1、手动调用Class.forName()
我们知道当我们连接MySQL数据库时,会使用如下代码:
1 try { 2 Class.forName("com.mysql.jdbc.Driver"); 3 connection = DriverManager.getConnection("jdbc:mysql://localhost:3306/test", "root", "123456"); 4 } catch (Exception e) { 5 e.printStackTrace(); 6 }
那么Class.forName()有什么作用呢?
首先我们知道Class.forName() 方法要求JVM查找并加载指定的类到内存中,此时将"com.mysql.jdbc.Driver" 当做参数传入,就是告诉JVM,去"com.mysql.jdbc"这个路径下找Driver类,将其加载到内存中。
由于加载类文件时会执行其中的静态代码块,其中Driver类的源码如下
public class Driver extends NonRegisteringDriver implements java.sql.Driver { public Driver() throws SQLException { } static { try { DriverManager.registerDriver(new Driver());//首先new一个Driver对象,并将它注册到DriverManage中 } catch (SQLException var1) { throw new RuntimeException("Can't register driver!"); } } }
接下来我们再看看这个DriverManager.registerDriver 方法:
public static synchronized void registerDriver(java.sql.Driver driver) throws SQLException { registerDriver(driver, null); }
继续看这个registerDriver(driver, null) 方法
private final static CopyOnWriteArrayList<DriverInfo> registeredDrivers = new CopyOnWriteArrayList<>();// registeredDrivers 是一个支持并发的arraylist ...... public static void registerDriver(java.sql.Driver driver, DriverAction da) throws SQLException { if (driver != null) { //如果该驱动尚未注册,那么将他添加到 registeredDrivers 中去。这是一个支持并发情况的特殊ArrayList registeredDrivers.addIfAbsent(new DriverInfo(driver, da)); } else { // This is for compatibility with the original DriverManager throw new NullPointerException(); } println("registerDriver: " + driver); }
此时,Class.forName(“com.mysql.jdbc.Driver”) 的工作就完成了,工作就是:将mysql驱动注册到DriverManager中去。接下来我们看是怎么进行调用的
2、DriverManager.getConnection方法分析
注册到DriverManager中之后,我们就可以通过DriverManager的getConnection方法获得mysql的连接了:
Connection conn = DriverManager.getConnection("jdbc:mysql://localhost:3306/test", "root", "123456");
接下来我们在看看这个getConnection方法:
@CallerSensitive public static Connection getConnection(String url, String user, String password) throws SQLException { .... return (getConnection(url, info, Reflection.getCallerClass())); }
同样,调用了自身的 getConnection方法,继续往下看
1 private static Connection getConnection( 2 String url, java.util.Properties info, Class<?> caller) throws SQLException { 3 ClassLoader callerCL = caller != null ? caller.getClassLoader() : null; 4 synchronized(DriverManager.class) { 5 // synchronize loading of the correct classloader. 6 if (callerCL == null) { 7 callerCL = Thread.currentThread().getContextClassLoader(); 8 } 9 } 10 // Walk through the loaded registeredDrivers attempting to make a connection. 11 // Remember the first exception that gets raised so we can reraise it. 12 SQLException reason = null; 13 14 for(DriverInfo aDriver : registeredDrivers) { 15 // If the caller does not have permission to load the driver then skip it. 16 if(isDriverAllowed(aDriver.driver, callerCL)) { 17 try { 18 Connection con = aDriver.driver.connect(url, info); 19 if (con != null) { 20 // Success! 21 return (con); 22 } 23 } catch (SQLException ex) { 24 if (reason == null) { 25 reason = ex; 26 } 27 } 28 } else { 29 println("skipping: " + aDriver.getClass().getName()); 30 } 31 } 32 33 // if we got here nobody could connect. 34 if (reason != null) { 35 println("getConnection failed: " + reason); 36 throw reason; 37 } 38 throw new SQLException("No suitable driver found for "+ url, "08001"); 39 }
可以看到它对上文提到的静态变量 registeredDrivers 进行了遍历,调用了connect(url, info)方法,这是一个接口,由各个不同的驱动自己实现。
/** * Attempts to make a database connection to the given URL. * The driver should return "null" if it realizes it is the wrong kind * of driver to connect to the given URL. This will be common, as when * the JDBC driver manager is asked to connect to a given URL it passes * the URL to each loaded driver in turn. */ Connection connect(String url, java.util.Properties info) throws SQLException;
到此为止,我们就获得了connection对象,现在就可以对数据库进行操作了。
3、不手动注册驱动也能使用JDBC [ 去除class.forName ]
在高版本的JDK,已经不需要手动调用class.forName方法了,在DriverManager的源码中可以看到一个静态块
/** * Load the initial JDBC drivers by checking the System property * jdbc.properties and then use the {@code ServiceLoader} mechanism */ static { loadInitialDrivers(); println("JDBC DriverManager initialized"); }
进入loadInitialDrivers()方法中看到以下一段代码:
