Java如何自定义异常打印非堆栈信息详解

2023-09-14 00:33:05 326

前言

在学习Java的过程中，想必大家都一定学习过异常这个篇章，异常的基本特性和使用这里就不再多讲了。什么是异常？我不知道大家都是怎么去理解的，我的理解很简单，那就是不正常的情况，比如我现在是个男的，但是我却有着女人所独有的东西，在我看来这尼玛肯定是种异常，简直不能忍。想必大家都能够理解看懂，并正确使用。

但是，光学会基本异常处理和使用不够的，在工作中出现异常并不可怕，有时候是需要使用异常来驱动业务的处理，例如:在使用唯一约束的数据库的时候，如果插入一条重复的数据，那么可以通过捕获唯一约束异常DuplicateKeyException来进行处理，这个时候，在server层中就可以向调用层抛出对应的状态，上层根据对应的状态再进行处理，所以有时候异常对业务来说，是一个驱动方式。

有的捕获异常之后会将异常进行输出，不知道细心的同学有没有注意到一点，输出的异常是什么东西呢？

下面来看一个常见的异常：

java.lang.ArithmeticException:/byzero
atgreenhouse.ExceptionTest.testException(ExceptionTest.java:16)
atsun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethod)
atsun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
atsun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
atjava.lang.reflect.Method.invoke(Method.java:597)
atorg.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:44)
atorg.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:15)
atorg.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:41)
atorg.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:20)
atorg.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:76)
atorg.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:50)
atorg.junit.runners.ParentRunner$3.run(ParentRunner.java:193)
atorg.junit.runners.ParentRunner$1.schedule(ParentRunner.java:52)
atorg.junit.runners.ParentRunner.runChildren(ParentRunner.java:191)
atorg.junit.runners.ParentRunner.access$000(ParentRunner.java:42)
atorg.junit.runners.ParentRunner$2.evaluate(ParentRunner.java:184)
atorg.junit.runners.ParentRunner.run(ParentRunner.java:236)
atorg.junit.runner.JUnitCore.run(JUnitCore.java:157)
atcom.intellij.junit4.JUnit4IdeaTestRunner.startRunnerWithArgs(JUnit4IdeaTestRunner.java:68)
atcom.intellij.rt.execution.junit.IdeaTestRunner$Repeater.startRunnerWithArgs(IdeaTestRunner.java:47)
atcom.intellij.rt.execution.junit.JUnitStarter.prepareStreamsAndStart(JUnitStarter.java:242)
atcom.intellij.rt.execution.junit.JUnitStarter.main(JUnitStarter.java:70)

一个空指针异常:

java.lang.NullPointerException
atgreenhouse.ExceptionTest.testException(ExceptionTest.java:16)
atsun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethod)
atsun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
atsun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
atjava.lang.reflect.Method.invoke(Method.java:597)
atorg.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:44)
atorg.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:15)
atorg.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:41)
atorg.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:20)
atorg.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:76)
atorg.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:50)
atorg.junit.runners.ParentRunner$3.run(ParentRunner.java:193)
atorg.junit.runners.ParentRunner$1.schedule(ParentRunner.java:52)
atorg.junit.runners.ParentRunner.runChildren(ParentRunner.java:191)
atorg.junit.runners.ParentRunner.access$000(ParentRunner.java:42)
atorg.junit.runners.ParentRunner$2.evaluate(ParentRunner.java:184)
atorg.junit.runners.ParentRunner.run(ParentRunner.java:236)
atorg.junit.runner.JUnitCore.run(JUnitCore.java:157)
atcom.intellij.junit4.JUnit4IdeaTestRunner.startRunnerWithArgs(JUnit4IdeaTestRunner.java:68)
atcom.intellij.rt.execution.junit.IdeaTestRunner$Repeater.startRunnerWithArgs(IdeaTestRunner.java:47)
atcom.intellij.rt.execution.junit.JUnitStarter.prepareStreamsAndStart(JUnitStarter.java:242)
atcom.intellij.rt.execution.junit.JUnitStarter.main(JUnitStarter.java:70)

大家有没有发现一个特点，就是异常的输出是中能够精确的输出异常出现的地点，还有后面一大堆的执行过程类调用，也都打印出来了，这些信息从哪儿来呢？这些信息是从栈中获取的，在打印异常日志的时候，会从栈中去获取这些调用信息。能够精确的定位异常出现的异常当然是好，但是我们有时候考虑到程序的性能，以及一些需求时，我们有时候并不需要完全的打印这些信息，并且去方法调用栈中获取相应的信息，是有性能消耗的，对于一些性能要求高的程序，我们完全可以在这一个方面为程序性能做一个提升。

