Hi there, quick question here : what's the difference between

s3

from

prefect.filesystems

and

S3Bucket

from

prefect_aws.s3

?

S3Bucket has many more methods available for working with data.

Alright thanks! I guess s3 is aimed at s3 storage deployments only whereas s3bucket is meant for data manipulation, right?

s3Bucket can handle flow code storage, too.

Alright, so

s3

is a bit redundant here. It seems I could get away with only

s3Bucket

blocks in most cases.

Yep. That’s true. The S3 is quicker to get going with and fine if just using for flow code storage.

Is it possible to have them work in combination with other

s3fs

libs, such as pandas? I did not test yet, but could I pass blocks as parameters for a pandas io function like

<http://df.to|df.to>_csv(s3_bucket_block, aws_credentials=aws_credentials_block)

. Is that a possible use case for blocks?