[JBoss JIRA] (TEIID-3817) HANA FLOOR and CELING functions not applicable to boolean
by Jan Stastny (JIRA)
Jan Stastny created TEIID-3817:
----------------------------------
Summary: HANA FLOOR and CELING functions not applicable to boolean
Key: TEIID-3817
URL: https://issues.jboss.org/browse/TEIID-3817
Project: Teiid
Issue Type: Bug
Components: Misc. Connectors
Affects Versions: 8.7.1.6_2
Reporter: Jan Stastny
Assignee: Steven Hawkins
SAP HANA doesn't support FLOOR and CEILING functions for BOOLEAN type.
Compared to postgresql, which also has BOOLEAN data type, it seems inconsistent.
When querying postgres directly:
{code:sql}
SELECT CEILING(BooleanValue) FROM BQT1.SmallA
{code}
is not valid query.
But through Teiid, it returns double values corresponding to the boolean value provided:
0.0 for CEILING(false)
1.0 for CELING(true)
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
9 years, 1 month
[JBoss JIRA] (TEIID-3805) HANA SUBSTRING function inconsistent with teiid behaviour
by Steven Hawkins (JIRA)
[ https://issues.jboss.org/browse/TEIID-3805?page=com.atlassian.jira.plugin... ]
Steven Hawkins resolved TEIID-3805.
-----------------------------------
Resolution: Deferred
Added a release note and updated the docs, will mark the issue as deferred until it's needed to be addressed by a customer case.
> HANA SUBSTRING function inconsistent with teiid behaviour
> ---------------------------------------------------------
>
> Key: TEIID-3805
> URL: https://issues.jboss.org/browse/TEIID-3805
> Project: Teiid
> Issue Type: Bug
> Components: Misc. Connectors
> Affects Versions: 8.7.1.6_2
> Reporter: Jan Stastny
> Assignee: Steven Hawkins
>
> When querying HANA using hana translator, there is an inconsistency in handling SUBSTRING with similar parameters:
> {code:sql}
> SELECT SUBSTRING('sth',20,1) FROM tablename
> {code}
> That is, when the starting offset supplied to SUBSTRING function is greater than the length of the string value, Teiid returns NULL in this case.
> But when I run:
> {code:sql}
> SELECT SUBSTRING(STRINGNUM,20,1) FROM bqt1.smalla
> {code}
> (it is pushed to HANA and STRINGNUM column contains strings of length <=2)
> I get empty strings instead of NULL values, which is default in Teiid.
> This inconsistency is caused by different handling of such situation on HANA's side.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
9 years, 1 month
[JBoss JIRA] (TEIID-3812) HANA MIN and MAX functions not applicable on boolean
by Steven Hawkins (JIRA)
[ https://issues.jboss.org/browse/TEIID-3812?page=com.atlassian.jira.plugin... ]
Steven Hawkins commented on TEIID-3812:
---------------------------------------
Floor/Ceiling expect the argument to be a double, so that would be a different issue. Can you log that separately?
> HANA MIN and MAX functions not applicable on boolean
> ----------------------------------------------------
>
> Key: TEIID-3812
> URL: https://issues.jboss.org/browse/TEIID-3812
> Project: Teiid
> Issue Type: Bug
> Components: Misc. Connectors
> Affects Versions: 8.7.1.6_2
> Reporter: Jan Stastny
> Assignee: Steven Hawkins
>
> SAP HANA doesn't support MIN and MAX functions for BOOLEAN type.
> Compared to postgresql, which also has BOOLEAN data type, it seems inconsistent, as postgresql translator mimics MIN and MAX functions by translating them using PostgreSQL specific constructs:
> min(BooleanValue) as bool_and(BooleanValue)
> max(BooleanValue) as bool_or(BooleanValue)
> Similar issue is with CEILING and FLOOR functions.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
9 years, 1 month
[JBoss JIRA] (TEIID-3796) It is much more slow to get the table record through TEIID from postgresql DB
by Steven Hawkins (JIRA)
[ https://issues.jboss.org/browse/TEIID-3796?page=com.atlassian.jira.plugin... ]
Steven Hawkins commented on TEIID-3796:
---------------------------------------
Is there any update on the details here? If not we'll resolve as incomplete description for now.
> It is much more slow to get the table record through TEIID from postgresql DB
> -----------------------------------------------------------------------------
>
> Key: TEIID-3796
> URL: https://issues.jboss.org/browse/TEIID-3796
> Project: Teiid
> Issue Type: Bug
> Reporter: Xian Liu
> Assignee: Steven Hawkins
>
> There is a table about 1000K+ record
> Postgresql Version: server 9.2.7
> Postgresql DB JDBC Driver Version: postgresql-9.3-1103.jdbc4.jar
> The table definition:
> {code}
> | Type | Modifiers
> ------------+-----------------------------+------------------------------------------------------------
> | integer | not null
> | integer |
> | integer | not null
> | timestamp without time zone | not null
> | integer | not null
> | text |
> | text |
> | bigint |
> | bigint | not null
> | integer |
> {code}
> I set driver connection parameter to disableLocalTxn=true;FetchSize=10;
> 1. When I make this table available in Mysql, and select * from it in Teiid, It will quickly get the data stream after around 30s.
> 2. But when I make this table in Postgresql DB, and query through Teiid, it is so slow and will get about 8min+ client can not get the data stream,
> when I try to directly connect this table in Postgresql table, it is much more faster, I do not know the reason.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
9 years, 1 month