所以如何避免输出这些堆栈信息呢？那么自定义异常就可以解决这个问题：

首先，自动异常需要继承RuntimeException,然后，再通过是重写fillInStackTrace,toString方法，例如，下面我定义一个AppException异常:

packagecom.green.monitor.common.exception;
importjava.text.MessageFormat;
/**
*自定义异常类
*/
publicclassAppExceptionextendsRuntimeException{
privatebooleanisSuccess=false;
privateStringkey;
privateStringinfo;
publicAppException(Stringkey){
super(key);
this.key=key;
this.info=key;
}
publicAppException(Stringkey,Stringmessage){
super(MessageFormat.format("{0}[{1}]",key,message));
this.key=key;
this.info=message;
}
publicAppException(Stringmessage,Stringkey,Stringinfo){
super(message);
this.key=key;
this.info=info;
}
publicbooleanisSuccess(){
returnisSuccess;
}
publicStringgetKey(){
returnkey;
}
publicvoidsetKey(Stringkey){
this.key=key;
}
publicStringgetInfo(){
returninfo;
}
publicvoidsetInfo(Stringinfo){
this.info=info;
}
@Override
publicThrowablefillInStackTrace(){
returnthis;
}
@Override
publicStringtoString(){
returnMessageFormat.format("{0}[{1}]",this.key,this.info);
}
}

那么为什么要重写fillInStackTrace,和toString方法呢？我们首先来看源码是怎么一回事.

publicclassRuntimeExceptionextendsException{
staticfinallongserialVersionUID=-7034897190745766939L;
/**Constructsanewruntimeexceptionwithnullasits
*detailmessage.Thecauseisnotinitialized,andmaysubsequentlybe
*initializedbyacallto{@link#initCause}.
*/
publicRuntimeException(){
super();
}
/**Constructsanewruntimeexceptionwiththespecifieddetailmessage.
*Thecauseisnotinitialized,andmaysubsequentlybeinitializedbya
*callto{@link#initCause}.
*
*@parammessagethedetailmessage.Thedetailmessageissavedfor
*laterretrievalbythe{@link#getMessage()}method.
*/
publicRuntimeException(Stringmessage){
super(message);
}
/**
*Constructsanewruntimeexceptionwiththespecifieddetailmessageand
*cause.Notethatthedetailmessageassociatedwith
*causeisnotautomaticallyincorporatedin
*thisruntimeexception'sdetailmessage.
*
*@parammessagethedetailmessage(whichissavedforlaterretrieval
*bythe{@link#getMessage()}method).
*@paramcausethecause(whichissavedforlaterretrievalbythe
*{@link#getCause()}method).(Anullvalueis
*permitted,andindicatesthatthecauseisnonexistentor
*unknown.)
*@since1.4
*/
publicRuntimeException(Stringmessage,Throwablecause){
super(message,cause);
}
/**Constructsanewruntimeexceptionwiththespecifiedcauseanda
*detailmessageof(cause==null?null:cause.toString())
*(whichtypicallycontainstheclassanddetailmessageof
*cause).Thisconstructorisusefulforruntimeexceptions
*thatarelittlemorethanwrappersforotherthrowables.
*
*@paramcausethecause(whichissavedforlaterretrievalbythe
*{@link#getCause()}method).(Anullvalueis
*permitted,andindicatesthatthecauseisnonexistentor
*unknown.)
*@since1.4
*/
publicRuntimeException(Throwablecause){
super(cause);
}
}

RuntimeException是继承Exception,但是它里面去只是调用了父类的方法，本身是没有做什么其余的操作。那么继续看Exception里面是怎么回事呢？

publicclassExceptionextendsThrowable{
staticfinallongserialVersionUID=-3387516993124229948L;
/**
*Constructsanewexceptionwithnullasitsdetailmessage.
*Thecauseisnotinitialized,andmaysubsequentlybeinitializedbya
*callto{@link#initCause}.
*/
publicException(){
super();
}
/**
*Constructsanewexceptionwiththespecifieddetailmessage.The
*causeisnotinitialized,andmaysubsequentlybeinitializedby
*acallto{@link#initCause}.
*
*@parammessagethedetailmessage.Thedetailmessageissavedfor
*laterretrievalbythe{@link#getMessage()}method.
*/
publicException(Stringmessage){
super(message);
}
/**
*Constructsanewexceptionwiththespecifieddetailmessageand
*cause.Notethatthedetailmessageassociatedwith
*causeisnotautomaticallyincorporatedin
*thisexception'sdetailmessage.
*
*@parammessagethedetailmessage(whichissavedforlaterretrieval
*bythe{@link#getMessage()}method).
*@paramcausethecause(whichissavedforlaterretrievalbythe
*{@link#getCause()}method).(Anullvalueis
*permitted,andindicatesthatthecauseisnonexistentor
*unknown.)
*@since1.4
*/
publicException(Stringmessage,Throwablecause){
super(message,cause);
}
/**
*Constructsanewexceptionwiththespecifiedcauseandadetail
*messageof(cause==null?null:cause.toString())(which
*typicallycontainstheclassanddetailmessageofcause).
*Thisconstructorisusefulforexceptionsthatarelittlemorethan
*wrappersforotherthrowables(forexample,{@link
*java.security.PrivilegedActionException}).
*
*@paramcausethecause(whichissavedforlaterretrievalbythe
*{@link#getCause()}method).(Anullvalueis
*permitted,andindicatesthatthecauseisnonexistentor
*unknown.)
*@since1.4
*/
publicException(Throwablecause){
super(cause);
}
}

