real loop indexes used in mbpf

[dkokron]

GDIT is testing a static analysis tool called Codee (https://www.codee.com/). Codee flagged some issues with mgbf.

codee: error: failed to analyze '/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90'

error: Semantic errors in /lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90
/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90:67:12: error: Must have INTEGER type, but is REAL(4)
V(n,m)=ampl*(mgimax0+ng)
^
/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90:67:14: error: Must have INTEGER type, but is REAL(4)
V(n,m)=ampl(mgimax0+ng)
^
/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90:145:12: error: Must have INTEGER type, but is REAL(4)
V(l,n,m)=ampl(mgimax0+ng) +(l-1)incrm
^
/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90:145:14: error: Must have INTEGER type, but is REAL(4)
V(l,n,m)=ampl(mgimax0+ng) +(l-1)incrm
^
/lfs/h1/hpc/support/USER/Projects/GSI/develop/src/mgbf/mg_input.f90:145:16: error: Must have INTEGER type, but is REAL(4)
V(l,n,m)=ampl(mg*imax0+ng) +(l-1)*incrm

The l,n,m variables are loop indexes but they are declared.
real(i_kind):: ng,mg,L,m,n

[TingLei-NOAA]

@dkokron Thanks for this validation effort. The apparent fix is to use " integer(i_kind)" to replace "real(i_kind)".
I think you can direct make a PR for this or I can do it if you prefer.
Please let me know how you want to proceed. Thanks.
BTW: it is mgbf rather than "mbpf".

[dkokron]

@TingLei-NOAA Please submit the PR for this issue.

[dkokron]

I thought Ting was handling this issue.

…

On Thu, Jul 11, 2024 at 8:24 AM RussTreadon-NOAA ***@***.***> wrote: Assigned #757 <#757> to @dkokron <https://github.com/dkokron>. — Reply to this email directly, view it on GitHub <#757 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACODV2CPO2DYYG464F5UHODZL2BSVAVCNFSM6AAAAABJQBPWM2VHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJTGQ3TCOBQGU4TMMA> . You are receiving this because you were assigned.Message ID: ***@***.***>

[TingLei-NOAA]

@dkokron You identified the problem and created the issue and you are also the coauthor for that PR. Finally, if you are satisfied the solution when the PR is finally merged into the develop branch, you could close this issue as the assignee.

[TingLei-daprediction]

An update: the PR with the quick fix for the issue had been closed and will be re-opened pending on future mgbf upgrading (#760).
It is also noted the program causing the safety concern identified in this would only be used when the mgbf package is used as the stand-alone version using its own driver and initialization blocks(which would invoke this code).

[RussTreadon-NOAA]

PR #760 was closed because there was no timeline for working on the PR. This issue will remain open until the bug is fixed. Clearly the declaration of integer(i_kind) variables as real(i_kind) in src/mgbf/mg_input.f90 is wrong.