Track S3 bucket sizes daily

from datetime import datetime, timedelta, timezone
import boto3
from clickruntime import on_trigger, credentials, notify

REGION = "us-east-1"

def aws(service):
    return boto3.client(service, region_name=REGION, aws_access_key_id=credentials.secret("AWS_ACCESS_KEY_ID"),
                        aws_secret_access_key=credentials.secret("AWS_SECRET_ACCESS_KEY"))

WARN_GB = 100

def metric(cw, bucket, name, storage):
    points = cw.get_metric_statistics(Namespace="AWS/S3", MetricName=name, Period=86400, Statistics=["Average"],
                                      StartTime=datetime.now(timezone.utc) - timedelta(days=2), EndTime=datetime.now(timezone.utc),
                                      Dimensions=[{"Name": "BucketName", "Value": bucket}, {"Name": "StorageType", "Value": storage}])["Datapoints"]
    return max(points, key=lambda p: p["Timestamp"])["Average"] if points else 0

@on_trigger
def run(event):
    cw, lines = aws("cloudwatch"), []
    for b in aws("s3").list_buckets()["Buckets"]:
        gb = metric(cw, b["Name"], "BucketSizeBytes", "StandardStorage") / 1e9
        objects = metric(cw, b["Name"], "NumberOfObjects", "AllStorageTypes")
        lines.append(f"{'WARNING ' if gb > WARN_GB else ''}{b['Name']}: {gb:,.1f} GB, {objects:,.0f} objects")
    notify.email_me("S3 bucket sizes", "\n".join(lines))
    return {"buckets": len(lines)}