[
https://issues.apache.org/jira/browse/ZOOKEEPER-1836?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dutch T. Meyer updated ZOOKEEPER-1836:
--------------------------------------
Attachment: ZOOKEEPER-1836.patch
Quick fix to an obvious code defect - unit test seems unnecessary to me.
> addrvec_next() fails to set next parameter if addrvec_hasnext() returns false
> -----------------------------------------------------------------------------
>
> Key: ZOOKEEPER-1836
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1836
> Project: ZooKeeper
> Issue Type: Bug
> Components: c client
> Reporter: Dutch T. Meyer
> Priority: Trivial
> Attachments: ZOOKEEPER-1836.patch
>
>
> There is a relatively innocuous but useless pointer assignment in
> addrvec_next():
> 195 void addrvec_next(addrvec_t *avec, struct sockaddr_storage *next)
> ....
> 203 if (!addrvec_hasnext(avec))
> 204 {
> 205 next = NULL;
> 206 return;
> That assignment on (205) has no point, as next is a local variable lost upon
> function return. Likely this should be a memset to zero out the actual
> parameter.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
