一、搜索API
1. 搜索API 端点地址
从索引tweet里面搜索字段user为kimchy的记录
GET /twitter/_search?q=user:kimchy
从索引tweet,user里面搜索字段user为kimchy的记录
GET /twitter/tweet,user/_search?q=user:kimchy
GET /kimchy,elasticsearch/_search?q=tag:wow
从所有索引里面搜索字段tag为wow的记录
GET /_all/_search?q=tag:wow GET /_search?q=tag:wow
说明:搜索的端点地址可以是多索引多mapping type的。搜索的参数可作为URI请求参数给出,也可用 request body 给出
2. URI Search
URI 搜索方式通过URI参数来指定查询相关参数。让我们可以快速做一个查询。
GET /twitter/_search?q=user:kimchy
可用的参数请参考: https://www.elastic.co/guide/en/elasticsearch/reference/current/search-uri-request.html
3. 查询结果说明
5. 特殊的查询参数用法
如果我们只想知道有多少文档匹配某个查询,可以这样用参数:
GET /bank/_search?q=city:b*&size=0
如果我们只想知道有没有文档匹配某个查询,可以这样用参数:
GET /bank/_search?q=city:b*&size=0&terminate_after=1
比较两个查询的结果可以知道第一个查询返回所有的命中文档数,第二个查询由于只需要知道有没有文档,所以只要有文档就立即返回
6. Request body Search
Request body 搜索方式以JSON格式在请求体中定义查询 query。请求方式可以是 GET 、POST 。
GET /twitter/_search { "query" : { "term" : { "user" : "kimchy" } } }
可用的参数:
timeout:请求超时时长,限定在指定时长内响应(即使没查完);
from: 分页的起始行,默认0;
size:分页大小;
request_cache:是否缓存请求结果,默认true。
terminate_after:限定每个分片取几个文档。如果设置,则响应将有一个布尔型字段terminated_early来指示查询执行是否实际已经terminate_early。缺省为no terminate_after;
search_type:查询的执行方式,可选值dfs_query_then_fetch or query_then_fetch ,默认: query_then_fetch ;
batched_reduce_size:一次在协调节点上应该减少的分片结果的数量。如果请求中的潜在分片数量可能很大,则应将此值用作保护机制以减少每个搜索请求的内存开销。
6.1 query 元素定义查询
query 元素用Query DSL 来定义查询。
GET /_search { "query" : { "term" : { "user" : "kimchy" } } }
6.2 指定返回哪些内容
6.2.1 source filter 对_source字段进行选择
GET /_search { "_source": false, "query" : { "term" : { "user" : "kimchy" } } }
通配符查询
GET /_search { "_source": [ "obj1.*", "obj2.*" ], "query" : { "term" : { "user" : "kimchy" } } } GET /_search { "_source": "obj.*", "query" : { "term" : { "user" : "kimchy" } } }
包含什么不包含什么
GET /_search { "_source": { "includes": [ "obj1.*", "obj2.*" ], "excludes": [ "*.description" ] }, "query" : { "term" : { "user" : "kimchy" } } }
6.2.2 stored_fields 来指定返回哪些stored字段
GET /_search { "stored_fields" : ["user", "postDate"], "query" : { "term" : { "user" : "kimchy" } } }
说明:* 可用来指定返回所有存储字段
6.2.3 docValue Field 返回存储了docValue的字段值
GET /_search { "query" : { "match_all": {} }, "docvalue_fields" : ["test1", "test2"] }
6.2.4 version 来指定返回文档的版本字段
GET /_search { "version": true, "query" : { "term" : { "user" : "kimchy" } } }
6.2.5 explain 返回文档的评分解释
GET /_search { "explain": true, "query" : { "term" : { "user" : "kimchy" } } }
6.2.6 Script Field 用脚本来对命中的每个文档的字段进行运算后返回
GET /bank/_search { "query": { "match_all": {} }, "script_fields": { "test1": { "script": { "lang": "painless", "source": "doc['balance'].value * 2" } }, "test2": { "script": { "lang": "painless", <!-- doc指文档--> "source": "doc['age'].value * params.factor", "params": { "factor": 2 } } } }}
搜索结果:
View Code
GET /bank/_search { "query": { "match_all": {} }, "script_fields": { "ffx": { "script": { "lang": "painless", "source": "doc['age'].value * doc['balance'].value" } }, "balance*2": { "script": { "lang": "painless", "source": "params['_source'].balance*2" } } } }
说明:
params _source 取 _source字段值
官方推荐使用doc,理由是用doc效率比取_source 高
搜索结果:
View Code
6.2.7 min_score 限制最低评分得分
GET /_search { "min_score": 0.5, "query" : { "term" : { "user" : "kimchy" } } }
6.2.8 post_filter 后置过滤:在查询命中文档、完成聚合后,再对命中的文档进行过滤。
如:要在一次查询中查询品牌为gucci且颜色为红色的shirts,同时还要得到gucci品牌各颜色的shirts的分面统计。
创建索引并指定mappping:
PUT /shirts { "mappings": { "_doc": { "properties": { "brand": { "type": "keyword"}, "color": { "type": "keyword"}, "model": { "type": "keyword"} } } } }
往索引里面放入文档即类似数据库里面的向表插入一行数据,并立即刷新
