[
https://issues.apache.org/jira/browse/PHOENIX-2125?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14632201#comment-14632201
]
Hudson commented on PHOENIX-2125:
---------------------------------
SUCCESS: Integrated in Phoenix-master #843 (See
[https://builds.apache.org/job/Phoenix-master/843/])
PHOENIX-2125 ORDER BY on full PK on salted table does not work (samarth.jain:
rev b329e85b697575fcebcde9555c991038d14e4a3c)
* phoenix-core/src/main/java/org/apache/phoenix/execute/ScanPlan.java
* phoenix-core/src/test/java/org/apache/phoenix/compile/QueryCompilerTest.java
*
phoenix-core/src/test/java/org/apache/phoenix/expression/ArrayToStringFunctionTest.java
* phoenix-core/src/main/java/org/apache/phoenix/util/ScanUtil.java
> ORDER BY on full PK on salted table does not work
> -------------------------------------------------
>
> Key: PHOENIX-2125
> URL: https://issues.apache.org/jira/browse/PHOENIX-2125
> Project: Phoenix
> Issue Type: Bug
> Affects Versions: 4.5.0
> Reporter: Brian Esserlieu
> Assignee: Samarth Jain
> Fix For: 4.5.0
>
> Attachments: PHOENIX-2125.patch
>
>
> Ordering by the full PK on a salted table does not work
> REPRO:
> --create a salted table with 2 pk fields
> CREATE TABLE IF NOT EXISTS TEST_TABLE (
> PK1 CHAR(15) NOT NULL,
> PK2 CHAR(15) NOT NULL
> CONSTRAINT PK PRIMARY KEY
> (
> PK1,
> PK2
> )
> ) SALT_BUCKETS=16;
> --insert data
> UPSERT INTO TEST_TABLE (PK1, PK2) VALUES ('FOO','BAR1-oiwnefin');
> UPSERT INTO TEST_TABLE (PK1, PK2) VALUES ('FOO','BAR2-qoosdfpin');
> UPSERT INTO TEST_TABLE (PK1, PK2) VALUES ('FOO','BAR3*)&&)*(@)');
> UPSERT INTO TEST_TABLE (PK1, PK2) VALUES ('FOO','BAR4 .');
> --do a quick query to verify the data in field PK2 is not completely in order
> --(salting + round robin should hopefully have thrown at least 2 rows out of
> order)
> SELECT * FROM TEST_TABLE
> --running this query, you can see that the rows are now properly ordered
> SELECT * FROM TEST_TABLE ORDER BY PK2
> --this query is logically identical to the previous SELECT statement with
> ordering,
> -- but as you can see it's actually identical to the first SELECT statement
> (without ordering)
> SELECT * FROM TEST_TABLE ORDER BY PK1, PK2
> --clean up if you like
> DROP TABLE IF EXISTS TEST_TABLE CASCADE;
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
