[
https://issues.apache.org/activemq/browse/CAMEL-618?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=43589#action_43589
]
Hadrian Zbarcea commented on CAMEL-618:
---------------------------------------
Thanks for riding the camel and for the patch!
We cannot apply patches if explicit approval was not granted by the author.
See the "Submitting patches"
[here|http://activemq.apache.org/camel/contributing.html]. Other than that I
reviewed the patch and it looks good. If you want this patch applied, say for
users who will still use the unpatched oracle versions, please grant us
license, otherwise you can close this issue with a 'won't fix'.
> Oracle AQ contains a typo "JMSXRecvTimeStamp" -- the JMS spec lists the
> correct spelling as "JMSXRcvTimeStamp". Camel's JmsBinding also has the
> correct spelling, but can be patched to prevent Oracle AQ from puking.
> -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CAMEL-618
> URL: https://issues.apache.org/activemq/browse/CAMEL-618
> Project: Apache Camel
> Issue Type: Bug
> Components: camel-jms
> Affects Versions: 1.3.0
> Environment: Windows XP, OC4J standalone Release 10.1.330, Oracle
> Database 10g Release 10.2.1.0
> Reporter: Seon Lee
> Attachments: oraclebugs.diff
>
>
> http://www.nabble.com/%28Fixed-%29-Updated---Help-w--Oracle-JMS-and-AQjmsException-%28JMS-147%29-td17989368s22882.html
> Oracle's Message implementation (AQjmsMessage) has a typo, I am pretty
> certain that it exists as "JMSXRecvTimeStamp". To workaround this typo I
> patched org.apache.camel.component.jms.JMSBinding to include the misspelled
> property name in set of ignored headers/properties.
> Additionally, I believe Oracle might be setting null values for empty or
> missing standard and optional JMS reserved properties/headers. Most notably
> JMSReplyTo. This is resulting in JMSBinding attempting to set a null
> Destination on the JMSReplyTo header which then results in an Oracle
> exception. A quick patch was to check for null values and ignore them in the
> method JmsBinding.appendJmsProperties() for the JMSReplyTo header. A better
> fix is probably to prevent empty/blank headers or properties from originating
> at the Oracle level... assuming nulls/blanks are not allowed by the JMS 1.1
> spec.
> Anyhoo, I will attach some diff patches to this issue for your review.
> Cheers!
> Seon
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
