Hadoop old API, mainly JobConf with Job, that there is anyrelationship between the two classes, can turn to each other?

 This post last edited by the virgo777 on 2013-05-04 12:39:06

is such that the new API is useful, I am currently in use,

But some old ways must only use the old API,

example

MultipleInputs.addInputPath (conf, path, inputFormatClass)

The first parameter is the JobConf type, whereas in the new API, has been using the Job class instead JobConf,

ie initialization and main methods are not JobConf definition.

Is there a solution?

PS: Hadoop said in the third edition has been achieved MultipleInputs.addInputPath (job, path, inputFormatClass)

But now I'm using an API is 0.20.2 or old.

Will the next API how to connect the old and new with it?

Is a program I use two API do?

Thank you reply.
------ Solution ---------------------------------------- ----
No, not butt
MR1.0 version of the database application

import org.apache.hadoop.mapred.lib.MultipleInputs;

MR2.0 version of the database application

import org.apache.hadoop.mapreduce.lib.input.MultipleInputs;

0.20.2 Although the main push has started MR2.0, but there are still some 1.0 interface is retained.
recommend LZ two solutions:
1, HADOOP upgrade to version 1.0.x Release
2, instead of using multiple FileInputFormat.addInputPath MultipleInputs.addInputPath

FileInputFormat.addInputPath(job, new Path(dfsPath[0]));
 FileInputFormat.addInputPath(job, new Path(dfsPath[1]));
 .....
 FileInputFormat.addInputPath(job, new Path(dfsPath[N]));

but this does not distinguish between different Map.class different input
only in the Map function to dynamically determine Iterable values ​​belong to that class file
------ For reference only --------------- ------------------------
their own roof, has an understanding of it?


------ For reference only ---------------------------------- -----
nobody know?
------ For reference only -------------------------------------- -
supplementary question, the general methods are MR2.0 mapreduce bag inside of it?

It is not meant to use the new API is in use MR2.0 of it?

there is a problem, 1.0.x Release version 0.23 of Hadoop Hadoop means you?
------ For reference only -------------------------------------- -
1, is the
2, is the
3,1.0. x is apache released stable version HADOOP, the highest stable version is 1.0.4
0.23 is 2.0 derivative, but the lack NameNode HA function, if it is recommended to use 1.0.x production environment
These are used MR2.0

can refer to: http://hadoop.apache.org/releases.html # Download
1.0.X - current stable version, 1.0 release
1.1.X - current beta version, 1.1 release
2.X.X - current alpha version
0.23.X - simmilar to 2.X.X but missing NN HA.
0.22.X - does not include security
0.20.203.X - old legacy stable version
0.20.X - old legacy version