从源码中可以看到,Exception里面也是直接调用了父类的方法，和RuntimeException一样，自己其实并没有做什么。那么直接来看Throwable里面是怎么一回事:

publicclassThrowableimplementsSerializable{
publicThrowable(Stringmessage){
fillInStackTrace();
detailMessage=message;
}

/**
*Fillsintheexecutionstacktrace.Thismethodrecordswithinthis
*Throwableobjectinformationaboutthecurrentstateof
*thestackframesforthecurrentthread.
*
*@returnareferencetothisThrowableinstance.
*@seejava.lang.Throwable#printStackTrace()
*/
publicsynchronizednativeThrowablefillInStackTrace();

/**
*Providesprogrammaticaccesstothestacktraceinformationprintedby
*{@link#printStackTrace()}.Returnsanarrayofstacktraceelements,
*eachrepresentingonestackframe.Thezerothelementofthearray
*(assumingthearray'slengthisnon-zero)representsthetopofthe
*stack,whichisthelastmethodinvocationinthesequence.Typically,
*thisisthepointatwhichthisthrowablewascreatedandthrown.
*Thelastelementofthearray(assumingthearray'slengthisnon-zero)
*representsthebottomofthestack,whichisthefirstmethodinvocation
*inthesequence.
*
*Somevirtualmachinesmay,undersomecircumstances,omitone
*ormorestackframesfromthestacktrace.Intheextremecase,
*avirtualmachinethathasnostacktraceinformationconcerning
*thisthrowableispermittedtoreturnazero-lengtharrayfromthis
*method.Generallyspeaking,thearrayreturnedbythismethodwill
*containoneelementforeveryframethatwouldbeprintedby
*printStackTrace.
*
*@returnanarrayofstacktraceelementsrepresentingthestacktrace
*pertainingtothisthrowable.
*@since1.4
*/
publicStackTraceElement[]getStackTrace(){
return(StackTraceElement[])getOurStackTrace().clone();
}
privatesynchronizedStackTraceElement[]getOurStackTrace(){
//Initializestacktraceifthisisthefirstcalltothismethod
if(stackTrace==null){
intdepth=getStackTraceDepth();
stackTrace=newStackTraceElement[depth];
for(inti=0;iindex<0||
*index>=getStackTraceDepth()
*/
nativeStackTraceElementgetStackTraceElement(intindex);

/**
*Returnsashortdescriptionofthisthrowable.
*Theresultistheconcatenationof:
*

*the{@linkplainClass#getName()name}oftheclassofthisobject
*
":"(acolonandaspace)
*
theresultofinvokingthisobject's{@link#getLocalizedMessage}
*method
*
*IfgetLocalizedMessagereturnsnull,thenjust
*theclassnameisreturned.
*
*@returnastringrepresentationofthisthrowable.
*/
publicStringtoString(){
Strings=getClass().getName();
Stringmessage=getLocalizedMessage();
return(message!=null)?(s+":"+message):s;
}

从源码中可以看到，到Throwable就几乎到头了，在fillInStackTrace()方法是一个native方法，这方法也就是会调用底层的C语言，返回一个Throwable对象,toString方法，返回的是throwable的简短描述信息，并且在getStackTrace方法和getOurStackTrace中调用的都是native方法getStackTraceElement,而这个方法是返回指定的栈元素信息，所以这个过程肯定是消耗性能的，那么我们自定义异常中的重写toString方法和fillInStackTrace方法就可以不从栈中去获取异常信息，直接输出，这样对系统和程序来说，相对就没有那么”重”,是一个优化性能的非常好的办法。那么如果出现自定义异常那么是什么样的呢？请看下面吧:

@Test
publicvoidtestException(){
try{
Stringstr=null;
System.out.println(str.charAt(0));
}catch(Exceptione){
thrownewAppException("000001","空指针异常");
}
}

那么在异常异常的时候，系统将会打印我们自定义的异常信息:

000001[空指针异常]
Processfinishedwithexitcode-1

所以特别简洁，优化了系统程序性能，让程序不这么“重”,所以对于性能要求特别要求的系统。赶紧自己的自定义异常吧！

总结

以上就是这篇文章的全部内容了，希望本文的内容对大家的学习或者工作具有一定的参考学习价值，如果有疑问大家可以留言交流，谢谢大家对毛票票的支持。

Java如何自定义异常打印非堆栈信息详解

热门推荐

随机推荐