Added serialization

.gitignore

@@ -21,6 +21,7 @@ log
 .classpath
 .project
 .settings/
+**/.factorypath
 
 # Latex and R related files
 _region_.prv/

benchmark/pom.xml

@@ -42,9 +42,9 @@
                 <artifactId>maven-compiler-plugin</artifactId>
                 <version>3.0</version>
                 <configuration>
-                    <compilerVersion>1.7</compilerVersion>
-                    <source>1.7</source>
-                    <target>1.7</target>
+                    <compilerVersion>${java.version}</compilerVersion>
+                    <source>${java.version}</source>
+                    <target>${java.version}</target>
                 </configuration>
             </plugin>
 

core/pom.xml

@@ -144,8 +144,8 @@
                 <configuration>
                     <verbose>true</verbose>
                     <compilerVersion>1.8</compilerVersion>
-                    <source>1.7</source>
-                    <target>1.7</target>
+                    <source>${java.version}</source>
+                    <target>${java.version}</target>
                 </configuration>
             </plugin>
             <plugin>

core/src/main/java/com/tdunning/math/stats/AbstractTDigestSerializer.java

@@ -0,0 +1,16 @@
+package com.tdunning.math.stats;
+
+import java.util.Properties;
+
+public abstract class AbstractTDigestSerializer<T extends TDigest, O extends Object> implements TDigestSerializer<T, O> {
+    private Properties properties;
+
+    public AbstractTDigestSerializer(Properties properties) {
+        this.properties = properties;
+    }    
+
+    public Properties getProperties() {
+        return properties;
+    }
+
+}

core/src/main/java/com/tdunning/math/stats/TDigestDefaultSerializer.java

@@ -0,0 +1,36 @@
+package com.tdunning.math.stats;
+
+import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.util.Properties;
+
+public class TDigestDefaultSerializer extends AbstractTDigestSerializer<TDigest, byte[]> {
+
+    public TDigestDefaultSerializer(Properties properties) {
+        super(properties);
+    }
+
+    @Override
+    public byte[] serialize(TDigest tDigest) throws TDigestSerializerException {
+        ByteArrayOutputStream baos = new ByteArrayOutputStream(5120);
+        try (ObjectOutputStream out = new ObjectOutputStream(baos)) {
+            out.writeObject(tDigest);
+            return baos.toByteArray();
+        } catch (IOException e) {
+            throw new TDigestSerializerException(e);
+        } 
+    }
+
+    @Override
+    public TDigest deserialize(byte[] object) throws TDigestSerializerException {
+        try (ObjectInputStream in = new ObjectInputStream(new ByteArrayInputStream(object))) {
+            return (TDigest) in.readObject();
+        } catch (ClassCastException | ClassNotFoundException | IOException e) {
+            throw new TDigestSerializerException(e);
+        }
+    }
+
+}

core/src/main/java/com/tdunning/math/stats/TDigestSerializer.java

@@ -0,0 +1,32 @@
+package com.tdunning.math.stats;
+
+import java.util.Properties;
+
+public interface TDigestSerializer<T extends TDigest, O extends Object> {
+
+    /**
+     * Serializes a TDigest object
+     * 
+     * @param tDigest
+     * @return
+     * @throws TDigestSerializerException
+     */
+    public O serialize(T tDigest) throws TDigestSerializerException;
+
+    /**
+     * De-serializes an Object into a TDigest
+     * 
+     * @param object
+     * @return
+     * @throws TDigestSerializerException
+     */
+    public T deserialize(O object) throws TDigestSerializerException;
+
+    /**
+     * Returns properties defined in serialization.properties file
+     * 
+     * @return
+     */
+    public Properties getProperties();
+
+}

core/src/main/java/com/tdunning/math/stats/TDigestSerializerException.java

@@ -0,0 +1,17 @@
+package com.tdunning.math.stats;
+
+public class TDigestSerializerException extends Exception {
+
+    public TDigestSerializerException(String error) {
+        super(error);
+    }
+
+    public TDigestSerializerException(String error, Exception e) {
+        super(error, e);
+    }
+
+    public TDigestSerializerException(Exception e) {
+        super(e);
+    }
+
+}

core/src/main/java/com/tdunning/math/stats/TDigestSerializerFactory.java

@@ -0,0 +1,46 @@
+package com.tdunning.math.stats;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.Properties;
+
+public class TDigestSerializerFactory {
+    private static final String PROPERTIES_FILE_NAME = "serialization.properties";
+    private static final String SERIALIZER_PROP_KEY = "tdigest.serializerClass";
+
+    private Class<? extends TDigestSerializer<? extends TDigest, ? extends Object>> clazz;
+    private Properties serializationProperties;
+
+    @SuppressWarnings("unchecked")
+    public TDigestSerializerFactory() {
+        Properties props = new Properties();
+        try {
+            File f = new File(PROPERTIES_FILE_NAME);
+            InputStream is = f.exists() ?
+                new FileInputStream(f)
+                : getClass().getClassLoader().getResourceAsStream(PROPERTIES_FILE_NAME);
+
+            props.load(is);
+            if(!props.containsKey(SERIALIZER_PROP_KEY))
+                throw new IllegalStateException(PROPERTIES_FILE_NAME + " does not contain key " + SERIALIZER_PROP_KEY);
+
+            clazz = (Class<? extends TDigestSerializer<? extends TDigest, ? extends Object>>) Class.forName(props.getProperty(SERIALIZER_PROP_KEY));
+            serializationProperties = props;
+        } catch (IOException | ClassNotFoundException e) {
+            clazz = TDigestDefaultSerializer.class;
+            e.printStackTrace(System.err);
+        }
+    }
+
+    @SuppressWarnings("rawtypes")
+    public TDigestSerializer create() throws TDigestSerializerException {
+        try {
+            return (TDigestSerializer)clazz.getConstructor(Properties.class).newInstance(serializationProperties);
+        } catch (Exception e) {
+            throw new TDigestSerializerException(e);
+        }
+    }
+
+}

