单个文档操作(增、改、删)
# ===================
# 1、创建索引
PUT es_document_db
# 2、PUT 新增文档
PUT /es_document_db/_doc/1
{
"name":"张三1",
"age":1,
"birthday":"2021-10-11",
"address":"中国上海长宁1"
}
# 3、POST 新增文档
POST /es_document_db/_doc
{
"name":"张三a",
"age":1,
"birthday":"2021-10-11",
"address":"中国上海长宁a"
}
# 4、获取索引下的所有文档 _search
GET /es_document_db/_doc/_search
# 执行的结果为:
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 1.0,
"hits" : [
{
"_index" : "es_document_db",
"_type" : "_doc",
"_id" : "1",
"_score" : 1.0,
"_source" : {
"name" : "张三1",
"age" : 1,
"birthday" : "2021-10-11",
"address" : "中国上海长宁1"
}
},
{
"_index" : "es_document_db",
"_type" : "_doc",
"_id" : "2uRil3wBDCx46Ou3gAno",
"_score" : 1.0,
"_source" : {
"name" : "张三a",
"age" : 1,
"birthday" : "2021-10-11",
"address" : "中国上海长宁a"
}
}
]
}
}
# 5、POST 更新id=1的文档
POST /es_document_db/_doc/1
{
"name":"张三a",
"age":1
}
# 6、删除id=1的文档
DELETE /es_document_db/_doc/1
说明:POST和PUT都能起到创建/更新的作用
1、需要注意的是==PUT==需要对一个具体的资源进行操作也就是要确定id才能进行==更新/创建,而==POST==是可以针对整个资源集合进行操作的,如果不写id就由ES生成一个唯一id进行==创建==新文档,如果填了id那就针对这个id的文档进行创建/更新
2、PUT/POST 会将json数据都进行替换
3、PUT与DELETE都是幂等性操作, 即不论操作多少次, 结果都一样
批量文档操作(增、改、删)
# 新建两个索引(es_index_one /es_index_two),用来做后面的文档搜索功能
PUT es_index_one
PUT es_index_two
批量对文档进行写操作是通过_bulk来实现的
- 请求方式:POST
- 请求地址:_bulk
- 请求参数:通过_bulk操作文档,一般至少有两行参数(或偶数行参数)
第一行参数为指定操作的类型及操作的对象(index,type和id)
第二行参数才是操作的数据
{"actionName":{"_index":"indexName", "_type":"typeName","_id":"id"}}
{"field1":"value1", "field2":"value2"}
- actionName:表示操作类型,主要有create,index,delete和update
# 2、分别对两个索引进行如下的操作
# 2.1、批量创建文档 (create)
POST _bulk
{"create":{"_index":"es_index_one","_type":"_doc","_id":1}}
{"name":"张三one-1","age":13,"birthday":"2021-10-13","address":"中国上海长宁one-1"}
{"create":{"_index":"es_index_one","_type":"_doc","_id":2}}
{"name":"张三one-2","age":23,"birthday":"2021-10-23","address":"中国上海长宁one-2"}
{"create":{"_index":"es_index_one","_type":"_doc","_id":3}}
{"name":"张三one-3","age":33,"birthday":"2021-10-30","address":"中国上海长宁one-3"}
POST _bulk
{"create":{"_index":"es_index_two","_type":"_doc","_id":1}}
{"name":"张三two-1","age":13,"birthday":"2021-10-13","address":"中国上海长宁two-1"}
{"create":{"_index":"es_index_two","_type":"_doc","_id":2}}
{"name":"张三two-2","age":23,"birthday":"2021-10-23","address":"中国上海长宁two-2"}
{"create":{"_index":"es_index_two","_type":"_doc","_id":3}}
{"name":"张三two-3","age":33,"birthday":"2021-10-30","address":"中国上海长宁two-3"}
# 2.2、批量创建或者更新文档 (index)
POST _bulk
{"index":{"_index":"es_index_one","_type":"_doc","_id":1}}
{"name":"张三one-11","age":13,"birthday":"2021-10-13","address":"中国上海长宁one-1"}
