Use hex string as the representation of byte array for queries (apach…

…e#4041) (#4) * Use hex string as the representation of byte array for queries * Adding test for indexing hexstring * Address comments
winedepot · Apr 1, 2019 · 80c633b · 80c633b
1 parent 2995f9c
commit 80c633b
Show file tree

Hide file tree

Showing 3 changed files with 63 additions and 9 deletions.
diff --git a/...in/java/org/apache/pinot/core/realtime/impl/dictionary/BytesOffHeapMutableDictionary.java b/...in/java/org/apache/pinot/core/realtime/impl/dictionary/BytesOffHeapMutableDictionary.java
@@ -21,6 +21,9 @@
 import java.io.IOException;
 import java.util.Arrays;
 import javax.annotation.Nonnull;
+import org.apache.commons.codec.DecoderException;
+import org.apache.commons.codec.binary.Hex;
+import org.apache.pinot.common.Utils;
 import org.apache.pinot.common.utils.primitive.ByteArray;
 import org.apache.pinot.core.io.readerwriter.PinotDataBufferMemoryManager;
 import org.apache.pinot.core.io.writer.impl.MutableOffHeapByteArrayStore;
@@ -30,6 +33,7 @@
  * OffHeap mutable dictionary for Bytes data type.
  */
 public class BytesOffHeapMutableDictionary extends BaseOffHeapMutableDictionary {
+
   private final MutableOffHeapByteArrayStore _byteStore;
 
   private ByteArray _min = null;
@@ -52,8 +56,19 @@ public BytesOffHeapMutableDictionary(int estimatedCardinality, int maxOverflowHa
 
   @Override
   public int indexOf(Object rawValue) {
-    assert rawValue instanceof byte[];
-    byte[] bytes = (byte[]) rawValue;
+    byte[] bytes = null;
+    // Convert hex string to byte[].
+    if (rawValue instanceof byte[]) {
+      bytes = (byte[]) rawValue;
+    } else if (rawValue instanceof String) {
+      try {
+        bytes = Hex.decodeHex(((String) rawValue).toCharArray());
+      } catch (DecoderException e) {
+        Utils.rethrowException(e);
+      }
+    } else {
+      assert rawValue instanceof byte[];
+    }
     return getDictId(new ByteArray(bytes), bytes);
   }
 
@@ -80,8 +95,18 @@ protected void setRawValueAt(int dictId, Object rawValue, byte[] serializedValue
 
   @Override
   public void index(@Nonnull Object rawValue) {
-    assert rawValue instanceof byte[];
-    byte[] bytes = (byte[]) rawValue;
+    byte[] bytes = null;
+    // Convert hex string to byte[].
+    if (rawValue instanceof String) {
+      try {
+        bytes = Hex.decodeHex(((String) rawValue).toCharArray());
+      } catch (DecoderException e) {
+        Utils.rethrowException(e);
+      }
+    } else {
+      assert rawValue instanceof byte[];
+      bytes = (byte[]) rawValue;
+    }
     ByteArray byteArray = new ByteArray(bytes);
     indexValue(byteArray, bytes);
     updateMinMax(byteArray);

diff --git a/...ain/java/org/apache/pinot/core/realtime/impl/dictionary/BytesOnHeapMutableDictionary.java b/...ain/java/org/apache/pinot/core/realtime/impl/dictionary/BytesOnHeapMutableDictionary.java
@@ -20,20 +20,36 @@
 
 import java.util.Arrays;
 import javax.annotation.Nonnull;
+import org.apache.commons.codec.DecoderException;
+import org.apache.commons.codec.binary.Hex;
+import org.apache.pinot.common.Utils;
 import org.apache.pinot.common.utils.primitive.ByteArray;
 
 
 /**
  * OnHeap mutable dictionary of Bytes type.
  */
 public class BytesOnHeapMutableDictionary extends BaseOnHeapMutableDictionary {
+
   private ByteArray _min = null;
   private ByteArray _max = null;
 
   @Override
   public int indexOf(Object rawValue) {
-    assert rawValue instanceof byte[];
-    return getDictId(new ByteArray((byte[]) rawValue));
+    byte[] bytes = null;
+    // Convert hex string to byte[].
+    if (rawValue instanceof byte[]) {
+      bytes = (byte[]) rawValue;
+    } else if (rawValue instanceof String) {
+      try {
+        bytes = Hex.decodeHex(((String) rawValue).toCharArray());
+      } catch (DecoderException e) {
+        Utils.rethrowException(e);
+      }
+    } else {
+      assert rawValue instanceof byte[];
+    }
+    return getDictId(new ByteArray(bytes));
   }
 
   @Override
@@ -48,8 +64,19 @@ public byte[] getBytesValue(int dictId) {
 
   @Override
   public void index(@Nonnull Object rawValue) {
-    assert rawValue instanceof byte[];
-    ByteArray byteArray = new ByteArray((byte[]) rawValue);
+    byte[] bytes = null;
+    // Convert hex string to byte[].
+    if (rawValue instanceof String) {
+      try {
+        bytes = Hex.decodeHex(((String) rawValue).toCharArray());
+      } catch (DecoderException e) {
+        Utils.rethrowException(e);
+      }
+    } else {
+      assert rawValue instanceof byte[];
+      bytes = (byte[]) rawValue;
+    }
+    ByteArray byteArray = new ByteArray(bytes);
     indexValue(byteArray);
     updateMinMax(byteArray);
   }

diff --git a/...e/src/test/java/org/apache/pinot/core/realtime/impl/dictionary/MutableDictionaryTest.java b/...e/src/test/java/org/apache/pinot/core/realtime/impl/dictionary/MutableDictionaryTest.java
@@ -30,6 +30,7 @@
 import java.util.concurrent.Executors;
 import java.util.concurrent.Future;
 import java.util.stream.Collectors;
+import org.apache.commons.codec.binary.Hex;
 import org.apache.commons.lang.RandomStringUtils;
 import org.apache.pinot.common.data.FieldSpec;
 import org.apache.pinot.common.utils.primitive.ByteArray;
@@ -174,7 +175,8 @@ private void testMutableDictionary(MutableDictionary dictionary, FieldSpec.DataT
       Comparable value =
           (i == 0 && dataType == FieldSpec.DataType.INT) ? Integer.MIN_VALUE : makeRandomObjectOfType(dataType);
 
-      Object rawValue = dataType == FieldSpec.DataType.BYTES ? ((ByteArray) value).getBytes() : value;
+      Object rawValue = dataType == FieldSpec.DataType.BYTES ? ((i % 2 == 0) ? ((ByteArray) value).getBytes()
+          : Hex.encodeHexString(((ByteArray) value).getBytes())) : value;
       if (valueToDictId.containsKey(value)) {
         Assert.assertEquals(dictionary.indexOf(rawValue), (int) valueToDictId.get(value));
       } else {