On Fri, Aug 25, 2023 at 12:53:26AM +0000, Shinichiro Kawasaki wrote: > This works, but a bit ugly. Another idea is to make one of the optional > arguments mandatory, a positional argument. I think the option --device_type > worth making mandatory and explicit, like, > > _nvmet_target_setup device > _nvmet_target_setup file Possible but as I said in the other mail, we have a lot more of default arguments to functions which I would like to drop too. > This will make it easier for me to review which test case uses which type. > (This might be against Sagi's comment, though.) > > Daniel, what do you think? I don't think it is good strategy to change blktests just to make ShellCheck happy, when we know it is broken. It rather have SC2119 ignored until ShellCheck is fixed.
