diff options
| author | felixcheung <felixcheung_m@hotmail.com> | 2015-10-03 22:42:36 -0700 |
|---|---|---|
| committer | Shivaram Venkataraman <shivaram@cs.berkeley.edu> | 2015-10-03 22:42:36 -0700 |
| commit | 721e8b5f35b230ff426c1757a9bdc1399fb19afa (patch) | |
| tree | e56ddb71163299a662b193425b2f67a8d154220b /R | |
| parent | ae6570ec2bf937e28bd1e7bada7813ac56a7b79d (diff) | |
| download | spark-721e8b5f35b230ff426c1757a9bdc1399fb19afa.tar.gz spark-721e8b5f35b230ff426c1757a9bdc1399fb19afa.tar.bz2 spark-721e8b5f35b230ff426c1757a9bdc1399fb19afa.zip | |
[SPARK-10904] [SPARKR] Fix to support `select(df, c("col1", "col2"))`
The fix is to coerce `c("a", "b")` into a list such that it could be serialized to call JVM with.
Author: felixcheung <felixcheung_m@hotmail.com>
Closes #8961 from felixcheung/rselect.
Diffstat (limited to 'R')
| -rw-r--r-- | R/pkg/R/DataFrame.R | 18 | ||||
| -rw-r--r-- | R/pkg/inst/tests/test_sparkSQL.R | 9 |
2 files changed, 21 insertions, 6 deletions
diff --git a/R/pkg/R/DataFrame.R b/R/pkg/R/DataFrame.R index 65e368c47d..14aea923fc 100644 --- a/R/pkg/R/DataFrame.R +++ b/R/pkg/R/DataFrame.R @@ -1075,12 +1075,20 @@ setMethod("subset", signature(x = "DataFrame"), #' select(df, c("col1", "col2")) #' select(df, list(df$name, df$age + 1)) #' # Similar to R data frames columns can also be selected using `$` -#' df$age +#' df[,df$age] #' } setMethod("select", signature(x = "DataFrame", col = "character"), function(x, col, ...) { - sdf <- callJMethod(x@sdf, "select", col, list(...)) - dataFrame(sdf) + if (length(col) > 1) { + if (length(list(...)) > 0) { + stop("To select multiple columns, use a character vector or list for col") + } + + select(x, as.list(col)) + } else { + sdf <- callJMethod(x@sdf, "select", col, list(...)) + dataFrame(sdf) + } }) #' @rdname select @@ -1853,13 +1861,13 @@ setMethod("crosstab", #' This function downloads the contents of a DataFrame into an R's data.frame. #' Since data.frames are held in memory, ensure that you have enough memory #' in your system to accommodate the contents. -#' +#' #' @title Download data from a DataFrame into a data.frame #' @param x a DataFrame #' @return a data.frame #' @rdname as.data.frame #' @examples \dontrun{ -#' +#' #' irisDF <- createDataFrame(sqlContext, iris) #' df <- as.data.frame(irisDF[irisDF$Species == "setosa", ]) #' } diff --git a/R/pkg/inst/tests/test_sparkSQL.R b/R/pkg/inst/tests/test_sparkSQL.R index 8f85eecbc4..faf42b7182 100644 --- a/R/pkg/inst/tests/test_sparkSQL.R +++ b/R/pkg/inst/tests/test_sparkSQL.R @@ -673,6 +673,13 @@ test_that("select with column", { expect_equal(columns(df3), c("x")) expect_equal(count(df3), 3) expect_equal(collect(select(df3, "x"))[[1, 1]], "x") + + df4 <- select(df, c("name", "age")) + expect_equal(columns(df4), c("name", "age")) + expect_equal(count(df4), 3) + + expect_error(select(df, c("name", "age"), "name"), + "To select multiple columns, use a character vector or list for col") }) test_that("subsetting", { @@ -1336,4 +1343,4 @@ test_that("Method as.data.frame as a synonym for collect()", { unlink(parquetPath) unlink(jsonPath) -unlink(jsonPathNa)
\ No newline at end of file +unlink(jsonPathNa) |