搜索引擎之Elastic Search的高级使用-51CTO.COM

es之索引别名的使用

别名有什么⽤

在开发中，随着业务需求的迭代，较⽼的业务逻辑就要⾯临更新甚⾄是重构，⽽对于es来说，为了适应新的业务逻辑，可能就要对原有的索引做⼀些修改，⽐如对某些字段做调整，甚⾄是重建索引。⽽做这些操作的时候，可能会对业务造成影响，甚⾄是停机调整等问题。由此，es提供了索引别名来解决这些问题。索引别名就像⼀个快捷⽅式或是软连接，可以指向⼀个或多个索引，也可以给任意⼀个需要索引名的API来使⽤。别名的应⽤为程序提供了极⼤地灵活性。

查询别名

GET /nba/_alias
GET /_alias

新增别名

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "nba",
        "alias": "nba_v1.0"
      }
    }
  ]
}

PUT /nba/_alias/nba_v1.1

删除别名

POST /_aliases
{
  "actions": [
    {
      "remove": {
        "index": "nba",
        "alias": "nba_v1.0"
      }
    }
  ]
}

DELETE /nba/_alias/nba_v1.1

重命名

POST /_aliases
{
  "actions": [
    {
      "remove": {
        "index": "nba",
        "alias": "nba_v1.0"
      }
    },
    {
      "add": {
        "index": "nba",
        "alias": "nba_v2.0"
      }
    }
  ]
}

为多个索引指定⼀个别名

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "nba",
        "alias": "national_player"
      }
    },
    {
      "add": {
        "index": "wnba",
        "alias": "national_player"
      }
    }
  ]
}

为同个索引指定多个别名

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "nba",
        "alias": "nba_v2.1"
      }
    },
    {
      "add": {
        "index": "nba",
        "alias": "nba_v2.2"
      }
    }
  ]
}

通过别名读索引

当别名指定了⼀个索引，则查出⼀个索引。

GET /nba_v2.1

当别名指定了多个索引，则查出多个索引。

GET /national_player

通过别名写索引

当别名指定了⼀个索引，则可以做写的操作。

POST /nba_v2.1/_doc/566
{
 "countryEn": "Croatia",
 "teamName": "快船",
 "birthDay": 858661200000,
 "country": "克罗地亚",
 "teamCityEn": "LA",
 "code": "ivica_zubac",
 "displayAffiliation": "Croatia",
 "displayName": "伊维察 祖巴茨哥哥",
 "schoolType": "",
 "teamConference": "⻄部",
 "teamConferenceEn": "Western",
 "weight": "108.9 公⽄",
 "teamCity": "洛杉矶",
 "playYear": 3,
 "jerseyNo": "40",
 "teamNameEn": "Clippers",
 "draft": 2016,
 "displayNameEn": "Ivica Zubac",
 "heightValue": 2.16,
 "birthDayStr": "1997-03-18",
 "position": "中锋",
 "age": 22,
 "playerId": "1627826"
}

当别名指定了多个索引，可以指定写某个索引。

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "nba",
        "alias": "national_player",
        "is_write_index": true
      }
    },
    {
      "add": {
        "index": "wnba",
        "alias": "national_player"
      }
    }
  ]
}

POST /national_player/_doc/566
{
 "countryEn": "Croatia",
 "teamName": "快船",
 "birthDay": 858661200000,
 "country": "克罗地亚",
 "teamCityEn": "LA",
 "code": "ivica_zubac",
 "displayAffiliation": "Croatia",
 "displayName": "伊维察 祖巴茨妹妹",
 "schoolType": "",
 "teamConference": "⻄部",
 "teamConferenceEn": "Western",
 "weight": "108.9 公⽄",
 "teamCity": "洛杉矶",
 "playYear": 3,
 "jerseyNo": "40",
 "teamNameEn": "Clippers",
 "draft": 2016,
 "displayNameEn": "Ivica Zubac",
 "heightValue": 2.16,
 "birthDayStr": "1997-03-18",
 "position": "中锋",
 "age": 22,
 "playerId": "1627826"
}

es之如何重建索引

背景

Elasticsearch是⼀个实时的分布式搜索引擎，为⽤户提供搜索服务，当我们决定存储某种数据时，在创建索引的时候需要将数据结构完整确定下来，于此同时索引的设定和很多固定配置将⽤不能改变。当需要改变数据结构时，就需要重新建⽴索引，为此，Elastic团队提供了很多辅助⼯具帮助开发⼈员进⾏重建索引。

步骤

nba取⼀个别名nba_latest, nba_latest作为对外使⽤。
新增⼀个索引nba_20220101，结构复制于nba索引，根据业务要求修改字段。
将nba数据同步到nba_20220101。
给nba_20220101添加别名nba_latest，删除nba别名nba_latest。
删除nba索引。

我们对外提供访问nba索引时使⽤的是nba_latest别名

1.新增⼀个索引(⽐如修改字段类型，jerseyNo改成keyword类型)

