I am trying to write a lambda function that will tag ec2 instances that will go from pending to running state. However, I have a problem reading the csv file that holds my ec2 instance tags. Currently, I have gone to where the lambda returns me the following result.

START RequestId: 6290699e-4018-4801-b7a8-6b5b46f26c2a Version: $LATEST
{'Key': 'Name1', 'Value': 'Machine-1'}
{'Key': 'Name2', 'Value': 'Machine-2'}
{'Key': 'Name3', 'Value': 'Machine-3'}
END RequestId: 6290699e-4018-4801-b7a8-6b5b46f26c2a
REPORT RequestId: 6290699e-4018-4801-b7a8-6b5b46f26c2a  Duration: 3306.40 ms    Billed Duration: 3307 ms    Memory Size: 128 MB Max Memory Used: 88 MB  Init Duration: 335.79 ms

However, I need a list of dictionaries.

myList = [{'Key': 'Name1', 'Value': 'Instance-1'}, {'Key': 'Name2', 'Value': 'Instance-2'}, {'Key': 'Name3', 'Value': 'Instance-3'}]

Because the rest of the code looks like the following

instance_id = event['detail']['instance-id']
response = ec2_client.create_tags(
  Resources=[
      instance_id,
    ],
  Tags=[
        {
            'Key': 'Name',
            'Value': 'event_bridge_lambda_tag'
        },
    ]
)

At the moment, my lambda code looks like the following

import csv
import boto3
from collections import OrderedDict

def lambda_handler(event, context):
    
    s3_client = boto3.client("s3")
    ec2_client = boto3.client("ec2")

    S3_BUCKET_NAME = "tag-holds-bucket"
    FILE_NAME = "tags.csv"

    s3_file = s3_client.get_object(Bucket=S3_BUCKET_NAME, Key=FILE_NAME)
    file_content = s3_file['Body'].read().decode('utf-8').splitlines()
    
    myList = list()

    records = csv.DictReader(file_content)
    for row in records:
       #print(dict(row))
       myList.append(row)

myList.append(row) -> gives me something like this as output

START RequestId: 618217d1-d1da-473f-b55e-77f1f7fe52dc Version: $LATEST
[OrderedDict([('Key', 'Name1'), ('Value', 'Instance-1')]), OrderedDict([('Key', 'Name2'), ('Value', 'Instance-2')]), OrderedDict([('Key', 'Name3'), ('Value', 'Instance-3')])]
END RequestId: 618217d1-d1da-473f-b55e-77f1f7fe52dc
REPORT RequestId: 618217d1-d1da-473f-b55e-77f1f7fe52dc  Duration: 3128.39 ms    Billed Duration: 3129 ms    Memory Size: 128 MB Max Memory Used: 88 MB  Init Duration: 330.02 ms

I don't know how to reach such a state

myList = [{'Key': 'Name1', 'Value': 'Instance-1'}, {'Key': 'Name2', 'Value': 'Instance-2'}, {'Key': 'Name3', 'Value': 'Instance-3'}]

My cvs file looks like this

tags.csv

And rest of the lambda code

instance_id = event['detail']['instance-id']
    response = ec2_client.create_tags(
    Resources=[
        instance_id,
    ],
    Tags=[
        {
            'Key': 'Name',
            'Value': 'event_bridge_lambda_tag'
        },
    ]
)

CodePudding user response：

csv.DictReader returns a dict or an OrderedDict depending on the Python version you are running.

As seen in the documentation:

Changed in version 3.6: Returned rows are now of type OrderedDict.

Changed in version 3.8: Returned rows are now of type dict.

So it seems you are running Python < 3.8 in the lambda function. So you have 2 possibilities to have the output as a dict:

1. Change your lambda function runtime to a higher one
2. Convert the OrderedDict to a dict in your code simply using dict method: my_dict = dict(myList)

CodePudding user response：

One way to serialize collections.OrderedDict into a list of dictionaries is by using json library.

import json

# ...

records = csv.DictReader(file_content)
for row in records:
    myList.append(row)
myList = json.loads(json.dumps(myList)) # add this line

or you can just use list comprehension

myList = [dict(item) for item in myList] # add this line