handles integer vector stored as matrix in i, closes #826

Author: jangorecki

R/data.table.R

@@ -546,7 +546,13 @@ chmatch2 <- function(x, table, nomatch=NA_integer_) {
             if (locked.N) lockBinding(".N", parent.frame())
         }
         if (remove.N) rm(list=".N", envir=parent.frame())
-        if (is.matrix(i)) stop("i is invalid type (matrix). Perhaps in future a 2 column matrix could return a list of elements of DT (in the spirit of A[B] in FAQ 2.14). Please let datatable-help know if you'd like this, or add your comments to FR #1611.")
+        if (is.matrix(i)) {
+            if (is.numeric(i) && ncol(i)==1L) { # #826 - subset DT on single integer vector stored as matrix
+                i = as.integer(i)
+            } else {
+                stop("i is invalid type (matrix). Perhaps in future a 2 column matrix could return a list of elements of DT (in the spirit of A[B] in FAQ 2.14). Please let datatable-help know if you'd like this, or add your comments to FR #657.")
+            }
+        }
         if (is.logical(i)) {
             if (notjoin) {
                 notjoin = FALSE

README.md

@@ -244,6 +244,9 @@
 
   65. `print.data.table` now accepts a `quote` argument defaulting to `FALSE` (_a la_ `print.data.frame` in `base`). This option surrounds all printed elements with quotes and, e.g., makes whitespace more evident. Closes [#1177](https://github.com/Rdatatable/data.table/issues/1177); thanks to @MichaelChirico for the PR.
 
+  66. `[.data.table` now accepts single column integer matrix in `i` argument the same way as data.frame. Closes [#826](https://github.com/Rdatatable/data.table/issues/826). Thanks to @jangorecki.
+
+
 #### NOTES
 
   1. Clearer explanation of what `duplicated()` does (borrowed from base). Thanks to @matthieugomez for pointing out. Closes [#872](https://github.com/Rdatatable/data.table/issues/872).

inst/tests/tests.Rraw

@@ -6878,6 +6878,17 @@ ans2 <- c("   s1 s2","1:  A  A","2:  B  B",
 test(1553.1, capture.output(print(DT1, quote = TRUE)), ans1)
 test(1553.2, capture.output(print(DT1)), ans2)
 
+# #826 - subset DT on single integer vector stored as matrix the same way as data.frame
+dt <- data.table(a=letters[1:10])
+idx <- c(2:4,7L,9:10)
+dim(idx) <- c(6L, 1L)
+dimnames(idx) <- list(NULL, "Resample1") # as in caret::createDataPartition
+test(1554.1, dt[idx], data.table(a=letters[idx]))
+test(1554.2, dt[-idx], data.table(a=letters[(1:10)[-idx]]))
+test(1554.3, dt[!idx], data.table(a=letters[(1:10)[-idx]]))
+test(1554.4, idx, structure(c(2L, 3L, 4L, 7L, 9L, 10L), .Dim = c(6L, 1L), .Dimnames = list(NULL, "Resample1")))
+
+
 ##########################
 
 

man/data.table.Rd

@@ -52,7 +52,7 @@ data.table(..., keep.rownames=FALSE, check.names=FALSE, key=NULL)
   \item{x}{ A \code{data.table}.
 
 }
-  \item{i}{ Integer, logical or character vector, expression of column names, \code{list} or \code{data.table}.
+  \item{i}{ Integer, logical or character vector, expression of column names, \code{list} or \code{data.table}, single column integer \code{matrix}.
 
   integer and logical vectors work the same way they do in \code{\link{[.data.frame}}. Other than \code{NA}s in logical \code{i} are treated as \code{FALSE} and a single \code{NA} logical is not recycled to match the number of rows, as it is in \code{[.data.frame}.