HADOOP-1821 Replace all String.getBytes() with String.getBytes("UTF-8")

git-svn-id: https://svn.apache.org/repos/asf/lucene/hadoop/trunk@571711 13f79535-47bb-0310-9956-ffa450edef68

src/contrib/hbase/CHANGES.txt

@@ -28,6 +28,7 @@ Trunk (unreleased changes)
                 (Ning Li via Stack)
     HADOOP-1800 output should default utf8 encoding
     HADOOP-1814	TestCleanRegionServerExit fails too often on Hudson
+    HADOOP-1821 Replace all String.getBytes() with String.getBytes("UTF-8")
 
   IMPROVEMENTS
     HADOOP-1737 Make HColumnDescriptor data publically members settable

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HConstants.java

@@ -19,7 +19,6 @@
  */
 package org.apache.hadoop.hbase;
 
-import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
 import org.apache.hadoop.io.Text;
 
 /**
@@ -86,11 +85,24 @@ public interface HConstants {
 
   // Always store the location of the root table's HRegion.
   // This HRegion is never split.
-
+  
   // region name = table + startkey + regionid. This is the row key.
   // each row in the root and meta tables describes exactly 1 region
   // Do we ever need to know all the information that we are storing?
 
+  // Note that the name of the root table starts with "-" and the name of the
+  // meta table starts with "." Why? it's a trick. It turns out that when we
+  // store region names in memory, we use a SortedMap. Since "-" sorts before
+  // "." (and since no other table name can start with either of these
+  // characters, the root region will always be the first entry in such a Map,
+  // followed by all the meta regions (which will be ordered by their starting
+  // row key as well), followed by all user tables. So when the Master is 
+  // choosing regions to assign, it will always choose the root region first,
+  // followed by the meta regions, followed by user regions. Since the root
+  // and meta regions always need to be on-line, this ensures that they will
+  // be the first to be reassigned if the server(s) they are being served by
+  // should go down.
+
   /** The root table's name. */
   static final Text ROOT_TABLE_NAME = new Text("-ROOT-");
 
@@ -133,11 +145,4 @@ public interface HConstants {
   /** When we encode strings, we always specify UTF8 encoding */
   static final String UTF8_ENCODING = "UTF-8";
 
-  /** Value stored for a deleted item */
-  static final ImmutableBytesWritable DELETE_BYTES =
-    new ImmutableBytesWritable("HBASE::DELETEVAL".getBytes());
-
-  /** Value written to HLog on a complete cache flush */
-  static final ImmutableBytesWritable COMPLETE_CACHEFLUSH =
-    new ImmutableBytesWritable("HBASE::CACHEFLUSH".getBytes());
 }

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HGlobals.java

@@ -19,14 +19,28 @@
  */
 package org.apache.hadoop.hbase;
 
+import java.io.UnsupportedEncodingException;
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
+
 /**
- * Global values used for finding and scanning the root and meta tables.
+ * Global values that require initialization that cannot be done in HConstants
  */
 public class HGlobals implements HConstants {
+
+  /** table descriptor for root table */
+  public static HTableDescriptor rootTableDesc = null;
+  
+  /** region info for the root region */
+  public static HRegionInfo rootRegionInfo = null;
+  
+  /** table descriptor for meta table */
+  public static HTableDescriptor metaTableDesc = null;
   
-  static HTableDescriptor rootTableDesc = null;
-  static HRegionInfo rootRegionInfo = null;
-  static HTableDescriptor metaTableDesc = null;
+  /** Value stored for a deleted item */
+  public static ImmutableBytesWritable deleteBytes = null;
+
+  /** Value written to HLog on a complete cache flush */
+  public static ImmutableBytesWritable completeCacheFlush = null;
 
   static {
     rootTableDesc = new HTableDescriptor(ROOT_TABLE_NAME.toString());
@@ -38,5 +52,17 @@ public class HGlobals implements HConstants {
     metaTableDesc = new HTableDescriptor(META_TABLE_NAME.toString());
     metaTableDesc.addFamily(new HColumnDescriptor(COLUMN_FAMILY, 1,
         HColumnDescriptor.CompressionType.NONE, false, Integer.MAX_VALUE, null));
+
+    try {
+      deleteBytes =
+        new ImmutableBytesWritable("HBASE::DELETEVAL".getBytes(UTF8_ENCODING));
+    
+      completeCacheFlush =
+        new ImmutableBytesWritable("HBASE::CACHEFLUSH".getBytes(UTF8_ENCODING));
+      
+    } catch (UnsupportedEncodingException e) {
+      assert(false);
+    }
+ 
   }
 }

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HLog.java

@@ -433,7 +433,7 @@ public class HLog implements HConstants {
     }
     
     writer.append(new HLogKey(regionName, tableName, HLog.METAROW, logSeqId),
-      new HLogEdit(HLog.METACOLUMN, COMPLETE_CACHEFLUSH.get(),
+      new HLogEdit(HLog.METACOLUMN, HGlobals.completeCacheFlush.get(),
         System.currentTimeMillis()));
     numEntries.getAndIncrement();
 

