Currently DCA does little checking of user input for validity or match with the built executable model etc. This needs to improve even for expert users.
If working directory is not present hdf5 output crashes with vague error.
Model sections for models other than the one the executable is built for are silently ignored.
Model sections with name typos are silently ignored and model is run with defaults.
Numeric inputs are not checkedd to be numeric data resulting in crashes due to random numerical errors later.
Currently DCA does little checking of user input for validity or match with the built executable model etc. This needs to improve even for expert users.