[
https://issues.apache.org/jira/browse/DERBY-2450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12483518
]
Anurag Shekhar commented on DERBY-2450:
---------------------------------------
This particular position method accepts a String so we are quite safe as
String can't grow beyond INTEGER.MAX_VALUE :). For larger search string
the other method position (Clob, Clob) should be used but only if the
Clob implementation support Clob of larger than INTEGER.MAX_VALUE.
Reusing the reader will be a great optimization but it's not possible
unless we modify the UTF8Reader to support mark and reset. 256 I
selected because the existing code was using it but changing it to
something larger is not a problem.Increasing the buffer size will
improve the performance when the Writer needs to be resettled back
previous read blocks. In other case it will add little value as the
Writer itself does a caching and reads in block from the underlying
stream and returns only the chars which it has in memory, subsequent
read will make it fetch another full block in memory.
There are good number of tests related to position in BlobClob4BlobTest
but some how they couldn't expose this bug. I suggest you to wait for
the new implementation of clob before you add new tests as it will be
nicer to know the implementation before writing the test so you can
write test which exercises the path which is ignored by the current tests.
> Clob.Position returning wrong value when operating on Reader
> ------------------------------------------------------------
>
> Key: DERBY-2450
> URL: https://issues.apache.org/jira/browse/DERBY-2450
> Project: Derby
> Issue Type: Bug
> Components: JDBC
> Affects Versions: 10.3.0.0
> Environment: all
> Reporter: Anurag Shekhar
> Assigned To: Anurag Shekhar
> Attachments: derby-2250.diff, PositionTest.java
>
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
