python-pachyderm
python-pachyderm copied to clipboard
Python client for Pachyderm
We want python-pachyderm to support both `v2.x` and `v1.x` of mainline pachyderm, and in general, we want to support one backward version in addition to the most current version of...
When @albscui created the initial python-pachyderm proto package, he set himself as the package author for expediency. However, to make sure that future contributors to the package have access to...
See https://github.com/pachyderm/pachyderm/issues/6116 Once support for generating a python module from Pachyderm's protobufs has been added to the main Pachyderm repo, we should migrate python-pachyderm to use that module instead of...
Some of our client methods accept various types for inputs, for example in `get_file(commit, path, URL)`, the `commit` can be a tuple, string, or `Commit` object. While we do some...
General feedback from users has been that python-pachyderm is hard to use. For example, we've heard "there's no way to parse a Pachyderm file as a CSV or a pandas...
Assumming I get an otp via pachctl, I run the following in python: `client.authenticate_one_time_password('')` this returns a string. It is unclear where and how this string is used from the...
To prevent users creating more than one SpoutManager object, it might be desirable for the SpoutManager to be a singleton object. This would avoid contention over the named pipe it...
Would be nice to have a good, color coded, non verbose way of viewing pipeline status over time without repeatedly calling `!pachctl list pipeline` etc.
`put_file_bytes` with an iterable of bytes or strings is deprecated, as any of the alternatives are more pythonic and likely more efficient. It should be remove eventually (perhaps coinciding with...
Currently `pachctl` support `.pachignore` files under the path in `transform.build.path`, but `python-pachyderm` doesn't. Ideally the two should be brought to parity