[ ] Would need to allow s3 database configuration.
[ ] Leave the output schema as is, but have the user configuration editable to flag which results would get pushed to s3 vs local.
[ ] For example: the output schema can define the option with s3 potential. but in the pipestat config file, maybe the config options in pipestat have includes, excludes, provide list of values.
[ ] We probably don't want to push ALL files. For example, some outputs are large and not helpful for results summary. Therefore the user define which results are to be pushed to s3
[ ] In the future have the output schema simply live remotely on the repository, so the end user doesn't need to change it.
Idea: Could pephub actually generate the summarized html page automatically?
Originally posted by @donaldcampbelljr in https://github.com/pepkit/pipestat/issues/191#issuecomment-2139516103
Notes from discussion:
We could allow Pipestat to push items to s3. We currently do this in bbclient: https://github.com/databio/geniml_dev/blob/4657e53d8de0720c7041aff5f6143a4694b19f40/geniml/bbclient/bbclient.py#L193-L205
[ ] Would need to allow s3 database configuration.
[ ] Leave the output schema as is, but have the user configuration editable to flag which results would get pushed to s3 vs local.
[ ] For example: the output schema can define the option with s3 potential. but in the pipestat config file, maybe the config options in pipestat have includes, excludes, provide list of values.
[ ] In the future have the output schema simply live remotely on the repository, so the end user doesn't need to change it.
[ ] pepkit/pephub#314