Merge pull request #1172 from DavidArenburg/test_branch

Edited data.table.R in order to fix setDT ignoring key for data.table…

R/data.table.R

@@ -2459,6 +2459,7 @@ setDT <- function(x, keep.rownames=FALSE, key=NULL) {
     if (is.data.table(x)) {
         # fix for #1078 and #1128, see .resetclass() for explanation.
         setattr(x, 'class', .resetclass(x, 'data.table'))
+        if (!is.null(key)) setkeyv(x, key) # fix for #1169
         if (selfrefok(x) > 0) return(invisible(x)) else alloc.col(x)
     } else if (is.data.frame(x)) {
         rn = if (!identical(keep.rownames, FALSE)) rownames(x) else NULL

README.md

@@ -207,6 +207,8 @@
   53. Internal function `fastmean()` retains column attributes. Closes [#1160](https://github.com/Rdatatable/data.table/issues/1160). Thanks to @renkun-ken.
 
   54. Using `.N` in `i`, for e.g., `DT[, head(.SD, 3)[1:(.N-1L)]]`  accessed incorrect value of `.N`. This is now fixed. Closes [#1145](https://github.com/Rdatatable/data.table/issues/1145). Thanks to @claytonstanley.
+
+  55. `setDT` is now setting keys on existing `data.table` object too. Closes [#1169](https://github.com/Rdatatable/data.table/issues/1169). Thanks to @DavidArenburg for the PR.
 
 #### NOTES
 

inst/tests/tests.Rraw

@@ -6415,6 +6415,12 @@ x = c(1, 2, 1)
 y = c(5, 8, 8, 4)
 test(1525, CJ(x, y, unique=TRUE), CJ(c(1,2), c(4,5,8)))
 
+# `key` argument is ignored within `setDT` if `DT` is already a `data.table` #1169
+DT <- data.table(A = 1:4, B = 5:8)
+setDT(DT, key = "A")
+test(1526, key(DT), "A")
+
+
 ##########################