Elasticsearch 7.x：3、文档管理

官方文档：
https://www.elastic.co/guide/en/elasticsearch/reference/current/docs.html

3.1 添加文档

（1）指定文档ID

PUT blog/_doc/1
{
  "title":"1、VMware Workstation虚拟机软件安装图解",
  "author":"chengyuqiang",
  "content":"1、VMware Workstation虚拟机软件安装图解...",
  "url":"http://x.co/6nc81"
}

Elasticsearch服务会返回一个JSON格式的响应。

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

响应结果说明：

_index：文档所在的索引名
_type：文档所在的类型名
_id：文档ID
_version：文档的版本
result：created已经创建
_shards： _shards表示索引操作的复制过程的信息。
- total：指示应在其上执行索引操作的分片副本（主分片和副本分片）的数量。
- successful：表示索引操作成功的分片副本数。
- failed：在副本分片上索引操作失败的情况下包含复制相关错误。

（2）不指定文档ID
添加文档时可以不指定文档id，则文档id是自动生成的字符串。注意，需要使用POST方法，而不是PUT方法。

POST blog/_doc
{
  "title":"2、Linux服务器安装图解",
  "author":"chengyuqiang",
  "content":"2、Linux服务器安装图解解...",
  "url":"http://x.co/6nc82"
}

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "5P2-O2gBNSQY7o-KMw2P",
  "_version" : 1,
  "result" : "created",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 1,
  "_primary_term" : 1
}

3.2 获取文档

（1）通过文档id获取指定的文档

GET blog/_doc/1

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 1,
  "found" : true,
  "_source" : {
    "title" : "1、VMware Workstation虚拟机软件安装图解",
    "author" : "chengyuqiang",
    "content" : "1、VMware Workstation虚拟机软件安装图解...",
    "url" : "http://x.co/6nc81"
  }
}

响应结果说明：

found值为true，表明查询到该文档
_source字段是文档的内容

（2）文档不存在的情况

GET blog/_doc/2

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "2",
  "found" : false
}

found字段值为false表明查询的文档不存在。
（3）判定文档是否存在

HEAD blog/_doc/1

200 - OK

3.3 更新文档

更改id为1的文档，删除了author，修改content字段。

PUT blog/_doc/1
{
  "title":"1、VMware Workstation虚拟机软件安装图解",
  "content":"下载得到VMware-workstation-full-15.0.2-10952284.exe可执行文件...",
  "url":"http://x.co/6nc81"
}

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 2,
  "result" : "updated",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 1,
  "_primary_term" : 1
}

_version更新为2

查看该文档

GET blog/_doc/1

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 2,
  "found" : true,
  "_source" : {
    "title" : "1、VMware Workstation虚拟机软件安装图解",
    "content" : "下载得到VMware-workstation-full-15.0.2-10952284.exe可执行文件...",
    "url" : "http://x.co/6nc81"
  }
}

（2）添加文档时，防止覆盖已存在的文档，可以通过_create加以限制。

PUT blog/_doc/1/_create
{
  "title":"1、VMware Workstation虚拟机软件安装图解",
  "content":"下载得到VMware-workstation-full-15.0.2-10952284.exe可执行文件...",
  "url":"http://x.co/6nc81"
}

该文档已经存在，添加失败。

{
  "error": {
    "root_cause": [
      {
        "type": "version_conflict_engine_exception",
        "reason": "[_doc][1]: version conflict, document already exists (current version [2])",
        "index_uuid": "GqC2fSqPS06GRfTLmh1TLg",
        "shard": "1",
        "index": "blog"
      }
    ],
    "type": "version_conflict_engine_exception",
    "reason": "[_doc][1]: version conflict, document already exists (current version [2])",
    "index_uuid": "GqC2fSqPS06GRfTLmh1TLg",
    "shard": "1",
    "index": "blog"
  },
  "status": 409
}

（3）更新文档的字段
通过脚本更新制定字段，其中ctx是脚本语言中的一个执行对象，先获取_source，再修改content字段

POST blog/_doc/1/_update
{
  "script": {
    "source": "ctx._source.content=\"从官网下载VMware-workstation，双击可执行文件进行安装...\""  
  } 
}

响应结果如下：

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 3,
  "result" : "updated",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 2,
  "_primary_term" : 1
}

再次获取文档GET blog/_doc/1，响应结果如下

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 3,
  "found" : true,
  "_source" : {
    "title" : "1、VMware Workstation虚拟机软件安装图解",
    "content" : "从官网下载VMware-workstation，双击可执行文件进行安装...",
    "url" : "http://x.co/6nc81"
  }
}