PUT /shirts/_doc/1?refresh { "brand": "gucci", "color": "red", "model": "slim" } PUT /shirts/_doc/2?refresh { "brand": "gucci", "color": "green", "model": "seec" }
执行查询:
GET /shirts/_search { "query": { "bool": { "filter": { "term": { "brand": "gucci" } } } }, "aggs": { "colors": { "terms": { "field": "color" } } }, "post_filter": { "term": { "color": "red" } } }
查询结果
{ "took": 109, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0, "hits": [ { "_index": "shirts", "_type": "_doc", "_id": "1", "_score": 0, "_source": { "brand": "gucci", "color": "red", "model": "slim" } } ] }, "aggregations": { "colors": { "doc_count_error_upper_bound": 0, "sum_other_doc_count": 0, "buckets": [ { "key": "green", "doc_count": 1 }, { "key": "red", "doc_count": 1 } ] } } }
6.2.9 sort 排序
可以指定按一个或多个字段排序。也可通过_score指定按评分值排序,_doc 按索引顺序排序。默认是按相关性评分从高到低排序。
GET /bank/_search { "query": { "match_all": {} }, "sort": [ { "age": { "order": "desc" } }, { "balance": { "order": "asc" } }, "_score" ] }
说明:
order 值:asc、desc。如果不给定,默认是asc,_score默认是desc
查询结果:
View Code
结果中每个文档会有排序字段值给出
"hits": { "total": 1000, "max_score": null, "hits": [ { "_index": "bank", "_type": "_doc", "_id": "549", "_score": 1, "_source": { "account_number": 549, "balance": 1932, "age": 40, "state": "OR" }, "sort": [ 40, 1932, 1 ] }
多值字段排序
对于值是数组或多值的字段,也可进行排序,通过mode参数指定按多值的:
PUT /my_index/_doc/1?refresh { "product": "chocolate", "price": [20, 4] } POST /_search { "query" : { "term" : { "product" : "chocolate" } }, "sort" : [ {"price" : {"order" : "asc", "mode" : "avg"}} ] }
Missing values 缺失该字段的文档
missing 的值可以是 _last, _first
GET /_search { "sort" : [ { "price" : {"missing" : "_last"} } ], "query" : { "term" : { "product" : "chocolate" } } }
地理空间距离排序
官方文档:
https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-sort.html#geo-sorting
GET /_search { "sort" : [ { "_geo_distance" : { "pin.location" : [-70, 40], "order" : "asc", "unit" : "km", "mode" : "min", "distance_type" : "arc" } } ], "query" : { "term" : { "user" : "kimchy" } } }
参数说明:
_geo_distance 距离排序关键字
pin.location是 geo_point 类型的字段
distance_type:距离计算方式 arc球面 、plane 平面。
unit: 距离单位 km 、m 默认m
Script Based Sorting 基于脚本计算的排序
GET /_search { "query" : { "term" : { "user" : "kimchy" } }, "sort" : { "_script" : { "type" : "number", "script" : { "lang": "painless", "source": "doc['field_name'].value * params.factor", "params" : { "factor" : 1.1 } }, "order" : "asc" } } }
6.3.0 折叠
用 collapse指定根据某个字段对命中结果进行折叠
GET /bank/_search { "query": { "match_all": {} }, "collapse" : { "field" : "age" }, "sort": ["balance"] }
查询结果:
View Code
高级折叠
GET /bank/_search { "query": { "match_all": {} }, "collapse" : { "field" : "age" , <!--指定inner_hits来解释折叠 --> "inner_hits": { "name": "details", <!-- 自命名 --> "size": 5, <!-- 指定每组取几个文档 --> "sort": [{ "balance": "asc" }] <!-- 组内排序 --> }, "max_concurrent_group_searches": 4 <!-- 指定组查询的并发数 --> }, "sort": ["balance"] }
查询结果:
View Code
在inner_hits 中返回多个角度的组内topN
GET /twitter/_search { "query": { "match": { "message": "elasticsearch" } }, "collapse" : { "field" : "user", "inner_hits": [ { "name": "most_liked", "size": 3, "sort": ["likes"] }, { "name": "most_recent", "size": 3, "sort": [{ "date": "asc" }] } ] }, "sort": ["likes"] }
说明:
most_liked:最像
most_recent:最近一段时间的
6.3.1 分页
from and size
GET /_search { "from" : 0, "size" : 10, "query" : { "term" : { "user" : "kimchy" } } }
注意:搜索请求耗用的堆内存和时间与 from + size 大小成正比。分页越深耗用越大,为了不因分页导致OOM或严重影响性能,ES中规定from + size 不能大于索引setting参数 index.max_result_window 的值,默认值为 10,000。
需要深度分页, 不受index.max_result_window 限制,怎么办?
Search after 在指定文档后取文档, 可用于深度分页
首次查询第一页
GET twitter/_search { "size": 10, "query": { "match" : { "title" : "elasticsearch" } }, "sort": [ {"date": "asc"}, {"_id": "desc"} ] }
后续页的查询
GET twitter/_search { "size": 10, "query": { "match" : { "title" : "elasticsearch" } }, "search_after": [1463538857, "654323"], "sort": [ {"date": "asc"}, {"_id": "desc"} ] }
注意:使用search_after,要求查询必须指定排序,并且这个排序组合值每个文档唯一(最好排序中包含_id字段)。 search_after的值用的就是这个排序值。 用search_after时 from 只能为0、-1。
6.3.2 高亮
准备数据:
PUT /hl_test/_doc/1 { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" }
查询高亮数据
GET /hl_test/_search { "query": { "match": { "title": "lucene" } }, "highlight": { "fields": { "title": {}, "content": {} } } }
查询结果:
{ "took": 113, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.2876821, "hits": [ { "_index": "hl_test", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" }, "highlight": { "title": [ "<em>lucene</em> solr and elasticsearch" ] } } ] } }
多字段高亮
GET /hl_test/_search { "query": { "match": { "title": "lucene" } }, "highlight": { "require_field_match": false, "fields": { "title": {}, "content": {} } } }
查询结果:
{ "took": 5, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.2876821, "hits": [ { "_index": "hl_test", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" }, "highlight": { "title": [ "<em>lucene</em> solr and elasticsearch" ], "content": [ "<em>lucene</em> solr and elasticsearch for search" ] } } ] } }
说明:
高亮结果在返回的每个文档中以hightlight节点给出
指定高亮标签
GET /hl_test/_search { "query": { "match": { "title": "lucene" } }, "highlight": { "require_field_match": false, "fields": { "title": { "pre_tags":["<strong>"], "post_tags": ["</strong>"] }, "content": {} } } }
查询结果:
{ "took": 5, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.2876821, "hits": [ { "_index": "hl_test", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" }, "highlight": { "title": [ "<strong>lucene</strong> solr and elasticsearch" ], "content": [ "<em>lucene</em> solr and elasticsearch for search" ] } } ] } }
高亮的详细设置请参考官网:https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-highlighting.html
6.3.3 Profile 为了调试、优化
对于执行缓慢的查询,我们很想知道它为什么慢,时间都耗在哪了,可以在查询上加入上 profile 来获得详细的执行步骤、耗时信息。
GET /twitter/_search { "profile": true, "query" : { "match" : { "message" : "some number" } } }
信息的说明请参考:
https://www.elastic.co/guide/en/elasticsearch/reference/current/search-profile.html
7. count api 查询数量
PUT /twitter/_doc/1?refresh { "user": "kimchy" } GET /twitter/_doc/_count?q=user:kimchy GET /twitter/_doc/_count { "query" : { "term" : { "user" : "kimchy" } } }
结果说明:
{ "count" : 1, "_shards" : { "total" : 5, "successful" : 5, "skipped" : 0, "failed" : 0 } }
8. validate api
用来检查我们的查询是否正确,以及查看底层生成查询是怎样的
GET twitter/_validate/query?q=user:foo
8.1 校验查询
GET twitter/_doc/_validate/query { "query": { "query_string": { "query": "post_date:foo", "lenient": false } } }
查询结果:
{ "valid": true, "_shards": { "total": 1, "successful": 1, "failed": 0 } }
8.2 获得查询解释
GET twitter/_doc/_validate/query?explain=true { "query": { "query_string": { "query": "post_date:foo", "lenient": false } } }
