Author: smarthi
Date: Mon Jun 24 15:04:47 2013
New Revision: 1496078
URL: http://svn.apache.org/r1496078
Log:
MAHOUT-833: Make conversion to sequence files map-reduce - fixed issue with not
reading a directory list
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java?rev=1496078&r1=1496077&r2=1496078&view=diff
==============================================================================
---
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java
(original)
+++
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java
Mon Jun 24 15:04:47 2013
@@ -133,7 +133,7 @@ public class SequenceFilesFromDirectory
Configuration jobConfig = job.getConfiguration();
jobConfig.set("keyPrefix", keyPrefix);
FileSystem fs = FileSystem.get(jobConfig);
- FileStatus fsFileStatus = HadoopUtil.listStatus(fs, input)[0];
+ FileStatus fsFileStatus = fs.getFileStatus(input);
String inputDirList = HadoopUtil.buildDirList(fs, fsFileStatus);
jobConfig.set("baseinputpath", input.toString());
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java?rev=1496078&r1=1496077&r2=1496078&view=diff
==============================================================================
---
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java
(original)
+++
mahout/trunk/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java
Mon Jun 24 15:04:47 2013
@@ -323,7 +323,7 @@ public final class SequenceFilesFromMail
}
FileSystem fs = FileSystem.get(jobConfig);
- FileStatus fsFileStatus = HadoopUtil.listStatus(fs, inputPath)[0];
+ FileStatus fsFileStatus = fs.getFileStatus(inputPath);
jobConfig.set("baseinputpath", inputPath.toString());
String inputDirList = HadoopUtil.buildDirList(fs, fsFileStatus);