Separate -run and -build

[rgerkin]

Currently -neuron -run will both compile the model files and run the model. However, in some cases I would only like to compile the model files (i.e. produce the entire generated x86_64 directory) but not run the model. Without -run I get a python script and some .mod files, but this appears insufficient to actually run the model. In pyNeuroML, pynml.py line 110 says:

                       -run\n'
                  '        compile NMODL files and run the main NEURON\n'
                  '        hoc file (Linux only currently)\n'

suggesting that -run is in fact doing both the compilation and the running.

If I understand this correctly, I would like to request that compilation and running be separated via a -build flag, so that I could compile but not run.

[JustasB]

@rgerkin as a workaround, under *NIX, you can compile the NEURON .mod files by running the nrnivmodl command in the folder with the mod files. nrnivmodl is located in the bin folder of the neuron install directory.

For more info and other OS's, see: https://www.neuron.yale.edu/neuron/faq#compilemod

[rgerkin]

@JustasB Thanks, I'll take a look.

[rgerkin]

@JustasB Actually, it appears that we are already doing that here.