Berkeley DB

Berkeley DB 是一个嵌入式数据库，它适合于管理海量的(256T)、简单的数据。BDB是以键值对(value/key)来存储和管理数据库的。键可以重复，数据值可以是任意类型的。BDB的底层是用B+树或者其他算法实现的。我用的jar包是B+树实现的版本。

Berkeley DB是用Environment对象来管理数据库，一个Environment可以管理多个database。每个database都存储键值对，而序列化到磁盘上是通过catalog实现的。BDB的操作是在内存和磁盘上的，最终BDB的存取结果集在程序中使用是通过容器实现的（数据库在程序中的视图）。

所以BDB的使用分为5步：

• 创建Environment

    //Open Environment
    private Environment environment;
    //同EnvironmentConfig来配置环境
    EnvironmentConfig environmentConfig=new EnvironmentConfig();
    environmentConfig.setTransactional(true);
    environmentConfig.setAllowCreate(true);
    //homeDirectory是数据库存放的目录
    environment=new Environment(new File(homeDirectory),environmentConfig);

• 打开database和catalog

    protected StoredClassCatalog catalog;//catalog
    protected Database database;//database
    private static final String CLASS_CATALOG="java_class_catalog";//数据库名
    protected Database catalogDatabase;//catalog存放处

    //open Database
    DatabaseConfig dbConfig=new DatabaseConfig();//数据库配置
    dbConfig.setTransactional(true);
    catalogDBConfig.setAllowCreate(true);
    dbConfig.setSortedDuplicates(false);//不存重复键值
    database=environment.openDatabase(null, "URL", dbConfig);//存放实际数据的数据库   

    //Open Catalog
    DatabaseConfig catalogDBConfig=new DatabaseConfig();//数据库配置
    catalogDBConfig.setTransactional(true);
    catalogDBConfig.setAllowCreate(true);
    catalogDatabase=environment.openDatabase(null, CLASS_CATALOG, catalogDBConfig);
    //用StoredClassCatalog类将catalog保存在db中并返回可操作对象。
    catalog=new StoredClassCatalog(catalogDatabase);

• 存储序列化类型绑定

    //键绑定
    EntryBinding<Integer> keyBinding=new SerialBinding<Integer>(catalog,Integer.class);
    //值绑定
    SerialBinding<Url> valueBinding=new SerialBinding<Url>(catalog,Url.class);

• 存储结果容器访问

    StoreMap<Integer,Url> urlMap;//结果集的操作容器（视图）
    urlMap=new StoredMap<Integer,Url>(database,keyBinding,valueBinding,true);

• 关闭资源

     database.close();
     catalog.close();
     environment.close();

我在爬虫项目中庸BDB作为内存数据库来保存未访问的URL。由于StoredMap元素存储无序，StoredsortedMap给元素排序后存储。但都不是队列的序列。故本例中以整数位主键存储，记录整数的值来模拟队列的头和尾。StoredMap实现了Map接口，可以使用其所有函数，例如：get()、remove()、put()等。

java代码如下：

//BDBFrontier.java
import java.io.File;
import java.io.FileNotFoundException;

import com.sleepycat.bind.serial.StoredClassCatalog;
import com.sleepycat.je.Database;
import com.sleepycat.je.DatabaseConfig;
import com.sleepycat.je.DatabaseException;
import com.sleepycat.je.Environment;
import com.sleepycat.je.EnvironmentConfig;

public abstract class BDBFrontier{
    public BDBFrontier(String homeDirectory) throws DatabaseException,FileNotFoundException{
        //Open Environment
        System.out.println("Opening environment in: "+homeDirectory);
        EnvironmentConfig environmentConfig=new EnvironmentConfig();
        environmentConfig.setTransactional(true);
        environmentConfig.setAllowCreate(true);
        environment=new Environment(new File(homeDirectory),environmentConfig);

        //Open Catalog
        DatabaseConfig catalogDBConfig=new DatabaseConfig();
        catalogDBConfig.setTransactional(true);
        catalogDBConfig.setAllowCreate(true);
        catalogDatabase=environment.openDatabase(null, CLASS_CATALOG, catalogDBConfig);
        catalog=new StoredClassCatalog(catalogDatabase);

        //open Database
        DatabaseConfig dbConfig=new DatabaseConfig();
        dbConfig.setTransactional(true);
        dbConfig.setAllowCreate(true);
        database=environment.openDatabase(null, "URL", dbConfig);
    }

    public void close() throws DatabaseException{
        database.close();
        catalog.close();//这句应该可以关闭与之相关的数据库，但是API上没有将
        environment.close();
    }

    protected abstract Object put(Object key,Object value);
    protected abstract Object get(Object key);
    protected abstract Object delete(Object key);

    private Environment environment;
    protected StoredClassCatalog catalog;
    protected Database database;

    private static final String CLASS_CATALOG="java_class_catalog";//name
    protected Database catalogDatabase;
}

//BDBFrontier.java
import java.io.FileNotFoundException;
import com.sleepycat.bind.EntryBinding;
import com.sleepycat.bind.serial.SerialBinding;
import com.sleepycat.collections.StoredMap;
import com.sleepycat.je.DatabaseException;

public class Frontier extends BDBFrontier implements UrlFrontier{

    private StoredMap urlMap=null;//a database view
    private Integer head;//URL队列头
    private Integer tail;//URL队列尾

    public Frontier(String homeDirectory) throws DatabaseException, FileNotFoundException {
        super(homeDirectory);
        // TODO Auto-generated constructor stub
        EntryBinding keyBinding=new SerialBinding(catalog,Integer.class);
        SerialBinding valueBinding=new SerialBinding(catalog,Url.class);
        //创建视图，并设置为可写
        urlMap=new StoredMap(database,keyBinding,valueBinding,true);//true可写
        head=0;
        tail=0;
    }

    @Override
    public Url getNext() throws Exception {
        // TODO Auto-generated method stub
        Url result=null;
        if(!urlMap.isEmpty()){
            result=urlMap.get(head);
            delete(head++);
        }
        return result;
    }

    @Override
    public boolean putUrl(Url url) throws Exception {
        // TODO Auto-generated method stub
        if( put(tail++,url) != null) return true;
        else return false;
    }

    @Override
    protected Object put(Object key, Object value) {
        // TODO Auto-generated method stub
        return urlMap.put((Integer)key, (Url)value);

    }

    @Override
    protected Object get(Object key) {
        // TODO Auto-generated method stub
        return urlMap.get(key);
    }

    @Override
    protected Object delete(Object key) {
        // TODO Auto-generated method stub
        return urlMap.remove(key);
    }

    public boolean isEmpty() {
        // TODO Auto-generated method stub
        return urlMap.isEmpty();
    }

    public boolean contains(Url url) {
        // TODO Auto-generated method stub
        return urlMap.containsValue(url);
    }

    public static void main(String[] args){
        try{
            Frontier frontier=new Frontier("D:\\workspace\\db");
            Url url=new Url();
            url.setOriUrl("http://www.163.com");
            frontier.putUrl(url);
            url.setOriUrl("http://www.164.com");
            frontier.putUrl(url);
            url.setOriUrl("http://www.165.com");
            frontier.putUrl(url);
            System.out.println(frontier.getNext().getOriUrl());
            System.out.println(frontier.getNext().getOriUrl());
            System.out.println(frontier.getNext().getOriUrl());
            frontier.close();
        }catch(Exception e){
            e.printStackTrace();
        }finally{

        }
    }
}

参考文献
自己动手写网络爬虫
嵌入式数据库系统Berkeley DB
Berkeley DB 使用经验总结