Merge pull request #3 from soundvibe/unlimited-keys

Unlimited keys

README.md

@@ -13,6 +13,17 @@ PalDB is an embeddable persistent key-value store with very fast read performanc
 
 PalDB's JAR is only 65K and has a single dependency (snappy, which isn't mandatory). It's also very easy to use with just a few configuration parameters.
 
+**This is separate evolution of original PalDB.**
+
+Improvements from PalDB 1.0.2
+-------------------
+- StoreReader is now fully thread-safe without any performance decrease.
+- StoreReader and StoreWriter can be used with generics (StoreReader<K<V> and StoreWriter<K,V>)
+- Typesafe ``PalDBConfigBuilder`` for easier configuration
+- There are no limits on how many keys you can store
+- Duplicates could be optionally allowed
+- Bloom filters could be enabled for even better read performance in some cases
+
 Performance
 -----------
 
@@ -74,12 +85,12 @@ PalDB is available on Maven Central, hence just add the following dependency:
 <dependency>
     <groupId>net.soundvibe</groupId>
     <artifactId>paldb</artifactId>
-    <version>2.0.1</version>
+    <version>2.0.2</version>
 </dependency>
 ```
 Scala SBT
 ```
-libraryDependencies += "net.soundvibe" % "paldb" % "2.0.1"
+libraryDependencies += "net.soundvibe" % "paldb" % "2.0.2"
 ```
 
 
@@ -92,7 +103,7 @@ No, the final binary file is created when `StoreWriter.close()` is called.
 
 **Are duplicate keys allowed?**
 
-No, duplicate keys aren't allowed and an exception will be thrown.
+Duplicates are not allowed by default. But it could be changed in writer configuration if needed.
 
 **Do keys have an order when iterating?**
 
@@ -131,6 +142,7 @@ Write parameters:
 + `compression.enabled`, enable compression (boolean) [default: false]
 + `bloom.filter.enabled`, enable bloom filter (boolean) [default: false]
 + `bloom.filter.error.factor`, bloom filter error rate  (double) [default: 0.01]
++ `duplicates.enabled`, allow duplicates  (boolean) [default: false]
 
 Read parameters:
 
@@ -204,7 +216,6 @@ Machine-learning applications often have complex binary model files created in t
 Limitations
 -----------
 + PalDB is optimal in replacing the usage of large in-memory data storage but still use memory (off-heap, yet much less) to do its job. Disabling memory mapping and relying on seeks is possible but is not what PalDB has been optimized for.
-+ The size of the index is limited to 2GB. There's no limitation in the data size however.
 + PalDB reader is thread-safe but writer is not thread-safe at the moment so synchronization should be done externally if multi-threaded.
 
 Contributions

pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>net.soundvibe</groupId>
     <artifactId>paldb</artifactId>
-    <version>2.0.1</version>
+    <version>2.0.2</version>
     <packaging>jar</packaging>
     <name>paldb</name>
     <description>Embeddable persistent write-once key-value store</description>
@@ -104,7 +104,7 @@
         <dependency>
             <groupId>org.rocksdb</groupId>
             <artifactId>rocksdbjni</artifactId>
-            <version>4.0</version>
+            <version>6.2.2</version>
             <scope>test</scope>
         </dependency>
 

src/main/java/com/linkedin/paldb/api/Configuration.java

@@ -51,6 +51,8 @@ public class Configuration {
   public static final String BLOOM_FILTER_ENABLED = "bloom.filter.enabled";
   //Bloom filter error rate
   public static final String BLOOM_FILTER_ERROR_FACTOR = "bloom.filter.error.factor";
+  //Enable duplicates (keep last key)
+  public static final String ALLOW_DUPLICATES = "duplicates.enabled";
 
   // Property map
   private final Map<String, String> properties = new HashMap<>();
@@ -72,6 +74,7 @@ public Configuration() {
     putWithSystemPropertyDefault(COMPRESSION_ENABLED, "false");
     putWithSystemPropertyDefault(BLOOM_FILTER_ENABLED, "false");
     putWithSystemPropertyDefault(BLOOM_FILTER_ERROR_FACTOR, "0.01");
+    putWithSystemPropertyDefault(ALLOW_DUPLICATES, "false");
 
     //Serializers
     serializers = new Serializers();
@@ -411,7 +414,7 @@ public boolean equals(Object o) {
 
   @Override
   public int hashCode() {
-    int result = properties != null ? properties.hashCode() : 0;
