Models where endpoints are defined after they are used

[john-harrold]

In this rxode2 issue it describes the problem. Essentially the endpoint is being defined after it's being used. Because of this the endpoint (Cc) is being flagged as a covariate.

nlmixr2/rxode2#829

There are a few models where this is happening. See the code below.

I'll fix them and submit a pull request over the weekend.

library(nlmixr2lib)
#> Warning: package 'nlmixr2lib' was built under R version 4.4.2

for(ridx in 1:nrow(modeldb)){

  fn = modeldb[ridx, ]$filename
  mn  = modeldb[ridx, ]$name
  ffn = file.path(system.file(package="nlmixr2lib", "modeldb"), fn)
  source(ffn)
  rx_fun = get(mn)
  rx_obj = rxode2::rxode2(rx_fun)

  if(length(rx_obj$allCovs) > 0){
    mod_ele = c(rx_obj$props$output$endpoint, rx_obj$props$output$secondary)

    if(any(rx_obj$allCovs %in% mod_ele)){
      badCovs = rx_obj$allCovs[rx_obj$allCovs %in% mod_ele]
      message(paste0("Bad covariate detected (", mn, "): ", badCovs))
    }
  }
}
#> Bad covariate detected (indirect_1cpt_inhi_kout_r0rmaxcrmax): Cc
#> Bad covariate detected (indirect_circ_1cpt_inhi_kin_kin_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_inhi_kin_kout_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_inhi_kout_kin_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_inhi_kout_kout_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_stim_kin_kin_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_stim_kin_kout_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_stim_kout_kin_t): Cc
#> Bad covariate detected (indirect_circ_1cpt_stim_kout_kout_t): Cc
#> Bad covariate detected (indirect_prec_1cpt_inhi_CLV): Cc
#> Bad covariate detected (indirect_prec_1cpt_inhi_r0rmaxcrmax): Cc
#> Bad covariate detected (indirect_prec_1cpt_stim_CLV): Cc
#> Bad covariate detected (indirect_prec_1cpt_stim_r0rmaxcrmax): Cc
#> ℹ parameter labels from comments will be replaced by 'label()'