RuntimeWarning: subpackages can technically be lazily loaded about pyjanitor HOT 16 CLOSED

Hmmm, I think we should just get rid of lazy loading. It was originally implemented out of a desire for speed, but I'm seeing a tradeoff.

I will work on this matter over the coming few days.

from pyjanitor.

@samukweku It does not. It has no effect.

I found a way to do so:

with warnings.catch_warnings():
    warnings.simplefilter("ignore")
    import janitor

from pyjanitor.

#1278 fixes this. @jppyykm kindly reinstall the Dev version.

from pyjanitor.

@jppyykm I have cut a new minor release. Give the bots a few minutes and we should be able to see the releases out in the wild.

from pyjanitor.

@samukweku I agree with you. That said, we are pre-1.0, so minor == major for us, and patch == minor.

from pyjanitor.

How can I ignore this warning?

from pyjanitor.

@ericmjl @thatlittleboy any ideas?

from pyjanitor.

thanks @jppyykm for the feedback. If you use a lower python version (3.8/3.9) do you get an error message? if you can, kindly create a new environment with a lower python version, and see if the error pops up again

from pyjanitor.

Of course. I ran it in Python 3.9 and got the same warning. I tried silencing it but couldn't, does anyone know how to do so?

from pyjanitor.

@jppyykm there is a pending PR that should fix this. in the mean time, you can use import warnings to silence it

from pyjanitor.

@samukweku Yes! Very grateful. I tried that and it doesn't work. Not sure it's relevant but it is perhaps more information to consider. Thanks!

from pyjanitor.

@jppyykm So the code below doesn't work?

import warnings
warnings.simplefilter("ignore", RuntimeWarning)

from pyjanitor.

@ericmjl it seems the release failed? https://github.com/pyjanitor-devs/pyjanitor/actions/runs/5649973510/job/15305429936#step:13:72

from pyjanitor.

Yeah it did. I’m gonna look at it after hours, doing kids stuff right now. Thanks for the heads up though.

from pyjanitor.

@ericmjl I feel it should be a major release cos of some possibly breaking changes

from pyjanitor.

@jppyykm closing this - it has been fixed with #1278; you should not have this issue with pyjanitor version 0.25. feel free to give more feedback if you still have this issue. thanks

from pyjanitor.