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HMemcache.java

@@ -258,7 +258,7 @@ public class HMemcache {
     for (Map.Entry<HStoreKey, byte []> es: tailMap.entrySet()) {
       HStoreKey itKey = es.getKey();
       if (itKey.matchesRowCol(curKey)) {
-        if(HConstants.DELETE_BYTES.compareTo(es.getValue()) == 0) {
+        if(HGlobals.deleteBytes.compareTo(es.getValue()) == 0) {
           // TODO: Shouldn't this be a continue rather than a break?  Perhaps
           // the intent is that this DELETE_BYTES is meant to suppress older
           // info -- see 5.4 Compactions in BigTable -- but how does this jibe

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HRegion.java

@@ -1103,7 +1103,7 @@ public class HRegion implements HConstants {
    * @throws IOException
    */
   public void put(long lockid, Text targetCol, byte [] val) throws IOException {
-    if (DELETE_BYTES.compareTo(val) == 0) {
+    if (HGlobals.deleteBytes.compareTo(val) == 0) {
       throw new IOException("Cannot insert value: " + val);
     }
     localput(lockid, targetCol, val);
@@ -1117,7 +1117,7 @@ public class HRegion implements HConstants {
    * @throws IOException
    */
   public void delete(long lockid, Text targetCol) throws IOException {
-    localput(lockid, targetCol, DELETE_BYTES.get());
+    localput(lockid, targetCol, HGlobals.deleteBytes.get());
   }
 
   /**

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HRegionServer.java

@@ -1064,7 +1064,7 @@ public class HRegionServer implements HConstants, HRegionInterface, Runnable {
       for(Map.Entry<Text, byte []> e: results.entrySet()) {
         HStoreKey k = new HStoreKey(key.getRow(), e.getKey(), key.getTimestamp());
         byte [] val = e.getValue();
-        if (DELETE_BYTES.compareTo(val) == 0) {
+        if (HGlobals.deleteBytes.compareTo(val) == 0) {
           // Column value is deleted. Don't return it.
           continue;
         }

src/contrib/hbase/src/java/org/apache/hadoop/hbase/HStore.java

@@ -1033,7 +1033,7 @@ class HStore implements HConstants {
             Text readcol = readkey.getColumn();
             if (results.get(readcol) == null
                 && key.matchesWithoutColumn(readkey)) {
-              if(readval.equals(HConstants.DELETE_BYTES)) {
+              if(readval.equals(HGlobals.deleteBytes)) {
                 break;
               }
               results.put(new Text(readcol), readval.get());
@@ -1086,14 +1086,14 @@ class HStore implements HConstants {
             continue;
           }
           if (readkey.matchesRowCol(key)) {
-            if(readval.equals(HConstants.DELETE_BYTES)) {
+            if(readval.equals(HGlobals.deleteBytes)) {
               break;
             }
             results.add(readval.get());
             readval = new ImmutableBytesWritable();
             while(map.next(readkey, readval) && readkey.matchesRowCol(key)) {
               if ((numVersions > 0 && (results.size() >= numVersions))
-                  || readval.equals(HConstants.DELETE_BYTES)) {
+                  || readval.equals(HGlobals.deleteBytes)) {
                 break;
               }
               results.add(readval.get());

src/contrib/hbase/src/java/org/apache/hadoop/hbase/Shell.java

@@ -40,8 +40,12 @@ public class Shell {
   /** audible keyboard bells */
   public static final boolean DEFAULT_BELL_ENABLED = true;
 
-  /** Main method */
-  public static void main(String args[]) throws IOException {
+  /** Main method
+   * 
+   * @param args not used
+   * @throws IOException
+   */
+  public static void main(@SuppressWarnings("unused") String args[]) throws IOException {
     Configuration conf = new HBaseConfiguration();
     ConsoleReader reader = new ConsoleReader();
     reader.setBellEnabled(conf.getBoolean("hbaseshell.jline.bell.enabled",
@@ -91,8 +95,14 @@ public class Shell {
     return (queryStr.toString().equals("")) ? "HBase > " : "    --> ";
   }
 
-  /** return a string of code execution time. */
+  /**
+   * @param watch true if execution time should be computed and returned
+   * @param start start of time interval
+   * @param end end of time interval
+   * @return a string of code execution time. */
   public static String executeTime(boolean watch, long start, long end) {
-    return (watch) ? "(" + String.format("%.2f", (end - start) * 0.001) + " sec)" : "";
+    return (watch) ?
+        "(" + String.format("%.2f", (end - start) * 0.001) + " sec)" :
+          "";
   }
 }

src/contrib/hbase/src/java/org/apache/hadoop/hbase/filter/RegExpRowFilter.java

@@ -33,7 +33,7 @@ import java.util.regex.Pattern;
 
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
-import org.apache.hadoop.hbase.HConstants;
+import org.apache.hadoop.hbase.HGlobals;
 import org.apache.hadoop.io.Text;
 
