search

GEOS-ESM / GOCART

GOCART Aerosol model including process library and framework interfaces (MAPL, NUOPC, and CCPP)
Apache License 2.0
15 stars 14 forks source link

Silence error messages in getChannel() and getWavelength(), and add header file in process library #249

Closed rmontuoro closed 8 months ago

rmontuoro commented 1 year ago

Resubmitting PR #248 against develop to address issue #247.

tclune commented 1 year ago

@amdasilva This PR exposes one of the downsides of having the process library not depending on MAPL. We lose some useful error handling utilities this way. At what point to we accept that CCPP is no longer something that needs support?

bbakernoaa commented 1 year ago

I really hope that we are not thinking of stopping support for this.

On Fri, Aug 4, 2023 at 11:52 AM Tom Clune @.***> wrote:

@amdasilva https://github.com/amdasilva This PR exposes one of the downsides of having the process library not depending on MAPL. We lose some useful error handling utilities this way. At what point to we accept that CCPP is no longer something that needs support?

— Reply to this email directly, view it on GitHub https://github.com/GEOS-ESM/GOCART/pull/249#issuecomment-1665837753, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFIUVN4HNOANWUWRT2N3QNLXTULDPANCNFSM6AAAAAA2Y2TPEU . You are receiving this because you are subscribed to this thread.Message ID: @.***>

--

Barry Baker

National Oceanic and Atmospheric Administration Air Resources Laboratory Physical Research Scientist Chemical Modeling and Emissions Group Leader NCWCP, R/ARL, Rm. 4204 5830 University Research Court College Park, Maryland 20740 Phone: ‪(301) 683-1395‬

tclune commented 1 year ago

@bbakernoaa Are you aware of any activity in that direction? I'm sure we're not changing anything without lots of review to be certain we are not closing any important doors.

bbakernoaa commented 1 year ago

Yes. There is going to be some work towards this in the near future.

Barry Baker

National Oceanic and Atmospheric Administration Air Resources Laboratory Physical Research Scientist Chemical Modeling and Emissions Group Leader NCWCP, R/ARL, Rm. 4204 5830 University Research Court College Park, Maryland 20740 Phone: ‪(301) 683-1395‬

On Fri, Aug 4, 2023 at 12:52 PM Tom Clune @.***> wrote:

@bbakernoaa https://github.com/bbakernoaa Are you aware of any activity in that direction? I'm sure we're not changing anything without lots of review to be certain we are not closing any important doors.

— Reply to this email directly, view it on GitHub https://github.com/GEOS-ESM/GOCART/pull/249#issuecomment-1665914327, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFIUVN76FGHJGBYH7YJVSKDXTUSFBANCNFSM6AAAAAA2Y2TPEU . You are receiving this because you were mentioned.Message ID: @.***>

tclune commented 1 year ago

@rmontuoro Unfortunately, it is not just the macros. There are a handful of procedures involved as well. We could create a "MAPL-lite" and make that a dependency. If we get frustrated enough, I suppose we could look into that.

I'd also like to use pflogger in that layer, but same issue ...

amdasilva commented 8 months ago

@amdasilva This PR exposes one of the downsides of having the process library not depending on MAPL. We lose some useful error handling utilities this way. At what point to we accept that CCPP is no longer something that needs support?

@tclune No, it is a bad idea to have any dependency on MAPL and ESMF. This defeats the purpose of isolating the ESMF code. Also, any simple offline utility that needs any of this functionally will need to depend on a gigantic library. Keep the Process Library self contained and basic.

tclune commented 8 months ago

I understand. Just illuminating the tradeoff. Dependencies are also bad. I hate shades of grey.