查询结果
{ "valid": true, "_shards": { "total": 1, "successful": 1, "failed": 0 }, "explanations": [ { "index": "twitter", "valid": true, "explanation": """+MatchNoDocsQuery("unmapped field [post_date]") #MatchNoDocsQuery("Type list does not contain the index type")""" } ] }
8.3 用rewrite获得比explain 更详细的解释
GET twitter/_doc/_validate/query?rewrite=true { "query": { "more_like_this": { "like": { "_id": "2" }, "boost_terms": 1 } } }
查询结果:
{ "valid": true, "_shards": { "total": 1, "successful": 1, "failed": 0 }, "explanations": [ { "index": "twitter", "valid": true, "explanation": """+(MatchNoDocsQuery("empty BooleanQuery") -ConstantScore(MatchNoDocsQuery("empty BooleanQuery"))) #MatchNoDocsQuery("Type list does not contain the index type")""" } ] }
8.4 获得所有分片上的查询解释
GET twitter/_doc/_validate/query?rewrite=true&all_shards=true { "query": { "match": { "user": { "query": "kimchy", "fuzziness": "auto" } } } }
查询结果:
{ "valid": true, "_shards": { "total": 3, "successful": 3, "failed": 0 }, "explanations": [ { "index": "twitter", "shard": 0, "valid": true, "explanation": """MatchNoDocsQuery("unmapped field [user]")""" }, { "index": "twitter", "shard": 1, "valid": true, "explanation": """MatchNoDocsQuery("unmapped field [user]")""" }, { "index": "twitter", "shard": 2, "valid": true, "explanation": """MatchNoDocsQuery("unmapped field [user]")""" } ] }
官网链接:
https://www.elastic.co/guide/en/elasticsearch/reference/current/search-validate.html
9. Explain api
获得某个查询的评分解释,及某个文档是否被这个查询命中
GET /twitter/_doc/0/_explain { "query" : { "match" : { "message" : "elasticsearch" } } }
官网链接:
https://www.elastic.co/guide/en/elasticsearch/reference/current/search-explain.html
10. Search Shards API
让我们可以了解可执行查询的索引分片节点情况
GET /twitter/_search_shards
查询结果:
View Code
想知道指定routing值的查询将在哪些分片节点上执行
GET /twitter/_search_shards?routing=foo,baz
查询结果:
{ "nodes": { "qkmtovyLRPWjXcfDTryNwA": { "name": "qkmtovy", "ephemeral_id": "sxgsvzsORraAnN7PIlMYpg", "transport_address": "127.0.0.1:9300", "attributes": {} } }, "indices": { "twitter": {} }, "shards": [ [ { "state": "STARTED", "primary": true, "node": "qkmtovyLRPWjXcfDTryNwA", "relocating_node": null, "shard": 1, "index": "twitter", "allocation_id": { "id": "8S88pnUkSSy8kiCcwBgb9Q" } } ] ] }
11. Search Template 查询模板
注册一个模板
POST _scripts/<templatename> { "script": { "lang": "mustache", "source": { "query": { "match": { "title": "{{query_string}}" } } } } }
使用模板进行查询
GET _search/template { "id": "<templateName>", "params": { "query_string": "search for these words" } }
查询结果:
{ "took": 11, "timed_out": false, "_shards": { "total": 38, "successful": 38, "skipped": 0, "failed": 0 }, "hits": { "total": 0, "max_score": null, "hits": [] } }
详细了解请参考官网:
https://www.elastic.co/guide/en/elasticsearch/reference/current/search-template.html
二、Query DSL
官网介绍链接:https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl.html
Query DSL 介绍
1. DSL是什么?
Domain Specific Language:领域特定语言
Elasticsearch基于JSON提供完整的查询DSL来定义查询。
一个查询可由两部分字句构成:
Leaf query clauses 叶子查询字句
Leaf query clauses 在指定的字段上查询指定的值, 如:match, term or range queries. 叶子字句可以单独使用.