 /**
@@ -176,7 +176,7 @@ public class RegExpRowFilter implements RowFilterInterface {
       }
     }
     if (nullColumns.contains(colKey)) {
-      if (data != null && !Arrays.equals(HConstants.DELETE_BYTES.get(), data)) {
+      if (data != null && !Arrays.equals(HGlobals.deleteBytes.get(), data)) {
         if (LOG.isDebugEnabled()) {
           LOG.debug("filter returning true for rowKey: " + rowKey + 
             " colKey: " + colKey);
@@ -198,7 +198,7 @@ public class RegExpRowFilter implements RowFilterInterface {
   public boolean filterNotNull(final TreeMap<Text, byte[]> columns) {
     for (Entry<Text, byte[]> col : columns.entrySet()) {
       if (nullColumns.contains(col.getKey())
-          && !Arrays.equals(HConstants.DELETE_BYTES.get(), col.getValue())) {
+          && !Arrays.equals(HGlobals.deleteBytes.get(), col.getValue())) {
         if (LOG.isDebugEnabled()) {
           LOG.debug("filterNotNull returning true for colKey: " + col.getKey()
             + ", column should be null.");

src/contrib/hbase/src/java/org/apache/hadoop/hbase/mapred/GroupingTableMap.java

@@ -20,9 +20,11 @@
 package org.apache.hadoop.hbase.mapred;
 
 import java.io.IOException;
+import java.io.UnsupportedEncodingException;
 import java.util.ArrayList;
 import java.util.Map;
 
+import org.apache.hadoop.hbase.HConstants;
 import org.apache.hadoop.hbase.HStoreKey;
 import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
 
@@ -147,7 +149,11 @@ public class GroupingTableMap extends TableMap {
       if(i > 0) {
         sb.append(" ");
       }
-      sb.append(new String(vals[i]));
+      try {
+        sb.append(new String(vals[i], HConstants.UTF8_ENCODING));
+      } catch (UnsupportedEncodingException e) {
+        throw new RuntimeException(e);
+      }
     }
     return new Text(sb.toString());
   }

src/contrib/hbase/src/test/org/apache/hadoop/hbase/HBaseTestCase.java

@@ -159,7 +159,7 @@ public abstract class HBaseTestCase extends TestCase {
       for (char d = secondCharStart; d <= LAST_CHAR; d++) {
         for (char e = thirdCharStart; e <= LAST_CHAR; e++) {
           byte [] bytes = new byte [] {(byte)c, (byte)d, (byte)e};
-          Text t = new Text(new String(bytes));
+          Text t = new Text(new String(bytes, HConstants.UTF8_ENCODING));
           if (endKey != null && endKey.getLength() > 0
               && endKey.compareTo(t) <= 0) {
             break EXIT;

src/contrib/hbase/src/test/org/apache/hadoop/hbase/PerformanceEvaluation.java

@@ -21,6 +21,7 @@ package org.apache.hadoop.hbase;
 
 import java.io.IOException;
 import java.io.PrintStream;
+import java.io.UnsupportedEncodingException;
 import java.text.SimpleDateFormat;
 import java.util.Arrays;
 import java.util.Date;
@@ -312,7 +313,13 @@ public class PerformanceEvaluation implements HConstants {
       while(val.length() < ROW_LENGTH) {
         val.append(Long.toString(this.rand.nextLong()));
       }
-      return val.toString().getBytes();
+      byte[] value = null;
+      try {
+        value = val.toString().getBytes(HConstants.UTF8_ENCODING);
+      } catch (UnsupportedEncodingException e) {
+        assert(false);
+      }
+      return value;
     }
     
     private String generateStatus(final int sr, final int i, final int lr) {

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestBatchUpdate.java

@@ -99,7 +99,7 @@ public class TestBatchUpdate extends HBaseClusterTestCase {
       while(scanner.next(key, results)) {
         for(Map.Entry<Text, byte[]> e: results.entrySet()) {
           System.out.println(key + ": row: " + e.getKey() + " value: " + 
-              new String(e.getValue()));
+              new String(e.getValue(), HConstants.UTF8_ENCODING));
         }
       }
     } catch (Exception e) {

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestGet.java

@@ -69,7 +69,7 @@ public class TestGet extends HBaseTestCase {
   }
   
   /** 
-   * Constructor
+   * the test
    * @throws IOException
    */
   public void testGet() throws IOException {
@@ -144,14 +144,14 @@ public class TestGet extends HBaseTestCase {
       lockid = r.startUpdate(ROW_KEY);
 
       r.put(lockid, new Text(HConstants.COLUMN_FAMILY + "region"),
-        "region2".getBytes());
+        "region2".getBytes(HConstants.UTF8_ENCODING));
 
       String otherServerName = "bar.foo.com:4321";
       r.put(lockid, HConstants.COL_SERVER, 
         Writables.stringToBytes(new HServerAddress(otherServerName).toString()));
       
       r.put(lockid, new Text(HConstants.COLUMN_FAMILY + "junk"),
-        "junk".getBytes());
+        "junk".getBytes(HConstants.UTF8_ENCODING));
       
       r.commit(lockid, System.currentTimeMillis());
 

