Create multiple columns of the same type

[mgirlich]

It would be great to make it easier to define multiple columns of the same type to avoid repetition. For example something along the line

    library(vroom)

    fct_cols <- c("success", "attempts")
    yr_cols <- c("birthyr", "firstyr")
    mdy_cols <- c("birthdy")
    chr_cols <- c("name")

    cols_of <- function(col_format, names) {
      col_spec <- rep_len(list(col_format), length.out = length(names))
      setNames(col_spec, names)
    }

    col_spec_list <- c(
      cols_of(col_character(), chr_cols),
      cols_of(col_factor(), fct_cols),
      cols_of(col_date(format = "%Y"), yr_cols),
      cols_of(col_date(format = "%m/%d/%Y"), mdy_cols)
    )
    cols_only(!!!col_spec_list)
    #> cols_only(
    #>   name = col_character(),
    #>   success = col_factor(levels = NULL, ordered = FALSE, include_na = FALSE),
    #>   attempts = col_factor(levels = NULL, ordered = FALSE, include_na = FALSE),
    #>   birthyr = col_date(format = "%Y"),
    #>   firstyr = col_date(format = "%Y"),
    #>   birthdy = col_date(format = "%m/%d/%Y")
    #> )

<sup>Created on 2020-10-26 by the [reprex package](https://reprex.tidyverse.org) (v0.3.0)</sup>

It is already quite easy to do but it would be great to export a function for this or at least document it. If you like the idea I'm happy to add a PR (and one might also want to add that to readr).

PS: I discovered an issue for using the tidyselect helpers: https://github.com/tidyverse/readr/issues/237 which would also be very nice to have. Unfortunately, it doesn't seem like it is currently worked on. Is this still something on the roadmap?

[jennybc]

I will be thinking about colspec matters soon, so I'm leaving this open as an interesting idea and example. But no promises re: exactly what will happen.