http://www.w3cschool.cc/design-pattern/abstract-factory-pattern.html,《hadoop技术内幕》
意图:提供一个创建一系列相关或相互依赖对象的接口,而无需指定它们具体的类。
主要解决:主要解决接口选择的问题。
何时使用:系统的产品有多于一个的产品族,而系统只消费其中某一族的产品。
如何解决:在一个产品族里面,定义多个产品。
关键代码:在一个工厂里聚合多个同类产品。
应用实例:工作了,为了参加一些聚会,肯定有两套或多套衣服吧,比如说有商务装(成套,一系列具体产品)、时尚装(成套,一系列具体产品),甚至对于一个家庭来说,可能有商务女装、商务男装、时尚女装、时尚男装,这些也都是成套的,即一系列具体产品。假设一种情况(现实中是不存在的,要不然,没法进入共产主义了,但有利于说明抽象工厂模式),在您的家中,某一个衣柜(具体工厂)只能存放某一种这样的衣服(成套,一系列具体产品),每次拿这种成套的衣服时也自然要从这个衣柜中取出了。用 OO 的思想去理解,所有的衣柜(具体工厂)都是衣柜类的(抽象工厂)某一个,而每一件成套的衣服又包括具体的上衣(某一具体产品),裤子(某一具体产品),这些具体的上衣其实也都是上衣(抽象产品),具体的裤子也都是裤子(另一个抽象产品)。
优点:当一个产品族中的多个对象被设计成一起工作时,它能保证客户端始终只使用同一个产品族中的对象。
缺点:产品族扩展非常困难,要增加一个系列的某一产品,既要在抽象的 Creator 里加代码,又要在具体的里面加代码。
使用场景: 1、QQ 换皮肤,一整套一起换。 2、生成不同操作系统的程序。
注意事项:产品族难扩展,产品等级易扩展。
总结:抽象工厂就是生产工厂的工厂,可以从里边取一类产品。hadoop序列化框架中,Serializer和Deserializer实例相反,相互依赖,必须通过抽象工厂Serialization,才能获得对应的实现。
相关代码如下(hadoop-1.0.0的代码,省略注释):
package org.apache.hadoop.io.serializer; public interface Serialization<T> { //Serialization抽象工厂,为Serializer和Deserializer对象创建抽象类来获取工厂。 boolean accept(Class<?> c); Serializer<T> getSerializer(Class<T> c); //得到Serializer,定义为接口 Deserializer<T> getDeserializer(Class<T> c); //得到Deserializer,定义为接口 }
package org.apache.hadoop.io.serializer; import java.io.IOException; import java.io.OutputStream; public interface Serializer<T> { //为Serializer创建一个接口 void open(OutputStream out) throws IOException; void serialize(T t) throws IOException; void close() throws IOException; }
package org.apache.hadoop.io.serializer; import java.io.IOException; import java.io.InputStream; public interface Deserializer<T> { //为Deserializer创建一个接口 void open(InputStream in) throws IOException; T deserialize(T t) throws IOException; void close() throws IOException; }
package org.apache.hadoop.io.serializer; import java.io.IOException; import java.io.InputStream; import java.io.ObjectInputStream; import java.io.ObjectOutputStream; import java.io.OutputStream; import java.io.Serializable; public class JavaSerialization implements Serialization<Serializable> { //创建扩展了Serialization的工厂类JavaSerialization ,基于给定的信息生成实体类的对象。 static class JavaSerializationDeserializer<T extends Serializable> implements Deserializer<T> { private ObjectInputStream ois; public void open(InputStream in) throws IOException { ois = new ObjectInputStream(in) { @Override protected void readStreamHeader() { // no header } }; } @SuppressWarnings("unchecked") public T deserialize(T object) throws IOException { try { // ignore passed-in object return (T) ois.readObject(); } catch (ClassNotFoundException e) { throw new IOException(e.toString()); } } public void close() throws IOException { ois.close(); } } static class JavaSerializationSerializer implements Serializer<Serializable> { private ObjectOutputStream oos; public void open(OutputStream out) throws IOException { oos = new ObjectOutputStream(out) { @Override protected void writeStreamHeader() { // no header } }; } public void serialize(Serializable object) throws IOException { oos.reset(); // clear (class) back-references oos.writeObject(object); } public void close() throws IOException { oos.close(); } } public boolean accept(Class<?> c) { return Serializable.class.isAssignableFrom(c); } public Deserializer<Serializable> getDeserializer(Class<Serializable> c) { return new JavaSerializationDeserializer<Serializable>(); } public Serializer<Serializable> getSerializer(Class<Serializable> c) { return new JavaSerializationSerializer(); } }
