Closed timothyas closed 4 years ago
I will go ahead and merge this later once it passes since it 99% does not change behavior, except for:
output_freq_code
options for load_ecco_vars_from_mds
... not even sure what this is forread_bin *
functions passed an empty list rather than returning an error, which would then cause other conversion routines to give weird errors. I replaced some of these to actually raise an error at the source because it seemed more expected and easier to understand what went wrong.Merging #106 into master will increase coverage by
27.48%
. The diff coverage is85.04%
.
@@ Coverage Diff @@
## master #106 +/- ##
===========================================
+ Coverage 33.58% 61.06% +27.48%
===========================================
Files 17 17
Lines 2102 2060 -42
Branches 465 454 -11
===========================================
+ Hits 706 1258 +552
+ Misses 1350 735 -615
- Partials 46 67 +21
Impacted Files | Coverage Δ | |
---|---|---|
ecco_v4_py/tile_plot_proj.py | 53.25% <75.00%> (+45.56%) |
:arrow_up: |
ecco_v4_py/llc_array_conversion.py | 98.38% <76.19%> (+14.02%) |
:arrow_up: |
ecco_v4_py/read_bin_llc.py | 76.41% <80.00%> (+56.94%) |
:arrow_up: |
ecco_v4_py/ecco_utils.py | 65.59% <85.71%> (+50.11%) |
:arrow_up: |
ecco_v4_py/calc_meridional_trsp.py | 97.22% <100.00%> (+74.14%) |
:arrow_up: |
ecco_v4_py/calc_section_trsp.py | 97.97% <100.00%> (+80.50%) |
:arrow_up: |
ecco_v4_py/get_basin.py | 97.01% <100.00%> (+77.83%) |
:arrow_up: |
ecco_v4_py/plot_utils.py | 100.00% <100.00%> (+30.76%) |
:arrow_up: |
ecco_v4_py/read_bin_gen.py | 100.00% <100.00%> (+46.51%) |
:arrow_up: |
ecco_v4_py/vector_calc.py | 100.00% <100.00%> (+75.00%) |
:arrow_up: |
... and 15 more |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 9d75399...f7de7ad. Read the comment docs.
It's been a long time coming. I've made a pretty big effort to finally increase our testing and address #1, which I think is pretty important! I was able to increased coverage to a whopping ... 60% or so. I am essentially now testing everything except:
netcdf_product_generation
tile_io
and additionally only some of the options and and routines are covered in:
ecco_utils
resample_to_latlon
tile_plot
tile_plot_proj
:plot_global/plot_stereo
read_bin_llc
:load_ecco_vars_from_mds
because I wanted to at least cover the routines that I wrote or modified. I also rewrote the tests to take advantage of the hard work that the xmitgcm crew has put in, to download xmitgcm global_oce_llc90 data to run tests. Because of this, I dropped everything from our binary_data/ except for basins since it's the only thing we need. This will make the package almost 100MB lighter.
While at it, I addressed #103 to make this function accept arrays, or basically whatever anyone wants to send at it.