参数验证失败：参数 Key 的类型无效。，值：，类型：<class 'str'> ，有效类型：<class 'dict'></class></class>

Question

I have 2 Lambda, 1 is doing a batch_write and put_item to ddb. The other lambda does the get_item from first lambda (It has permissions to get_item).

ERROR:

[ERROR] ParamValidationError: Parameter validation failed:
Invalid type for parameter Key.active_employee, value: jen, type: <class 'str'>, valid types: <class 'dict'>
Traceback (most recent call last):
  File "/var/task/my_lambda/checks.py", line 100, in lambda_handler
    response = ddb.get_item(TableName="testtable", Key={"active_employee": user})

Lambda 1:

        with gzip.open(response["Body"], "rt") as file:
        try:

            with table.batch_writer(overwrite_by_pkeys=["active_employee"]) as batch:
                for active_users in file:
                    user_dict = json.loads(active_users)
             
                    manager = user_dict["manager"]
                    user = user_dict["user"]


                    if not manager:
                        continue


                    if not user:
                        continue
                 
        
                    else:
                        batch.put_item(
                            Item={
                                "active_employee": user,
                                "mgr_email": mgr_email
            
                            },
                        )

            logger.info("Loaded data into table %s.", table.name)

        except ClientError:
            logger.exception("Couldn't load data into table %s.", table.name)
            raise

Lambda 2

user = "jen"

    ddb = boto3.client("dynamodb")
    response = ddb.get_item(TableName="testtable", Key={"active_employee": user})
    employee_data = json.loads(response["Item"])
    if employee_data and employee_data["active_employee"] == user:
        manager = employee_data["mgr_email"]
        print(f"{user} is active")
        print(f"{manager}")
    else:
        print("user not in ddb")

I am expecting to get in Lambda jen is active then the manager email. I do not know the manager value. Say the DDB has a million in it and I cannot use scan or query. I've read that get_item is a lot faster when getting a single item.

• How can I fix the error?
• How do I get_item as dictionary? should the user = "jen" be made into dictionary? What is the syntax?
• When doing get_item can I only use the pk and expect to also get the other key (mgr_email). I can only get_item the user and I need it to look for the manager email for me of that alias too if it exists and return both.

Answer 1

In Lambda 2 you are using the low level client, which expects DynamoDB JSON such as:

{'active_employee':{'S':'jen'}}

Now, for you to make it work in your current context, you would be better using the Resource client, as you do in Lambda 1.

dynamodb = boto3.resource("dynamodb", region_name='us-west-2')

table = dynamodb.Table('testtable')

try:
    response = table.get_item(
        Key={
            'active_employee': "jen"
        }
    )
except ClientError as e:
    print(e.response['Error']['Message'])

Be careful not to mix your clients up, and always refer to the documentation for the specific client you are using.

Resource Client: https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/dynamodb.html#DynamoDB.Table.get_item

Low Level Client: https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/dynamodb.html#DynamoDB.Client.get_item