Package on Anaconda via conda-forge? about pystow HOT 6 CLOSED

pystow only relies on a small number of super popular packages so i'm not super worried about pinning their versions. Also, this sounds like it's catering to end-users who aren't really creating reusable software

from pystow.

I'm going to close this as pystow is now on conda-forge. I'm not exactly sure what the update instructions are so i will burn that bridge when we need to come to it

from pystow.

if i can figure this out, sure. I'm traveling tomorrow and will try to get it set up then

(though personally I'm a huge conda hater - it's a massive problem for reproducibility)

from pystow.

@cthoyt thanks for being willing to take a look!

from pystow.

@sgbaird I think it's up, https://anaconda.org/conda-forge/pystow. Let me know if you're able to use this in your package now

from pystow.

@cthoyt I'm still waiting on the conda-forge devs to merge mine conda-forge/staged-recipes#19864, but I will let you know if I run into any issues with incorporating it. In the meantime, I did a local installation in a fresh conda env and was able to run a few examples from the README and nothing seems incompatible based on the meta.yaml recipe you uploaded, so I don't think it will pose a problem. I'll keep the reproducibility issue in mind. One of the recommendations from PyScaffold's data science extension template is:

Dependency Management & Reproducibility

1. Always keep your abstract (unpinned) dependencies updated in environment.yml and eventually
   in setup.cfg if you want to ship and install your package via pip later on.
2. Create concrete dependencies as environment.lock.yml for the exact reproduction of your
   environment with:

   conda env export -n matbench-genmetrics -f environment.lock.yml

   For multi-OS development, consider using --no-builds during the export.
3. Update your current environment with respect to a new environment.lock.yml using:

   conda env update -f environment.lock.yml --prune

I also came across https://github.com/conda-incubator/conda-lock. Do either of these address that issue? I'm sure each comes with its slew of issues.

from pystow.