github/sqoop
5
0
Fork 0
mirror of https://github.com/apache/sqoop.git synced 2025-05-04 23:42:14 +08:00
Code Issues Projects Releases Wiki Activity

SQOOP-844: Sqoop2: HdfsExportPartitioner is not always respecting maximal number of partitions

Browse Source 
(Vasanth kumar RJ via Jarek Jarcec Cecho)
This commit is contained in:
Jarek Jarcec Cecho 2013-04-13 11:41:04 -07:00
parent 3865f7deec
commit ae075f2e37
2 changed files with 33 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
execution/mapreduce/src/main/java/org/apache/sqoop/job/etl/HdfsExportPartitioner.java
View File

@ -73,6 +73,10 @@ public List<Partition> getPartitions(PartitionerContext context,
long numInputBytes = getInputSize(conf);
maxSplitSize = numInputBytes / context.getMaxPartitions();
if(numInputBytes % context.getMaxPartitions() != 0 ) {
maxSplitSize += 1;
}
long minSizeNode = 0;
long minSizeRack = 0;
long maxSize = 0;

29
execution/mapreduce/src/test/java/org/apache/sqoop/job/TestHdfsExtract.java
View File

@ -21,6 +21,7 @@
import java.io.IOException;
import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.util.List;
import junit.framework.TestCase;
@ -38,6 +39,8 @@
import org.apache.sqoop.job.etl.HdfsSequenceImportLoader;
import org.apache.sqoop.job.etl.Loader;
import org.apache.sqoop.job.etl.LoaderContext;
import org.apache.sqoop.job.etl.Partition;
import org.apache.sqoop.job.etl.PartitionerContext;
import org.apache.sqoop.job.io.Data;
import org.apache.sqoop.job.mr.SqoopFileOutputFormat;
import org.junit.Test;
@ -54,6 +57,32 @@ public TestHdfsExtract() {
indir = INPUT_ROOT + getClass().getSimpleName();
}
/**
* Test case for validating the number of partitions creation
* based on input.
* Success if the partitions list size is less or equal to
* given max partition.
* @throws Exception
*/
@Test
public void testHdfsExportPartitioner() throws Exception {
FileUtils.delete(indir);
FileUtils.mkdirs(indir);
createTextInput(null);
Configuration conf = new Configuration();
conf.set(JobConstants.HADOOP_INPUTDIR, indir);
HdfsExportPartitioner partitioner = new HdfsExportPartitioner();
PrefixContext prefixContext = new PrefixContext(conf, "");
int[] partitionValues = {2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 17};
for(int maxPartitions : partitionValues) {
PartitionerContext partCont = new PartitionerContext(prefixContext, maxPartitions);
List<Partition> partitionList = partitioner.getPartitions(partCont, null, null);
assertTrue(partitionList.size()<=maxPartitions);
}
}
@Test
public void testUncompressedText() throws Exception {
FileUtils.delete(indir);