AWS DynamoDB - Load data with Boto3 using JSON file as input

5
votes

I have multiple tables in Amazon DynamoDB, JSON Data is currently uploaded into the tables using the batch-write-item command that is available as part of AWS CLI - this works well.

However I would like to use just Python + Boto3 but have not been able to execute the Boto BatchWriteItem request with an external data file as input. I envision if there is a Boto3 script it would look like this shown below, but I have not been able to find documentation/examples for it.

Example (Pseudo Code)

table = dynamodb.Table(‘my_table’)
table.BatchWriteItem(RequestItems=file://MyData.json)

Reference: http://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_BatchWriteItem.html

Pointers appreciated.

2
pythonamazon-web-servicesamazon-dynamodbbotoboto3

2 Answers

9
votes

The best place to look would be Boto3's readthedocs here: https://boto3.readthedocs.org/en/latest/reference/services/dynamodb.html#DynamoDB.Client.batch_write_item

As long as your JSON was formatted correctly for the request as in the example you could use:

f = open('MyData.json')
request_items = json.loads(f.read())
client = boto3.client('dynamodb')
response = client.batch_write_item(RequestItems=request_items)
0
votes

I loaded the JSON this way 

import boto3
import json

dynamodbclient=boto3.resource('dynamodb')
sample_table = dynamodbclient.Table('ec2metadata')

with open('/samplepath/spotec2interruptionevent.json', 'r') as myfile:
    data=myfile.read()

# parse file
obj = json.loads(data)

#instance_id and cluster_id is the Key in dynamodb table 

    response=sample_table.put_item(
                              Item={
                                  'instance_id': instanceId,
                                  'cluster_id': clusterId,
                                  'event':obj

                              }
                              )