Cannot handle missing values

[ShannonS-3791]

I am running the historical_exchange_rates function for multiple currencies; back to 2010. While the function works fine most of the time, I occasionally go this I get this error:

Error in pmap():
ℹ In index: 1.
Caused by error in purrr::map_dbl():
ℹ In index: 1.
ℹ With name: 2010-01-01.
Caused by error in .x[[index]]:
! subscript out of bounds
Backtrace:
1. priceR::historical_exchange_rates(...)
13. purrr::map_dbl(...)
14. purrr:::map_("double", .x, .f, ..., .progress = .progress)
18. priceR (local) .f(.x[[i]], ...)

I investigated this using the code from git, and I believe the issue occurs when running the 'get_values' function when there is no data available for a certain period. The list created via fromJSON has a line with the date, but no additional piece with the data to apply the purrr::map_dbl to.

Thank you for your attention to this matter.

[ShannonS-3791]

I came up with a temporary fix on my end; and hoping my work can help. I'm not the best with apply/map, so I'm confident this (for loop) is not the most efficient code, but it gets to the root of the problem. I inserted the following code immediately after the dat <- endpoint %>% fromJSON step

dat_fix <- dat[[8]]
  
  rows_to_delete <- c()
  
  for (l in 1:length(dat_fix)) {
    if (length(dat_fix[[l]]) == 0) {
      rows_to_delete <- append(rows_to_delete,l)
    }
  }
  
  dat2 <- if(length(rows_to_delete)>0){
    dat_fix[-rows_to_delete]
  } else {dat_fix}

then in subsequent code, the dat[[8]] or response[[8]] had to be replaced with just dat2 or response