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestHBaseCluster.java

@@ -90,8 +90,10 @@ public class TestHBaseCluster extends HBaseClusterTestCase {
 
     for (int k = FIRST_ROW; k <= NUM_VALS; k++) {
       long writeid = table.startUpdate(new Text("row_" + k));
-      table.put(writeid, CONTENTS_BASIC, (CONTENTSTR + k).getBytes());
-      table.put(writeid, new Text(ANCHORNUM + k), (ANCHORSTR + k).getBytes());
+      table.put(writeid, CONTENTS_BASIC,
+          (CONTENTSTR + k).getBytes(HConstants.UTF8_ENCODING));
+      table.put(writeid, new Text(ANCHORNUM + k),
+          (ANCHORSTR + k).getBytes(HConstants.UTF8_ENCODING));
       table.commit(writeid);
     }
     System.out.println("Write " + NUM_VALS + " rows. Elapsed time: "
@@ -107,14 +109,14 @@ public class TestHBaseCluster extends HBaseClusterTestCase {
 
       byte bodydata[] = table.get(rowlabel, CONTENTS_BASIC);
       assertNotNull(bodydata);
-      String bodystr = new String(bodydata).toString().trim();
+      String bodystr = new String(bodydata, HConstants.UTF8_ENCODING).trim();
       String teststr = CONTENTSTR + k;
       assertEquals("Incorrect value for key: (" + rowlabel + "," + CONTENTS_BASIC
           + "), expected: '" + teststr + "' got: '" + bodystr + "'",
           bodystr, teststr);
       collabel = new Text(ANCHORNUM + k);
       bodydata = table.get(rowlabel, collabel);
-      bodystr = new String(bodydata).toString().trim();
+      bodystr = new String(bodydata, HConstants.UTF8_ENCODING).trim();
       teststr = ANCHORSTR + k;
       assertEquals("Incorrect value for key: (" + rowlabel + "," + collabel
           + "), expected: '" + teststr + "' got: '" + bodystr + "'",
@@ -145,7 +147,7 @@ public class TestHBaseCluster extends HBaseClusterTestCase {
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte val[] = curVals.get(col);
-          String curval = new String(val).trim();
+          String curval = new String(val, HConstants.UTF8_ENCODING).trim();
 
           if(col.compareTo(CONTENTS_BASIC) == 0) {
             assertTrue("Error at:" + curKey.getRow() + "/" + curKey.getTimestamp()

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestHLog.java

@@ -84,7 +84,7 @@ public class TestHLog extends HBaseTestCase implements HConstants {
           assertEquals(tableName, key.getTablename());
           assertEquals(HLog.METAROW, key.getRow());
           assertEquals(HLog.METACOLUMN, val.getColumn());
-          assertEquals(0, COMPLETE_CACHEFLUSH.compareTo(val.getVal()));
+          assertEquals(0, HGlobals.completeCacheFlush.compareTo(val.getVal()));
           System.out.println(key + " " + val);
         }
       } finally {

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestHMemcache.java

@@ -20,6 +20,7 @@
 package org.apache.hadoop.hbase;
 
 import java.io.IOException;
+import java.io.UnsupportedEncodingException;
 import java.util.Iterator;
 import java.util.Map;
 import java.util.TreeMap;
@@ -85,7 +86,11 @@ public class TestHMemcache extends TestCase {
       TreeMap<Text, byte []> columns = new TreeMap<Text, byte []>();
       for (int ii = 0; ii < COLUMNS_COUNT; ii++) {
         Text k = getColumnName(i, ii);
-        columns.put(k, k.toString().getBytes());
+        try {
+          columns.put(k, k.toString().getBytes(HConstants.UTF8_ENCODING));
+        } catch (UnsupportedEncodingException e) {
+          fail();
+        }
       }
       hmc.add(getRowName(i), columns, System.currentTimeMillis());
     }
@@ -147,7 +152,7 @@ public class TestHMemcache extends TestCase {
   }
   
   private void isExpectedRow(final int rowIndex,
-      TreeMap<Text, byte []> row) {
+      TreeMap<Text, byte []> row) throws UnsupportedEncodingException {
     int i = 0;
     for (Text colname: row.keySet()) {
       String expectedColname =
@@ -159,13 +164,15 @@ public class TestHMemcache extends TestCase {
       // for BytesWriteable.  For comparison, comvert bytes to
       // String and trim to remove trailing null bytes.
       byte [] value = row.get(colname);
-      String colvalueStr = new String(value).trim();
+      String colvalueStr = new String(value, HConstants.UTF8_ENCODING).trim();
       assertEquals("Content", colnameStr, colvalueStr);
     }
   }
 
-  /** Test getFull from memcache */
-  public void testGetFull() {
+  /** Test getFull from memcache
+   * @throws UnsupportedEncodingException
+   */
+  public void testGetFull() throws UnsupportedEncodingException {
     addRows(this.hmemcache);
     for (int i = 0; i < ROW_COUNT; i++) {
       HStoreKey hsk = new HStoreKey(getRowName(i));

