Boto3：将数据插入dynamodb表

Question

I have a DynamoDB table called events and I want to put the data into the table.

My dynamodb table structure is given below

{
 "partition_key": "user_id1111",
 "sort_key": "version_1",
 "attributes": {
  "events": [
   {
    "t": "1614712316",  
    "a": "product_view",   
    "i": "1275"
   },
   ...
   ...
   ...
  ]
 }
}

I have 2 entries of data:

entry1 = {
     "t": 1607208938861,
     "a": "product_add",
     "i": "142320"
}

entry2 = 
    {
      "M": {
       "t": {
        "N": "1607208938862"
       },
       "a": {
        "S": "product_purchase"
       },
       "i": {
        "S": "142318"
       }
      }
     }

I want to insert there two entries into the dynamodb table.

I know we can use BOTO3-Dynamodb-resource to insert entry1 into the table

I know we can use BOTO3-Dynamodb-client to insert entry2 into the table

NOTE:

entry1 ==> the data will always be in the format similar to dynamodb resource

entry2 ==> the data will always be in the format similar to dynamodb client

GOAL:

using single boto3-resource method( put-item ), I want to insert these two records into dynamodb table.

MY FINAL OUTPUT IN THE DYNAMODB TABLE WILL BE SIMILAR TO BELOW

{
 "partition_key": "user_id1111",
 "sort_key": "version_1",
 "attributes": {
  "events": [
     {
     "t": 1607208938861,
     "a": "product_add",
     "i": "142320"
     },
     {
     "t": 1607208938862,
     "a": "product_purchase",
     "i": "142318"
     },

   
  ]
 }
}

Can anyone suggest a solution for this?

Answer 1

Basically I found a python library called dynamodb-json : https://pypi.org/project/dynamodb-json/

this library helps us to convert dynamodb-json format into python representation . then, we can directly use boto3-dynamodb-resource to put python representation of an entry into dynamodb.

import boto3
import json
from dynamodb_json import json_util as dbjson


dynamodb = boto3.resource('dynamodb')
dbtable = dynamodb.Table('events')


entry1 = json.dumps({
     "t": 1607208938861,
     "a": "product_add",
     "i": "142320"
})

entry2 = 
    {
      "M": {
       "t": {
        "N": "1607208938862"
       },
       "a": {
        "S": "product_purchase"
       },
       "i": {
        "S": "142318"
       }
      }
     }


# convert entry2 into python representation
obj = json.dumps(dbjson.loads(entry2))

in_list = []
in_list.append(json.loads(entry1))
in_list.append(json.loads(obj))

response = dbtable.put_item(
           Item={
              "cache_key": "user_id1111",
              "sort_key": "version_1",
              "attributes": {"events":in_list},
     }
        )

The above code solves my problem.