全文检索字段用match， 其他非text字段匹配用term。 不要使用term来进行文本字段查询

match_all query

  1. POST /lagou-company-index/_search
  2. {
  3.   "query": {
  4.     "match_all": {}
  5.   }
  6. }

全文搜索(full-text query)

匹配搜索 match query

数据准备

  1. PUT /lagou-property
  2. {
  3.   "settings": {},
  4.   "mappings": {
  5.     "properties": {
  6.       "title": {
  7.         "type": "text",
  8.         "analyzer": "ik_max_word"
  9.       },
  10.       "images": {
  11.         "type": "keyword"
  12.       },
  13.       "price": {
  14.         "type": "float"
  15.       }
  16.     }
  17.   }
  18. }
  20. POST /lagou-property/_doc/
  21. {
  22.   "title": "小米电视4A",
  23.   "images": "http://image.lagou.com/12479122.jpg",
  24.   "price": 4288
  25. }
  27. POST /lagou-property/_doc/
  28. {
  29.   "title": "小米手机",
  30.   "images": "http://image.lagou.com/12479622.jpg",
  31.   "price": 2699
  32. }
  33. POST /lagou-property/_doc/
  34. {
  35.   "title": "华为手机",
  36.   "images": "http://image.lagou.com/12479922.jpg",
  37.   "price": 5699
  38. }

  • or ： match默认就是or关系

    1. POST /lagou-property/_search
    2. {
    3. "query": {
    4.   "match": {
    5.     "title": "小米电视4A"
    6.   }
    7. }
    8. }

  • and 分词后，多个词之间是and关系， 小米、电视 、4A

    1. POST /lagou-property/_search
    2. {
    3. "query": {
    4.   "match": {
    5.     "title": {
    6.       "query": "小米电视4A",
    7.       "operator": "and"
    8.     }
    9.   }
    10. }
    11. }

    短语匹配 **match phrase query**

    短语中的空格去掉后，不需要移动作为一个整体就能查询到 ```yaml GET /lagou-property/_search { “query”: { “match_phrase”: {

    1. "title": "小米电视"

    } } }

slop移动因子， 可以指定跨动分词个数

GET /lagou-property/_search { “query”: { “match_phrase”: { “title”: { “query”: “小米 4A”, “slop”: 2 } } } }

  1. <a name="HWzL7"></a>
  2. #### query string 
  3. 无需指定某字段而对文档全文进行匹配查询的一个高级查询
  4. ```yaml
  5. # 默认 全字段搜索
  6. GET /lagou-property/_search
  7. {
  8.   "query": {
  9.     "query_string": {
  10.       "query": "2699"
  11.     }
  12.   }
  13. }
  15. # 指定字段
  16. GET /lagou-property/_search
  17. {
  18.   "query": {
  19.     "query_string": {
  20.       "query": "2699",
  21.       "default_field": "price"
  22.     }
  23.   }
  24. }
  26. # 逻辑查询
  27. GET /lagou-property/_search
  28. {
  29.   "query": {
  30.     "query_string": {
  31.       "query": "手机 or 小米",
  32.       "default_field": "title"
  33.     }
  34.   }
  35. }
  37. # 模糊查询 ~1 表示带有修正的查询
  38. GET /lagou-property/_search
  39. {
  40.   "query": {
  41.     "query_string": {
  42.       "query": "大米~1",
  43.       "default_field": "title"
  44.     }
  45.   }
  46. }
  48. # 多字段支持
  49. GET /lagou-property/_search
  50. {
  51.   "query": {
  52.     "query_string": {
  53.       "query": "2699",
  54.       "fields": [
  55.         "title",
  56.         "price"
  57.       ]
  58.     }
  59.   }
  60. }

