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
This is mainly for @bena-nasa to ponder. Today, he helped me try to figure out why my Spack CI tests here with MAPL were failing. Eventually, we figured out it was because the timeString bits of ESMF_TimeGet were returning garbage.
This was actually the @atrayanosprintf bug fixed by @billsacks in esmf-org/esmf#298. So when I finagled Spack to build ESMF 8.8.0 (which is always fun with a "not-quite-released" package 😄 ), all worked, see:
I agree. We have good code in several places in MAPL that handles date and time information. Some of it uses ESMF, and some of it uses our own date and time code. We should consolidate it in one place. My experience is that code that parsing strings is fragile, and so it should only be used when necessary. Further, it's best to consolidate it one place with variations together using a common core code.
This is mainly for @bena-nasa to ponder. Today, he helped me try to figure out why my Spack CI tests here with MAPL were failing. Eventually, we figured out it was because the timeString bits of ESMF_TimeGet were returning garbage.
This was actually the @atrayano
sprintf
bug fixed by @billsacks in esmf-org/esmf#298. So when I finagled Spack to build ESMF 8.8.0 (which is always fun with a "not-quite-released" package 😄 ), all worked, see:https://github.com/GEOS-ESM/MAPL/actions/runs/12697215473/job/35393142177?pr=3123
But, before @bena-nasa and I realized that, we actually figured out how to fix up bits of MAPL to let it work. And we saw some oddities. For example:
MAPL/gridcomps/History/MAPL_HistoryGridComp.F90
Lines 456 to 483 in 116fb6f
We realized that this could be simplfied to:
So, I wonder if there are places
timeString
is being used that are unnecessary?The text was updated successfully, but these errors were encountered: