master update from dtc/develop 2019/12/17 #246
Conversation
… name in suite definition file
Bugfix #2 for travis test in schemes/check/ccpp_prebuild_config.py
…s instead of auto-detecting them
…fs-weather-model/gmtb-scm
…e_unit_conversion Bugfix libxml2 library location macOS, logging message for unit conversions, bugfix for metadata2html.py
…nd basedir/builddir argument, use default value from CCPP prebuild config if not provided on the command line (ccpp_prebuild.py), always use default value from CCPP prebuild config (metadata2html.py)
dtc/develop: update logic around builddir (ccpp_prebuild.py) / basedir (metadata2html.py)
…necessary files for static mode
…_mode dtc/develop: compile only required source files in static mode
climbfuji
requested review from
gold2718,
grantfirl and
llpcarson
as code owners
| @@ -183,7 +183,7 @@ class API(object): | |||
| {suite_switch} | |||
| else | |||
|
|
|||
| write({ccpp_var_name}%errmsg,'(*(a))'), 'Invalid suite ' // trim(suite_name) | |||
| write({ccpp_var_name}%errmsg,'(*(a))') 'Invalid suite ' // trim(suite_name) | |||
There was a problem hiding this comment.
I do not understand the asterisk here. It appears to be something l like a 'general repeat count' but my reading of the standard says that repeat counts are always positive literal integers. On the other hand, Nag accepts this syntax, even for strict Fortran 2003 processing so I must be missing something. Do you have a format description of this syntax anywhere?
There was a problem hiding this comment.
No, I don't. I have seen the same as you, it works with all compilers I have tested and with N=1...MANY string variables.
There was a problem hiding this comment.
Right. Nag is the gold standard when it comes to being really picky about Fortran standards so the fact that it accepts this syntax means I am missing something. I have to go find a Fortran uber-expert . . .
This PR contains:
Associated PRs:
NOAA-EMC/GFDL_atmos_cubed_sphere#11
#246
NCAR/ccpp-physics#372
NOAA-EMC/fv3atm#30
ufs-community/ufs-weather-model#18
For regression testing, see ufs-community/ufs-weather-model#18.