How to delete a file from a S3 Bucket folder

PBM

Hi, 
We use Amazon S3 to load data into Amazon Redshift Database. After data is loaded we want to clean up the S3 text files. I see the Read, Loop, and Write S3 operators. Does RM allow Deleting files from S3, considering that I have access to delete in S3?  Otherwise any workaround suggestions?
Thank You

sgenzer

hi @PBM yep that is pretty much the only way to do it. You can also install the AWS CLI and use the Execute Program operator to run shell script commands.

It would be a logical feature request...

Scott

PBM

Hi all,

Ok it seems that this is a case where RapidMiner does not have the object but then you can extend it by using external scripts. In this case I have used the Execute Python script and boto library. 

from boto.s3.connection import S3Connection

S3_BUCKET_NAME='myBucket'
AWS_ACCESS_KEY='myAccesskey'
AWS_SECRET_KEY='mySecretKey'
path_to_file='mysubFolderPath'

def rm_main():
  # Create connection
  conn = S3Connection(AWS_ACCESS_KEY, AWS_SECRET_KEY)
         
  # Connet to my  bucket
  bucket = conn.get_bucket(S3_BUCKET_NAME)
     
  # Get subdirectory info and delete files (except the subdirectory itself)
  for key in bucket.list(prefix=path_to_file, delimiter='/'):
    if key.name != path_to_file:
      bucket.delete_key(key)

  return

if __name__ == "__main__":
   rm_main()

Thank You