1 private static void loadInitialDrivers() { 2 String drivers; 3 try { 4 drivers = AccessController.doPrivileged(new PrivilegedAction<String>() { 5 public String run() { 6 return System.getProperty("jdbc.drivers"); 7 } 8 }); 9 } catch (Exception ex) { 10 drivers = null; 11 } 12 // If the driver is packaged as a Service Provider, load it. 13 // Get all the drivers through the classloader 14 // exposed as a java.sql.Driver.class service. 15 // ServiceLoader.load() replaces the sun.misc.Providers() 16 17 AccessController.doPrivileged(new PrivilegedAction<Void>() { 18 public Void run() { 19 20 ServiceLoader<Driver> loadedDrivers = ServiceLoader.load(Driver.class); 21 Iterator<Driver> driversIterator = loadedDrivers.iterator(); 22 23 /* Load these drivers, so that they can be instantiated. 24 * It may be the case that the driver class may not be there 25 * i.e. there may be a packaged driver with the service class 26 * as implementation of java.sql.Driver but the actual class 27 * may be missing. In that case a java.util.ServiceConfigurationError 28 * will be thrown at runtime by the VM trying to locate 29 * and load the service. 30 * 31 * Adding a try catch block to catch those runtime errors 32 * if driver not available in classpath but it's 33 * packaged as service and that service is there in classpath. 34 */ 35 try{ 36 while(driversIterator.hasNext()) { 37 driversIterator.next(); 38 } 39 } catch(Throwable t) { 40 // Do nothing 41 } 42 return null; 43 } 44 });
重点是第20行,ServiceLoader.load(Driver.class)
上面这行代码可以把类路径下所有jar包中META-INF/services/java.sql.Driver文件中定义的类加载上来,此类必须继承自java.sql.Driver。
最后我们看一下第37行最后我们看一下Iterator的next()方法做了什么就完全懂了,通过next()方法调用了:
private S nextService() { if (!hasNextService()) throw new NoSuchElementException(); String cn = nextName; nextName = null; Class<?> c = null; try { c = Class.forName(cn, false, loader); //看这里,Class.forName() } catch (ClassNotFoundException x) { fail(service, "Provider " + cn + " not found"); } if (!service.isAssignableFrom(c)) { fail(service, "Provider " + cn + " not a subtype"); } try { S p = service.cast(c.newInstance()); providers.put(cn, p); return p; } catch (Throwable x) { fail(service, "Provider " + cn + " could not be instantiated", x); } throw new Error(); // This cannot happen }
我们看到了Class.forName,这样是通过SPI的方式把用户手动做的动作变成框架做。
4、SPI是什么?
-
在jar包的META-INF/services目录下创建一个以"接口全限定名"为命名的文件,内容为实现类的全限定名
-
接口实现类所在的jar包在classpath下
-
主程序通过java.util.ServiceLoader动态状态实现模块,它通过扫描META-INF/services目录下的配置文件找到实现类的全限定名,把类加载到JVM
-
SPI的实现类必须带一个无参构造方法
接着我们看一下具体例子,首先定义一个SpiService,它是一个接口:
package org.xrq.test.spi; public interface SpiService { public void hello(); }
两个实现类,分别为SpiServiceA与SpiServiceB
package org.xrq.test.spi; public class SpiServiceA implements SpiService {
@Override public void hello() { System.out.println("SpiServiceA.Hello"); } } package org.xrq.test.spi; public class SpiServiceB implements SpiService { @Override public void hello() { System.out.println("SpiServiceB.hello"); } }
接着我们建一个META-INF/services的文件夹,里面建一个file,file的名字是接口的全限定名org.xrq.test.spi.SpiService:
文件的内容是SpiService实现类SpiServiceA、SpiServiceB的全限定名:
org.xrq.test.spi.SpiServiceA
org.xrq.test.spi.SpiServiceB
这样就大功告成了!然后我们写个测试类自动加载一下这两个类:
public class SpiTest { @Test public void testSpi() { ServiceLoader<SpiService> serviceLoader = ServiceLoader.load(SpiService.class); Iterator<SpiService> iterator = serviceLoader.iterator(); while (iterator.hasNext()) { SpiService spiService = iterator.next(); spiService.hello(); } } }
结果一目了然,调用了hello()方法:
SpiServiceA.Hello
SpiServiceB.hello
这就是SPI的使用示例,接着我们看一下SPI在实际场景中的应用。
5、对SPI的理解
我理解的SPI其实是一种可插拔技术的总称,最简单的例子就是USB,厂商提供了USB的标准,厂家根据USB的标准制造自己的外设,例如鼠标、键盘、游戏手柄等等,但是USB标准具体在电脑中是怎么用的,厂家就不需要管了。
回到我们的代码中也是一样的道理。当我们开发一个框架的时候,除了保证基本的功能外,最重要的一个点是什么?我认为最重要的应该是松耦合,即对扩展开放、对修改关闭,保证框架实现对于使用者来说是黑盒。
框架不可能做好所有的事情,只能把共性的部分抽离出来进行流程化,松耦合实现的核心就是定义好足够松散的接口,或者可以理解是扩展点,具体的扩展点让使用者去实现,这样不同的扩展就不需要修改源代码或者对框架进行定制,这就是面向接口编程的好处。
回到我们框架的部分来说:
-
JDK对于SPI的实现是通过META-INF/services这个目录 + ServiceLoader
-
Spring实现SPI的方式是留了N多的接口,例如BeanPostProcessor、InitializingBean、DisposableBean,我们只需要实现这些接口然后注入即可
对已有框架而言,我们可以通过框架给我们提供的扩展点扩展框架功能。对自己写框架而言,记得SPI这个事情,留好足够的扩展点,这将大大加强你写的框架的扩展性。
转载链接: https://blog.csdn.net/zt928815211/article/details/83420828
https://www.zhihu.com/question/22925738/answer/23088255
https://mp.weixin.qq.com/s/I1Nf8-sQ8wk5_RGnupJoSg