Can't import wodnc

[BecCowley]

@BillMills Maybe a bug, maybe something I've done wrong, but getting the following when trying to import the wodnc from wodpy:

from wodpy import wodnc
Traceback (most recent call last):
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/code.py", line 90, in runcode
exec(code, self.locals)
File "", line 1, in
ImportError: cannot import name 'wodnc' from 'wodpy' (/Users/xxx/venvs/iquod/lib/python3.8/site-packages/wodpy/init.py)

Tried in python3.8 and 3.10
Any help appreciated!

[BecCowley]

I think it didn't install at all - perhaps an issue at the pip install side of things which I know nothing about!
I got around it by manually adding the wodnc.py to the library folder.

[bkatiemills]

Hi @BecCowley, sorry for the super slow reply here - can you show me the result of your pip freeze? This functionality isn't in the releases on pypi since there's still some issues to sort out before this is 100% ready for production (see the points in https://github.com/IQuOD/wodpy/releases/tag/2.0.0b0, would love your input there). For now, you can install the 2.0.0 beta by cloning this repo and doing:

cd dist
pip install wodpy-2.0.0b0.tar.gz

and then all should be well. Let me know if not?

[BecCowley]

Hi @BillMills - I understand now!
Output from pip freeze:

aiobotocore==2.5.4
aiohttp==3.8.5
aioitertools==0.11.0
aiosignal==1.3.1
async-timeout==4.0.3
attrs==23.1.0
botocore==1.31.17
certifi==2023.7.22
cftime==1.0.3.4
charset-normalizer==3.2.0
contourpy==1.1.0
cycler==0.11.0
exceptiongroup==1.1.3
fonttools==4.42.1
frozenlist==1.4.0
fsspec==2023.6.0
idna==3.4
importlib-resources==6.0.1
iniconfig==2.0.0
jmespath==1.0.1
kiwisolver==1.4.5
matplotlib==3.7.2
multidict==6.0.4
netCDF4==1.6.4
numpy==1.24.4
packaging==23.1
pandas==2.0.3
Pillow==10.0.0
pluggy==1.3.0
pyparsing==3.0.9
pytest==7.4.0
python-dateutil==2.8.2
pytz==2023.3
s3fs==2023.6.0
six==1.16.0
tomli==2.0.1
typing_extensions==4.7.1
tzdata==2023.3
urllib3==1.26.16
wodpy==1.6.2
wrapt==1.15.0
yarl==1.9.2
zipp==3.16.2

I ended up adding the code to my forked version and made my own branch. I've made some improvements (or created new bugs?).
Happy to contribute these edits. I think I accidentally pushed my branch to the iquod repo before I'd forked. If you can advise me how best you would like me to share, that would be great.

[bkatiemills]

right, the wodpy=1.6.2 there doesn't have the nc functionality, is why you saw what you saw. In future if you want a smoother experience, you can install the beta per my comment above.

Great, I'd love to see what you were working on! I generally get the best results from a fork-and-pull workflow:

• fork the repo (you already did this, no need to do it ever again)
• push your feature branch to your fork (sounds like you pushed IQUODflagTesting here and to your fork - that's fine, but I would normally just push your feature branch to your fork, and keep the main repo tidy with to-be-released content).
• set up a PR from your fork's feature branch to the IQuOD repo's main branch and tag me in it so I can give feedback before we merge. You very nearly did this in your PR Specify integer type for offset and nentries to allow for zero levels… #38, but you merged your fork's IQUODflagTesting into IQuOD's IQUODflagTesting rather than main.

Not the only way to do it, but that keeps things nice and tidy and is how I manage my other projects, let me know any questions.

[BecCowley]

OK - I did stuff around a bit with the branch in the iquod repo. I wasn't sure if it could be just deleted?

I'll do a PR now from my fork to iquod's main.

[mylesmc123]

Hello, just tacking on my recent experience. I installed the 2.0 release in a conda environment, by cloning the repo and following the steps:

cd dist
pip install wodpy-2.0.0b0.tar.gz 

However, I then got errors related to HDF5, and then netCDF4 complained about not having a path variable. So I ran the following sequence to get everything up and running:

conda install hdf5
conda install netcdf4
cd dist
pip install wodpy-2.0.0b0.tar.gz