wt_tidy_species usage is confusing, have to download wt_get_species twice

[see24]

In the data wrangling vignette it recommends to first do an inner join between your data and the species table downloaded with wt_get_species and then to do wt_tidy_species to remove species that are not of interest. In fact this is required because if you don't have a species_class column in the data passed to wt_tidy_species you get an error.

Error in `dplyr::select()`:
! Can't subset columns that don't exist.
✖ Column `species_class` doesn't exist.

But then when you run wt_tidy_species it runs wt_get_species internally and re-downloads the same table. This isn't a big deal because it doesn't take long, but seems unnecessary.

It looks like the error above is just because if zerofill = TRUE there is a line to remove the species_class column which gives an error if it is not present. If that select statement was changed to select the desired columns or use select(-any_of(c("species_code", "species_class", ...) then there would be no need to download the species table and join it before running wt_tidy_species which perhaps is the intended process. If that's the case you could update the vignette since the join to the species table is not really necessary.

[see24]

It seems like the zerofill = TRUE option is not working very well in general. It is not filtering out anything for me. I think it is because there are columns in my data (like scientific_name) that are not removed in the select statement before unique so they are all being kept and then added back. It might work better to do dplyr::distinct on the columns that should uniquely id a visit.