package org.apache.hadoop.io.serializer; import java.io.DataInputStream; import java.io.DataOutputStream; import java.io.IOException; import java.io.InputStream; import java.io.OutputStream; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.conf.Configured; import org.apache.hadoop.io.Writable; import org.apache.hadoop.util.ReflectionUtils; public class WritableSerialization extends Configured implements Serialization<Writable> { //创建扩展了Serialization的工厂类WritableSerialization,基于给定的信息生成实体类的对象。 static class WritableDeserializer extends Configured implements Deserializer<Writable> { private Class<?> writableClass; private DataInputStream dataIn; public WritableDeserializer(Configuration conf, Class<?> c) { setConf(conf); this.writableClass = c; } public void open(InputStream in) { if (in instanceof DataInputStream) { dataIn = (DataInputStream) in; } else { dataIn = new DataInputStream(in); } } public Writable deserialize(Writable w) throws IOException { Writable writable; if (w == null) { writable = (Writable) ReflectionUtils.newInstance(writableClass, getConf()); } else { writable = w; } writable.readFields(dataIn); return writable; } public void close() throws IOException { dataIn.close(); } } static class WritableSerializer implements Serializer<Writable> { private DataOutputStream dataOut; public void open(OutputStream out) { if (out instanceof DataOutputStream) { dataOut = (DataOutputStream) out; } else { dataOut = new DataOutputStream(out); } } public void serialize(Writable w) throws IOException { w.write(dataOut); } public void close() throws IOException { dataOut.close(); } } public boolean accept(Class<?> c) { return Writable.class.isAssignableFrom(c); } public Deserializer<Writable> getDeserializer(Class<Writable> c) { return new WritableDeserializer(getConf(), c); } public Serializer<Writable> getSerializer(Class<Writable> c) { return new WritableSerializer(); } }
package org.apache.hadoop.io.serializer; import java.util.ArrayList; import java.util.List; import org.apache.commons.logging.Log; import org.apache.commons.logging.LogFactory; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.conf.Configured; import org.apache.hadoop.util.ReflectionUtils; import org.apache.hadoop.util.StringUtils; public class SerializationFactory extends Configured { //创建一个工厂创造器/生成器类,通过传递类(Class<T> c)信息来获取工厂。 private static final Log LOG = LogFactory.getLog(SerializationFactory.class.getName()); private List<Serialization<?>> serializations = new ArrayList<Serialization<?>>(); public SerializationFactory(Configuration conf) { super(conf); for (String serializerName : conf.getStrings("io.serializations", new String[]{"org.apache.hadoop.io.serializer.WritableSerialization"})) { add(conf, serializerName); } } @SuppressWarnings("unchecked") private void add(Configuration conf, String serializationName) { try { Class<? extends Serialization> serializionClass = (Class<? extends Serialization>) conf.getClassByName(serializationName); serializations.add((Serialization) ReflectionUtils.newInstance(serializionClass, getConf())); } catch (ClassNotFoundException e) { LOG.warn("Serilization class not found: " + StringUtils.stringifyException(e)); } } public <T> Serializer<T> getSerializer(Class<T> c) { return getSerialization(c).getSerializer(c); } public <T> Deserializer<T> getDeserializer(Class<T> c) { return getSerialization(c).getDeserializer(c); } @SuppressWarnings("unchecked") public <T> Serialization<T> getSerialization(Class<T> c) { for (Serialization serialization : serializations) { if (serialization.accept(c)) { return (Serialization<T>) serialization; } } return null; } }
原文地址:http://blog.csdn.net/kakarot5/article/details/40376131