Compound query clauses 复合查询字句
以逻辑方式组合多个叶子、复合查询为一个查询
2. Query and filter context
一个查询字句的行为取决于它是用在query context 还是 filter context 中 。
Query context 查询上下文
用在查询上下文中的字句回答“这个文档有多匹配这个查询?”。除了决定文档是否匹配,字句匹配的文档还会计算一个字句评分,来评定文档有多匹配。查询上下文由 query 元素表示。
Filter context 过滤上下文
过滤上下文由 filter 元素或 bool 中的 must not 表示。用在过滤上下文中的字句回答“这个文档是否匹配这个查询?”,不参与相关性评分。
被频繁使用的过滤器将被ES自动缓存,来提高查询性能。
示例:
GET /_search { <!--查询 --> "query": { "bool": { "must": [ { "match": { "title": "Search" }}, { "match": { "content": "Elasticsearch" }} ], <!--过滤 --> "filter": [ { "term": { "status": "published" }}, { "range": { "publish_date": { "gte": "2015-01-01" }}} ] } } }
说明:查询和过滤都是对所有文档进行查询,最后两个结果取交集
提示:在查询上下文中使用查询子句来表示影响匹配文档得分的条件,并在过滤上下文中使用所有其他查询子句。
查询分类介绍
1. Match all query 查询所有
GET /_search { "query": { "match_all": {} } }
相反,什么都不查
GET /_search { "query": { "match_none": {} } }
2. Full text querys
全文查询,用于对分词的字段进行搜索。会用查询字段的分词器对查询的文本进行分词生成查询。可用于短语查询、模糊查询、前缀查询、临近查询等查询场景
官网链接:
https://www.elastic.co/guide/en/elasticsearch/reference/current/full-text-queries.html
3. match query
全文查询的标准查询,它可以对一个字段进行模糊、短语查询。 match queries 接收 text/numerics/dates, 对它们进行分词分析, 再组织成一个boolean查询。可通过operator 指定bool组合操作(or、and 默认是 or ), 以及minimum_should_match 指定至少需多少个should(or)字句需满足。还可用ananlyzer指定查询用的特殊分析器。
GET /_search { "query": { "match" : { "message" : "this is a test" } } }
说明:message是字段名
官网链接:https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-match-query.html
示例:
构造索引和数据:
PUT /ftq/_doc/1 { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } PUT /ftq/_doc/2 { "title": "java spring boot", "content": "lucene is writerd by java" }
执行查询1
GET ftq/_doc/_validate/query?rewrite=true { "query": { "match": { "title": "lucene java" } } }
查询结果1:
{ "valid": true, "_shards": { "total": 1, "successful": 1, "failed": 0 }, "explanations": [ { "index": "ftq", "valid": true, "explanation": "title:lucene title:java" } ] }
执行查询2:
GET ftq/_search { "query": { "match": { "title": "lucene java" } } }
查询结果2:
{ "took": 6, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 2, "max_score": 0.2876821, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "2", "_score": 0.2876821, "_source": { "title": "java spring boot", "content": "lucene is writerd by java" } }, { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
执行查询3:指定操作符
GET ftq/_search { "query": { "match": { "title": { "query": "lucene java", "operator": "and" } } } }
查询结果3:
{ "took": 4, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 0, "max_score": null, "hits": [] } }
模糊查询,最大编辑数为2
GET ftq/_search { "query": { "match": { "title": { "query": "ucen elatic", "fuzziness": 2 } } } }
模糊查询结果:
{ "took": 280, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.14384104, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.14384104, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
指定最少需满足两个词匹配
GET ftq/_search { "query": { "match": { "content": { "query": "ucen elatic java", "fuzziness": 2, "minimum_should_match": 2 } } } }
查询结果:
{ "took": 19, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.43152314, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "2", "_score": 0.43152314, "_source": { "title": "java spring boot", "content": "lucene is writerd by java" } } ] } }
可用max_expansions 指定模糊匹配的最大词项数,默认是50。比如:反向索引中有 100 个词项与 ucen 模糊匹配,只选用前50 个。
