[
https://issues.apache.org/jira/browse/HBASE-16984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15682747#comment-15682747
]
stack commented on HBASE-16984:
-------------------------------
Add to getRawAsyncTable that it is for advanced users only on commit.
AsyncResultScanner needs a class comment. Add it on commit.
All looks good except these things that I've moaned about before...
{code]
103 /**
104 * Set timeout of a single operation in a scan, such as openScanner
and next. Will override the
105 * value {@code hbase.client.scanner.timeout.period} in configuration.
106 * <p>
107 * Generally a scan will never timeout after we add heartbeat support
unless the region is
108 * crashed. The {@code scanTimeout} works like the {@code
operationTimeout} for each single
109 * operation in a scan.
110 */
111 void setScanTimeout(long timeout, TimeUnit unit);
112
113 /**
114 * Get the timeout of a single operation in a scan.
115 */
116 long getScanTimeout(TimeUnit unit);
{code}
... would be cool if could set it per scan but if too hard ingore my moaning...
Is Scan only thing that is 'raw' when I get a RawAsyncTable? If so, is that
enough reason to have two AsyncTables? Should we rather just have a single
AsycnTable that returns a raw scan and a simple scan?
Is doScan in the new TestAsyncTableScanner an example of how to use the
'simple' Interface? If so, looks reasaonable.
Thanks [~Apache9]
> Implement getScanner
> --------------------
>
> Key: HBASE-16984
> URL: https://issues.apache.org/jira/browse/HBASE-16984
> Project: HBase
> Issue Type: Sub-task
> Components: Client
> Affects Versions: 2.0.0
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-16984-v1.patch, HBASE-16984.patch
>
>
> It will just return the old ResultScanner and work like the
> AsyncPrefetchClientScanner. I think we still need this as we can not do time
> consuming work in the ScanObserver introduced in HBASE-16838.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