多字段匹配搜索 multi_match

  1. GET /lagou-property/_search
  2. {
  3.   "query": {
  4.     "multi_match": {
  5.       "query": "2699",
  6.       "fields": ["title","price"]
  7.     }
  8.   }
  9. }
  11. #可以用通配符来表示字段
  12. GET /lagou-property/_search
  13. {
  14.   "query": {
  15.     "multi_match": {
  16.       "query": "http://image.lagou.com/12479622.jpg",
  17.       "fields": [
  18.         "title",
  19.         "ima*"
  20.       ]
  21.     }
  22.   }
  23. }

词条级搜索 term_level queries

不分析词条，精确级搜索

  1. PUT /book
  2. {
  3.   "settings": {},
  4.   "mappings": {
  5.     "properties": {
  6.       "description": {
  7.         "type": "text",
  8.         "analyzer": "ik_max_word"
  9.       },
  10.       "name": {
  11.         "type": "text",
  12.         "analyzer": "ik_max_word"
  13.       },
  14.       "price": {
  15.         "type": "float"
  16.       },
  17.       "timestamp": {
  18.         "type": "date",
  19.         "format": "yyyy-MM-dd HH:mm:ss||yyyy-MM-dd||epoch_millis"
  20.       }
  21.     }
  22.   }
  23. }
  26. PUT /book/_doc/1
  27. {
  28.   "name": "lucene",
  29.   "description": "Lucene Core is a Java library providing powerful indexing and search features, as well as spellchecking, hit highlighting and advanced analysis/tokenization capabilities. The PyLucene sub project provides Python bindings for Lucene Core. ",
  30.   "price": 100.45,
  31.   "timestamp": "2020-08-21 19:11:35"
  32. }
  34. PUT /book/_doc/2
  35. {
  36.   "name": "solr",
  37.   "description": "Lucene Core is a Java library providing powerful indexing and search features, as well as spellchecking, hit highlighting and advanced analysis/tokenization capabilities. The PyLucene sub project provides Python bindings for Lucene Core. ",
  38.   "price": 100.45,
  39.   "timestamp": "2020-08-21 19:11:35"
  40. }
  42. PUT /book/_doc/3
  43. {
  44.   "name": "Hadoop",
  45.   "description": "The Apache Hadoop software library is a framework that allows for the distributed processing of large data sets across clusters of computers using simple programming models.",
  46.   "price": 620.45,
  47.   "timestamp": "2020-08-22 19:18:35"
  48. }
  49. PUT /book/_doc/4
  50. {
  51.   "name": "ElasticSearch",
  52.   "description": "Elasticsearch是一个基于Lucene的搜索服务器。它提供了一个分布式多用户能力 的全文搜索引擎，基于RESTful web接口。Elasticsearch是用Java语言开发的，并作为Apache许可条 款下的开放源码发布，是一种流行的企业级搜索引擎。Elasticsearch用于云计算中，能够达到实时搜 索，稳定，可靠，快速，安装使用方便。官方客户端在Java、.NET（C#）、PHP、Python、Apache Groovy、Ruby和许多其他语言中都是可用的。根据DB-Engines的排名显示，Elasticsearch是最受欢 迎的企业搜索引擎，其次是Apache Solr，也是基于Lucene。",
  53.   "price": 999.99,
  54.   "timestamp": "2020-08-15 10:11:35"
  55. }

term query

用于查询指定字段包含某个词项的文档，类似于%xxx%

  1. POST /book/_search
  2. {
  3.   "query": {
  4.     "term": {
  5.       "name": "solr"
  6.     }
  7.   }
  8. }
  10. POST /book/_search
  11. {
  12.   "query": {
  13.     "term": {
  14.       "description": "solr"
  15.     }
  16.   }
  17. }

terms query

用于查询 指定字段 包含某些词项 ，类似于in %xxx%, %xxx%

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "terms": {
  5.       "name": [
  6.         "solr",
  7.         "elasticsearch"
  8.       ]
  9.     }
  10.   }
  11. }

