[
https://issues.apache.org/jira/browse/CALCITE-2299?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16512540#comment-16512540
]
Sergey Nuyanzin commented on CALCITE-2299:
------------------------------------------
currently faced with another issue relating to timestampadd and nanosecond
{code} tester.checkScalar(
"timestampadd(MICROSECOND, 2000000, timestamp '2016-02-24 12:42:25')",
"2016-02-24 12:42:27",
"TIMESTAMP(0) NOT NULL");{code} fails like
{noformat}java.lang.AssertionError:
Expected :[2016-02-24 12:42:27]
Actual :[2016-02-24 12:42:25]{noformat}
Unfortunately the provided patch from CALCITE-2281 does not help with any of
issues. As far as I understand the issue with precision is common for both
timestampadd and timestampdiff - not clear for me how to fix, may be later will
deep dive.
At the same time for my issue and for nanosecond/SQL_TSI_FRAC_SECOND I have a
fix in 2 parts
Avatica part: https://github.com/apache/calcite-avatica/pull/58
Calcite part: https://github.com/apache/calcite/pull/731
> TIMESTAMPADD(SQL_TSI_FRAC_SECOND) should be nanoseconds not microseconds
> ------------------------------------------------------------------------
>
> Key: CALCITE-2299
> URL: https://issues.apache.org/jira/browse/CALCITE-2299
> Project: Calcite
> Issue Type: Bug
> Components: core
> Affects Versions: 1.16.0
> Reporter: James Duong
> Assignee: Julian Hyde
> Priority: Minor
>
> When calling TIMESTAMPADD with the first parameter SQL_TSI_FRAC_SECOND and
> the JDBC escape sequence, this gets interpreted as microseconds when it
> should be interpreted as nanoseconds.
>
> From the ODBC spec on MSDN:
> [https://docs.microsoft.com/en-us/sql/odbc/reference/appendixes/time-date-and-interval-functions?view=sql-server-2017]
> 'where fractional seconds are expressed in billionths of a second.'
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
