1、java标准异常概述
Throwable表示任何可以作为异常被抛出的类,有两个子类Error和Exception。从这两个类的源代码中可以看出,这两个类并没有添加新的方法,Throwable提供了所以方法的实现。Error表示编译时和系统错误。Exception是可以被抛出的异常类。RuntimeException继承自Exception(如NullPointerException),表示运行时异常,JVM会自动抛出.
2、自定义异常类
自定义异常类方法: 通过继承Throwable或Exception。异常类的所有实现都是基类Throwable实现的,所以构造自定义异常类完全可以参考Exception和Error类。我们只要添加上自定义异常类的构造方法就可以了
-
package demo.others; /** * 自定义异常类方法 * 1、通过继承Throwable * 2、通过继承Exception * * @author Touch */ public class MyExceptionDemo extends Exception { private static final long serialVersionUID = 1L; public MyExceptionDemo() { super(); } public MyExceptionDemo(String message) { super(message); } public MyExceptionDemo(String message, Throwable cause) { super(message, cause); } public MyExceptionDemo(Throwable cause) { super(cause); } }
3、异常栈及异常处理方式
可以通过try、catch来捕获异常。捕获到的异常。下面的示例演示了几种常用异常处理方式
-
package demo.others; import mine.util.exception.MyException; public class ExceptionDemo1 { public void f() throws MyException { throw new MyException("自定义异常"); } public void g() throws MyException { f(); } public void h() throws MyException { try { g(); } catch (MyException e) { //1、通过获取栈轨迹中的元素数组来显示异常抛出的轨迹 for (StackTraceElement ste : e.getStackTrace()) System.out.println(ste.getMethodName()); //2、直接将异常栈信息输出至标准错误流或标准输出流 e.printStackTrace();//输出到标准错误流 e.printStackTrace(System.err); e.printStackTrace(System.out); //3、将异常信息输出到文件中 //e.printStackTrace(new PrintStream("file/exception.txt")); //4、重新抛出异常,如果直接抛出那么栈路径是完整的,如果用fillInStackTrace() //那么将会从这个方法(当前是h()方法)作为异常发生的原点。 //throw e; throw (MyException)e.fillInStackTrace(); } } public static void main(String[] args) { try { new ExceptionDemo1().h(); } catch (MyException e) { // TODO Auto-generated catch block e.printStackTrace(); } } }
运行结果:
f
g
h
main
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo1.f(ExceptionDemo1.java:7)
at demo.others.ExceptionDemo1.g(ExceptionDemo1.java:11)
at demo.others.ExceptionDemo1.h(ExceptionDemo1.java:16)
at demo.others.ExceptionDemo1.main(ExceptionDemo1.java:35)
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo1.f(ExceptionDemo1.java:7)
at demo.others.ExceptionDemo1.g(ExceptionDemo1.java:11)
at demo.others.ExceptionDemo1.h(ExceptionDemo1.java:16)
at demo.others.ExceptionDemo1.main(ExceptionDemo1.java:35)
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo1.f(ExceptionDemo1.java:7)
at demo.others.ExceptionDemo1.g(ExceptionDemo1.java:11)
at demo.others.ExceptionDemo1.h(ExceptionDemo1.java:16)
at demo.others.ExceptionDemo1.main(ExceptionDemo1.java:35)
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo1.h(ExceptionDemo1.java:30)
at demo.others.ExceptionDemo1.main(ExceptionDemo1.java:35)
分析上面的程序,首先main函数被调用,然后是调用h函数,再g函数、f函数,f函数抛出异常,并在h函数捕获,这时将依次从栈顶到栈底输出异常栈路径。
4、异常链
有时候我们会捕获一个异常后在抛出另一个异常,如下代码所示:
-
package demo.others; import java.io.IOException; import mine.util.exception.MyException; public class ExceptionDemo2 { public void f() throws MyException { throw new MyException("自定义异常"); } public void g() throws Exception { try { f(); } catch (MyException e) { e.printStackTrace(); throw new Exception("重新抛出的异常1"); } } public void h() throws IOException { try { g(); } catch (Exception e) { // TODO Auto-generated catch block e.printStackTrace(); throw new IOException("重新抛出异常2"); } } public static void main(String[] args) { try { new ExceptionDemo2().h(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } } }
运行结果:
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo2.f(ExceptionDemo2.java:9)
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:14)
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:23)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:32)
java.lang.Exception: 重新抛出的异常1
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:17)
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:23)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:32)
