0


1.elasticsearch文档存储(保存|修改|删除)

【README】

0.本文部分内容(数据)总结自 es 开发文档, Document APIs | Elasticsearch Guide [7.2] | Elastic ;

1.本文的es版本是7.2.1;

2.elasticsearch 是一个数据存储,检索和分析引擎;本文介绍的是 es数据存储开发方式;

  • es是以文档为单位存储数据的,数据被序列化为json文档进行存储;

3.文档存储包括文档保存,修改,删除;(文档查询的开发方式比较复杂,单独新开一篇阐述)

  • 保存文档:使用 put 或 post请求;
  • 修改文档:使用put和post请求; 1. put请求更新文档是全量替换;2. post请求路径不带 _update 更新文档是 全量替换;3. post请求路径带 _update 更新文档是 部分更新;且 带了 _update 时,会比较新老数据是否相同,若相同则不执行更新,返回noop;
  • 删除文档:用delete请求(包括删除文档,删除索引);

【1】保存文档

保存文档可以用put请求或post请求;

  • Put 与 post请求保存文档的区别:put必须带文档id,post可以带 或者不带id;

【1.1】put请求保存文档

1) 在customer索引下的 external类型下保存2号文档,文档id设置为2

Put localhost:9200/customer/external/2

{
  "name":"zhangsan2"
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 1,
    "result": "created", // 这里表示创建
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 2,
    "_primary_term": 1
} 

2)再执行一次,result就是更新 updated 的了;

{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 2,
    "result": "updated", // 第2次执行就是更新 
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 3,
    "_primary_term": 1
}

【1.2】post请求保存文档

1) 在customer索引下的 external类型下保存文档(不带文档id);

Post localhost:9200/customer/external 

{
  "name":"zhangsan_post_1"
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "Qb1Gq4MBJc7j47GNcnTa", // 自动生成id,请求路径不需要设置id 
    "_version": 1,
    "result": "created",  // 新增数据
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 4,
    "_primary_term": 1
}

【Post请求小结】

  • 不带id就是永远新增,带了id的规则是无则新增,有则更新;

2)post请求保存文档,带文档id

Post  localhost:9200/customer/external/Qr1Jq4MBJc7j47GNp3RF 
{
  "name":"zhangsan_post_3"
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "Qr1Jq4MBJc7j47GNp3RF",
    "_version": 5,  // 版本号递增 
    "result": "updated",  // 更新 
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 9,
    "_primary_term": 1
}

【2】es文档简单查询

为了方便验证es文档保存,修改,删除的效果,本文介绍了一个简单查询。

【2.1】get请求带文档id查询文档

Get localhost:9200/customer/external/1
{}

{
    "_index": "customer", // 在哪个索引 
    "_type": "external", // 在哪个类型 
    "_id": "1",  // 记录id 
    "_version": 2,  // 表明数据被更新过,因为新建数据的版本号是1 
    "_seq_no": 1,  // 并发控制字段,每次更新就会+1;用来做乐观锁;
    "_primary_term": 1,  // 同上,主分片重新分配,如重启,就会变化 
    "found": true,  // 
    "_source": {  // 真正的内容 
        "name": "zhangsan"
    }
}

【2.2】seq_no 和 primary_term 做并发控制(乐观锁)

步骤1)查询id等于1的文档

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 4,
    "_seq_no": 11,
    "_primary_term": 1,
    "found": true,
    "_source": {
        "name": "zhangsan2_update_seq_11"
    }
}

步骤2)把 seq_no 和 primary_term 作为更新条件

客户端1:(修改成功

Post localhost:9200/customer/external/1?if_seq_no=11&if_primary_term=1
{
  "name":"zhangsan_post_update_c1"
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 5,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 12,
    "_primary_term": 1
}

客户端2(修改失败),因为 seq_no 因客户端1修改文档的事件而递增了,即不等于11 :

Post  localhost:9200/customer/external/1?if_seq_no=11&if_primary_term=1 
{
  "name":"zhangsan_post_update_c2"
}

{
    "error": {
        "root_cause": [
            {
                "type": "version_conflict_engine_exception",
                "reason": "[1]: version conflict, required seqNo [11], primary term [1]. current document has seqNo [12] and primary term [1]",
                "index_uuid": "IIq46lyCQMisw9_iPDheiw",
                "shard": "0",
                "index": "customer"
            }
        ],
        ......
    },
    "status": 409
}

【3】修改文档

1)post 或 put请求都可以修改文档;

  • post请求修改文档又分为 请求路径带 _update 和 不带_update;(post不带_update更新文档是全量替换,带_update 是部分更新)
  • put请求修改文档 时的请求路径不能带 _update ;(put更新文档是全量替换)

【3.1】post请求路径带_update更新文档

1)更新文档id为1的文档

Post localhost:9200/customer/external/1/_update
{
    "doc":{ // 注意 post请求路径带update的更新的请求体不同
        "name":"post_update_202210061135"
    }
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 6,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 13,
    "_primary_term": 1
}

2)再发送一次相同请求,且报文体也相同,则得到 noop ;

Post localhost:9200/customer/external/1/_update
{
    "doc":{
        "name":"post_update_202210061135"
    }
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 6,
    "result": "noop", // 没有做任何操作 
    "_shards": {
        "total": 0,
        "successful": 0,
        "failed": 0
    }
}

