果，检索结果信息。total- 搜索到的总条数。max_score- 所有结果中文档得分的最高分。hits- 搜索结果的文档对象数组，每个元素是一条搜索到的文档信息。_index- 索引库。_type- 文档类型。_id- 文档 id。_score- 评分；索引库的一个概念；关联度。_source- 原始数据。

文档得分：使用 ES 时，对于查询出的文档无疑会有文档相似度之别；而理想的排序是和查询条件相关性越高排序越靠前，而这个排序的依据就是_score。

匹配查询（match）

加入一条数据用于测试：

PUT /renda/goods/3
{"title":"小米电视4A","images":"","price": 3899.00
}

索引库中有 3 部手机，1 台电视。

match类型查询，会把查询条件进行分词，然后进行查询，多个词条之间是or的关系：

GET /renda/_search
{"query": {"match": {"title":"小米电视"}
}
}

响应结果：

{"took":15,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":2,"max_score":0.5753642,"hits": [
{"_index":"renda","_type":"goods","_id":"3","_score":0.5753642,"_source": {"title":"小米电视4A","images":"","price":3899}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":0.2876821,"_source": {"title":"小米手机","images":"","price":2699}
}
]
}
}

在上面的案例中，不仅会查询到电视，而且与小米相关的都会查询到，多个词之间是 or 的关系。

某些情况下，需要更精确查找，即 and 关系。比如在电商平台精确搜索商品时，希望这个关系（查询条件切分词之后的关系）变成 and，可以这样做：

GET /renda/_search
{"query": {"match": {"title": {"query":"小米电视","operator":"and"}
}
}
}

响应结果：

{"took":8,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":1,"max_score":0.5753642,"hits": [
{"_index":"renda","_type":"goods","_id":"3","_score":0.5753642,"_source": {"title":"小米电视4A","images":"","price":3899}
}
]
}
}

此时，只有同时包含小米和电视的词条才会被搜索到。

词条匹配（term）

term查询被用于精确值匹配，这些精确值可能是数字、时间、布尔，或者那些未分词的字符串、keyword 类型的字符串。

效果类似于：select * from tableName where colName=value;

GET /renda/_search
{"query":{"term":{"price": 2699.00
}
}
}

响应结果：

{"took":6,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":2,"max_score":1,"hits": [
{"_index":"renda","_type":"goods","_id":"2","_score":1,"_source": {"title":"白米手机","images":"","price":2699}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":1,"_source": {"title":"小米手机","images":"","price":2699}
}
]
}
}

布尔组合（bool）

bool把各种其它查询通过must - 与、must_not -非、should - 或的方式进行组合。

GET /renda/_search
{"query":{"bool":{"must": {"match": {"title":"小米"}
},"must_not": {"match": {"title":"电视"}
},"should": {"match": {"title":"手机"}
}
}
}
}

响应结果：

{"took":2,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":1,"max_score":0.5753642,"hits": [
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":0.5753642,"_source": {"title":"小米手机","images":"","price":2699}
}
]
}
}

范围查询（range）

range查询找出那些落在指定区间内的数字或者时间。

GET/renda/_search{"query":{"range":{"price":{"gte":3000.0,"lt":4000.00}}}}

响应结果：

{"took":1,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":1,"max_score":1,"hits": [
{"_index":"renda","_type":"goods","_id":"3","_score":1,"_source": {"title":"小米电视4A","images":"","price":3899}
}
]
}
}

range查询允许以下字符：

gt- 大于gte- 大于等于lt- 小于lte- 小于等于

模糊查询（fuzzy）

fuzzy查询是term查询的模糊等价，很少直接使用它。

新增一个商品：

POST /renda/goods/5
{"title":"Apple手机","images":"","price": 6899.00
}

响应结果：

{"_index":"renda","_type":"goods","_id":"5","_version":1,"result":"created","_shards": {"total":2,"successful":1,"failed":0},"_seq_no":0,"_primary_term":2}

fuzzy查询是term查询的模糊等价，它允许用户搜索词条与实际词条的拼写出现偏差，但是偏差的编辑距离不得超过2：

GET /renda/_search
{"query": {"fuzzy": {"title":"applas"}
}
}

上面的查询，也能查询到 apple 手机：

{"took":4,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":1,"max_score":0.17260925,"hits": [
{"_index":"renda","_type":"goods","_id":"5","_score":0.17260925,"_source": {"title":"Apple手机","images":"","price":6899}
}
]
}
}

结果过滤

默认情况下，Elasticsearch 在搜索的结果中，会把文档中保存在_source的所有字段都返回。

如果只想获取其中的部分字段，可以添加_source的过滤。

直接指定字段

示例：

GET /renda/_search
{"_source": ["title","price"],"query": {"term": {"price": 2699
}
}
}

返回的结果：

{"took":2,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":2,"max_score":1,"hits": [
{"_index":"renda","_type":"goods","_id":"2","_score":1,"_source": {"price":2699,"title":"白米手机"}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":1,"_source": {"price":2699,"title":"小米手机"}
}
]
}
}

