Download from S3 Browser to File Browser and vice versa

[gogakoreli]

Feature Request

As a user I would like to download files from S3 Browser to my local File Browser environment and vice versa if I am in the cloud environment (for example Sagemaker JupyterLab). There is a workaround to download from S3 Browser to my local computer and upload to the File Browser. However to improve the UX ideally we want this to be seamlessly done only within the JupyterLab itself.

[ellisonbg]

I think this already work. Upload is done by dragging a local file to the file browser and download is available as an option in the context menu for a file. We should confirm this all works as expected though.

[DenisaCG]

Thanks for opening this issue!

The linked PR adds the workflow of directly copying a file from an S3 bucket to the local fillebrowser, it also contains a screenshare to showcase the operation.

Users can still download a file from S3 locally and then manually uploaded it to the default filebrowser and vice versa to directly upload a file to the bucket from the local storage using the Upload toolbar button.

The downloading of files using the context menu operation works as expected, it was tested on text, notebook and PDF files, but there are some problems with certain uploads, the issue is documented here and will be addressed in a later PR.