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestHRegion.java

@@ -121,8 +121,10 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
 
     for (int k = FIRST_ROW; k <= NUM_VALS; k++) {
       long writeid = region.startUpdate(new Text("row_" + k));
-      region.put(writeid, CONTENTS_BASIC, (CONTENTSTR + k).getBytes());
-      region.put(writeid, new Text(ANCHORNUM + k), (ANCHORSTR + k).getBytes());
+      region.put(writeid, CONTENTS_BASIC,
+          (CONTENTSTR + k).getBytes(HConstants.UTF8_ENCODING));
+      region.put(writeid, new Text(ANCHORNUM + k),
+          (ANCHORSTR + k).getBytes(HConstants.UTF8_ENCODING));
       region.commit(writeid, System.currentTimeMillis());
     }
     System.out.println("Write " + NUM_VALS + " rows. Elapsed time: "
@@ -147,14 +149,14 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
 
       byte [] bodydata = region.get(rowlabel, CONTENTS_BASIC);
       assertNotNull(bodydata);
-      String bodystr = new String(bodydata).toString().trim();
+      String bodystr = new String(bodydata, HConstants.UTF8_ENCODING).trim();
       String teststr = CONTENTSTR + k;
       assertEquals("Incorrect value for key: (" + rowlabel + "," + CONTENTS_BASIC
           + "), expected: '" + teststr + "' got: '" + bodystr + "'",
           bodystr, teststr);
       collabel = new Text(ANCHORNUM + k);
       bodydata = region.get(rowlabel, collabel);
-      bodystr = new String(bodydata).toString().trim();
+      bodystr = new String(bodydata, HConstants.UTF8_ENCODING).trim();
       teststr = ANCHORSTR + k;
       assertEquals("Incorrect value for key: (" + rowlabel + "," + collabel
           + "), expected: '" + teststr + "' got: '" + bodystr + "'",
@@ -170,7 +172,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
     // Try put with bad lockid.
     boolean exceptionThrown = false;
     try {
-      region.put(-1, CONTENTS_BASIC, "bad input".getBytes());
+      region.put(-1, CONTENTS_BASIC,
+          "bad input".getBytes(HConstants.UTF8_ENCODING));
     } catch (LockException e) {
       exceptionThrown = true;
     }
@@ -183,7 +186,7 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
       lockid = region.startUpdate(new Text("Some old key"));
       String unregisteredColName = "FamilyGroup:FamilyLabel";
       region.put(lockid, new Text(unregisteredColName),
-        unregisteredColName.getBytes());
+        unregisteredColName.getBytes(HConstants.UTF8_ENCODING));
     } catch (IOException e) {
       exceptionThrown = true;
     } finally {
@@ -276,8 +279,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
       String kLabel = String.format("%1$03d", k);
 
       long lockid = region.startUpdate(new Text("row_vals1_" + kLabel));
-      region.put(lockid, cols[0], vals1[k].getBytes());
-      region.put(lockid, cols[1], vals1[k].getBytes());
+      region.put(lockid, cols[0], vals1[k].getBytes(HConstants.UTF8_ENCODING));
+      region.put(lockid, cols[1], vals1[k].getBytes(HConstants.UTF8_ENCODING));
       region.commit(lockid, System.currentTimeMillis());
       numInserted += 2;
     }
@@ -300,10 +303,12 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
           for(int j = 0; j < cols.length; j++) {
             if(col.compareTo(cols[j]) == 0) {
-              assertEquals("Error at:" + curKey.getRow() + "/" + curKey.getTimestamp()
+              assertEquals("Error at:" + curKey.getRow() + "/"
+                  + curKey.getTimestamp()
                   + ", Value for " + col + " should be: " + k
                   + ", but was fetched as: " + curval, k, curval);
               numFetched++;
@@ -345,10 +350,12 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
           for(int j = 0; j < cols.length; j++) {
             if(col.compareTo(cols[j]) == 0) {
-              assertEquals("Error at:" + curKey.getRow() + "/" + curKey.getTimestamp()
+              assertEquals("Error at:" + curKey.getRow() + "/"
+                  + curKey.getTimestamp()
                   + ", Value for " + col + " should be: " + k
                   + ", but was fetched as: " + curval, k, curval);
               numFetched++;
@@ -375,8 +382,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
       String kLabel = String.format("%1$03d", k);
       
       long lockid = region.startUpdate(new Text("row_vals1_" + kLabel));
-      region.put(lockid, cols[0], vals1[k].getBytes());
-      region.put(lockid, cols[1], vals1[k].getBytes());
+      region.put(lockid, cols[0], vals1[k].getBytes(HConstants.UTF8_ENCODING));
+      region.put(lockid, cols[1], vals1[k].getBytes(HConstants.UTF8_ENCODING));
       region.commit(lockid, System.currentTimeMillis());
       numInserted += 2;
     }
@@ -398,10 +405,12 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
           for(int j = 0; j < cols.length; j++) {
             if(col.compareTo(cols[j]) == 0) {
-              assertEquals("Error at:" + curKey.getRow() + "/" + curKey.getTimestamp()
+              assertEquals("Error at:" + curKey.getRow() + "/"
+                  + curKey.getTimestamp()
                   + ", Value for " + col + " should be: " + k
                   + ", but was fetched as: " + curval, k, curval);
               numFetched++;
@@ -443,7 +452,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
           for (int j = 0; j < cols.length; j++) {
             if (col.compareTo(cols[j]) == 0) {
               assertEquals("Value for " + col + " should be: " + k
@@ -480,7 +490,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
           for (int j = 0; j < cols.length; j++) {
             if (col.compareTo(cols[j]) == 0) {
               assertEquals("Value for " + col + " should be: " + k
@@ -529,7 +540,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
 