{"index":{"_index":"es_index_one","_type":"_doc","_id":2}}
{"name":"张三one-21","age":23,"birthday":"2021-10-23","address":"中国上海长宁one-2"}
{"index":{"_index":"es_index_one","_type":"_doc","_id":3}}
{"name":"张三one-31","age":33,"birthday":"2021-10-30","address":"中国上海长宁one-3"}
create 和index的区别:如果数据存在,使用create操作失败,会提示文档已经存在,使用index则可以成功执行。
# 2.3、批量修改文档 (update)
POST _bulk
{"update":{"_index":"es_index_one","_type":"_doc","_id":1}}
{"doc":{"name":"张三one-11q","age":131,"address":"中国重庆one-001"}}
{"update":{"_index":"es_index_one","_type":"_doc","_id":2}}
{"doc":{"name":"张三one-21q","age":231,"sex":"男"}}
{"update":{"_index":"es_index_one","_type":"_doc","_id":3}}
{"doc":{"name":"张三one-31q","age":331}}
update: 如果指定的_id不存在,则那一条更新会报错,其他的数据不会报错,如果更新的字段存在于已有的文档中,则进行更新,如果加入了新的字段也一样的会加入到文档中,update 并不会将在update中不存在的字段进行删除。
# 2.4、批量删除文档 (delete)
POST _bulk
{"delete":{"_index":"es_index_one","_type":"_doc","_id":1}}
{"delete":{"_index":"es_index_one","_type":"_doc","_id":2}}
{"delete":{"_index":"es_index_one","_type":"_doc","_id":3}}
_bulk 一次最大处理多少数据量?
bulk会把将要处理的数据载入内存中,所以数据量是有限制的,最佳的数据量不是一个确定的数值,它取决于你的硬件,你的文档大小以及复杂性,你的索引以及搜索的负载。
一般建议是1000-5000个文档,如果你的文档很大,可以适当减少队列,大小建议是5-15MB,默认不能超过100M,可以在es的配置文件(即$ES_HOME下的config下的elasticsearch.yml)中。
文档批量获取
批量获取文档数据是通过_mget 来实现的
- 请求方式:GET
- 请求地址:_mget
- 功能说明 : 可以通过ID批量获取不同index和type的数据
- 请求参数:
docs : 文档数组参数
_index : 指定index
_source : 指定要查询的字段
_id : 指定id_type : 指定type
# 首先将之前进行批量删除数据加回去
# 3.1、 获取指定索引下所有的文档数据
GET /es_index_one/_doc/_search
GET /es_index_two/_doc/_search
# 3.2、通过 _mget 批量获取
GET _mget
{
"docs": [
{
"_index": "es_index_one",
"_type": "_doc",
"_id": 1
},
{
"_index": "es_index_one",
"_type": "_doc",
"_id": 2,
"_source":[
"name","age"
]
},
{
"_index": "es_index_two",
"_type": "_doc",
"_id": 5
}
]
}
# 得到的响应是:
{
"docs" : [
{
"_index" : "es_index_one",
"_type" : "_doc",
"_id" : "1",
"_version" : 1,
"_seq_no" : 14,
"_primary_term" : 1,
"found" : true,
"_source" : {
"name" : "张三one-1",
"age" : 13,
"birthday" : "2021-10-13",
"address" : "中国上海长宁one-1"
}
},
{
"_index" : "es_index_one",
"_type" : "_doc",
"_id" : "2",
"_version" : 1,
"_seq_no" : 15,
"_primary_term" : 1,
"found" : true,
"_source" : {
"name" : "张三one-2",
"age" : 23
}
},
{
"_index" : "es_index_two",
"_type" : "_doc",
"_id" : "5",
"found" : false
}
]
}
# 3.3、在 URL中指定 index
GET /es_index_one/_mget
{
"docs": [
{
"_type": "_doc",
"_id": 1
},
{
"_type": "_doc",
"_id": 5
}
]
}
# 3.4、在 URL中指定 index 与 type
GET /es_index_one/_doc/_mget
{
"docs": [
{
"_id": 1
},
{
"_id": 5
}
]
}