Closed edwardhartnett closed 5 months ago
Hi, ipolates and ipolatev need a method for C to call the fortran routines. Much like the gdswiz_c routine. Wesley
On Sat, Mar 30, 2024 at 9:35 AM Edward Hartnett @.***> wrote:
@webisu https://github.com/webisu needs C wrappers for some more of the fortran subroutines, including ipolatesv() @.*** https://github.com/webisu what others?)
This is needed for wgrib2 to use NCEPLIBS-ip instead of an internal copy of the ip code.
@webisu https://github.com/webisu has suggested that he will submit a PR. He was concerned with the level of unit testing required; only the correct calling needs to be tested in such a PR, not the entire functionality of the ipolatesv() function. I told @webisu https://github.com/webisu that he could assume the Fortran code was already tested, and just writes a test to ensure the C code is correctly calling the Fortran code, and getting correct results.
— Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NCEPLIBS-ip/issues/235, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEIB7ZUKKO4CRRFUEQIFHY3Y225Q5AVCNFSM6AAAAABFPSDTICVHI2DSMVQWIX3LMV43ASLTON2WKOZSGIYTMNBVGU2TGNA . You are receiving this because you were assigned.Message ID: @.***>
@AlexanderRichert-NOAA we need to get some C wrappers into ip, then do a release, so that wgrib2 can use them and we can do a release of wgrib2.
@webisu do you have the wrappers written?
FYI @AlysonStahl-NOAA is working on wgrib2 as well now...
Hi, My mistake. I was looking at the ip2 library. Yesterday I was examining the ip library and they have different entry points but (ex ipolates_grib2_single_field) but the new entry points have bind(c) on them. So wgrib2 can use ipolates(s|v)_grib2_singlefield instead of current ipolate(s|v). Did one test with the new ip library. I seemed to work but there is something I don't understand about the build. Wesley
On Tue, Apr 16, 2024 at 4:33 PM Edward Hartnett @.***> wrote:
@AlexanderRichert-NOAA https://github.com/AlexanderRichert-NOAA we need to get some C wrappers into ip, then do a release, so that wgrib2 can use them and we can do a release of wgrib2.
@webisu https://github.com/webisu do you have the wrappers written?
FYI @AlysonStahl-NOAA https://github.com/AlysonStahl-NOAA is working on wgrib2 as well now...
— Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NCEPLIBS-ip/issues/235#issuecomment-2059499132, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEIB7ZR6Q4YRJK5M3QLUXGDY5VHGHAVCNFSM6AAAAABFPSDTICVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJZGQ4TSMJTGI . You are receiving this because you were mentioned.Message ID: @.***>
issue is closed, see previous comment
@webisu needs C wrappers for some more of the fortran subroutines, including ipolatesv() (@webisu what others?)
This is needed for wgrib2 to use NCEPLIBS-ip instead of an internal copy of the ip code.
@webisu has suggested that he will submit a PR. He was concerned with the level of unit testing required; only the correct calling needs to be tested in such a PR, not the entire functionality of the ipolatesv() function. I told @webisu that he could assume the Fortran code was already tested, and just writes a test to ensure the C code is correctly calling the Fortran code, and getting correct results.