【3.2】post请求路径不带_update更新文档

1)更新id等于1的文档

Post localhost:9200/customer/external/1
{
  "name":"zhangsan_post_update_noupdate"
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 7,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 14,
    "_primary_term": 1
}

2)再发一次相同请求,相同报文体,如下:

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 8,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 15,
    "_primary_term": 1
}

【小结】

  • Post请求路径带update,会检查新老数据是否相同,若相同,则无需更新;即version, seq_no 都不会增加;
  • Post请求路径不带update,则不会检查新老数据是否相同,无论是否相同,都会更新;即 version 和 seq_no 都会增加;

【3.3】put请求路径带_update更新文档(API不存在)

put请求更新文档时,其请求路径不能带 _update ;

【3.4】put请求路径不带_update更新文档 (全量替换)

0)初始数据:id为2的文档;

post localhost:9200/customer/external/2
{}
{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 6,
    "_seq_no": 8,
    "_primary_term": 2,
    "found": true,
    "_source": {
        "name": "post_zhangsan_02",
        "addr": "post_addr_02"
    }
}

1)更新id为2的文档

Put localhost:9200/customer/external/2
{
    "doc":{
        "name":"put_update_202210071519"        
    }
}
// 更新后的结果 
{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 7,
    "_seq_no": 9,
    "_primary_term": 2,
    "found": true,
    "_source": {
        "doc": {
            "name": "put_update_202210071519"
        }
    }
}

【小结】

  • put请求路径不带_update可以更新成功; 不会检查新老数据是否相同,都会更新;
  • put请求更新文档是全量覆盖(虽然put请求仅更新了name字段,但结果是addr字段被删除了);

【3.5】post请求为索引新增字段

1)为id为1的文档新增 addr 字段;

Post localhost:9200/customer/external/1 
{
    "doc":{
        "name":"post_noupdate_202210061151"
        , "addr":"成都高新区01_post_新增属性"
    }
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 11,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 18,
    "_primary_term": 1
}

【3.6】put请求为索引新增字段

1)为id为1的文档新增 addr 字段;

Put localhost:9200/customer/external/1  
{
    "doc":{
        "name":"post_noupdate_202210061154"
        , "addr":"成都高新区02_put_新增属性"
    }
}

{
    "_index": "customer",
    "_type": "external",
    "_id": "1",
    "_version": 12,
    "result": "updated",
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 19,
    "_primary_term": 1
}

【3.7】post请求更新文档时带_update与不带_update的区别

1)post请求路径带_update,仅部分更新(我想大部分场景应该选择带 _update)

  • 如文档有字段1,字段2;
  • 通过post带update的更新,可以仅更新字段1 或 字段2;

2)post请求路径不带update,是全量覆盖

  • 如文档有字段1,字段2;
  • 通过post不带update的更新,是全量覆盖;

3)例子(post请求带与不带_update 更新文档的区别):

3.0)初始数据

get localhost:9200/customer/external/2 
{}
{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 6,
    "_seq_no": 24,
    "_primary_term": 1,
    "found": true,
    "_source": {
        "name": "zhangsan2_insert",
        "addr": "post_带update_成都04"
    }
}

3.1)Post请求路径带update的部分更新

Post localhost:9200/customer/external/2/_update 
{
    "doc":{
        "addr":"post_带update_成都05"
    }
}

// 更新后的结果(部分更新): 
{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 7,
    "_seq_no": 25,
    "_primary_term": 1,
    "found": true,
    "_source": {
        "name": "zhangsan2_insert",
        "addr": "post_带update_成都05"
    }
}

3.2)post请求路径不带_update 的**全量替换 **;

Post localhost:9200/customer/external/2
{    
    "addr":"post_不带update_成都06"    
}

// 更新后的结果(全量替换):
{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 8,
    "_seq_no": 26,
    "_primary_term": 1,
    "found": true,
    "_source": {
        "addr": "post_不带update_成都06" // (显然,这里是全量替换)
    }
}

【4】删除数据(删除文档和索引)

【4.1】删除文档

1)删除文档id为2的文档

Delete localhost:9200/customer/external/2

{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "_version": 9,
    "result": "deleted", // 删除成功 
    "_shards": {
        "total": 2,
        "successful": 1,
        "failed": 0
    },
    "_seq_no": 27,
    "_primary_term": 1
}

2)删除完成后,再次查询id为2的文档,如下:

Get localhost:9200/customer/external/2 
{}

{
    "_index": "customer",
    "_type": "external",
    "_id": "2",
    "found": false  // 显然没有找到 
}

【4.2】删除索引

注意: es中没有提供删除类型的api ;

1)删除customer索引;

Delete localhost:9200/customer

{
    "acknowledged": true
}

2)删除索引文档后,,再次查询(报索引不存在);

Get localhost:9200/customer/external/2
{}
{
    "error": {
        "root_cause": [
            {
                "type": "index_not_found_exception",
                "reason": "no such index [customer]",
                "resource.type": "index_expression",
                "resource.id": "customer",
                "index_uuid": "_na_",
                "index": "customer"
            }
        ],
       ...... 
    },
    "status": 404
}

本文转载自: https://blog.csdn.net/PacosonSWJTU/article/details/127193918
版权归原作者 PacosonSWJTU 所有, 如有侵权,请联系我们删除。

“1.elasticsearch文档存储(保存|修改|删除)”的评论:

还没有评论