HADOOP-5138. Fixing failing test cases by separating TestCharFileTailing and TestFileTailing adaptors. Contributed by Jerome Boulon.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/core/trunk@740237 13f79535-47bb-0310-9956-ffa450edef68

src/contrib/chukwa/CHANGES.txt

@@ -10,6 +10,9 @@ Trunk (unreleased changes)
 
   BUG FIXES
 
+    HADOOP-5138. Fixing failing test cases by separating TestCharFileTailing
+    and TestFileTailing adaptors. (Jerome Boulon via acmurthy)
+
 Release 0.19.0 - Unreleased
 
   IMPROVEMENTS

src/contrib/chukwa/src/java/org/apache/hadoop/chukwa/datacollection/adaptor/filetailer/FileTailingAdaptor.java

@@ -84,16 +84,15 @@ public class FileTailingAdaptor implements Adaptor
 	    this.dest = dest;
 	    this.attempts = 0;
 			
-	    Pattern cmd = Pattern.compile("(\\d+)\\s+(.+)");
+	    Pattern cmd = Pattern.compile("(\\d+)\\s+(.+)\\s");
 	    Matcher m = cmd.matcher(params);
 	    if(m.matches()) {
 	        offsetOfFirstByte = Long.parseLong(m.group(1));
 	        toWatch = new File(m.group(2));
 	    } else {
-	        toWatch = new File(params);
+	        toWatch = new File(params.trim());
 	    }
 	  
-	    
 		this.fileReadOffset= bytes;
 		tailer.startWatchingFile(this);
 	}

src/contrib/chukwa/src/java/org/apache/hadoop/chukwa/datacollection/agent/ChukwaAgent.java

@@ -267,7 +267,7 @@ public class ChukwaAgent
   // but can be arbitrarily many space
   // delimited agent specific params )
   // 4) offset
-  Pattern addCmdPattern = Pattern.compile("[aA][dD][dD]\\s+(\\S+)\\s+(\\S+)\\s+(.*\\S)?\\s*(\\d+)\\s*");
+  Pattern addCmdPattern = Pattern.compile("[aA][dD][dD]\\s+(\\S+)\\s+(\\S+)\\s+(.*\\s)?\\s*(\\d+)\\s*");
   // FIXME: should handle bad lines here
   public long processCommand(String cmd)
   {

src/contrib/chukwa/src/java/org/apache/hadoop/chukwa/util/ConstRateAdaptor.java

@@ -39,26 +39,26 @@ public class ConstRateAdaptor  extends Thread implements Adaptor {
   
   private volatile boolean stopping = false;
   public String getCurrentStatus() throws AdaptorException {
-    return Integer.toString(bytesPerSec);
+    return type.trim()  + " " + bytesPerSec + " " + offset;
   }
 
-  public void start(long adaptor, String type, String status, long offset, ChunkReceiver dest) throws AdaptorException
+  public void start(long adaptor, String type, String bytesPerSecParam, long offset, ChunkReceiver dest) throws AdaptorException
   {
     try{
-      bytesPerSec = Integer.parseInt(status);
+      bytesPerSec = Integer.parseInt(bytesPerSecParam.trim());
     } catch(NumberFormatException e) {
-      throw new AdaptorException("bad argument to const rate adaptor: " + status);
+      throw new AdaptorException("bad argument to const rate adaptor: [" + bytesPerSecParam + "]");
     }
     this.adaptorID = adaptor;
     this.offset = offset;
     this.type = type;
     this.dest = dest;
-    this.setName("ConstRate Adaptor");
+    this.setName("ConstRate Adaptor_" + type);
     super.start();  //this is a Thread.start
   }
   
   public String getStreamName() {
-	  return ""+bytesPerSec;
+    return this.type;
   }
   
   public void run()

src/contrib/chukwa/src/test/org/apache/hadoop/chukwa/datacollection/adaptor/filetailer/TestCharFileTailingAdaptorUTF8.java

@@ -0,0 +1,63 @@
+package org.apache.hadoop.chukwa.datacollection.adaptor.filetailer;
+
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.PrintWriter;
+
+import junit.framework.TestCase;
+
+import org.apache.hadoop.chukwa.Chunk;
+import org.apache.hadoop.chukwa.datacollection.agent.ChukwaAgent;
+import org.apache.hadoop.chukwa.datacollection.connector.ChunkCatcherConnector;
+
+public class TestCharFileTailingAdaptorUTF8 extends TestCase {
+  ChunkCatcherConnector chunks;
+
+  public TestCharFileTailingAdaptorUTF8() {
+    chunks = new ChunkCatcherConnector();
+    chunks.start();
+  }
+
+  public void testCrSepAdaptor() throws IOException, InterruptedException,
+      ChukwaAgent.AlreadyRunningException {
+    ChukwaAgent agent = new ChukwaAgent();
+    File testFile = makeTestFile("/tmp/chukwaTest", 80);
+    long adaptorId = agent
+        .processCommand("add org.apache.hadoop.chukwa.datacollection.adaptor.filetailer.CharFileTailingAdaptorUTF8"
+            + " lines " + testFile + " 0");
+    assertTrue(adaptorId != -1);
+    System.out.println("getting a chunk...");
+    Chunk c = chunks.waitForAChunk();
+    System.out.println("got chunk");
+    assertTrue(c.getSeqID() == testFile.length());
+
+    assertTrue(c.getRecordOffsets().length == 80);
+    int recStart = 0;
+    for (int rec = 0; rec < c.getRecordOffsets().length; ++rec) {
+      String record = new String(c.getData(), recStart,
+          c.getRecordOffsets()[rec] - recStart + 1);
+      System.out.println("record " + rec + " was: " + record);
+      assertTrue(record.equals(rec + " abcdefghijklmnopqrstuvwxyz\n"));
+      recStart = c.getRecordOffsets()[rec] + 1;
+    }
+    assertTrue(c.getDataType().equals("lines"));
+    agent.stopAdaptor(adaptorId, false);
+    agent.shutdown();
+  }
+
+  private File makeTestFile(String name, int size) throws IOException {
+    File tmpOutput = new File(name);
+    FileOutputStream fos = new FileOutputStream(tmpOutput);
+
+    PrintWriter pw = new PrintWriter(fos);
+    for (int i = 0; i < size; ++i) {
+      pw.print(i + " ");
+      pw.println("abcdefghijklmnopqrstuvwxyz");
+    }
+    pw.flush();
+    pw.close();
+    return tmpOutput;
+  }
+
+}