AWS S3にSQSを全部出力するLambda書いてみた

AWS SQS

Python で記載

import boto3
import json
import os

# クライアントを作成
sqs_client = boto3.client('sqs')
s3_client = boto3.client('s3')

def process_sqs_queue_to_s3(queue_url, bucket_name):
    """指定されたSQSキューの全メッセージをS3に保存する関数"""
    processed_count = 0
    while True:
        response = sqs_client.receive_message(
            QueueUrl=queue_url,
            MaxNumberOfMessages=10,
            WaitTimeSeconds=10
        )

        if 'Messages' not in response:
            break

        for message in response['Messages']:
            message_body = message['Body']
            s3_key = f"messages/{message['MessageId']}.json"
            s3_client.put_object(
                Bucket=bucket_name,
                Key=s3_key,
                Body=json.dumps({"message_body": message_body})
            )
            sqs_client.delete_message(
                QueueUrl=queue_url,
                ReceiptHandle=message['ReceiptHandle']
            )
            processed_count += 1

    return processed_count

def lambda_handler(event, context):
    # キューリストとS3バケット名を環境変数から取得
    queue_urls = os.environ['SQS_QUEUE_URLS'].split(',')
    s3_bucket_name = os.environ['S3_BUCKET_NAME']

    total_processed = 0

    for queue_url in queue_urls:
        processed_count = process_sqs_queue_to_s3(queue_url.strip(), s3_bucket_name)
        total_processed += processed_count
        print(f"Queue {queue_url} processed {processed_count} messages.")

    return {'statusCode': 200, 'body': f"Total {total_processed} messages processed from all queues."}

環境変数 SQS_QUEUE_URLS にすべてのキューURLをカンマ区切りで入力（例: https://sqs.region.amazonaws.com/123456789012/Queue1,https://sqs.region.amazonaws.com/123456789012/Queue2