（4）添加字段

POST blog/_doc/1/_update
{
  "script": {
    "source": "ctx._source.author=\"chengyuqiang\""  
  } 
}

再次获取文档GET blog/_doc/1，响应结果如下

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 4,
  "found" : true,
  "_source" : {
    "title" : "1、VMware Workstation虚拟机软件安装图解",
    "content" : "从官网下载VMware-workstation，双击可执行文件进行安装...",
    "url" : "http://x.co/6nc81",
    "author" : "chengyuqiang"
  }
}

（5）删除字段

POST blog/_doc/1/_update
{
  "script": {
    "source": "ctx._source.remove(\"url\")"  
  } 
}

再次获取文档GET blog/_doc/1，响应结果如下

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 5,
  "found" : true,
  "_source" : {
    "title" : "1、VMware Workstation虚拟机软件安装图解",
    "content" : "从官网下载VMware-workstation，双击可执行文件进行安装...",
    "author" : "chengyuqiang"
  }
}

3.4 删除文档

DELETE blog/_doc/1

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 6,
  "result" : "deleted",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 6,
  "_primary_term" : 1
}

再次判定该文档是否存在，执行HEAD blog/_doc/1，响应结果404 - Not Found

3.5 批量操作

如果文档数量非常庞大，商业运维中都是海量数据，一个一个操作文档显然不合实际。幸运的是ElasticSearch提供了文档的批量操作机制。我们已经知道mget允许一次性检索多个文档，ElasticSearch提供了Bulk API，可以执行批量索引、批量删除、批量更新等操作，也就是说Bulk API允许使用在单个步骤中进行多次 create 、 index 、 update 或 delete 请求。

bulk 与其他的请求体格式稍有不同，bulk请求格式如下：

{ action: { metadata }}\n
{ request body        }\n
{ action: { metadata }}\n
{ request body        }\n
...

这种格式类似一个有效的单行 JSON 文档流，它通过换行符(\n)连接到一起。注意两个要点：

每行一定要以换行符(\n)结尾，包括最后一行。这些换行符被用作一个标记，可以有效分隔行。
这些行不能包含未转义的换行符，因为他们将会对解析造成干扰。这意味着这个 JSON 不能使用 pretty 参数打印。
action/metadata 行指定哪一个文档做什么操作。metadata 应该指定被索引、创建、更新或者删除的文档的 _index 、 _type 和 _id 。
request body 行由文档的 _source 本身组成–文档包含的字段和值。它是 index 和 create 操作所必需的。
（1）批量导入

POST /_bulk
{ "create": { "_index": "blog", "_type": "_doc", "_id": "1" }}
{ "title": "1、VMware Workstation虚拟机软件安装图解" ,"author":"chengyuqiang","content":"官网下载VMware-workstation，双击可执行文件进行安装" , "url":"http://x.co/6nc81" }
{ "create": { "_index": "blog", "_type": "_doc", "_id": "2" }}
{ "title":  "2、Linux服务器安装图解" ,"author":  "chengyuqiang" ,"content": "VMware模拟Linux服务器安装图解" , "url": "http://x.co/6nc82" }
{ "create": { "_index": "blog", "_type": "_doc", "_id": "3" }}
{ "title":  "3、Xshell 6 个人版安装与远程操作连接服务器" , "author": "chengyuqiang" ,"content": "Xshell 6 个人版安装与远程操作连接服务器..." , "url": "http://x.co/6nc84" }

这个 Elasticsearch 响应包含 items 数组，这个数组的内容是以请求的顺序列出来的每个请求的结果。

{
  "took" : 132,
  "errors" : false,
  "items" : [
    {
      "create" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "1",
        "_version" : 7,
        "result" : "created",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 7,
        "_primary_term" : 1,
        "status" : 201
      }
    },
    {
      "create" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "2",
        "_version" : 1,
        "result" : "created",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 8,
        "_primary_term" : 1,
        "status" : 201
      }
    },
    {
      "create" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "3",
        "_version" : 1,
        "result" : "created",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 0,
        "_primary_term" : 1,
        "status" : 201
      }
    }
  ]
}

（2）批量操作，包括删除、更新、新增

POST /_bulk
{ "delete": { "_index": "blog", "_type": "_doc", "_id": "1" }}
{ "update": { "_index": "blog", "_type": "_doc", "_id": "3", "retry_on_conflict" : 3} }
{ "doc" : {"title" : "Xshell教程"} }
{ "index": { "_index": "blog", "_type": "_doc", "_id": "4" }}
{ "title": "4、CentOS 7.x基本设置" ,"author":"chengyuqiang","content":"CentOS 7.x基本设置","url":"http://x.co/6nc85" }
{ "create": { "_index": "blog", "_type": "_doc", "_id": "5" }}
{ "title": "5、图解Linux下JDK安装与环境变量配置","author":"chengyuqiang" ,"content": "图解JDK安装配置" , "url": "http://x.co/6nc86" }

在7.0版本中，retry_on_conflict参数取代了之前的_retry_on_conflict

{
  "took" : 125,
  "errors" : false,
  "items" : [
    {
      "delete" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "1",
        "_version" : 2,
        "result" : "deleted",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 3,
        "_primary_term" : 1,
        "status" : 200
      }
    },
    {
      "update" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "3",
        "_version" : 2,
        "result" : "updated",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 4,
        "_primary_term" : 1,
        "status" : 200
      }
    },
    {
      "index" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "4",
        "_version" : 1,
        "result" : "created",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 1,
        "_primary_term" : 1,
        "status" : 201
      }
    },
    {
      "create" : {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "5",
        "_version" : 1,
        "result" : "created",
        "_shards" : {
          "total" : 2,
          "successful" : 1,
          "failed" : 0
        },
        "_seq_no" : 5,
        "_primary_term" : 1,
        "status" : 201
      }
    }
  ]
}

3.6 批量获取

GET blog/_doc/_mget
{
    "ids" : ["1", "2","3"]
}

id为1的文档已经删除，所以没有搜索到

{
  "docs" : [
    {
      "_index" : "blog",
      "_type" : "_doc",
      "_id" : "1",
      "found" : false
    },
    {
      "_index" : "blog",
      "_type" : "_doc",
      "_id" : "2",
      "_version" : 1,
      "found" : true,
      "_source" : {
        "title" : "2、Linux服务器安装图解",
        "author" : "chengyuqiang",
        "content" : "VMware模拟Linux服务器安装图解",
        "url" : "http://x.co/6nc82"
      }
    },
    {
      "_index" : "blog",
      "_type" : "_doc",
      "_id" : "3",
      "_version" : 2,
      "found" : true,
      "_source" : {
        "title" : "Xshell教程",
        "author" : "chengyuqiang",
        "content" : "Xshell 6 个人版安装与远程操作连接服务器...",
        "url" : "http://x.co/6nc84"
      }
    }
  ]
}

3.7 简单搜索

这里介绍一下简单的文档搜索操作，后面章节会详细介绍。
（1）词项查询，也称term查询

GET blog/_search
{
  "query": {
    "term": {
      "title": "centos"
    }
  }
}

{
  "took" : 1,
  "timed_out" : false,
  "_shards" : {
    "total" : 2,
    "successful" : 2,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 0.71023846,
    "hits" : [
      {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "4",
        "_score" : 0.71023846,
        "_source" : {
          "title" : "4、CentOS 7.x基本设置",
          "author" : "chengyuqiang",
          "content" : "CentOS 7.x基本设置",
          "url" : "http://x.co/6nc85"
        }
      }
    ]
  }
}

GET blog/_search
{
  "query": {
    "term": {
      "title": "远程"
    }
  }
}

{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 2,
    "successful" : 2,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 0,
      "relation" : "eq"
    },
    "max_score" : null,
    "hits" : [ ]
  }
}

GET blog/_search
{
  "query": {
    "term": {
      "title": "程"
    }
  }
}

{
  "took" : 2,
  "timed_out" : false,
  "_shards" : {
    "total" : 2,
    "successful" : 2,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 1.3486402,
    "hits" : [
      {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "3",
        "_score" : 1.3486402,
        "_source" : {
          "title" : "Xshell教程",
          "author" : "chengyuqiang",
          "content" : "Xshell 6 个人版安装与远程操作连接服务器...",
          "url" : "http://x.co/6nc84"
        }
      }
    ]
  }
}

（2）匹配查询，也称match查询
与term精确查询不同，对于match查询，只要被查询字段中存在任何一个词项被匹配，就会搜索到该文档。

GET blog/_search
{
  "query": {
    "match": {
      "title": {
        "query": "远程"
      }
    }
  }
}

{
  "took" : 9,
  "timed_out" : false,
  "_shards" : {
    "total" : 2,
    "successful" : 2,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 1.3486402,
    "hits" : [
      {
        "_index" : "blog",
        "_type" : "_doc",
        "_id" : "3",
        "_score" : 1.3486402,
        "_source" : {
          "title" : "Xshell教程",
          "author" : "chengyuqiang",
          "content" : "Xshell 6 个人版安装与远程操作连接服务器...",
          "url" : "http://x.co/6nc84"
        }
      }
    ]
  }
}