core/src/main/resources/serialization.properties

@@ -0,0 +1 @@
+tdigest.serializerClass=com.tdunning.math.stats.TDigestDefaultSerializer

core/src/test/java/com/tdunning/math/stats/LogHistogramTest.java

@@ -21,7 +21,6 @@
 import org.junit.Test;
 
 import java.io.FileNotFoundException;
-import java.io.IOException;
 
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertTrue;

core/src/test/java/com/tdunning/math/stats/ScaleFunctionTests.java

@@ -167,6 +167,7 @@ public void testK() throws FileNotFoundException {
                 }
             }
         }
+        out.close();
     }
 
     @Test

core/src/test/java/com/tdunning/math/stats/TDigestSerializationTest.java

@@ -17,66 +17,55 @@
 
 package com.tdunning.math.stats;
 
-import org.junit.Test;
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
+import static org.junit.Assert.assertNotNull;
 
-import java.io.ByteArrayInputStream;
-import java.io.ByteArrayOutputStream;
-import java.io.IOException;
-import java.io.ObjectInputStream;
-import java.io.ObjectOutputStream;
-import java.io.Serializable;
 import java.util.Iterator;
 import java.util.Random;
 
-import static org.junit.Assert.assertEquals;
-import static org.junit.Assert.assertFalse;
-import static org.junit.Assert.assertNotNull;
+import org.junit.Before;
+import org.junit.Test;
 
 /**
  * Verifies that the various TDigest implementations can be serialized.
  * <p>
  * Serializability is important, for example, if we want to use t-digests with Spark.
  */
 public class TDigestSerializationTest {
+
+    private static TDigestSerializer<TDigest, byte[]> tDigestSerializer = null;
+
+    @Before
+    @SuppressWarnings("unchecked")
+    public void setup() throws TDigestSerializerException {
+        tDigestSerializer = new TDigestSerializerFactory().create();
+    }
+
     @Test
-    public void testMergingDigest() throws IOException {
+    public void testMergingDigest() throws TDigestSerializerException {
         assertSerializesAndDeserializes(new MergingDigest(100));
     }
 
     @Test
-    public void testAVLTreeDigest() throws IOException {
+    public void testAVLTreeDigest() throws TDigestSerializerException {
         assertSerializesAndDeserializes(new AVLTreeDigest(100));
     }
+
 
-    private <T extends TDigest> void assertSerializesAndDeserializes(T tdigest) throws IOException {
-        assertNotNull(deserialize(serialize(tdigest)));
+    @SuppressWarnings("unchecked")
+    private <T extends TDigest> void assertSerializesAndDeserializes(T tdigest) throws TDigestSerializerException {
+        assertNotNull(tDigestSerializer.deserialize(tDigestSerializer.serialize(tdigest)));
 
         final Random gen = new Random();
         for (int i = 0; i < 100000; i++) {
             tdigest.add(gen.nextDouble());
         }
-        T roundTrip = deserialize(serialize(tdigest));
+        T roundTrip = (T) tDigestSerializer.deserialize(tDigestSerializer.serialize(tdigest));
 
         assertTDigestEquals(tdigest, roundTrip);
     }
 
-    private static byte[] serialize(Serializable obj) throws IOException {
-        ByteArrayOutputStream baos = new ByteArrayOutputStream(5120);
-        try (ObjectOutputStream out = new ObjectOutputStream(baos)){
-            out.writeObject(obj);
-            return baos.toByteArray();
-        }
-    }
-
-    private static <T> T deserialize(byte[] objectData) throws IOException {
-        try (ObjectInputStream in = new ObjectInputStream(new ByteArrayInputStream(objectData))) {
-            //noinspection unchecked
-            return (T) in.readObject();
-        } catch (ClassCastException | ClassNotFoundException | IOException e) {
-            throw new IOException(e);
-        }
-    }
-
     private void assertTDigestEquals(TDigest t1, TDigest t2) {
         assertEquals(t1.getMin(), t2.getMin(), 0);
         assertEquals(t1.getMax(), t2.getMax(), 0);

pom.xml

@@ -9,6 +9,10 @@
     <packaging>pom</packaging>
     <version>3.3-SNAPSHOT</version>
     <name>T-digest Parent</name>
+
+    <properties>
+        <java.version>1.7</java.version>
+    </properties>
 
     <modules>
         <module>core</module>