Base date should be a tuple of ints and units should be in days.
The original code was OK but got units of time[:] changed to hours in May 2024 so this is a regression issue. Comparing to a previous download (say April 1 2024) would be an acceptable test. It would be good to test within schism using wind speed and comparing to a known station.
This bug has the potential to "silently succeed but be wrong".
There is a data component: archived files in Modeling_Data and on hpc5 will have to be overwritten.
Base date should be a tuple of ints and units should be in days.
The original code was OK but got units of time[:] changed to hours in May 2024 so this is a regression issue. Comparing to a previous download (say April 1 2024) would be an acceptable test. It would be good to test within schism using wind speed and comparing to a known station.
This bug has the potential to "silently succeed but be wrong".
There is a data component: archived files in Modeling_Data and on hpc5 will have to be overwritten.