[
https://issues.apache.org/jira/browse/DERBY-2465?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Suresh Thalamati updated DERBY-2465:
------------------------------------
Attachment: derby.log
derby2465_v1.diff
Following the similar code in the SQLBlob data type, I made small changes to
SQLClob.java as in the attached patch to make clob datatype read using
getCharacterStream().
When I ran the tests, lang/TriggerTest Failied with following error:
1) testTypesInActionStatement(org.apache.derbyTesting.functionTests.tests.lang.T
riggerTest)ERROR XSDA7: Restore of a serializable or SQLData object of class , a
ttempted to read more data than was originally stored
I wonder if I am doing something stupid or just hit a bug ?
> Clob data type should use getCharacterStream() to read a column data from a
> VTI instead of getString().
> ---------------------------------------------------------------------------------------------------------
>
> Key: DERBY-2465
> URL: https://issues.apache.org/jira/browse/DERBY-2465
> Project: Derby
> Issue Type: Improvement
> Affects Versions: 10.3.0.0
> Reporter: Suresh Thalamati
> Attachments: derby.log, derby2465_v1.diff
>
>
> Currently clob data is read as String. This can consume lot of memory when
> lot
> of rows read from a VTI resultset. I think it would consume less memory if
> data is read using streams.
> org.apache.derby.iapi.types.SQLClob.java does not implement
> setValueFromResultSet() that is used by the VTI to read the data, it defaults
> to super class SQLChar.java implementation, which reads the data using
> getString().
> One case I noticed the current implementation uses lot of memory is when
> doing import of clobs(DERBY-378).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
