Upadate of transit spectra calculator

[Kiefersv]

Hey all,

I updated the transit spectrum calculator. Instead of using a fixed longitudinal opening angle for all latitudes, it now uses a coordinate transformation to average over the full terminator grid cell, even a the pols.

I still need to adjust the tests, but the code is arleady tested and should work.

[Kiefersv]

I checked the tests and I am not sure if I broke the unit tests or not because to me they seem unrelated to my changes. @AaronDavidSchneider could you have a look at it?

[AaronDavidSchneider]

I raised this as an issue in the h5py repo. Lets see if they reply.

[AaronDavidSchneider]

I tried to fix it in this commit:
8fc113a

Edit: not working yet: I will fix it in the next days in #58

[codecov]

Codecov Report

Merging #57 (cd24e5f) into main (d462282) will increase coverage by 0.34%.
The diff coverage is 91.22%.

❗ Current head cd24e5f differs from pull request most recent head 4367309. Consider uploading reports for the commit 4367309 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #57      +/-   ##
==========================================
+ Coverage   78.30%   78.65%   +0.34%     
==========================================
  Files          14       14              
  Lines        1549     1593      +44     
==========================================
+ Hits         1213     1253      +40     
- Misses        336      340       +4     

Flag	Coverage Δ
unittests	78.65% <91.22%> (+0.34%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
gcm_toolkit/utils/interface.py	50.51% <91.22%> (+3.26%)	⬆️

[Kiefersv]

Got it to work! Thanks again @AaronDavidSchneider for your help with the h5py problems.

[AaronDavidSchneider]

looks good to me! Thanks @Kiefersv

[AaronDavidSchneider]

in an ideal world, we would add these hardcoded values in the test to the test configuration in test_gcmtools_common.py

[AaronDavidSchneider]

... but I am also fine keeping it like this.

[Kiefersv]

There is something wired in the test values, i will have closer look

[AaronDavidSchneider]

any update @Kiefersv?