范围搜索 range query

  • gte：大于等于
  • gt：大于
  • lte：小于等于
  • lt：小于
  • boost：查询权重 ```yaml GET /book/_search { “query”: { “range”: {
    1. "price": {
    2.   "gte": 10,
    3.   "lte": 200,
    4.   "boost": 2
    5. }
    } } }

GET /book/_search { “query”: { “range”: { “timestamp”: { “gte”: “now-10d/d”, “lt”: “now/d” } } } }

指定日期格式

GET book/_search { “query”: { “range”: { “timestamp”: { “gte”: “18/08/2020”, “lte”: “2021”, “format”: “dd/MM/yyyy||yyyy” } } } }

  1. <a name="eFWzD"></a>
  2. #### exists query 不为空搜索
  3. 相当于: column is not null
  4. ```yaml
  5. GET /book/_search
  6. {
  7.   "query": {
  8.     "exists": {
  9.       "field": "price"
  10.     }
  11.   }
  12. }

词项前缀搜索(prefix query)

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "prefix": {
  5.       "name": "so"
  6.     }
  7.   }
  8. }

通配符搜索(wildcard query)

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "wildcard": {
  5.       "name": "so*r"
  6.     }
  7.   }
  8. }
  10. GET /book/_search
  11. {
  12.   "query": {
  13.     "wildcard": {
  14.       "name": {
  15.         "value": "lu*",
  16.         "boost": 2
  17.       }
  18.     }
  19.   }
  20. }

模糊搜索(fuzzy query)

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "fuzzy": {
  5.       "name": "so"
  6.     }
  7.   }
  8. }
  10. #fuzziness模糊值
  11. GET /book/_search
  12. {
  13.   "query": {
  14.     "fuzzy": {
  15.       "name": {
  16.         "value": "so",
  17.         "boost": 1,
  18.         "fuzziness": 2
  19.       }
  20.     }
  21.   }
  22. }
  25. GET /book/_search
  26. {
  27.   "query": {
  28.     "fuzzy": {
  29.       "name": {
  30.         "value": "sorl",
  31.         "boost": 1,
  32.         "fuzziness": 2
  33.       }
  34.     }
  35.   }
  36. }

ids搜索(id集合查询)

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "ids": {
  5.       "values": [
  6.         "1",
  7.         "3"
  8.       ]
  9.     }
  10.   }
  11. }

复合搜索(compound query)

constant_score query

用来包装一个查询，将查询的文档评分设为常值

  1. GET /book/_search
  2. {
  3.   "query": {
  4.     "term": {
  5.       "description": "solr"
  6.     }
  7.   }
  8. }
  10. GET /book/_search
  11. {
  12.   "query": {
  13.     "constant_score": {
  14.       "filter": {
  15.         "term": {
  16.           "description": "solr"
  17.         }
  18.       },
  19.       "boost": 1.0
  20.     }
  21.   }
  22. }

布尔搜索(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

  • must：必须满足
  • should：或
  • filter：必须满足
  • must_not：必须不满足 ```yaml POST /book/_search { “query”: { “bool”: {
    1. "should": {
    2.   "match": {
    3.     "description": "java"
    4.   }
    5. },
    6. "filter": {
    7.   "term": {
    8.     "name": "solr"
    9.   }
    10. },
    11. "must_not": {
    12.   "range": {
    13.     "price": {
    14.       "gte":300,
    15.       "lte":400
    16.     }
    17.   }
    18. },
    19. "minimum_should_match": 1,
    20. "boost": 1
    } } }
  1. minimum_should_match代表了最小匹配精度，如果设置minimum_should_match=1，那么should 语句中至少需要有一个条件满足。
  2. <a name="jG4xV"></a>
  3. #### 排序
  5. - 元数据分数，按相关性得分升序
  6. ```yaml
  7. POST /book/_search
  8. {
  9.   "query": {
  10.     "match": {
  11.       "description": "solr"
  12.     }
  13.   },
  14.   "sort": [
  15.     {
  16.       "_score": {
  17.         "order": "asc"
  18.       }
  19.     }
  20.   ]
  21. }

