Docs

[zmoon]

Update docs, configure some things that were not yet configured, ...

• clean up Conda env
• fix GH issue links in whats-new
• RTD theme set in conf
• add intersphinx/crossreferences
• clean up docstrings
• main API page organized; use top-level namespace for the generated documentation of the functions
• edit other text a bit

Remaining:

• example nbs
• add * to other signatures

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[roxyboy]

Thanks for these edits @zmoon ! :)

[roxyboy]

@zmoon Can you run black on detrend.py? The pre-commit tests seems to be failing...

[zmoon]

@roxyboy, when are you planning to make the true_* arguments True by default? Just wondering, as currently the examples generate warnings related to that.

Also would you mind if I switch from nbsphinx to myst_nb to render the nbs? nbsphinx is not doing well in the cases with inline code inside strong or links, e.g. in the main example. And with myst_nb, we can easily cross-reference API and use admonitions etc.

[zmoon]

Should this still be FutureWarning given that it has already been changed?

[roxyboy]

The defaults is still False for these flags I think in the current version.

[zmoon]

The default lag is still None, but the behavior has changed slightly (ca652c9), since it now tries the default_lag attr before defaulting to 0.0, whereas before pretty much nothing was done for lag=None, making it effectively zero. So maybe just normal UserWarning instead of FutureWarning?

[roxyboy]

Ok sure, I'm ok with UserWarning

[zmoon]

Also how big is the dataset for the MITgcm example? Could it be included in the repo or somewhere else where it could be downloaded?

[rabernat]

Thanks so much for this much needed maintenance @zmoon! We really appreciate your efforts! 🏆

I am fine with changing any of the configurations in the docs to make things easier going forward. myst_nb is definitely fine.

Also how big is the dataset for the MITgcm example? Could it be included in the repo or somewhere else where it could be downloaded?

We could put it in Zenodo and use Pooch. Or we could use one one of the Pangeo Cloud datasets - https://catalog.pangeo.io/browse/master/ocean/channel/

[roxyboy]

@roxyboy, when are you planning to make the true_* arguments True by default? Just wondering, as currently the examples generate warnings related to that.

This is a good question. What's the normal period/cycles packages allow for a deprecation...?

[KevinYanesG]

Great work! I've found some small typos in the notebooks of zmoon:docs. I cannot comment it directly here, because the diff is too large. What is the best-practice-procedure here? Should I fork zmoon:docs and pull a request into his repo?


--> I think these should be xrft.fft and xrft.ifft, respectively.

One more thing: What about the xrft.fft FutureWarnings, shouldn't they be triggered only when using xrft.dft?

[KevinYanesG]

I wanted to correct the other two notebooks that haven't been refactored in this PR: #185.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

KevinYanesG commented on 2022-04-28T14:53:14Z
----------------------------------------------------------------

Now I learned that we can review the notebooks cell by cell with this tool. As I said, I think in the last sentence we should have xrft.fft and xrft.ifft.

[DWesl]

@roxyboy, when are you planning to make the true_* arguments True by default? Just wondering, as currently the examples generate warnings related to that.

This is a good question. What's the normal period/cycles packages allow for a deprecation...?

• NumPy: The longer of two releases or one year
• SciPy: Six months, usually two releases
• Python: At least two minor versions (~2 years), ideally five years