[
https://issues.apache.org/jira/browse/DRILL-5898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16219336#comment-16219336
]
Robert Hou commented on DRILL-5898:
-----------------------------------
Boaz, you are correct. The order shown in the output is not correct. The test
framework handles order by verification separately, so what you see is not the
order in which the results are displayed. We do this because on a distributed
cluster, if a query does not have an order by clause (majority of our tests),
then results can be returned in any order, and can differ from run to run. So
we can display the order of the results in various orders.
I ran the query through sqlline, and the order is displayed correctly. If we
detected an incorrect order, then a different error message would have appeared.
> Query returns columns in the wrong order
> ----------------------------------------
>
> Key: DRILL-5898
> URL: https://issues.apache.org/jira/browse/DRILL-5898
> Project: Apache Drill
> Issue Type: Bug
> Components: Execution - Relational Operators
> Affects Versions: 1.11.0
> Reporter: Robert Hou
> Assignee: Robert Hou
> Priority: Blocker
> Fix For: 1.12.0
>
>
> This is a regression. It worked with this commit:
> {noformat}
> f1d1945b3772bb782039fd6811e34a7de66441c8 DRILL-5582: C++ Client: [Threat
> Modeling] Drillbit may be spoofed by an attacker and this may lead to data
> being written to the attacker's target instead of Drillbit
> {noformat}
> It fails with this commit, although there are six commits total between the
> last good one and this one:
> {noformat}
> b0c4e0486d6d4620b04a1bb8198e959d433b4840 DRILL-5876: Use openssl profile
> to include netty-tcnative dependency with the platform specific classifier
> {noformat}
> Query is:
> {noformat}
> select * from
> dfs.`/drill/testdata/tpch100_dir_partitioned_50000files/lineitem` where
> dir0=2006 and dir1=12 and dir2=15 and l_discount=0.07 order by l_orderkey,
> l_extendedprice limit 10
> {noformat}
> Columns are returned in a different order. Here are the expected results:
> {noformat}
> foxes. furiously final ideas cajol 1994-05-27 0.07 1731.42 4
> F 653442 4965666.0 1.0 1994-06-23 A 1994-06-22
> NONE SHIP 215671 0.07 2006 12 15 (1 time(s))
> lly final account 1994-11-09 0.07 45881.78 3 F
> 653412 1.320809E7 46.0 1994-11-24 R 1994-11-08 TAKE
> BACK RETURN REG AIR 458104 0.08 2006 12 15 (1 time(s))
> the asymptotes 1997-12-29 0.07 60882.8 6 O 653413
> 1.4271413E7 44.0 1998-02-04 N 1998-01-20 DELIVER IN
> PERSON MAIL 21456 0.05 2006 12 15 (1 time(s))
> carefully a 1996-09-23 0.07 5381.88 2 O 653378
> 1.6702792E7 3.0 1996-11-14 N 1996-10-15 NONE REG
> AIR 952809 0.05 2006 12 15 (1 time(s))
> ly final requests. boldly ironic theo 1995-09-04 0.07 2019.94 2
> O 653380 2416094.0 2.0 1995-11-14 N 1995-10-18
> COLLECT COD FOB 166101 0.02 2006 12 15 (1 time(s))
> alongside of the even, e 1996-02-14 0.07 86140.32 2
> O 653409 5622872.0 48.0 1996-05-02 N 1996-04-22
> NONE SHIP 372888 0.04 2006 12 15 (1 time(s))
> es. regular instruct 1996-10-18 0.07 25194.0 1 O 653382
> 6048060.0 25.0 1996-08-29 N 1996-08-20 DELIVER IN
> PERSON AIR 798079 0.0 2006 12 15 (1 time(s))
> en package 1993-09-19 0.07 18718.32 2 F 653440
> 1.372054E7 12.0 1993-09-12 A 1993-09-09 DELIVER IN
> PERSON TRUCK 970554 0.0 2006 12 15 (1 time(s))
> ly regular deposits snooze. unusual, even 1998-01-18 0.07
> 12427.92 1 O 653413 2822631.0 8.0 1998-02-09
> N 1998-02-05 TAKE BACK RETURN REG AIR 322636 0.01
> 2006 12 15 (1 time(s))
> ironic ideas. bra 1996-10-13 0.07 64711.53 3 O
> 653383 6806672.0 41.0 1996-12-06 N 1996-11-10 TAKE
> BACK RETURN AIR 556691 0.01 2006 12 15 (1 time(s))
> {noformat}
> Here are the actual results:
> {noformat}
> 2006 12 15 653383 6806672 556691 3 41.0 64711.53
> 0.07 0.01 N O 1996-11-10 1996-10-13 1996-12-06
> TAKE BACK RETURN AIR ironic ideas. bra
> 2006 12 15 653378 16702792 952809 2 3.0 5381.88
> 0.07 0.05 N O 1996-10-15 1996-09-23 1996-11-14
> NONE REG AIR carefully a
> 2006 12 15 653380 2416094 166101 2 2.0 2019.94 0.07
> 0.02 N O 1995-10-18 1995-09-04 1995-11-14
> COLLECT COD FOB ly final requests. boldly ironic theo
> 2006 12 15 653413 2822631 322636 1 8.0 12427.92
> 0.07 0.01 N O 1998-02-05 1998-01-18 1998-02-09
> TAKE BACK RETURN REG AIR ly regular deposits snooze. unusual, even
> 2006 12 15 653382 6048060 798079 1 25.0 25194.0 0.07
> 0.0 N O 1996-08-20 1996-10-18 1996-08-29
> DELIVER IN PERSON AIR es. regular instruct
> 2006 12 15 653442 4965666 215671 4 1.0 1731.42 0.07
> 0.07 A F 1994-06-22 1994-05-27 1994-06-23 NONE
> SHIP foxes. furiously final ideas cajol
> 2006 12 15 653413 14271413 21456 6 44.0 60882.8
> 0.07 0.05 N O 1998-01-20 1997-12-29 1998-02-04
> DELIVER IN PERSON MAIL the asymptotes
> 2006 12 15 653412 13208090 458104 3 46.0
> 45881.78 0.07 0.08 R F 1994-11-08 1994-11-09
> 1994-11-24 TAKE BACK RETURN REG AIR lly final account
> 2006 12 15 653440 13720540 970554 2 12.0
> 18718.32 0.07 0.0 A F 1993-09-09 1993-09-19
> 1993-09-12 DELIVER IN PERSON TRUCK en package
> 2006 12 15 653409 5622872 372888 2 48.0 86140.32
> 0.07 0.04 N O 1996-04-22 1996-02-14 1996-05-02
> NONE SHIP alongside of the even, e
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