4. match phrase query
match_phrase 查询用来对一个字段进行短语查询,可以指定 analyzer、slop移动因子。
对字段进行短语查询1:
GET ftq/_search { "query": { "match_phrase": { "title": "lucene solr" } } }
结果1:
{ "took": 3, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.5753642, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.5753642, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
对字段进行短语查询2:
GET ftq/_search { "query": { "match_phrase": { "title": "lucene elasticsearch" } } }
结果2:
{ "took": 3, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 0, "max_score": null, "hits": [] } }
对查询指定移动因子:
GET ftq/_search { "query": { "match_phrase": { "title": { "query": "lucene elasticsearch", "slop": 2 } } } }
查询结果:
{ "took": 2174, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 0.27517417, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.27517417, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
5. match phrase prefix query
match_phrase_prefix 在 match_phrase 的基础上支持对短语的最后一个词进行前缀匹配
GET /_search { "query": { "match_phrase_prefix" : { "message" : "quick brown f" } } }
指定前缀匹配选用的最大词项数量
GET /_search { "query": { "match_phrase_prefix" : { "message" : { "query" : "quick brown f", "max_expansions" : 10 } } } }
6. Multi match query
如果你需要在多个字段上进行文本搜索,可用multi_match 。 multi_match在 match的基础上支持对多个字段进行文本查询。
查询1:
GET ftq/_search { "query": { "multi_match" : { "query": "lucene java", "fields": [ "title", "content" ] } } }
结果1:
{ "took": 1973, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 2, "max_score": 0.5753642, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "2", "_score": 0.5753642, "_source": { "title": "java spring boot", "content": "lucene is writerd by java" } }, { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
查询2:字段通配符查询
GET ftq/_search { "query": { "multi_match" : { "query": "lucene java", "fields": [ "title", "cont*" ] } } }
结果2:
{ "took": 5, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 2, "max_score": 0.5753642, "hits": [ { "_index": "ftq", "_type": "_doc", "_id": "2", "_score": 0.5753642, "_source": { "title": "java spring boot", "content": "lucene is writerd by java" } }, { "_index": "ftq", "_type": "_doc", "_id": "1", "_score": 0.2876821, "_source": { "title": "lucene solr and elasticsearch", "content": "lucene solr and elasticsearch for search" } } ] } }
查询3:给字段的相关性评分加权重
GET ftq/_search?explain=true { "query": { "multi_match" : { "query": "lucene elastic", "fields": [ "title^5", "content" ] } } }
结果3:
View Code
7. Common terms query
common 常用词查询
问1、什么是停用词?索引时做停用词处理的目的是什么?
不再使用的词,做停用词处理的目的是提高索引的效率,去掉不需要的索引操作,即停用词不需要索引
问2、如果在索引时应用停用词处理,下面的两个查询会查询什么词项?
the brown fox—— brown fox
not happy——happy
问3、索引时应用停用词处理对搜索精度是否有影响?如果不做停用词处理又会有什么影响?如何协调这两个问题?如何保证搜索的精确度又兼顾搜索性能?
索引时应用停用词处理对搜索精度有影响,不做停用词处理又会影响索引的效率,要协调这两个问题就必须要使用tf-idf 相关性计算模型
7.1 tf-idf 相关性计算模型简介
tf:term frequency 词频 :指一个词在一篇文档中出现的频率。
如“世界杯”在文档A中出现3次,那么可以定义“世界杯”在文档A中的词频为3。请问在一篇3000字的文章中出现“世界杯”3次和一篇150字的文章中出现3词,哪篇文章更是与“世界杯”有关的。也就是说,简单用出现次数作为频率不够准确。那就用占比来表示:
问:tf值越大是否就一定说明这个词更相关?
不是,出现太多了说明不重要
说明:tf的计算不一定非是这样的,可以定义不同的计算方式。
df:document frequency 词的文档频率 :指包含某个词的文档数(有多少文档中包含这个词)。 df越大的词越常见,哪些词会是高频词?
问1:词的df值越大说明这个词在这个文档集中是越重要还是越不重要?
越不重要
问2:词t的tf高,在文档集中的重要性也高,是否说明文档与该词越相关?举例:整个文档集中只有3篇文档中有“世界杯”,文档A中就出现了“世界杯”好几次。
不能说明文档与该词越相关
问3:如何用数值体现词t在文档集中的重要性?df可以吗?