  • 按字段排序

    1. POST /book/_search
    2. {
    3. "query": {
    4.   "match_all": {}
    5. },
    6. "sort": [
    7.   {
    8.     "price": {
    9.       "order": "desc"
    10.     }
    11.   }
    12. ]
    13. }

  • 多级排序

    1. POST /book/_search
    2. {
    3. "query": {
    4.   "match_all": {}
    5. },
    6. "sort": [
    7.   {
    8.     "price": {
    9.       "order": "desc"
    10.     }
    11.   },
    12.   {
    13.     "timestamp": {
    14.       "order": "desc"
    15.     }
    16.   }
    17. ]
    18. }

    分页

    1. POST /book/_search
    2. {
    3. "query": {
    4.   "match_all": {}
    5. },
    6. "sort": [
    7.   {
    8.     "price": {
    9.       "order": "desc"
    10.     }
    11.   }
    12. ],
    13. "size": 2,
    14. "from": 2
    15. }

    高亮

    1. POST /book/_search
    2. {
    3. "query": {
    4.   "match": {
    5.     "name": "solr"
    6.   }
    7. },
    8. "highlight": {
    9.   "pre_tags": "<font color='pink'>",
    10.   "post_tags": "</font>",
    11.   "fields": [
    12.     {
    13.       "name": {}
    14.     }
    15.   ]
    16. },
    17. "size": 2,
    18. "from": 0
    19. }
    20. # 因为match的 字段是name,所以只能高亮name
    21. POST /book/_search
    22. {
    23. "query": {
    24.   "match": {
    25.     "name": "elasticsearch"
    26.   }
    27. },
    28. "highlight": {
    29.   "pre_tags": "<font color='pink'>",
    30.   "post_tags": "</font>",
    31.   "fields": [
    32.     {
    33.       "name": {}
    34.     },
    35.     {
    36.       "description": {}
    37.     }
    38.   ]
    39. }
    40. }
    41. # 匹配所有字段
    42. POST /book/_search
    43. {
    44. "query": {
    45.   "query_string": {
    46.     "query": "elasticsearch"
    47.   }
    48. },
    49. "highlight": {
    50.   "pre_tags": "<font color='pink'>",
    51.   "post_tags": "</font>",
    52.   "fields": [
    53.     {
    54.       "name": {}
    55.     },
    56.     {
    57.       "description": {}
    58.     }
    59.   ]
    60. }
    61. }

    文档批量操作

  • mget ```yaml

    mget 不同索引

    GET /_mget { “docs”: [ {

    1. "_index": "book",
    2. "_id": 1

    }, {

    1. "_index": "book",
    2. "_id": 2

    } ] }

mget同一个索引

GET /book/_mget { “docs”: [ { “_id”: 2 }, { “_id”: 3 } ] }

简化写法

POST /book/_search { “query”: { “ids”: { “values”: [ “1”, “4” ] } } }

  2. - bulk
  4. 批量操作增删改
  5. ```yaml
  6. POST /_bulk
  7. {"delete":{"_index":"book","_id":"1"}}
  8. {"create":{"_index":"book","_id":"5"}}
  9. {"name":"test14","price":100.99}
  10. {"update":{"_index":"book","_id":"2"}}
  11. {"doc":{"name":"test"}}
  • delete：删除一个文档，只要1个json串就可以了 删除的批量操作不需要请求体
  • create：相当于强制创建 PUT /index/type/id/_create
  • index：普通的put操作，可以是创建文档，也可以是全量替换文档
  • update：执行的是局部更新partial update操作

    格式：每个json不能换行。相邻json必须换行。 隔离：每个操作互不影响。操作失败的行会返回其失败信息。 一般建议是1000-5000个文档，大小建议是5-15MB，默认不能超过100M，可以在es的配置文件（ES的 confifig下的elasticsearch.yml）中配置。 http.max_content_length: 10mb