between support missing bounds for character input, closes #3667

NEWS.md

@@ -198,6 +198,8 @@
 
 24. `column not found` could incorrectly occur in rare non-equi-join cases, [#3635](https://github.com/Rdatatable/data.table/issues/3635). Thanks to @UweBlock for the report.
 
+25. Function `between` and operator `%between%` now handles missing bounds for character input as documented, [#3667](https://github.com/Rdatatable/data.table/issues/3667). Thanks to @AnonymousBoba for the report.
+
 #### NOTES
 
 1. `rbindlist`'s `use.names="check"` now emits its message for automatic column names (`"V[0-9]+"`) too, [#3484](https://github.com/Rdatatable/data.table/pull/3484). See news item 5 of v1.12.2 below.

R/between.R

@@ -51,8 +51,23 @@ between = function(x,lower,upper,incbounds=TRUE) {
   } else {
     if (isTRUE(getOption("datatable.verbose"))) cat("optimised between not available for this data type, fallback to slow R routine\n")
     # now just for character input. TODO: support character between in Cbetween and remove this branch
-    if (incbounds) x>=lower & x<=upper
-    else x>lower & x<upper
+    # support NAs as missing bounds also for character #3667
+    lower_na = is.na(lower)
+    upper_na = is.na(upper)
+    if (lower_na || upper_na) {
+      if (lower_na && upper_na) {
+        rep(TRUE, length(x))
+      } else if (lower_na) {
+        if (incbounds) x<=upper else x<upper
+      } else if (upper_na) {
+        if (incbounds) x>=lower else x>lower
+      } else {
+        stop("internal error in between, all cases of lower/upper NAs for character should be handled already, please report") # nocov
+      }
+    } else {
+      if (incbounds) x>=lower & x<=upper
+      else x>lower & x<upper
+    }
   }
 }
 

inst/tests/tests.Rraw

@@ -15428,6 +15428,16 @@ test(2071.10, dcast(data.table(a=1, b=1, l=list(list(1))), a ~ b, value.var='l')
 test(2071.11, dcast(data.table(a = 1, b = 2, c = 3), a ~ b, value.var = 'c', fill = '2'),
      data.table(a=1, `2`=3, key='a'))
 
+# between NAs in character bounds #3667
+test(2072.01, letters %between% c("m", NA_character_), c(rep(FALSE, 12L), rep(TRUE, 14L)))
+test(2072.02, between(letters, "m", NA_character_, incbounds=TRUE), c(rep(FALSE, 12L), rep(TRUE, 14L)))
+test(2072.03, between(letters, "m", NA_character_, incbounds=FALSE), c(rep(FALSE, 13L), rep(TRUE, 13L)))
+test(2072.04, between(letters, NA_character_, "m", incbounds=TRUE), c(rep(TRUE, 13L), rep(FALSE, 13L)))
+test(2072.05, between(letters, NA_character_, "m", incbounds=FALSE), c(rep(TRUE, 12L), rep(FALSE, 14L)))
+test(2072.06, between(letters, NA_character_, NA_character_, incbounds=TRUE), rep(TRUE, 26L))
+test(2072.07, between(letters, NA_character_, NA_character_, incbounds=FALSE), rep(TRUE, 26L))
+
+
 ###################################
 #  Add new tests above this line  #
 ###################################