[
https://issues.apache.org/jira/browse/PHOENIX-7630?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hari Krishna Dara resolved PHOENIX-7630.
----------------------------------------
Release Note: When executing an UPSERT query with ON DUPLICATE KEY, the
JDBC statement's upsert() and executeUpsert() APIs will result in a ResultSet
to retrieve the new or old (when used with IGNORE) state of the row. This will
facilitate an atomic update followed by a read.
Resolution: Fixed
> Standard JDBC support for UPSERT returning ResultSet
> ----------------------------------------------------
>
> Key: PHOENIX-7630
> URL: https://issues.apache.org/jira/browse/PHOENIX-7630
> Project: Phoenix
> Issue Type: New Feature
> Reporter: Hari Krishna Dara
> Assignee: Hari Krishna Dara
> Priority: Major
>
> PHOENIX-7462 and PHOENIX-7398 added support for executing an UPSERT with ON
> {{DUPLICATE KEY}} and get back a {{{}ResultSet{}}}, but to be able access the
> {{{}ResultSet{}}}, one has to currently go outside the JDBC API and execute a
> direct Phoenix API (PhoeixPreparedStatement.executeAtomicUpdateReturnRow()).
> This is not ideal as it adds a compile time dependency on Phoenix. Also, it
> assumes that {{ON DUPLICATE KEY}} can only be used with prepared statements.
> JDBC has an established pattern for executing UPDATE statements that return a
> result set which involves calling {{executeUpdate()}} or {{execute()}}
> followed by a call to {{{}getResultSet(){}}}.
> Phoenix should support this same usage pattern instead of forcing the use of
> non-JDBC specific API.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
