svn commit: r1486388 - in /mahout/trunk: CHANGELOG core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java

[smarthi]

Author: smarthi
Date: Sun May 26 03:26:12 2013
New Revision: 1486388

URL: http://svn.apache.org/r1486388
Log:
MAHOUT-1213: SSVD job doesn't clean it's temp dir, and fails when seeing it 
again

Modified:
    mahout/trunk/CHANGELOG
    
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java

Modified: mahout/trunk/CHANGELOG
URL: 
http://svn.apache.org/viewvc/mahout/trunk/CHANGELOG?rev=1486388&r1=1486387&r2=1486388&view=diff
==============================================================================
--- mahout/trunk/CHANGELOG (original)
+++ mahout/trunk/CHANGELOG Sun May 26 03:26:12 2013
@@ -2,6 +2,8 @@ Mahout Change Log
 
 Release 0.8 - unreleased
 
+  MAHOUT-1213: SSVD job doesn't clean it's temp dir, and fails when seeing it 
again (smarthi)
+
 __MAHOUT-1223: Fixed point skipped in StreamingKMeans when iterating through 
centroids from a reducer (dfilimon)
 
 __MAHOUT-1222: Fix total weight in FastProjectionSearch (dfilimon)

Modified: 
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java
URL: 
http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java?rev=1486388&r1=1486387&r2=1486388&view=diff
==============================================================================
--- 
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java
 (original)
+++ 
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java
 Sun May 26 03:26:12 2013
@@ -25,6 +25,7 @@ import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.util.ToolRunner;
 import org.apache.mahout.common.AbstractJob;
+import org.apache.mahout.common.HadoopUtil;
 import org.apache.mahout.common.commandline.DefaultOptionCreator;
 import org.apache.mahout.math.Vector;
 import org.apache.mahout.math.hadoop.MatrixColumnMeansJob;
@@ -108,8 +109,7 @@ public class SSVDCli extends AbstractJob
     Path xiPath = xiPathStr == null ? null : new Path(xiPathStr);
     boolean pca = Boolean.parseBoolean(getOption("pca")) || xiPath != null;
 
-    boolean overwrite =
-      pargs.containsKey(keyFor(DefaultOptionCreator.OVERWRITE_OPTION));
+    boolean overwrite = hasOption(DefaultOptionCreator.OVERWRITE_OPTION);
 
     Configuration conf = getConf();
     if (conf == null) {
@@ -118,7 +118,17 @@ public class SSVDCli extends AbstractJob
 
     Path[] inputPaths = { getInputPath() };
     Path tempPath = getTempPath();
-    FileSystem fs = FileSystem.get(getOutputPath().toUri(), conf);
+    FileSystem fs = FileSystem.get(getTempPath().toUri(), conf);
+
+    // housekeeping
+    if (overwrite) {
+      // clear the output path
+      HadoopUtil.delete(getConf(), getOutputPath());
+      // clear the temp path
+      HadoopUtil.delete(getConf(), getTempPath());
+    }
+
+    fs.mkdirs(getOutputPath());
 
     // MAHOUT-817
     if (pca && xiPath == null) {
@@ -156,13 +166,6 @@ public class SSVDCli extends AbstractJob
 
     solver.run();
 
-    // housekeeping
-    if (overwrite) {
-      fs.delete(getOutputPath(), true);
-    }
-
-    fs.mkdirs(getOutputPath());
-
     Vector svalues = solver.getSingularValues().viewPart(0, k);
     SSVDHelper.saveVector(svalues, getOutputPath("sigma"), conf);
 
@@ -184,6 +187,10 @@ public class SSVDCli extends AbstractJob
         && !fs.rename(new Path(solver.getvHalfSigmaPath()), getOutputPath())) {
       throw new IOException("Unable to move V*Sigma^0.5 results to the output 
path.");
     }
+
+    // Delete the temp path on exit
+    fs.deleteOnExit(getTempPath());
+
     return 0;
   }