       // Write to the HRegion
       long writeid = region.startUpdate(new Text("row_" + k));
-      region.put(writeid, CONTENTS_BODY, buf1.toString().getBytes());
+      region.put(writeid, CONTENTS_BODY,
+          buf1.toString().getBytes(HConstants.UTF8_ENCODING));
       region.commit(writeid, System.currentTimeMillis());
       if (k > 0 && k % (N_ROWS / 100) == 0) {
         System.out.println("Flushing write #" + k);
@@ -656,7 +668,7 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          String curval = new String(val).trim();
+          String curval = new String(val, HConstants.UTF8_ENCODING).trim();
 
           if(col.compareTo(CONTENTS_BASIC) == 0) {
             assertTrue("Error at:" + curKey.getRow() + "/" + curKey.getTimestamp()
@@ -709,7 +721,8 @@ public class TestHRegion extends HBaseTestCase implements RegionUnavailableListe
         for(Iterator<Text> it = curVals.keySet().iterator(); it.hasNext(); ) {
           Text col = it.next();
           byte [] val = curVals.get(col);
-          int curval = Integer.parseInt(new String(val).trim());
+          int curval =
+            Integer.parseInt(new String(val, HConstants.UTF8_ENCODING).trim());
 
           for (int j = 0; j < cols.length; j++) {
             if (col.compareTo(cols[j]) == 0) {

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestHStoreFile.java

@@ -79,7 +79,7 @@ public class TestHStoreFile extends TestCase {
     try {
       for (char d = FIRST_CHAR; d <= LAST_CHAR; d++) {
         byte[] b = new byte[] {(byte)d};
-        Text t = new Text(new String(b));
+        Text t = new Text(new String(b, HConstants.UTF8_ENCODING));
         writer.append(new HStoreKey(t, t, System.currentTimeMillis()),
             new ImmutableBytesWritable(t.getBytes()));
       }
@@ -101,7 +101,7 @@ public class TestHStoreFile extends TestCase {
       for (char d = FIRST_CHAR; d <= LAST_CHAR; d++) {
         for (char e = FIRST_CHAR; e <= LAST_CHAR; e++) {
           byte[] b = new byte[] { (byte) d, (byte) e };
-          Text t = new Text(new String(b));
+          Text t = new Text(new String(b, HConstants.UTF8_ENCODING));
           writer.append(new HStoreKey(t, t, System.currentTimeMillis()),
             new ImmutableBytesWritable(t.getBytes()));
         }
@@ -248,7 +248,7 @@ public class TestHStoreFile extends TestCase {
       LOG.info("Last in top: " + key.toString());
       top.getClosest(midkey, value);
       // Assert value is same as key.
-      assertEquals(new String(value.get()),
+      assertEquals(new String(value.get(), HConstants.UTF8_ENCODING),
         ((HStoreKey) midkey).getRow().toString());
 
       // Next test using a midkey that does not exist in the file.

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestScanner2.java

@@ -20,6 +20,7 @@
 package org.apache.hadoop.hbase;
 
 import java.io.IOException;
+import java.io.UnsupportedEncodingException;
 import java.util.ArrayList;
 import java.util.HashSet;
 import java.util.List;
@@ -58,9 +59,18 @@ public class TestScanner2 extends HBaseClusterTestCase {
   final char LAST_ROWKEY = 'z';
   final char FIRST_COLKEY = '0';
   final char LAST_COLKEY = '3';
-  final byte[] GOOD_BYTES = "goodstuff".getBytes();
-  final byte[] BAD_BYTES = "badstuff".getBytes();
+  static byte[] GOOD_BYTES = null;
+  static byte[] BAD_BYTES = null;
 
+  static {
+    try {
+      GOOD_BYTES = "goodstuff".getBytes(HConstants.UTF8_ENCODING);
+      BAD_BYTES = "badstuff".getBytes(HConstants.UTF8_ENCODING);
+    } catch (UnsupportedEncodingException e) {
+      fail();
+    }
+  }
+  
   /**
    * Test the scanner's handling of various filters.  
    * 
@@ -260,7 +270,8 @@ public class TestScanner2 extends HBaseClusterTestCase {
     HTable t = new HTable(conf, table);
     try {
       long lockid = t.startUpdate(region.getRegionName());
-      t.put(lockid, HConstants.COL_REGIONINFO, Writables.getBytes(region.getRegionInfo()));
+      t.put(lockid, HConstants.COL_REGIONINFO,
+          Writables.getBytes(region.getRegionInfo()));
       t.put(lockid, HConstants.COL_SERVER,
         Writables.stringToBytes(serverAddress.toString()));
       t.put(lockid, HConstants.COL_STARTCODE, Writables.longToBytes(startCode));

