[
https://issues.apache.org/jira/browse/HBASE-18621?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16137212#comment-16137212
]
Chia-Ping Tsai commented on HBASE-18621:
----------------------------------------
The changes looks like this.
{code}
enum Option {
HBASE_VERSION = 0;
LIVE_SERVERS = 1;
DEAD_SERVERS = 2;
REGIONS_STATE = 3;
MASTER_COPROCESSORS = 4;
MASTER = 5;
BACKUP_MASTERS = 6;
BALANCER_ON = 7;
}
message Options {
repeated Option options = 1;
}
{code}
As far as I know, the message size will increase because the *repeated* will
introduce a List object. But on the other hand, we can use Switch statement in
converting the proto's object to our object.
{code}
for (ClusterStatusProtos.Option option : options.getOptionsList()) {
switch (option) {
}
}
{code}
Personally, i prefer this coding style. This change is not, however, required.
> Refactor ClusterOptions before applying to code base
> ----------------------------------------------------
>
> Key: HBASE-18621
> URL: https://issues.apache.org/jira/browse/HBASE-18621
> Project: HBase
> Issue Type: Improvement
> Reporter: Reid Chan
> Assignee: Reid Chan
> Attachments: HBASE-18621.master.001.patch,
> HBASE-18621.master.002.patch, HBASE-18621.master.003.patch
>
>
> So far, ClusterStatus.Options is not so clean that can be applied to code
> base.
> Refactoring it before next move.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
