added fct_lump_inorder() function

NAMESPACE

@@ -17,6 +17,7 @@ export(fct_infreq)
 export(fct_inorder)
 export(fct_inseq)
 export(fct_lump)
+export(fct_lump_inorder)
 export(fct_lump_min)
 export(fct_match)
 export(fct_other)

R/lump_inorder.R

@@ -0,0 +1,52 @@
+#' Lump together into "other" levels of a factor appearing after the N first (or prior to the N last)
+#'
+#' @param f A factor (or character vector).
+#' @param n
+#'   If `n` is missing, `fct_lump_inorder` lumps
+#'   together all but the first appearing level into "other".
+#'
+#'   Positive `n` preserves the first appearing `n` levels.
+#'   Negative `n` preserves the last appearing `n` levels.
+#'
+#' @param other_level Value of level used for "other" values.
+#' @export
+#' @seealso [fct_other()] to convert specified levels to other.
+#' @seealso [fct_lump()] to convert low-frequency levels to other.
+#' @examples
+#' library(magrittr)
+#' library(forcats)
+#' f <- c("hello","goodbye","ciao","arrivederci","ciao","hello") %>% fct_inorder
+#' f
+#' #> [1] hello       goodbye     ciao        arrivederci
+#' #> [5] ciao        hello
+#' #> Levels: hello goodbye ciao arrivederci
+#' # Preserve the first two appearing levels:
+#' f %>% fct_lump_inorder(2)
+#' #> [1] hello       goodbye     Other       Other
+#' #> [5] Other        hello
+#' #>Levels: hello goodbye Other
+#' # Preserve the last two appearing levels, reversing order of levels:
+#' f %>% fct_lump_inorder(-2)
+#' #> [1] Other       Other     ciao       arrivederci
+#' #> [5] ciao        Other
+#' #>Levels: arrivederci ciao Other
+#'
+fct_lump_inorder <- function(f, n=1, other_level="Other") {
+  if (!is.factor(f)) f <- as.factor(f) # levels will be alphabetical
+  # if zero n or n is greater than f's levels, no lumping
+  if (n==0|(abs(n)>length(levels(f)))) return(f)
+  f_io <- fct_inorder(f) # levels ordered by appearance
+  if(n>0) {
+    levs <- levels(f_io)
+    fct_other(f_io,
+              keep=levs[1:n],
+              other_level=other_level)
+  }
+  else {
+    f_io <- fct_rev(f_io) # reverses level order
+    levs <- levels(f_io)
+    fct_other(f_io,
+              keep=levs[1:abs(n)], # n is negative
+              other_level=other_level)
+  }
+}

README.Rmd

@@ -27,7 +27,8 @@ R uses __factors__ to handle categorical variables, variables that have a fixed
 * `fct_reorder()`: Reordering a factor by another variable.
 * `fct_infreq()`: Reordering a factor by the frequency of values.
 * `fct_relevel()`: Changing the order of a factor by hand.
-* `fct_lump()`: Collapsing the least/most frequent values of a factor into "other".
+* `fct_lump()`: Collapsing the least/most frequent levels of a factor into "other".
+* `fct_lump_inorder():`: Collapsing the first/last appearing levels of a factor into "other".
 
 You can learn more about each of these in `vignette("forcats")`. If you're new to factors, the best place to start is the [chapter on factors](http://r4ds.had.co.nz/factors.html) in R for Data Science.