[ 
https://issues.apache.org/jira/browse/HDFS-8156?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14500447#comment-14500447
 ] 

Tsz Wo Nicholas Sze commented on HDFS-8156:
-------------------------------------------

- Should chunkSize be renamed to "cellSize"?
- The javadoc below is wrong.  A javadoc block does not work for multiple 
fields.  We need to have a javadoc block for each field.
{code}
+  /**
+   * The most common and general parameters found in typical erasure codes.
+   */
   private int numDataUnits;
   private int numParityUnits;
   private int chunkSize;
 {code}
- Why do we need initWith?  Could initWith be changed to a constructor and 
change all the fields to final, i.e. making ECSchema immutable?
- A schemas have a name and a set of logical parameters (i.e. numDataUnits, 
numParityUnits, chunkSize, extraOptions). 
-* Could two schemas having the same name but different set of logical 
parameters? 
-* Should two schemas having the same set of logical parameters but different 
names be considered the same?

> Add/implement necessary APIs even we just have the system default schema
> ------------------------------------------------------------------------
>
>                 Key: HDFS-8156
>                 URL: https://issues.apache.org/jira/browse/HDFS-8156
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Kai Zheng
>            Assignee: Kai Zheng
>         Attachments: HDFS-8156-v1.patch, HDFS-8156-v2.patch, 
> HDFS-8156-v3.patch, HDFS-8156-v4.patch
>
>
> According to the discussion here, this issue was repurposed and modified.
> This is to add and implement some necessary APIs even we just have the system 
> default schema, to resolve some TODOs left for HDFS-7859 and HDFS-7866 as 
> they're still subject to further discussion.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to