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestSplit.java

@@ -134,7 +134,8 @@ public class TestSplit extends HBaseTestCase {
     // of each.
     int interval = (LAST_CHAR - FIRST_CHAR) / 3;
     for (HRegion r: sortedMap.values()) {
-      assertGet(r, COLFAMILY_NAME3, new Text(new String(b)));
+      assertGet(r, COLFAMILY_NAME3,
+          new Text(new String(b, HConstants.UTF8_ENCODING)));
       b[0] += interval;
     }
   }

src/contrib/hbase/src/test/org/apache/hadoop/hbase/TestTableMapReduce.java

@@ -20,6 +20,7 @@
 package org.apache.hadoop.hbase;
 
 import java.io.IOException;
+import java.io.UnsupportedEncodingException;
 import java.util.Map;
 import java.util.TreeMap;
 
@@ -60,12 +61,20 @@ public class TestTableMapReduce extends HBaseTestCase {
   private Path dir;
   private MiniHBaseCluster hCluster = null;
   
-  private byte[][] values = {
-      "0123".getBytes(),
-      "abcd".getBytes(),
-      "wxyz".getBytes(),
-      "6789".getBytes()
-  };
+  private static byte[][] values = null;
+  
+  static {
+    try {
+      values = new byte[][] {
+          "0123".getBytes(HConstants.UTF8_ENCODING),
+          "abcd".getBytes(HConstants.UTF8_ENCODING),
+          "wxyz".getBytes(HConstants.UTF8_ENCODING),
+          "6789".getBytes(HConstants.UTF8_ENCODING)
+      };
+    } catch (UnsupportedEncodingException e) {
+      fail();
+    }
+  }
 
   /**
    * {@inheritDoc}
@@ -144,7 +153,8 @@ public class TestTableMapReduce extends HBaseTestCase {
       // Get the original value and reverse it
       
       String originalValue =
-        new String(((ImmutableBytesWritable)value.get(keys[0])).get());
+        new String(((ImmutableBytesWritable)value.get(keys[0])).get(),
+            HConstants.UTF8_ENCODING);
       StringBuilder newValue = new StringBuilder();
       for(int i = originalValue.length() - 1; i >= 0; i--) {
         newValue.append(originalValue.charAt(i));
@@ -153,8 +163,8 @@ public class TestTableMapReduce extends HBaseTestCase {
       // Now set the value to be collected
 
       MapWritable outval = new MapWritable();
-      outval.put(TEXT_OUTPUT_COLUMN,
-          new ImmutableBytesWritable(newValue.toString().getBytes()));
+      outval.put(TEXT_OUTPUT_COLUMN, new ImmutableBytesWritable(
+          newValue.toString().getBytes(HConstants.UTF8_ENCODING)));
       
       output.collect(tKey, outval);
     }
@@ -297,7 +307,7 @@ public class TestTableMapReduce extends HBaseTestCase {
         
         for(Map.Entry<Text, byte[]> e: results.entrySet()) {
           LOG.info(" column: " + e.getKey() + " value: "
-              + new String(e.getValue()));
+              + new String(e.getValue(), HConstants.UTF8_ENCODING));
         }
       }
       

src/contrib/hbase/src/test/org/apache/hadoop/hbase/filter/TestRegExpRowFilter.java

@@ -23,12 +23,14 @@ import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.DataInputStream;
 import java.io.DataOutputStream;
+import java.io.UnsupportedEncodingException;
 import java.util.Map;
 import java.util.TreeMap;
 
 import junit.framework.TestCase;
 
 import org.apache.hadoop.hbase.HConstants;
+import org.apache.hadoop.hbase.HGlobals;
 import org.apache.hadoop.io.Text;
 
 /**
@@ -39,9 +41,16 @@ public class TestRegExpRowFilter extends TestCase {
   RowFilterInterface mainFilter;
   final char FIRST_CHAR = 'a';
   final char LAST_CHAR = 'e';
-  byte [] GOOD_BYTES = "abc".getBytes();
   final String HOST_PREFIX = "org.apache.site-";
-  
+  static byte [] GOOD_BYTES = null;
+
+  static {
+    try {
+      GOOD_BYTES = "abc".getBytes(HConstants.UTF8_ENCODING);
+    } catch (UnsupportedEncodingException e) {
+      fail();
+    }
+  }
   /** {@inheritDoc} */
   @Override
   protected void setUp() throws Exception {
@@ -112,7 +121,9 @@ public class TestRegExpRowFilter extends TestCase {
       yahooSite, filter.filter(new Text(yahooSite)));
   }
   
-  private void regexRowColumnTests(RowFilterInterface filter) {
+  private void regexRowColumnTests(RowFilterInterface filter)
+    throws UnsupportedEncodingException {
+    
     for (char c = FIRST_CHAR; c <= LAST_CHAR; c++) {
       Text t = createRow(c);
       for (Map.Entry<Text, byte []> e: this.colvalues.entrySet()) {
@@ -129,7 +140,7 @@ public class TestRegExpRowFilter extends TestCase {
     
     // Do same but with bad bytes.
     assertTrue("Failed with character " + c,
-      filter.filter(r, col, "badbytes".getBytes()));
+      filter.filter(r, col, "badbytes".getBytes(HConstants.UTF8_ENCODING)));
     
