Fix multiple temporal avg calls on same dataset breaking

[tomvothecoder]

Description

This was caused by initializing the ds.temporal class once with self._time_bounds. Instead, it needs to be set every time the user runs temporal averaging because this self._time_bounds can change between each call depending on user settings (e.g., drop incomplete djf).

• Closes [Bug]: error on climatology calculation for custom seasons if after running for default seasons #325
• Closes [Bug]: Error from climatology calculation when getting it multiple times #327

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules

If applicable:

• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass with my changes (locally and CI/CD build)
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have noted that this is a breaking change for a major release (fix or feature that would cause existing functionality to not work as expected)

[tomvothecoder]

Hi @lee1043 and @gleckler1, this PR fixes the issue where multiple temporal averaging calls on the same dataset object breaks.

This is what happens:

import xcdat
infile = "/p/css03/esgf_publish/CMIP6/CMIP/CSIRO-ARCCSS/ACCESS-CM2/historical/r1i1p1f1/Amon/ts/gn/v20191108/ts_Amon_ACCESS-CM2_historical_r1i1p1f1_gn_185001-201412.nc"
d = xcdat.open_dataset(lf, data_var=var)

# `self._time_bounds is initialized in the `.temporal` class
d_ac = d.temporal.climatology(var, freq="month", weighted=True)  # okay

# `self._time_bounds is reused since it is initialized already, and now have been adjusted to drop incomplete DJF seasons
d_clim = d.temporal.climatology(var, freq="season", weighted=True, season_config={"dec_mode": "DJF", "drop_incomplete_djf": True},)  # okay

# `self._time_bounds is reused again, but with dropped incomplete DJF seasons. This results in mismatching bounds shape with the time coordinates.
d_ac = d.temporal.climatology(var, freq="month", weighted=True)  # error when repeat after get seasonal climatology

The solution: Initialize self._time_bounds every time temporal averaging methods to avoid reusing previously initialized values (which could have been modified).

[tomvothecoder]

I had this line here originally, but made the silly mistake of moving it to the class__init__ method in another PR which caused this bug.

[codecov-commenter]

Codecov Report

Merging #329 (b7d8b1f) into main (e38fe72) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##              main      #329   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           14        14           
  Lines         1191      1191           
=========================================
  Hits          1191      1191           

Impacted Files	Coverage Δ
xcdat/temporal.py	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[lee1043]

Thank you @tomvothecoder for the quick update. I am taking this PR a look for verifying followings:

• Minimal code from [Bug]: Error from climatology calculation when getting it multiple times #327 (comment)
• @gleckler1's example from convert pcmdi_compute_climatologies.py to pcmdi_compute_climatologies-xcdat.py PCMDI/pcmdi_metrics#859 (comment)
• Minimal code from [Bug]: error on climatology calculation for custom seasons if after running for default seasons #325 (comment)