四十二 Python分布式爬虫打造搜索引擎Scrapy精讲—elasticsearch(搜索引擎)的mget和bulk批量操作

注意：前面讲到的各种操作都是一次http请求操作一条数据，如果想要操作多条数据就会产生多次请求，所以就有了mget和bulk批量操作，mget和bulk批量操作是一次请求可以操作多条数据

1、mget批量操作（查询）

批量操作(同一个索引同一个表里的批量查询)

说明：

#mget批量操作(同一个表里的批量查询)
GET _mget
{
  "docs":[
    {
      "_index":"索引名称",
      "_type":"表名称",
      "_id":id号
    },
    {
      "_index":"索引名称",
      "_type":"表名称",
      "_id":id号
    }
    ]
}

代码：

#mget批量操作(同一个表里的批量查询)
GET _mget
{
  "docs":[
    {
      "_index":"jobbole",
      "_type":"job",
      "_id":1
    },
    {
      "_index":"jobbole",
      "_type":"job",
      "_id":2
    }
    ]
}

 

批量操作(同一个索引同一个表里的不同id批量查询)

#批量操作(同一个索引同一个表里的不同id批量查询)
GET jobbole/job/_mget
{
  "ids":[1,2]
}

批量操作(同一个索引不同表里的批量查询)

#mget批量操作(同一个索引不同表里的批量查询)
GET jobbole/_mget
{
  "docs":[
    {
      "_type":"job",
      "_id":1
    },
    {
      "_type":"job2",
      "_id":1
    }
    ]
}

批量操作(不同索引不同表里的批量查询，相当于数据库的组合查询)

#mget批量操作(不同索引不同表里的批量查询)
GET _mget
{
  "docs":[
    {
      "_index":"jobbole",
      "_type":"job",
      "_id":1
    },
    {
      "_index":"yuxiou",
      "_type":"biao",
      "_id":2
    }
    ]
}

2、bulk批量操作(增删改)
批量导入可以合并多个操作，比如index,delete,update,create等等。也可以帮助从一个索引导入到另一个索引

bulk批量操作批量添加数据

说明：添加一条数据由两行代码实现，第一行设置添加数据的索引名称、表、id，第二行设置添加数据的字段和值

#_bulk批量添加数据
POST _bulk
#设置添加数据的索引名称、表、id
{"index":{"_index":"jobbole","_type":"job","_id":"4"}}
#设置添加数据的字段和值
{"title": "爬虫开发","salary_min": 15000,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}
#设置添加数据的索引名称、表、id
{"index":{"_index":"jobbole","_type":"job","_id":"5"}}
#设置添加数据的字段和值
{"title": "微信开发","salary_min": 15000,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}

代码：

POST _bulk
{"index":{"_index":"jobbole","_type":"job","_id":"4"}}
{"title": "爬虫开发","salary_min": 15000,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}
{"index":{"_index":"jobbole","_type":"job","_id":"5"}}
{"title": "微信开发","salary_min": 15000,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}

bulk批量操作批量创建数据(添加)

POST _bulk
{"create":{"_index":"jobbole","_type":"job","_id":"6"}}
{"title": "开发","salary_min": 100,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}

bulk批量操作批量删除数据

POST _bulk
{"delete":{"_index":"jobbole","_type":"job","_id":"5"}}
{"delete":{"_index":"jobbole","_type":"job","_id":"6"}}

bulk批量操作批量修改数据

POST _bulk
{"update":{"_index":"jobbole","_type":"job","_id":"1"}}
{"doc":{"title": "开发","salary_min": 100,"city": "北京","company": {"name": "百度","company_addr": "北京市软件园"},"publish_date": "2017-4-16","comments": 15}}