You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 30, 2022. It is now read-only.
@mdpiper There are some variables that are conditionally allocated. After talking with Parker I think I'm going to just take them out of the BMI. However for the sake of argument would something like this work fine in a bmi function?
case('gw_upslope')
if(this%model%control_data%cascadegw_flag%value > 0) then !Mark will this conditional work here?
size = sizeof(this%model%model_simulation%groundwater%gw_upslope(1))
bmi_status = BMI_SUCCESS
else
size =-1
bmi_status = BMI_FAILURE
endif
The text was updated successfully, but these errors were encountered:
Thanks Mark, I'm adding @pnorton-usgs for the sake of conversation. In this case gw_upslope does not have the allocatable attribute so allocated() doesn't work. so using cascadegw_flag would be the test.
Aside: The cascade module is not implemented in PRMS6 yet and Parker suggests removing vars associated with the cascade module for now and perhaps that's the best solution. However if the cascade module is implemented we'd still need conditional to return the value.
Regardless of the aside - I'm wondering, fully acknowledging my Fortran limitations, if gw_upslope, and others like it should have an allocatable attribute? Intel's take is that unless a var needs to be represented as just a pointer, if the var can use ALLOCATABLE it should (https://software.intel.com/en-us/forums/intel-fortran-compiler/topic/276711). If I'm off base here I'm sorry and apologize for taking your time :)
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
@mdpiper There are some variables that are conditionally allocated. After talking with Parker I think I'm going to just take them out of the BMI. However for the sake of argument would something like this work fine in a bmi function?
The text was updated successfully, but these errors were encountered: