I tried providing the 2015_PHD_VERHELST_COD deposit at https://doi.org/10.14284/435 to write_dwc(). But the columns in our tables have changed slightly since, such as adding tag_serial_number in animals, which is not present in the deposited data.
Hence, I got:
Error in `dplyr::select()` at etn/R/create_animals_occurrence.R:67:3:
! Can't select columns that don't exist.
✖ Column `tag_serial_number` doesn't exist.
Run `rlang::last_trace()` to see where the error occurred.
Warning message:
One or more parsing issues, call `problems()` on your data frame for details,
e.g.:
dat <- vroom(...)
problems(dat)
I think it would be good to capture required columns and provide a nice error. In movepub, we do this with check_ref() and check_gps(), see https://github.com/inbo/movepub/blob/main/R/check_ref.R
Adding on function per table would add quite a lot of functions, so I would opt for a generic check_table(name = "animals") and call it from write_dwc() (cf. movepub).
This is an enhancement, the current functionality is ok, just a bit hard to debug for the user.
I tried providing the 2015_PHD_VERHELST_COD deposit at https://doi.org/10.14284/435 to
write_dwc(). But the columns in our tables have changed slightly since, such as addingtag_serial_numberin animals, which is not present in the deposited data.Hence, I got:
I think it would be good to capture required columns and provide a nice error. In movepub, we do this with
check_ref()andcheck_gps(), see https://github.com/inbo/movepub/blob/main/R/check_ref.RAdding on function per table would add quite a lot of functions, so I would opt for a generic
check_table(name = "animals")and call it fromwrite_dwc()(cf. movepub).This is an enhancement, the current functionality is ok, just a bit hard to debug for the user.