search

UMD-AOSC / UMD-SST

Testbed sea surface analysis system
Apache License 2.0
1 stars 2 forks source link

Finished Feature/dirac.x #43

Closed loganchen39 closed 3 years ago

loganchen39 commented 3 years ago

@travissluka Passed all tests. Some points below.

  1. I set 8 grid points which are across the domain for testing purpose, see dirac.yml.
  2. As I checked, there's no easy and direct mapping from our specification (e.g. (i=20, j=10)) to 1D local_index, we need to first compute its 1D global_index and then find the corresponding local_index by search. We could use more efficient search algorithm (e.g. binary search) in the future if certain conditions are met for sure, e.g. the array is increasing. For now I used the simplest linear search algo.
  3. The output netCDF file consists of two values, 273.15 and 274.15, which corresponds to 0 and 1. Bear it for now. The Fields::write() function converts from Celsius to Kelvin, as when we were implementing class State, its input netCDF file is in Kelvin, so the Fields::write() function has to output Kelvin as well, or it won't pass tests for class State. We may need to change it in the future, e.g. convert all of our SST netCDF files from Kelvin to Celsius.

close #35 .

travissluka commented 3 years ago

@loganchen39 otherwise, looks good :+1:, and i'll merge once the above issues are addressed so that the ctests pass

loganchen39 commented 3 years ago

@travissluka I've fixed these minor problems. But it seems the Cheyenne environment has been changed and I can not run tests or use some git commands (e.g. "git push"). Will let you know once it's solved. Merry Christmas and Happy New Year!

loganchen39 commented 3 years ago

@travissluka Mark just fixed the Cheyenne problem. I re-compiled and re-run the tests, it seems ok. Please try to approve the PR. Thanks,