java.io.IOException: 重新抛出异常2
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:27)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:32)
从结果中我们可以看出,异常栈变小了。也就是说丢失了最原始的异常信息。怎样保存最原始的异常信息呢?Throwable类中有个Throwable cause属性,表示原始异常。通过接收cause参数的构造器可以把原始异常传递给新异常,或者通过initCause()方法。如下示例:
-
package demo.others; import java.io.IOException; import mine.util.exception.MyException; public class ExceptionDemo2 { public void f() throws MyException { throw new MyException("自定义异常"); } public void g() throws Exception { try { f(); } catch (MyException e) { e.printStackTrace(); throw new Exception("重新抛出的异常1",e); } } public void h() throws IOException { try { g(); } catch (Exception e) { // TODO Auto-generated catch block e.printStackTrace(); IOException io=new IOException("重新抛出异常2"); io.initCause(e); throw io; } } public static void main(String[] args) { try { new ExceptionDemo2().h(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } } }
结果:
mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo2.f(ExceptionDemo2.java:9)
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:14)
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:23)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:34)
java.lang.Exception: 重新抛出的异常1
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:17)
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:23)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:34)
Caused by: mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo2.f(ExceptionDemo2.java:9)
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:14)
... 2 more
java.io.IOException: 重新抛出异常2
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:27)
at demo.others.ExceptionDemo2.main(ExceptionDemo2.java:34)
Caused by: java.lang.Exception: 重新抛出的异常1
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:17)
at demo.others.ExceptionDemo2.h(ExceptionDemo2.java:23)
... 1 more
Caused by: mine.util.exception.MyException: 自定义异常
at demo.others.ExceptionDemo2.f(ExceptionDemo2.java:9)
at demo.others.ExceptionDemo2.g(ExceptionDemo2.java:14)
... 2 more
从结果中看出当获取到“重新抛出异常2的时候,同时可以输出原始异常“重新抛出的异常1“和原始异常”自定义异常,这就是异常链。
5、finally的使用
finally子句总是执行的,通常用来做一些清理工作,如关闭文件,关闭连接等
下面举几个finally的例子:
-
// 读取指定路径文本文件 public static String read(String filePath) { StringBuilder str = new StringBuilder(); BufferedReader in = null; try { in = new BufferedReader(new FileReader(filePath)); String s; try { while ((s = in.readLine()) != null) str.append(s + '\n'); } finally { in.close(); } } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return str.toString(); }
分析:如果调用in = new BufferedReader(new FileReader(filePath));时发生异常,这时是一个文件路径不存在的异常,也就是说并没有打开文件,这时将会直接跳到catch块,而不会执行try...finally块(并不是finally子句)里面的语句in.close();此时不需要关闭文件。
再看一个例子,会导致异常的丢失
-
package demo.others; import mine.util.exception.MyException; public class ExceptionDemo3 { public void f() throws MyException { throw new MyException("异常1"); } public void g() throws MyException { throw new MyException("异常2"); } public static void main(String[] args) { try { ExceptionDemo3 ex = new ExceptionDemo3(); try { ex.f(); } finally { ex.g();//此时捕获g方法抛出的异常,f方法抛出的异常丢失了 } } catch (MyException e) { System.out.println(e); } } }
结果:mine.util.exception.MyException: 异常2
此时异常1就丢失了
或者这样写:
-
import mine.util.exception.MyException; public class ExceptionDemo3 { public void g() throws MyException { throw new MyException("异常2"); } public static void main(String[] args) { ExceptionDemo3 ex = new ExceptionDemo3(); try { ex.g(); } finally { //直接return会丢失所以抛出的异常 return; } } }
6、异常的限制
(1)当覆盖方法时,只能抛出在基类方法的异常说明里列出的那些异常,有些基类的方法声明抛出异常其实并没有抛出异常,这是因为可能在其子类的覆盖方法中会抛出异常
(2)构造器可以抛出任何异常而不必理会基类构造器所抛出的异常,派生类构造器异常说明必须包含基类构造器异常说明,因为构造派生类对象时会调用基类构造器。此外,派生类构造器不能捕获基类构造器抛出的异常。
转载至 http://blog.csdn.net/touch_2011/article/details/6860043