[
https://issues.apache.org/jira/browse/HADOOP-5721?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12702567#action_12702567
]
Konstantin Shvachko commented on HADOOP-5721:
---------------------------------------------
# Several includes in FSEditLog became redundant after the input and output
streams have been factored out.
{code}
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.RandomAccessFile;
import java.nio.ByteBuffer;
import java.nio.channels.FileChannel;
import org.apache.hadoop.hdfs.server.namenode.EditLogFileInputStream;
import org.apache.hadoop.io.DataOutputBuffer;
{code}
# Could you please move the import of {{EditLogFileOutputStream}} down in the
respective group of imports. The imports are lexicographically ordered in the
file.
# {{FSEditLog.sizeFlushBuffer}} should be a private static field of
{{EditLogFileOutputStream}} together with {{FSEditLog.setBufferCapacity()}}
method.
Please resubmit the patch (cancel and then submit again) so that Hudson could
run it again.
> Provide EditLogFileInputStream and EditLogFileOutputStream as independent
> classes
> ----------------------------------------------------------------------------------
>
> Key: HADOOP-5721
> URL: https://issues.apache.org/jira/browse/HADOOP-5721
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Affects Versions: 0.19.1
> Reporter: Luca Telloli
> Priority: Minor
> Fix For: 0.21.0
>
> Attachments: HADOOP-5721.patch, HADOOP-5721.patch, HADOOP-5721.patch
>
>
> EditLogFileInputStream and EditLogFileOutputStream are currently part of
> FSEditLog. With this patch we want to extract them and provide as independent
> classes.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
