[jira] [Commented] (PHOENIX-1686) Data length should be checked in KeyValueSchema.next() and maxOffset should be set correctly by the caller

["Hudson (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/PHOENIX-1686?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14340954#comment-14340954 ] 

Hudson commented on PHOENIX-1686:
---------------------------------

SUCCESS: Integrated in Phoenix-master #601 (See [https://builds.apache.org/job/Phoenix-master/601/])
PHOENIX-1686 Data length should be checked in KeyValueSchema.next() and maxOffset should be set correctly by the caller (maryannxue: rev 93f560575319b7f26f5fc5db618bc2d8a09be930)
* phoenix-core/src/main/java/org/apache/phoenix/schema/KeyValueSchema.java
* phoenix-core/src/main/java/org/apache/phoenix/expression/ProjectedColumnExpression.java


> Data length should be checked in KeyValueSchema.next() and maxOffset should be set correctly by the caller
> ----------------------------------------------------------------------------------------------------------
>
>                 Key: PHOENIX-1686
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-1686
>             Project: Phoenix
>          Issue Type: Bug
>    Affects Versions: 4.2
>            Reporter: Maryann Xue
>            Priority: Minor
>             Fix For: 4.3
>
>         Attachments: 1686.patch
>
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> When decoding values from a KeyValueSchema data object, we should check that the length specified (either by its type of fixed length or by the length flag) does not exceed maxOffset.
> Meanwhile, maxOffset should not be the total length of the ImmutableBytesWritable object, but instead should be the total length minus the length of the tailing valueSetBit.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)