Inconsistent naming in argument patterns

[benjaminhlina]

Arguments either start with nblank (e.g., ntime which is now n_time) or they start with the object they are focused on recblank (e.g., recX which is now rec_x). This naming convention switches for extents which are ylim or y_lim. This is inconsistent and needs to adhere to a consistent style. IMO this should be the variable_object which would make rec_x be x_rec or is this confusing. Either way the inconsistency in arguments needs to be rectified.

[mhpob]

Let me have at this one if you haven't already started.

Two ways forward I see:

1. add ... as an argument, which will feed into a check function stored in R/utils.R for the old name and switch them with a warning.
2. move the old names to the back of the argument list so they're still accepted.

Do you have a preference between the two? I prefer the first option at this point since we can add the check function to utils and do the "change in one place" thing. and I worry that adding arguments to accommodate old names may make the help docs long and confusing since there are so many arguments.

[benjaminhlina]

I'm in favour of option 1 as that seems the easiest. As of today the naming-conventions fork all functions and tests use arguments that are snake_case. Maybe I did more work than needed here as I may not fully understand your suggestion, are you saying old names as in ntest or n_test.

[mhpob]

Something like:

(old)
function(ntest){}

vs (new)
function(n_test, ...){if(...) check_old_names()}

[benjaminhlina]

That sounds great to me

[benjaminhlina]

... will have to be mapped to those arguments as right now ... is mapped specifically to rstan::sampling() so it knows how to handle additional arguments for both.