[ 
https://issues.apache.org/jira/browse/PIG-4923?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15725747#comment-15725747
 ] 

Adam Szita commented on PIG-4923:
---------------------------------

Thanks for all your comments on this. I've updated the diff both here 
([^PIG-4923.2.patch]) and reviewboard.

I tested the build with hadoop 3.0.0-alpha1 jars and there doesn't seem like a 
lot of changes needed to make it pass:
{code}
compile:
     [echo] *** Building Main Sources ***
     [echo] *** To compile with all warnings enabled, supply -Dall.warnings=1 
on command line ***
     [echo] *** Else, you will only be warned about deprecations ***
     [echo] *** Hadoop version used: 2 ; HBase version used: 1 ***
    [javac] Compiling 1007 source files to 
/Users/szita/shadow/CDH/pig/build/classes
    [javac] warning: [options] bootstrap class path not set in conjunction with 
-source 1.7
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/datastorage/HDataStorage.java:34:
 error: cannot find symbol
    [javac] import org.apache.hadoop.hdfs.DistributedFileSystem;
    [javac]                              ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: package org.apache.hadoop.hdfs
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/executionengine/HExecutionEngine.java:34:
 error: cannot find symbol
    [javac] import org.apache.hadoop.hdfs.DistributedFileSystem;
    [javac]                              ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: package org.apache.hadoop.hdfs
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/datastorage/HDataStorage.java:117:
 error: cannot find symbol
    [javac]         if (fs instanceof DistributedFileSystem) {
    [javac]                           ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: class HDataStorage
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/datastorage/HDataStorage.java:118:
 error: cannot find symbol
    [javac]             DistributedFileSystem dfs = (DistributedFileSystem) fs;
    [javac]             ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: class HDataStorage
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/datastorage/HDataStorage.java:118:
 error: cannot find symbol
    [javac]             DistributedFileSystem dfs = (DistributedFileSystem) fs;
    [javac]                                          ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: class HDataStorage
    [javac] 
/Users/szita/shadow/CDH/pig/src/org/apache/pig/backend/hadoop/executionengine/HExecutionEngine.java:198:
 error: cannot find symbol
    [javac]             new DistributedFileSystem();
    [javac]                 ^
    [javac]   symbol:   class DistributedFileSystem
    [javac]   location: class HExecutionEngine
    [javac] Note: Some input files use or override a deprecated API.
    [javac] Note: Recompile with -Xlint:deprecation for details.
    [javac] Note: Some input files use unchecked or unsafe operations.
    [javac] Note: Recompile with -Xlint:unchecked for details.
    [javac] 6 errors
    [javac] 1 warning
{code}

Actually I went ahead and applied some small adjustments for Hadoop3 just for a 
try, see: [^PIG-4923.2.IncrementalHadoop3.patch]
Are we really sure that we want to keep the shims layer? I think considering 
what Hadoop3 will need it might not be a bad idea to remove the additional 
complexity of using shims.
Let me know what you think.

> Drop Hadoop 1.x support in Pig 0.17
> -----------------------------------
>
>                 Key: PIG-4923
>                 URL: https://issues.apache.org/jira/browse/PIG-4923
>             Project: Pig
>          Issue Type: Improvement
>            Reporter: Daniel Dai
>            Assignee: Adam Szita
>             Fix For: 0.17.0
>
>         Attachments: PIG-4923.1.patch, PIG-4923.2.IncrementalHadoop3.patch, 
> PIG-4923.2.patch, ant-contrib-1.0b3.jar
>
>
> To facilitate the future development, we want to get rid of the legacy Hadoop 
> 1.x support and reduce the code complexity.



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

Reply via email to