Hello, currently, getting S3 to save/load in a specific folder path is impossible due to the bucket field. The bucket field is necessary to function, so I would not remove that. I would instead append the path to the endpoint after the bucket.
Say for example, I do not want to save my blobs in the root folder of my AWS storage. Such as https://cdn.my.cool.domain/pds/blobs/*.
You can technically do it whilst also appending the bucket's name via the endpoint url, but that's really messy and doesn't exactly work as intended. As it always results in smth like https://cdn.my.cool.domain/pds/blobsbucket-name/*
This honestly seems like a simple PR that I could do at some point, so I'll probably look into that.
Completed by #251