diff options
| author | Wenchen Fan <wenchen@databricks.com> | 2016-01-13 22:43:28 -0800 |
|---|---|---|
| committer | Reynold Xin <rxin@databricks.com> | 2016-01-13 22:43:28 -0800 |
| commit | 962e9bcf94da6f5134983f2bf1e56c5cd84f2bf7 (patch) | |
| tree | fa7174220efa51f56287d32bc82a379508ee4c17 /R/pkg/inst/tests | |
| parent | e2ae7bd046f6d8d6a375c2e81e5a51d7d78ca984 (diff) | |
| download | spark-962e9bcf94da6f5134983f2bf1e56c5cd84f2bf7.tar.gz spark-962e9bcf94da6f5134983f2bf1e56c5cd84f2bf7.tar.bz2 spark-962e9bcf94da6f5134983f2bf1e56c5cd84f2bf7.zip | |
[SPARK-12756][SQL] use hash expression in Exchange
This PR makes bucketing and exchange share one common hash algorithm, so that we can guarantee the data distribution is same between shuffle and bucketed data source, which enables us to only shuffle one side when join a bucketed table and a normal one.
This PR also fixes the tests that are broken by the new hash behaviour in shuffle.
Author: Wenchen Fan <wenchen@databricks.com>
Closes #10703 from cloud-fan/use-hash-expr-in-shuffle.
Diffstat (limited to 'R/pkg/inst/tests')
| -rw-r--r-- | R/pkg/inst/tests/testthat/test_sparkSQL.R | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/R/pkg/inst/tests/testthat/test_sparkSQL.R b/R/pkg/inst/tests/testthat/test_sparkSQL.R index 97625b94a0..40d5066a93 100644 --- a/R/pkg/inst/tests/testthat/test_sparkSQL.R +++ b/R/pkg/inst/tests/testthat/test_sparkSQL.R @@ -1173,7 +1173,7 @@ test_that("group by, agg functions", { expect_equal(3, count(mean(gd))) expect_equal(3, count(max(gd))) - expect_equal(30, collect(max(gd))[1, 2]) + expect_equal(30, collect(max(gd))[2, 2]) expect_equal(1, collect(count(gd))[1, 2]) mockLines2 <- c("{\"name\":\"ID1\", \"value\": \"10\"}", |