不可以
idf:inverse document frequency 词的逆文档频率 :用来表示词在文档集中的重要性。文档总数/ df ,df越小,词越重要,这个值会很大,那就对它取个自然对数,将值映射到一个较小的取值范围。
说明: +1 是为了避免除0(即词t在文档集中未出现的情况)
tf-idf 相关性性计算模型:tf-idf t = tf t,d * idf t
说明: tf-idf 相关性性计算模型的值为词频( tf t,d)乘以词的逆文档频率(idf t)
7.2 Common terms query
common 区分常用(高频)词查询让我们可以通过cutoff_frequency来指定一个分界文档频率值,将搜索文本中的词分为高频词和低频词,低频词的重要性高于高频词,先对低频词进行搜索并计算所有匹配文档相关性得分;然后再搜索和高频词匹配的文档,这会搜到很多文档,但只对和低频词重叠的文档进行相关性得分计算(这可保证搜索精确度,同时大大提高搜索性能),和低频词累加作为文档得分。实际执行的搜索是 必须包含低频词 + 或包含高频词。
思考:这样处理下,如果用户输入的都是高频词如 “to be or not to be”结果会是怎样的?你希望是怎样的?
优化:如果都是高频词,那就对这些词进行and 查询。
进一步优化:让用户可以自己定对高频词做and/or 操作,自己定对低频词进行and/or 操作;或指定最少得多少个同时匹配
示例1:
GET /_search { "query": { "common": { "message": { "query": "this is bonsai cool", "cutoff_frequency": 0.001 } } } }
说明:
cutoff_frequency : 值大于1表示文档数,0-1.0表示占比。 此处界定 文档频率大于 0.1%的词为高频词。
示例2:
GET /_search { "query": { "common": { "body": { "query": "nelly the elephant as a cartoon", "cutoff_frequency": 0.001, "low_freq_operator": "and" } } } }
说明:low_freq_operator指定对低频词做与操作
可用参数:minimum_should_match (high_freq, low_freq), low_freq_operator (default “or”) and high_freq_operator (default “or”)、 boost and analyzer
示例3:
GET /_search { "query": { "common": { "body": { "query": "nelly the elephant as a cartoon", "cutoff_frequency": 0.001, "minimum_should_match": 2 } } } }
示例4:
GET /_search { "query": { "common": { "body": { "query": "nelly the elephant not as a cartoon", "cutoff_frequency": 0.001, "minimum_should_match": { "low_freq" : 2, "high_freq" : 3 } } } } }
示例5:
8. Query string query
query_string 查询,让我们可以直接用lucene查询语法写一个查询串进行查询,ES中接到请求后,通过查询解析器解析查询串生成对应的查询。使用它要求掌握lucene的查询语法。
示例1:指定单个字段查询
GET /_search { "query": { "query_string" : { "default_field" : "content", "query" : "this AND that OR thus" } } }
示例2:指定多字段通配符查询
GET /_search { "query": { "query_string" : { "fields" : ["content", "name.*^5"], "query" : "this AND that OR thus" } } }
可与query同用的参数,如 default_field、fields,及query 串的语法请参考:
https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-query-string-query.html
9. 查询描述规则语法(查询解析语法)
Term 词项:
单个词项的表示: 电脑
短语的表示: "联想笔记本电脑"
Field 字段:
字段名:
示例: name:“联想笔记本电脑” AND type:电脑
如果name是默认字段,则可写成: “联想笔记本电脑” AND type:电脑
如果查询串是:type:电脑 计算机 手机
注意:只有第一个是type的值,后两个则是使用默认字段。
Term Modifiers 词项修饰符:
10. Simple Query string query
simple_query_string 查同 query_string 查询一样用lucene查询语法写查询串,较query_string不同的地方:更小的语法集;查询串有错误,它会忽略错误的部分,不抛出错误。更适合给用户使用。
示例:
GET /_search { "query": { "simple_query_string" : { "query": "\"fried eggs\" +(eggplant | potato) -frittata", "fields": ["title^5", "body"], "default_operator": "and" } } }
语法请参考:
https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-simple-query-string-query.html
11. Term level querys
官网链接:
https://www.elastic.co/guide/en/elasticsearch/reference/current/term-level-queries.html
11.1 Term query
term 查询用于查询指定字段包含某个词项的文档。
示例1:
POST _search { "query": { "term" : { "user" : "Kimchy" } } }
示例2:加权重
GET _search { "query": { "bool": { "should": [ { "term": { "status": { "value": "urgent", "boost": 2 } } }, { "term": { "status": "normal" } } ] } } }
11.2 Terms query
terms 查询用于查询指定字段包含某些词项的文档。
