Add lookup().

NAMESPACE

@@ -28,6 +28,7 @@ export(fieldforms_output)
 export(fieldforms_prepare)
 export(fill_na)
 export(hide_data_of_class)
+export(lookup)
 export(ls_csv_df)
 export(ls_join_df)
 export(ls_list_spreadsheets)

R/lookup.R

@@ -0,0 +1,43 @@
+#' `recode()` by looking up `old` and `new` values (from a lookup table).
+#'
+#' Use this function inside `dplyr::recode()` to recode a vector based on values
+#' from two other vectors, where `old` and `new` codes are looked up. These
+#' lookup vectors are commonly stored in a dataframe and come from a .csv or
+#' spreadsheet file.
+#'
+#' @param old,new Vectors of equal length giving old and new codes.
+#'
+#' @seealso `dplyr::recode()`
+#'
+#' @return A "spliced" list with names from `old` and values from `new`. The
+#' kind of data structure that you can feed to `...` in dplyr::recode()`.
+#'
+#' @export
+#'
+#' @examples
+#' set.seed(1)
+#' library(dplyr, warn.conflicts = FALSE)
+#' library(rlang)
+#'
+#' look <- tibble(
+#'   old = c("spp1", "unknown"),
+#'   new = c("spp3", "spp4")
+#' )
+#'
+#' lookup(look$old, look$new)
+#'
+#' x <- c("spp1", "spp2", "spp3", "unknown", "spp3", "unknown", "spp1", "spp1")
+#' x
+#' recode(x, lookup(look$old, look$new))
+#' # Same
+#' recode(x, !!!as.list(set_names(look$new, look$old)))
+#'
+#' dfm <- tibble(x = x)
+#' mutate(dfm, new_x = recode(x, lookup(look$old, look$new)))
+lookup <- function(old, new) {
+  if (length(old) != length(new)) {
+    stop("`old` and `new` must be of equal length.")
+  }
+  lookup <- rlang::set_names(new, old)
+  rlang::splice(as.list(lookup))
+}

_pkgdown.yml

@@ -53,6 +53,7 @@ reference:
 
   - title: Manipulate dataframes
     contents:
+    - lookup
     - recode_subquad
 
   - title: Manipulate strings