3.8 路由机制

当你索引（动词，对该文档建立倒排索引）一个文档，它被存储到master节点上的一个主分片上。

Elasticsearch是如何知道文档属于哪个分片的呢？当你创建一个新文档，它是如何知道是应该存储在分片1还是分片2上的呢？
解答这个问题，我们需要了解Elasticsearch的路由机制。
简单地说，Elasticsearch将具有相关Hash值的文档存放到同一个主分片中，分片位置计算算法如下：

shard = hash(routing) % number_of_primary_shards

算法说明：

routing值是一个字符串，它默认是文档_id，也可以自定义。这个routing字符串通过哈希函数生成一个数字，然后除以主切片的数量得到一个余数(remainder)，余数的范围是[0 , number_of_primary_shards-1]，这个数字就是特定文档所在的分片。
之前我们介绍过，创建索引时需要指定主分片数量，该不能修改。这是因为如果主分片的数量在未来改变了，所有先前的路由值就失效了，文档也就永远找不到了。
该算法基本可以保证所有文档在所有分片上平均分布，不会导致数据分布不均（数据倾斜）的情况。
默认情况下，routing值是文档的_id。我们创建文档时可以指定id的值；如果不指定id时，Elasticsearch将随机生成文档的_id值。这将导致在查询文档时，Elasticsearch不能确定文档的位置，需要将请求广播到所有的分片节点上。

假设我们有一个10个分片的索引。当一个请求在集群上执行时基本过程如下：

这个搜索的请求会被发送到一个节点。
接收到这个请求的节点，将这个查询广播到这个索引的每个分片上（可能是主分片，也可能是复制分片）。
每个分片执行这个搜索查询并返回结果。
结果在通道节点上合并、排序并返回给用户。

了解Elasticsearch的路由机制后，我们可以在创建某一类文档时指定文档的路由值，这样ElasticSearch就知道在处理这一类文档时，如何定位到正确的分片。比如，把某一特定类型的书籍存储到特定的分片上去，这样在搜索这一类书籍的时候就可以避免搜索其它的分片，也就避免了多个分片搜索结果的合并。路由机制向 Elasticsearch提供一种信息来决定哪些分片用于存储和查询。同一个路由值将映射到同一个分片。这基本上就是在说：“通过使用用户提供的路由值，就可以做到定向存储，定向搜索。

所有的文档API（GET、INDEX、DELETE、BULK、UPDATE、MGET）都接收一个routing参数，它用来自定义文档到分片的映射。添加routing参数形式与URL参数形式相同url?参数名=参数值。

PUT blog/_doc/1?routing=haron
{
  "title":"1、VMware安装",
  "author":"hadron",
  "content":"VMware Workstation虚拟机软件安装图解...",
  "url":"http://x.co/6nc81"
}

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 1,
  "result" : "created",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 12,
  "_primary_term" : 1
}

GET blog/_doc/1?routing=hardon

{
  "_index" : "blog",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 1,
  "_routing" : "hardon",
  "found" : true,
  "_source" : {
    "title" : "1、VMware安装",
    "author" : "hadron",
    "content" : "VMware Workstation虚拟机软件安装图解...",
    "url" : "http://x.co/6nc81"
  }
}

注意：自定义routing值可以造成数据分布不均的情况。例如用户hadron的文档非常多，有数十万个，而其他大多数用户的文档只有数个到数十个，这样将导致hadron所在的分片较大。

3.9 版本控制

参考文档
https://www.elastic.co/guide/en/elasticsearch/guide/2.x/version-control.html
https://www.elastic.co/guide/en/elasticsearch/guide/2.x/optimistic-concurrency-control.html

https://elasticsearch.cn/book/elasticsearch_definitive_guide_2.x/version-control.html
https://elasticsearch.cn/book/elasticsearch_definitive_guide_2.x/optimistic-concurrency-control.html

PUT website
{
    "settings" : {
        "index" : {
            "number_of_shards" : 1,
            "number_of_replicas" : 1
        }
    }
}

PUT /website/_doc/1/_create
{
  "title": "My first blog entry",
  "text":  "Just trying this out..."
}

GET website/_doc/1

{
  "_index" : "website",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 1,
  "found" : true,
  "_source" : {
    "title" : "My first blog entry",
    "text" : "Just trying this out..."
  }
}

PUT website/_doc/1?version=1
{
  "title": "My first blog entry",
  "text":  "Starting to get the hang of this..."
}

{
  "_index" : "website",
  "_type" : "_doc",
  "_id" : "1",
  "_version" : 2,
  "result" : "updated",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 1,
  "_primary_term" : 1
}

