[
https://issues.apache.org/jira/browse/CASSANDRA-3157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13100344#comment-13100344
]
Sylvain Lebresne commented on CASSANDRA-3157:
---------------------------------------------
bq. is while-on-newline really an improvement?
Felt more coherent with the coding style to have the bracket on its own line.
But I don't have to commit it.
> After a "short read", the wrong read command may be used
> --------------------------------------------------------
>
> Key: CASSANDRA-3157
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3157
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.0
> Reporter: Sylvain Lebresne
> Assignee: Sylvain Lebresne
> Priority: Minor
> Fix For: 1.0
>
> Attachments: 3157.patch
>
>
> In fetchRows, there is this code:
> {noformat}
> for (int i = 0; i < commandsToSend.size(); i++)
> {
> ReadCallback<Row> handler = readCallbacks.get(i);
> ReadCommand command = commands.get(i);
> {noformat}
> On the first iteration of fetchRows, commands == commandsToSend so this is
> ok, but on a short read, commandsToSend will only contain the command to
> retry so we'll pick up the wrong command on the last line.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