+    int result = properties.hashCode();
     result = 31 * result + (serializers != null ? serializers.hashCode() : 0);
     return result;
   }

src/main/java/com/linkedin/paldb/api/PalDBConfigBuilder.java

@@ -60,6 +60,18 @@ public PalDBConfigBuilder withEnableCompression(boolean enabled) {
         return this;
     }
 
+    /**
+     * <i>PalDB configuration property.</i>
+     * <p>
+     * <code>duplicates.enabled</code> - allow duplicates [default: false]
+     * @param enabled flag
+     * @return this {@code CachemeerConfigBuilder} instance (for chaining)
+     */
+    public PalDBConfigBuilder withEnableDuplicates(boolean enabled) {
+        palDbConfiguration.set(Configuration.ALLOW_DUPLICATES, String.valueOf(enabled));
+        return this;
+    }
+
     /**
      * <i>PalDB configuration property.</i>
      * <p>

src/main/java/com/linkedin/paldb/api/Serializer.java

@@ -29,17 +29,17 @@
  * configuration.registerSerializer(new PointSerializer());
  * </pre>
  *
- * @param <K> class type
+ * @param <T> class type
  */
-public interface Serializer<K> extends Serializable {
+public interface Serializer<T> extends Serializable {
 
   /**
    * Writes the instance <code>input</code> to the data output.
    * @param dataOutput data output
    * @param input instance
    * @throws IOException if an io error occurs
    */
-  void write(DataOutput dataOutput, K input) throws IOException;
+  void write(DataOutput dataOutput, T input) throws IOException;
 
   /**
    * Reads the data input and creates the instance.
@@ -48,7 +48,7 @@ public interface Serializer<K> extends Serializable {
    * @return new instance of type <code>K</code>.
    * @throws IOException if an io error occurs
    */
-  K read(DataInput dataInput) throws IOException;
+  T read(DataInput dataInput) throws IOException;
 
-  Class<K> serializedClass();
+  Class<T> serializedClass();
 }

src/main/java/com/linkedin/paldb/api/errors/DuplicateKeyException.java

@@ -1,6 +1,6 @@
 package com.linkedin.paldb.api.errors;
 
-public class DuplicateKeyException extends RuntimeException {
+public class DuplicateKeyException extends PalDBException {
 
     public DuplicateKeyException(String message) {
         super(message);

src/main/java/com/linkedin/paldb/api/errors/MissingClass.java

@@ -0,0 +1,7 @@
+package com.linkedin.paldb.api.errors;
+
+public class MissingClass extends PalDBException {
+    public MissingClass(String message) {
+        super(message);
+    }
+}

src/main/java/com/linkedin/paldb/api/errors/MissingSerializer.java

@@ -0,0 +1,7 @@
+package com.linkedin.paldb.api.errors;
+
+public class MissingSerializer extends PalDBException {
+    public MissingSerializer(String message) {
+        super(message);
+    }
+}

src/main/java/com/linkedin/paldb/api/errors/OutOfDiskSpace.java

@@ -1,6 +1,6 @@
 package com.linkedin.paldb.api.errors;
 
-public class OutOfDiskSpace extends RuntimeException {
+public class OutOfDiskSpace extends PalDBException {
 
     public OutOfDiskSpace(String message) {
         super(message);

src/main/java/com/linkedin/paldb/api/errors/PalDBException.java

@@ -0,0 +1,8 @@
+package com.linkedin.paldb.api.errors;
+
+public abstract class PalDBException extends RuntimeException {
+
+    public PalDBException(String message) {
+        super(message);
+    }
+}

src/main/java/com/linkedin/paldb/api/errors/UnsupportedTypeException.java

@@ -21,7 +21,7 @@
  *
  * @see StoreReader
  */
-public class UnsupportedTypeException extends RuntimeException {
+public class UnsupportedTypeException extends PalDBException {
 
   public UnsupportedTypeException(Object obj) {
     super("The type '" + obj.getClass() + "' isn't supported");

src/main/java/com/linkedin/paldb/api/errors/VersionMismatch.java

@@ -0,0 +1,8 @@
+package com.linkedin.paldb.api.errors;
+
+public class VersionMismatch extends PalDBException {
+
+    public VersionMismatch(String message) {
+        super(message);
+    }
+}

src/main/java/com/linkedin/paldb/impl/ReaderImpl.java

@@ -110,8 +110,6 @@ public V get(K key, V defaultValue) {
       }
     } catch (IOException ex) {
       throw new UncheckedIOException(ex);
-    } catch (ClassNotFoundException e) {
-      throw new RuntimeException(e);
     }
   }