{
  "error": {
    "root_cause": [
      {
        "type": "version_conflict_engine_exception",
        "reason": "[_doc][1]: version conflict, current version [2] is different than the one provided [1]",
        "index_uuid": "hFH2LLumS9a-FBb-SY1LrA",
        "shard": "0",
        "index": "website"
      }
    ],
    "type": "version_conflict_engine_exception",
    "reason": "[_doc][1]: version conflict, current version [2] is different than the one provided [1]",
    "index_uuid": "hFH2LLumS9a-FBb-SY1LrA",
    "shard": "0",
    "index": "website"
  },
  "status": 409
}

例如，要创建一个新的具有外部版本号 5 的博客文章，我们可以按以下方法进行：

PUT /website/_doc/2?version=5&version_type=external
{
  "title": "My first external blog entry",
  "text":  "Starting to get the hang of this..."
}

在响应中，我们能看到当前的 _version 版本号是 5 ：

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

现在我们更新这个文档，指定一个新的 version 号是 10 ：

PUT /website/_doc/2?version=10&version_type=external
{
  "title": "My first external blog entry",
  "text":  "This is a piece of cake..."
}

请求成功并将当前 _version 设为 10 ：

{
  "_index" : "website",
  "_type" : "_doc",
  "_id" : "2",
  "_version" : 10,
  "result" : "updated",
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "_seq_no" : 3,
  "_primary_term" : 1
}

如果你要重新运行此请求时，它将会失败，并返回像我们之前看到的同样的冲突错误，因为指定的外部版本号不大于 Elasticsearch 的当前版本号。

{
  "error": {
    "root_cause": [
      {
        "type": "version_conflict_engine_exception",
        "reason": "[_doc][2]: version conflict, current version [10] is higher or equal to the one provided [10]",
        "index_uuid": "5616aEUkQ7yvQIYUDyLudg",
        "shard": "0",
        "index": "website"
      }
    ],
    "type": "version_conflict_engine_exception",
    "reason": "[_doc][2]: version conflict, current version [10] is higher or equal to the one provided [10]",
    "index_uuid": "5616aEUkQ7yvQIYUDyLudg",
    "shard": "0",
    "index": "website"
  },
  "status": 409
}

3.10 ?refresh

Index，Update，Delete和Bulk API支持设置刷新，以控制此请求所做的更改何时对搜索可见。这些是允许的值：

空字符串或true
Refresh the relevant primary and replica shards (not the whole index) immediately after the operation occurs, so that the updated document appears in search results immediately. This should ONLY be done after careful thought and verification that it does not lead to poor performance, both from an indexing and a search standpoint.
在操作发生后立即刷新相关的主分片和副本分片（而不是整个索引），以便更新的文档立即显示在搜索结果中。只有在从索引和搜索角度进行仔细考虑并验证它不会导致性能不佳之后，才能进行此操作。
wait_for
Wait for the changes made by the request to be made visible by a refresh before replying. This doesn’t force an immediate refresh, rather, it waits for a refresh to happen. Elasticsearch automatically refreshes shards that have changed every index.refresh_interval which defaults to one second. That setting is dynamic. Calling the Refresh API or setting refresh to true on any of the APIs that support it will also cause a refresh, in turn causing already running requests with refresh=wait_for to return.
在回复之前，请等待刷新请求所做的更改。这不会强制立即刷新，而是等待刷新发生。 Elasticsearch会自动刷新已更改每个index.refresh_interval的分片，默认为一秒。那个设置是动态的。在任何支持它的API上调用Refresh API或将refresh设置为true也会导致刷新，从而导致已经运行的请求返回refresh = wait_for。
false（默认值）
Take no refresh related actions. The changes made by this request will be made visible at some point after the request returns.
不采取与刷新相关的操作。此请求所做的更改将在请求返回后的某个时间点显示。

（1）立即刷新，文档可见
These will create a document and immediately refresh the index so it is visible:
这些将创建一个文档并立即刷新索引，使其可见：

DELETE test
PUT test/_doc/1?refresh
{"message": "测试文档1"}
PUT test/_doc/2?refresh=true
{"message": "测试文档2"}

（2）不刷新
These will create a document without doing anything to make it visible for search:
这些将创建一个文档而不做任何使搜索可见的内容：

PUT test/_doc/3
{"message": "测试文档3"}
PUT test/_doc/4?refresh=false
{"message": "测试文档4"}

（3）等待刷新可见
This will create a document and wait for it to become visible for search:
这将创建一个文档并等待它对搜索可见：

PUT test/_doc/5?refresh=wait_for
{"message": "测试文档5"}