GET /_search { "query": { "terms" : { "user" : ["kimchy", "elasticsearch"]} } }
Terms 查询支持嵌套查询的方式来获得查询词项,相当于 in (select term from other)
示例1:Terms query 嵌套查询示例
PUT /users/_doc/2 { "followers" : ["1", "3"] } PUT /tweets/_doc/1 { "user" : "1" } GET /tweets/_search { "query": { "terms": { "user": { "index": "users", "type": "_doc", "id": "2", "path": "followers" } } } }
查询结果:
{ "took": 14, "timed_out": false, "_shards": { "total": 5, "successful": 5, "skipped": 0, "failed": 0 }, "hits": { "total": 1, "max_score": 1, "hits": [ { "_index": "tweets", "_type": "_doc", "_id": "1", "_score": 1, "_source": { "user": "1" } } ] } }
嵌套查询可用参数说明:
11.3 range query
范围查询示例1:
GET _search { "query": { "range" : { "age" : { "gte" : 10, "lte" : 20, "boost" : 2.0 } } } }
范围查询示例2:
GET _search { "query": { "range" : { "date" : { "gte" : "now-1d/d", "lt" : "now/d" } } } }
范围查询示例3:
GET _search { "query": { "range" : { "born" : { "gte": "01/01/2012", "lte": "2013", "format": "dd/MM/yyyy||yyyy" } } } }
范围查询参数说明:
范围查询时间舍入 ||说明:
时间数学计算规则请参考:
https://www.elastic.co/guide/en/elasticsearch/reference/current/common-options.html#date-math
11.4 exists query
查询指定字段值不为空的文档。相当 SQL 中的 column is not null
GET /_search { "query": { "exists" : { "field" : "user" } } }
查询指定字段值为空的文档
GET /_search { "query": { "bool": { "must_not": { "exists": { "field": "user" } } } } }
11.5 prefix query 词项前缀查询
示例1:
GET /_search { "query": { "prefix" : { "user" : "ki" } } }
示例2:加权
GET /_search { "query": { "prefix" : { "user" : { "value" : "ki", "boost" : 2.0 } } } }
11.6 wildcard query 通配符查询: ? *
示例1:
GET /_search { "query": { "wildcard" : { "user" : "ki*y" } } }
示例2:加权
GET /_search { "query": { "wildcard": { "user": { "value": "ki*y", "boost": 2 } } }}
11.7 regexp query 正则查询
示例1:
GET /_search { "query": { "regexp":{ "name.first": "s.*y" } } }
示例2:加权
GET /_search { "query": { "regexp":{ "name.first":{ "value":"s.*y", "boost":1.2 } } } }
正则语法请参考:
https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-regexp-query.html#regexp-syntax
11.8 fuzzy query 模糊查询
示例1:
GET /_search { "query": { "fuzzy" : { "user" : "ki" } } }
示例2:
GET /_search { "query": { "fuzzy" : { "user" : { "value": "ki", "boost": 1.0, "fuzziness": 2, "prefix_length": 0, "max_expansions": 100 } } } }
11.9 type query mapping type 查询
GET /_search { "query": { "type" : { "value" : "_doc" } } }
11.10 ids query 根据文档id查询
GET /_search { "query": { "ids" : { "type" : "_doc", "values" : ["1", "4", "100"] } } }
12. Compound querys 复合查询
官网链接:
https://www.elastic.co/guide/en/elasticsearch/reference/current/compound-queries.html
12.1 Constant Score query
用来包装另一个查询,将查询匹配的文档的评分设为一个常值。
GET /_search { "query": { "constant_score" : { "filter" : { "term" : { "user" : "kimchy"} }, "boost" : 1.2 } } }
12.2 Bool query
Bool 查询用bool操作来组合多个查询字句为一个查询。 可用的关键字:
示例:
POST _search { "query": { "bool" : { "must" : { "term" : { "user" : "kimchy" } }, "filter": { "term" : { "tag" : "tech" } }, "must_not" : { "range" : { "age" : { "gte" : 10, "lte" : 20 } } }, "should" : [ { "term" : { "tag" : "wow" } }, { "term" : { "tag" : "elasticsearch" } } ], "minimum_should_match" : 1, "boost" : 1.0 } } }
说明:should满足一个或者两个或者都不满足