[
https://issues.apache.org/jira/browse/BEAM-6810?focusedWorklogId=211890&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-211890
]
ASF GitHub Bot logged work on BEAM-6810:
----------------------------------------
Author: ASF GitHub Bot
Created on: 12/Mar/19 17:25
Start Date: 12/Mar/19 17:25
Worklog Time Spent: 10m
Work Description: amaliujia commented on issue #8035: [BEAM-6810] Disable
CalcRemoveRule to fix trivial projections
URL: https://github.com/apache/beam/pull/8035#issuecomment-472100139
LGTM. Thanks for finding this issue.
The issue is not because of BeamSQL as this rule is implemented in Calcite.
It might not be a bug. Maybe same other rules needed to help alias not dropped.
We are lacking of a rich sets of queries to verify and protect semantic
correctness for the long time. The rich query set should define what an
implementation behave if it uses Calcite as its SQL dialect. I have no idea if
there is such set exists.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 211890)
Time Spent: 0.5h (was: 20m)
> [SQL] Plan optimizer eliminates aliasing in trivial projections
> ---------------------------------------------------------------
>
> Key: BEAM-6810
> URL: https://issues.apache.org/jira/browse/BEAM-6810
> Project: Beam
> Issue Type: Bug
> Components: dsl-sql
> Affects Versions: 2.11.0
> Reporter: Gleb Kanterov
> Assignee: Gleb Kanterov
> Priority: Major
> Fix For: 2.12.0
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> Trivial programs project precisely their input fields, without dropping or
> re-ordering them. The simplest example is:
> {code}
> SELECT * FROM PCOLLECTION
> {code}
> Given input PCollection with a schema `c_int64: BIGINT` following query
> should result in PCollection with a schema `abc: BIGINT`, however, due to a
> bug in query optimization, it results in `c_int64: BIGINT`
> {code}
> SELECT c_int64 as abc FROM PCOLLECTION
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
