run_glm requires that the nml control file be named glm2.nml

[jsta]

I am interested in passing an arbitrarily named nml file ( glm2_foo.nml) to run_glm:

run_glm(sim_folder, args = "--nml glm2_foo.nml")

It seems that the GLM executable allows this:

$ glm --help

       ------------------------------------------------
       |  General Lake Model (GLM)   Version 2.0.4    |
       ------------------------------------------------
--help  : show this blurb
--nml <nmlfile> : get parameters from nmlfile
--xdisp : display temp/salt and selected others in x-window
--xdisp <plotsfile> : like --xdisp, but use <plotsfile> instead of plots.nml
--saveall : save plots to png files
--save-all-in-one : save all plots to png file

[jsta]

Might be difficult to otherwise validate the nml file without creating a circular dependency with glmtools...

[lawinslow]

Does this combo work?
run_glm(sim_folder, args = "--nml glm2_foo.nml")

[jsta]

It only works if you also have a "glm2.nml" in the same folder because of the first line in the run_glm source:

if(!file.exists(file.path(sim_folder, 'glm2.nml'))){
    stop('You must have a valid glm2.nml file in your sim_folder: ', sim_folder)
}

[lawinslow]

Ah, I think we did that to give the user more meaningful feedback. Otherwise they get a 127 return code or something and no other info.

We could add a parameter to that command which then either alters or skips that check. glm_file parameter or something like that. Then it would feed directly in to the paired --nml parameter.