MR-279: org.apache.hadoop.mapred.CombineFileInputFormat.getSplits() throws
ArrayStoreException
----------------------------------------------------------------------------------------------
Key: MAPREDUCE-2660
URL: https://issues.apache.org/jira/browse/MAPREDUCE-2660
Project: Hadoop Map/Reduce
Issue Type: Bug
Reporter: Amareshwari Sriramadasu
In branch MR-279, org.apache.hadoop.mapred.getSplits() throws
RuntimeException:ArrayStoreException
The following code in trunk:
{noformat}
public InputSplit[] getSplits(JobConf job, int numSplits)
throws IOException {
List<org.apache.hadoop.mapreduce.InputSplit> newStyleSplits =
super.getSplits(new Job(job));
InputSplit[] ret = new InputSplit[newStyleSplits.size()];
for(int pos = 0; pos < newStyleSplits.size(); ++pos) {
org.apache.hadoop.mapreduce.lib.input.CombineFileSplit newStyleSplit =
(org.apache.hadoop.mapreduce.lib.input.CombineFileSplit)
newStyleSplits.get(pos);
ret[pos] = new CombineFileSplit(job, newStyleSplit.getPaths(),
newStyleSplit.getStartOffsets(), newStyleSplit.getLengths(),
newStyleSplit.getLocations());
}
return ret;
}
{noformat}
got changed to
{noformat}
public InputSplit[] getSplits(JobConf job, int numSplits)
throws IOException {
return super.getSplits(new Job(job)).toArray(new InputSplit[0]);
}
{noformat}
Code in trunk works fine. We should change the code in MR-279 to the same in
trunk.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
