jastang

{
   "Version": "2012-10-17",
   "Statement": [
      {
         "Effect": "Allow",
         "Action": [
             "rds-db:connect"
         ],
         "Resource": [
             "arn:aws:rds-db:your-region:XXXXXXXXXX:dbuser:*/benchsci"

jastang

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Principal": {
        "AWS": [
          "arn:aws:iam::XXXXXXXXX:role/your-lambda-function-role"
        ],
        "Service": "lambda.amazonaws.com"

jastang

provider:
  name: aws
  runtime: python3.7
  stage: 'production'
  region: us-east-1
  iamRoleStatements:
    - Effect: "Allow"
      Action:
        - "rds:*"
        - "sts:*"

jastang

CREATE USER benchsci WITH LOGIN; 
GRANT rds_iam TO benchsci;
-- optional
GRANT USAGE, SELECT ON ALL SEQUENCES IN SCHEMA <schema> TO benchsci;

jastang

"""
First, start the crawler if it is not already running. 
"""
crawler = glue.get_crawler(Name='%s-%s' % (client, DATA_FREQUENCY))

if crawler['Crawler']['State'] == 'READY':
    response = glue.start_crawler(Name='%s-%s' % (client, DATA_FREQUENCY))
else:
    print('Crawler is not in a READY state or already running! Checking the catalog instead.')

jastang

def lambda_handler(event, context):
    records = event['Records']

    for record in records:
        b = record['s3']['bucket']['name']
        s3_file = record['s3']['object']['key']

        # Load the metadata validation configuration for this client
        conf = json.loads(s3.meta.client.get_object(Bucket=b, Key=VALIDATION_CONFIG)['Body'].read(),
                          object_pairs_hook=OrderedDict)

jastang

def lambda_handler(event, context):
    if event is None:
        print("Couldn't find the object or byte range!")
        return

    # The byte range is passed from the Chunking function
    start_byte = event['start_byte']
    end_byte = event['end_byte']
    bucket = event['bucket']
    key = event['key']

jastang

while end_byte <= objectsize:

            # Ensure the end_byte is a carriage return, so the line splits work.
            end_byte = scan_to_eol(end_byte, objectsize, bucket, key)

            # Invoke the cleaning function
            ctx = {
                "start_byte": start_byte,
                "end_byte": end_byte,
                "bucket": bucket,

jastang

def lambda_handler(event, context):
    sb = event['Src_Bucket']
    sk = event['Src_Key']
    db = event['Dest_Bucket']
    dk = event['Dest_Key']

    src = {
        'Bucket': sb,
        'Key': sk
    }

jastang

s3 = boto3.resource('s3')

conf = json.loads(s3.meta.client.get_object(Bucket=b, Key='my-config.json')['Body'].read(), object_pairs_hook=OrderedDict)

bucket = s3.Bucket(name=b)
# This is just metadata
src = bucket.objects.filter(Prefix='sftp')
# If we have N mandatory files, are most recent N uploads representative of the mandatory files?
mandatory_objects = [o.key for o in sorted(src, key=lambda x: x.last_modified, reverse=True)][:len(conf['mandatory_files'])]
# You can now validate whatever you want in mandatory_objects