     // Do with good bytes but bad column name.  Should not filter out.
     assertFalse("Failed with character " + c,
@@ -175,7 +186,7 @@ public class TestRegExpRowFilter extends TestCase {
     // that maps to a null value.
     // Testing row with columnKeys: a-e, e maps to null
     colvalues.put(new Text(new String(new char[] { LAST_CHAR })), 
-      HConstants.DELETE_BYTES.get());
+      HGlobals.deleteBytes.get());
     assertFalse("Failed with last columnKey " + LAST_CHAR + " mapping to null.", 
       filter.filterNotNull(colvalues));
   }

src/contrib/hbase/src/test/org/apache/hadoop/hbase/filter/TestRowFilterSet.java

@@ -23,10 +23,12 @@ import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.DataInputStream;
 import java.io.DataOutputStream;
+import java.io.UnsupportedEncodingException;
 import java.util.HashSet;
 import java.util.Set;
 import java.util.TreeMap;
 
+import org.apache.hadoop.hbase.HConstants;
 import org.apache.hadoop.io.Text;
 
 
@@ -42,10 +44,19 @@ public class TestRowFilterSet extends TestCase {
   static final int MAX_PAGES = 5;
   final char FIRST_CHAR = 'a';
   final char LAST_CHAR = 'e';
-  final byte[] GOOD_BYTES = "abc".getBytes();
-  final byte[] BAD_BYTES = "def".getBytes();
   TreeMap<Text, byte[]> colvalues;
+  static byte[] GOOD_BYTES = null;
+  static byte[] BAD_BYTES = null;
 
+  static {
+    try {
+      GOOD_BYTES = "abc".getBytes(HConstants.UTF8_ENCODING);
+      BAD_BYTES = "def".getBytes(HConstants.UTF8_ENCODING);
+    } catch (UnsupportedEncodingException e) {
+      fail();
+    }
+  }
+  
   /** {@inheritDoc} */
   @Override
   protected void setUp() throws Exception {

src/contrib/hbase/src/test/org/onelab/test/StringKey.java

@@ -48,6 +48,8 @@
  */
 package org.onelab.test;
 
+import java.io.UnsupportedEncodingException;
+import org.apache.hadoop.hbase.HConstants;
 import org.onelab.filter.Key;
 
 /**
@@ -70,9 +72,10 @@ public class StringKey extends Key {
    * Construct a Key using the specified String and default weight
    * 
    * @param key String key value
+   * @throws UnsupportedEncodingException
    */
-  public StringKey(String key){
-    super(key.getBytes());
+  public StringKey(String key) throws UnsupportedEncodingException {
+    super(key.getBytes(HConstants.UTF8_ENCODING));
   }
 
   /**
@@ -80,9 +83,12 @@ public class StringKey extends Key {
    * 
    * @param key - String key value
    * @param weight key weight
+   * @throws UnsupportedEncodingException
    */
-  public StringKey(String key, double weight){
-    super(key.getBytes(), weight);
+  public StringKey(String key, double weight)
+    throws UnsupportedEncodingException {
+    
+    super(key.getBytes(HConstants.UTF8_ENCODING), weight);
   }
 
 }

src/contrib/hbase/src/test/org/onelab/test/TestFilter.java

@@ -48,6 +48,7 @@
  */
 package org.onelab.test;
 
+import java.io.UnsupportedEncodingException;
 import junit.framework.TestCase;
 
 import org.onelab.filter.*;
@@ -61,8 +62,10 @@ import org.onelab.filter.*;
  */
 public class TestFilter extends TestCase {
   
-  /** Test a BloomFilter */
-  public void testBloomFilter() {
+  /** Test a BloomFilter
+   * @throws UnsupportedEncodingException
+   */
+  public void testBloomFilter() throws UnsupportedEncodingException {
     Filter bf = new BloomFilter(8, 2);
     Key key = new StringKey("toto");
     Key k2 = new StringKey("lulu");
@@ -76,8 +79,10 @@ public class TestFilter extends TestCase {
     assertTrue(bf.membershipTest(new StringKey("abcd")));       // False positive
   }
   
-  /** Test a CountingBloomFilter */
-  public void testCountingBloomFilter() {
+  /** Test a CountingBloomFilter
+   * @throws UnsupportedEncodingException
+   */
+  public void testCountingBloomFilter() throws UnsupportedEncodingException {
     Filter bf = new CountingBloomFilter(8, 2);
     Key key = new StringKey("toto");
     Key k2 = new StringKey("lulu");
@@ -91,8 +96,10 @@ public class TestFilter extends TestCase {
     assertTrue(bf.membershipTest(new StringKey("abcd")));       // False positive
   }
   
-  /** Test a DynamicBloomFilter */
-  public void testDynamicBloomFilter() {
+  /** Test a DynamicBloomFilter
+   * @throws UnsupportedEncodingException
+   */
+  public void testDynamicBloomFilter() throws UnsupportedEncodingException {
     Filter bf = new DynamicBloomFilter(8, 2, 2);
     Key key = new StringKey("toto");
     Key k2 = new StringKey("lulu");