I think it would be useful if ncdump could also be called with a string (a NetCDF file path, hopefully existing), instead of just a dataset, in order to avoid the following error:
>>> pmip_exp_anm = clim_average(pmip_exp_resolved, 'ANM')
>>> pmip_exp_anm
time_average(ds('CMIP6%%pr%1850-2399%global%/bdd%IPSL-CM6A-LR%IPSL%PMIP%Amon%midHolocene%r1i1p1f1%gr%latest'))
>>> d_anm = cfile(pmip_exp_anm)
>>> d_anm
'/data/jypmce/climafcache/a8/24b33e3d31c270a22d64f244bdafe3add84942c63819d1b775917b.nc'
>>> ncdump(d_anm)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "<string>", line 3, in ncdump
File "/ciclad-home/jservon/Evaluation/CliMAF/climaf_installs/climaf_V2.0.0/climaf/driver.py", line 69, in capply
res = capply_script(climaf_operator, *operands, **parameters)
File "/ciclad-home/jservon/Evaluation/CliMAF/climaf_installs/climaf_V2.0.0/climaf/driver.py", line 146, in capply_script
return maketree(script_name, script, *operands, **parameters)
File "/ciclad-home/jservon/Evaluation/CliMAF/climaf_installs/climaf_V2.0.0/climaf/driver.py", line 162, in maketree
defaultVariable = classes.varOf(operands[0])
File "/ciclad-home/jservon/Evaluation/CliMAF/climaf_installs/climaf_V2.0.0/climaf/classes.py", line 2148, in varOf
def varOf(cobject): return attributeOf(cobject, "variable")
File "/ciclad-home/jservon/Evaluation/CliMAF/climaf_installs/climaf_V2.0.0/climaf/classes.py", line 2200, in attributeOf
raise Climaf_Classes_Error("Unknown class for argument " + repr(cobject))
climaf.utils.Climaf_Classes_Error: "Unknown class for argument '/data/jypmce/climafcache/a8/24b33e3d31c270a22d64f244bdafe3add84942c63819d1b775917b.nc'"
>>> ncdump(pmip_exp_anm)
netcdf \24b33e3d31c270a22d64f244bdafe3add84942c63819d1b775917b {
dimensions:
time = UNLIMITED ; // (1 currently)
bnds = 2 ;
lon = 144 ;
lat = 143 ;
variables:
double time(time) ;
[...]
I think it would be useful if ncdump could also be called with a string (a NetCDF file path, hopefully existing), instead of just a dataset, in order to avoid the following error: