diff options
author | Prashant Sharma <prashant.s@imaginea.com> | 2014-04-03 15:42:17 -0700 |
---|---|---|
committer | Matei Zaharia <matei@databricks.com> | 2014-04-03 15:42:17 -0700 |
commit | c1ea3afb516c204925259f0928dfb17d0fa89621 (patch) | |
tree | 7d8a3485101dd9e191d858058d40e11ec3a7461e /sql | |
parent | 5d1feda217d25616d190f9bb369664e57417cd45 (diff) | |
download | spark-c1ea3afb516c204925259f0928dfb17d0fa89621.tar.gz spark-c1ea3afb516c204925259f0928dfb17d0fa89621.tar.bz2 spark-c1ea3afb516c204925259f0928dfb17d0fa89621.zip |
Spark 1162 Implemented takeOrdered in pyspark.
Since python does not have a library for max heap and usual tricks like inverting values etc.. does not work for all cases.
We have our own implementation of max heap.
Author: Prashant Sharma <prashant.s@imaginea.com>
Closes #97 from ScrapCodes/SPARK-1162/pyspark-top-takeOrdered2 and squashes the following commits:
35f86ba [Prashant Sharma] code review
2b1124d [Prashant Sharma] fixed tests
e8a08e2 [Prashant Sharma] Code review comments.
49e6ba7 [Prashant Sharma] SPARK-1162 added takeOrdered to pyspark
Diffstat (limited to 'sql')
0 files changed, 0 insertions, 0 deletions