指定 includes 和 excludes

includes：来指定想要显示的字段。

excludes：来指定不想要显示的字段。

二者都是可选的。

示例：

GET /renda/_search
{"_source": {"includes":["title","price"]
},"query": {"term": {"price": 2699
}
}
}

与下面的结果将是一样的：

GET /renda/_search
{"_source": {"excludes": ["images"]
},"query": {"term": {"price": 2699
}
}
}

响应结果：

{"took":1,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":2,"max_score":1,"hits": [
{"_index":"renda","_type":"goods","_id":"2","_score":1,"_source": {"price":2699,"title":"白米手机"}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":1,"_source": {"price":2699,"title":"小米手机"}
}
]
}
}

过滤（Filter）

Elasticsearch 使用的查询语言（DSL）拥有一套查询组件，这些组件可以以无限组合的方式进行搭配。

这套组件可以在以下两种情况下使用：过滤情况 - filtering context 和查询情况 - query context。

如何选择查询与过滤：

通常的规则是，使用查询（query）语句来进行全文搜索或者其它任何需要影响相关性得分的搜索；除此以外的情况都使用过滤（filters)。

条件查询中进行过滤：

所有的查询都会影响到文档的评分及排名。如果需要在查询结果中进行过滤，并且不希望过滤条件影响评分，那么就不要把过滤条件作为查询条件来用，而是使用 filter 方式：

GET /renda/_search
{"query":{"bool":{"must":{"match": {"title":"小米手机"}},"filter":{"range":{"price":{"gt":2000.00,"lt":3800.00}}
}
}
}
}

无查询条件，直接过滤：

如果一次查询只有过滤，没有查询条件，不希望进行评分，可以使用 constant_score 取代只有 filter 语句的 bool 查询。在性能上是完全相同的，但对于提高查询简洁性和清晰度有很大帮助。

GET /renda/_search
{"query":{"constant_score": {"filter": {"range": {"price":{"gt":2000.00,"lt":3000.00}}
}
}
}
}

排序

单字段排序

sort可以按照不同的字段进行排序，并且通过order指定排序的方式。

GET /renda/_search
{"query": {"match": {"title":"小米手机"}
},"sort": [
{"price": {"order":"desc"}
}
]
}

多字段排序

假定想要结合使用price和_score进行查询，并且匹配的结果首先按照价格排序，然后按照相关性得分排序：

GET /renda/_search
{"query":
{"bool":
{"must":
{"match":
{"title":"小米手机"}
},"filter":{"range":
{"price":
{"gt":2000,"lt":4000
}
}
}
}
},"sort": [
{"price":
{"order":"desc"}
},
{"_score":
{"order":"desc"}

}
]
}

分页

Elasticsearch 中数据都存储在分片中，当执行搜索时每个分片独立搜索后，数据再经过整合返回。那么，如何实现分页查询呢？

Elasticsearch 的分页与 MySQL 数据库非常相似，都是指定两个值：

from- 目标数据的偏移值（开始位置），默认 from 为 0。size- 每页大小。GET/renda/_search{"query":{"match_all":{}},"sort":[{"price":{"order":"asc"}}],"from":3,"size":3}

结果：

{"took":1,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":4,"max_score":null,"hits": [
{"_index":"renda","_type":"goods","_id":"5","_score":null,"_source": {"title":"Apple手机","images":"","price":6899},"sort": [6899]
}
]
}
}

高亮

高亮原理：

服务端搜索数据，得到搜索结果。把搜索结果中，搜索关键字都加上约定好的标签。前端页面提前写好标签的 CSS 样式，即可高亮。

Elasticsearch 中实现高亮的语法比较简单：

GET /renda/_search
{"query": {"match": {"title":"手机"}
},"highlight": {"pre_tags":"","post_tags":"","fields": {"title": {}
}
}
}

在使用 match 查询的同时，加上一个 highlight 属性：

pre_tags：前置标签
post_tags：后置标签
fields：需要高亮的字段
title：这里声明 title 字段需要高亮

结果：

{"took":2,"timed_out":false,"_shards": {"total":5,"successful":5,"skipped":0,"failed":0},"hits": {"total":3,"max_score":0.2876821,"hits": [
{"_index":"renda","_type":"goods","_id":"5","_score":0.2876821,"_source": {"title":"Apple手机","images":"","price":6899},"highlight": {"title": ["Apple手机"]
}
},
{"_index":"renda","_type":"goods","_id":"2","_score":0.2876821,"_source": {"title":"白米手机","images":"","price":2699},"highlight": {"title": ["白米手机"]
}
},
{"_index":"renda","_type":"goods","_id":"gPeQqHUB-UTJAEEuqOm9","_score":0.2876821,"_source": {"title":"小米手机","images":"","price":2699},"highlight": {"title": ["小米手机"]
}
}
]
}
}

想了解更多，欢迎关注我的微信公众号：Renda_Zhang

上一篇:

下一篇:

  推荐阅读

分享