PUT /nba_20220101
{
  "mappings": {
    "properties": {
      "age": {
        "type": "integer"
      },
      "birthDay": {
        "type": "date"
      },
      "birthDayStr": {
        "type": "keyword"
      },
      "code": {
        "type": "text"
      },
      "country": {
        "type": "keyword"
      },
      "countryEn": {
        "type": "keyword"
      },
      "displayAffiliation": {
        "type": "text"
      },
      "displayName": {
        "type": "text"
      },
      "displayNameEn": {
        "type": "text"
      },
      "draft": {
        "type": "long"
      },
      "heightValue": {
        "type": "float"
      },
      "jerseyNo": {
        "type": "keyword"
      },
      "playYear": {
        "type": "long"
      },
      "playerId": {
        "type": "keyword"
      },
      "position": {
        "type": "text"
      },
      "schoolType": {
        "type": "text"
      },
      "teamCity": {
        "type": "text"
      },
      "teamCityEn": {
        "type": "text"
      },
      "teamConference": {
        "type": "keyword"
      },
      "teamConferenceEn": {
        "type": "keyword"
      },
      "teamName": {
        "type": "keyword"
      },
      "teamNameEn": {
        "type": "keyword"
      },
      "weight": {
        "type": "text"
      }
    }
  }
}

2.将旧索引数据copy到新索引

同步等待，接⼝将会在 reindex 结束后返回。

POST /_reindex
{
  "source": {
    "index": "nba"
  },
  "dest": {
    "index": "nba_20220101"
  }
}

异步执⾏，如果 reindex 时间过⻓，建议加上 wait_for_completion=false 的参数条件，这样 reindex 将直接返回 taskId

POST /_reindex?wait_for_completion=false
{
  "source": {
    "index": "nba"
  },
  "dest": {
    "index": "nba_20220101"
  }
}

3.替换别名

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "nba_20220101",
        "alias": "nba_latest"
      }
    },
    {
      "remove": {
        "index": "nba",
        "alias": "nba_latest"
      }
    }
  ]
}

4.删除旧索引

DELETE /nba

5.通过别名访问新索引

POST /nba_latest/_search
{
  "query": {
    "match": {
      "displayNameEn": "james"
    }
  }
}

es之refresh操作

理想的搜索：

新的数据⼀添加到索引中⽴⻢就能搜索到，但是真实情况不是这样的。我们使⽤链式命令请求，先添加⼀个⽂档，再⽴刻搜索。

curl -X PUT localhost:9200/star/_doc/888 -H 'Content-Type:
application/json' -d '{ "displayName": "蔡徐坤" }'
curl -X GET localhost:9200/star/_doc/_search?pretty

强制刷新

curl -X PUT localhost:9200/star/_doc/666?refresh -H 'Content-Type:
application/json' -d '{ "displayName": "杨超越" }'
curl -X GET localhost:9200/star/_doc/_search?pretty

修改默认更新时间(默认时间是1s)

PUT /star/_settings
{
  "index": {
    "refresh_interval": "5s"
  }
}

将refresh关闭

PUT /star/_settings
{
  "index": {
    "refresh_interval": "-1"
  }
}

es之高亮查询

前⾔

如果返回的结果集中很多符合条件的结果，那怎么能⼀眼就能看到我们想要的那个结果呢？⽐如下⾯⽹站所示的那样，我们搜索⼩d课堂，在结果集中，将所有⼩d课堂⾼亮显示？

高亮查询

POST /nba_latest/_search
{
  "query": {
    "match": {
      "displayNameEn": "james"
    }
  },
  "highlight": {
    "fields": {
      "displayNameEn": {}
    }
  }
}

自定义高亮查询

POST /nba_latest/_search
{
  "query": {
    "match": {
      "displayNameEn": "james"
    }
  },
  "highlight": {
    "fields": {
      "displayNameEn": {
        "pre_tags": [
          "<h1>"
        ],
        "post_tags": [
          "</h1>"
        ]
      }
    }
  }
}

es之查询建议

查询建议是什么

查询建议，是为了给⽤户提供更好的搜索体验。包括：词条检查，⾃动补全。
词条检查

⾃动补全

Suggester

Term suggester
Phrase suggester
Completion suggester

字段

Term suggester

term 词条建议器，对给输⼊的⽂本进⾏分词，为每个分词提供词项建议。

POST /nba_latest/_search
{
  "suggest": {
    "my-suggestion": {
      "text": "jamse hardne",
      "term": {
        "suggest_mode": "missing",
        "field": "displayNameEn"
      }
    }
  }
}

Phrase suggester

phrase 短语建议，在term的基础上，会考量多个term之间的关系，⽐如是否同时出现在索引的原⽂⾥，相邻程度，以及词频等。

POST /nba_latest/_search
{
  "suggest": {
    "my-suggestion": {
      "text": "jamse harden",
      "phrase": {
        "field": "displayNameEn"
      }
    }
  }
}

Completion suggester

Completion 完成建议。

POST /nba_latest/_search
{
  "suggest": {
    "my-suggestion": {
      "text": "Miam",
      "completion": {
        "field": "teamCityEn"
      }
    }
  }
}