Started by an SCM change7289be3d5fd401ed3393a0a251b98752e552010ab989be3d5fd401ed3393a0a251b98752e552010ab9refs/remotes/origin/master89be3d5fd401ed3393a0a251b98752e552010ab989be3d5fd401ed3393a0a251b98752e552010ab9refs/remotes/origin/master89be3d5fd401ed3393a0a251b98752e552010ab989be3d5fd401ed3393a0a251b98752e552010ab9refs/remotes/origin/masterhttps://github.com/larson-group/clubb.gitfalse#7216287743345594clubb_BFB_varying_flags_gfortran_test #7272false729002FAILURE1790670727411http://carson.math.uwm.edu/jenkins/view/all/job/clubb_BFB_varying_flags_gfortran_test/72/clubb_jax/src/CLUBB_core/precipitation_fraction.pysrc/CLUBB_core/precipitation_fraction.F90a4826465aeec11af80e51e711d1cc0b8287e96c01790609573000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comRemove upsilon 0 and 1 if branches (#1365) * Removed unnecessary upsilon_precip_frac_rat == 0 or 1 checks The read in parameter upsilon_precip_frac_rat is set in .in files. upsilon_precip_frac_rat describes the precipitation ratio between the first and second PDF component. The edge cases, all perticipitation in one PDF (upsilon=0 or 1) had their own if branches. These edge case branches are not strictly necessary as upsilon should never be set to 0 or 1 and we can live with a tiny precipitation fraction of precip_frac_tol in such a case. Furthermore a logic error was present which would never execute the if branch for the upsilon=0 case anyway. Per default upsilon is not set to 0 and afaik is not tested in the jenkins tests. However if upsilon is read in as 0 or 1 this commit would be technically BIT_CHANGING. A mirrored code removal in clubb-jax was also done. * changed comment about role of upsilon a bit 2026-09-28 10:32:53 -0500a4826465aeec11af80e51e711d1cc0b8287e96c0Remove upsilon 0 and 1 if branches (#1365)editsrc/CLUBB_core/precipitation_fraction.F90editclubb_jax/src/CLUBB_core/precipitation_fraction.pyjenkins_tests/clubb_restart_multicol_gfortran_test/Jenkinsfilesrc/clubb_driver.F90tests/run_restart_test.pyjenkins_tests/clubb_restart_gfortran_test/Jenkinsfile89be3d5fd401ed3393a0a251b98752e552010ab91790614619000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comRestart clubb 2d arr slicing fix (#1366) * Fixed 2d arr passing into restart_clubb subroutine which expects a 1d arr This only affects behaviour upon clubb restarts. Because restarts are more of a debug feature and disabled in all current "..._model.in" files I don't see the necessity to mark with bit changing. * Added multicol support to restart tests script Forwarding any extra cli options to the run_scm.py script which run_restart.py calls. E.g. -multicol * Changed the bomex jenkins restart test to run with multicol. Changed the jenkins test name 2026-09-28 11:56:59 -050089be3d5fd401ed3393a0a251b98752e552010ab9Restart clubb 2d arr slicing fix (#1366)deletejenkins_tests/clubb_restart_gfortran_test/Jenkinsfileeditsrc/clubb_driver.F90addjenkins_tests/clubb_restart_multicol_gfortran_test/Jenkinsfileedittests/run_restart_test.pygithttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse71http://carson.math.uwm.edu/jenkins/view/all/job/clubb_BFB_varying_flags_gfortran_test/71/