Exception handling in Java isn’t an easy topic. Beginners find it hard to understand and even experienced developers can spend hours discussing how and which Java exceptions should be thrown or handled.
That’s why most development teams have their own set of rules on how to use them.
And if you’re new to a team, you might be surprised how different these rules can be to the ones you’ve used before.
Nevertheless, there are several best practices that are used by most teams. Today’s post is going to show you nine of the most important ones you can use to get started or improve your exception handling.
Before we dive into that, though, we’ll quickly cover the concept of exceptions itself.
After answering these questions we’ll be ready to talk about the best practices. Of course, if you’d like to skip right to the best practices themselves, you can just click here.
Let’s get started!
What is exception handling? What are exceptions?
Before diving into our list of exceptions best practices, we need to understand what those things are and what we use them for.
So, what are exceptions?
In a nutshell, exceptions are anomalous situations during a program’s execution.
An exception happens when something goes wrong. Tried to open a file but it doesn’t exist? You’ve got an exception. Attempted to call a method on an object but the variable was null?
Boom! Exception.
Exceptions are bad things that weren’t supposed to happen. But since they do happen we need to deal with them. And that “dealing” is what we call exception handling.
Exception handling is an error-handling mechanism. When something goes wrong, an exception is thrown. If you do nothing, the exception causes your application to crash.
Or you can choose to handle the exception. That is, you acknowledge that the problem happened, prevent the application from crashing, and take the steps necessaries in order to either recover from the error or fail gracefully.
As you’ve just read, exception handling is an error-handling mechanism, but it’s not the only one.
Nowadays, the vast majority of programming languages offer exception handling as a feature, but things weren’t always this way.
How do languages that don’t have exceptions deal with errors?
One popular approach is to return error codes. For instance, suppose you have a C function that takes an array of integers and a single integer and then searches and return the index of the first item that matches the specified number.
What to do when the item isn’t found in the array? A popular choice would be to return -1.
This approach has the advantage of keeping the code easy to reason about: there’s no interrupting and jumping around the code. Just good old functions returning values.
On the other hand, this error-handling strategy encourages the consumer of a function to always check for error values.
The sheer number of those verifications can make the code polluted. Lots of defensive checks might hide important business logic, making the code less readable and maintainable.
Another serious disavantage of error codes is their lack of context. You might know that code “-5” means that some file wasn’t found…but which file? An integer can’t easily carry that information.
Yet another problem of this approach is the fact that, sometimes, you just can’t return a value (e.g. think of constructors). Something went wrong while initializing an object? With error codes, you’re helpless.
Exceptions came about to solve the problems you’ve just read about.
When an exception is thrown, the control flow of the program is interrupted. If no one handles the exception, it causes the program to crash.
The user will see the exception’s message, which they most likely won’t understand. The message might not even be localized to their language.
In short, that’s not a great user experience.
So, you’ll want to handle exception (even if it’s just for logging the problem, displaying a more user-friendly error message, and then exiting the application anyways).
To handle the exception that was thrown, we have to catch it. We do this by using an exception-handling block.
When we catch the exception, the program’s flow control is handled to the exception-handling block. There, we can make the necessary arrangement in order to deal with the exception.
Still sounds too abstract? Keep reading. You’ll see practical examples when we get to the specifics of Java exceptions.
Consider the following excerpt of code:
package com.company; import java.io.*; public class Main { public static void main(String[] args){ System.out.println("First line"); System.out.println("Second line"); System.out.println("Third line"); int[] myIntArray = new int[]{1, 2, 3}; print4hItemInArray(myIntArray); System.out.println("Fourth line"); System.out.println("Fith line"); } private static void print4thItemInArray(int[] arr) { System.out.println(arr[3]); System.out.println("Fourth element successfully displayed!"); } }
The code above prints some innocuous messages that only say which line they are.
After printing the third line, the code initializes an array with three integers, and pass it as an argument to a private method. The method tries to print the fourth item in the array, which doesn’t exist. This causes the ArrayIndexOutOfBoundsException exception to be thrown.
When that happens, the program’s execution is stopped and the exception’s message is shown. The fourth and fifth messages are never displayed. Neither is executed the second line in of the print4thItemInArray() method.
This is the resulting output:
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 3 at com.company.Main.print4hItemInArray(Main.java:26) at com.company.Main.main(Main.java:13) First line Second line Third line
Now, let’s change the example, adding some exception handling:
package com.company; import java.io.*; public class Main { public static void main(String[] args) { // write your code here System.out.println("First line"); System.out.println("Second line"); System.out.println("Third line"); try { int[] myIntArray = new int[]{1, 2, 3}; print4thItemInArray(myIntArray); } catch (ArrayIndexOutOfBoundsException e){ System.out.println("The array doesn't have four items!"); } System.out.println("Fourth line"); System.out.println("Fith line"); } private static void print4thItemInArray(int[] arr) { System.out.println(arr[3]); } }
Now, after running the code, this is the output we get:
First line Second line Third line The array doesn't have four items! Fourth line Fith line
This time, the exception still happens, same as before. The execution of the private method stops immediately—that’s why the line “Fourth element successfully displayed!” isn’t shown.
The flow control of the program is then handed to the catch block. Inside the catch block, the code just prints a message explaining that the array doesn’t have four items. Then, execution is resumed.
What we’ve just covered is just the tip of the iceberg.
When it comes to handling exceptions in Java, there’s a lot more to it than our brief introduction.
Fortunately, for you, we have another post in which we cover at length the details of how to handle exceptions. Go check that out if you need to learn more!
And without further ado, here are the list of best practices we promised you.
It happens quite often that you use a resource in your try block, like an InputStream, which you need to close afterward. A common mistake in these situations is to close the resource at the end of the try block.
public void doNotCloseResourceInTry() { FileInputStream inputStream = null; try { File file = new File("./tmp.txt"); inputStream = new FileInputStream(file); // use the inputStream to read a file // do NOT do this inputStream.close(); } catch (FileNotFoundException e) { log.error(e); } catch (IOException e) { log.error(e); } }
The problem is that this approach seems to work perfectly fine as long as no exception gets thrown. All statements within the try block will get executed, and the resource gets closed.
But you added the try block for a reason.
You call one or more methods which might throw an exception, or maybe you throw the exception yourself. That means you might not reach the end of the try block. And as a result, you will not close the resources.
You should, therefore, put all your clean up code into the finally block or use a try-with-resource statement.
In contrast to the last few lines of your try block, the finally block gets always executed. That happens either after the successful execution of the try block or after you handled an exception in a catch block. Due to this, you can be sure that you clean up all the opened resources.
public void closeResourceInFinally() { FileInputStream inputStream = null; try { File file = new File("./tmp.txt"); inputStream = new FileInputStream(file); // use the inputStream to read a file } catch (FileNotFoundException e) { log.error(e); } finally { if (inputStream != null) { try { inputStream.close(); } catch (IOException e) { log.error(e); } } } }
Another option is the try-with-resource statement which I explained in more detail in my introduction to Java exception handling.
You can use it if your resource implements the AutoCloseable interface. That’s what most Java standard resources do.
When you open the resource in the try clause, it will get automatically closed after the try block got executed, or an exception handled.
public void automaticallyCloseResource() { File file = new File("./tmp.txt"); try (FileInputStream inputStream = new FileInputStream(file);) { // use the inputStream to read a file } catch (FileNotFoundException e) { log.error(e); } catch (IOException e) { log.error(e); } }
The more specific the exception that you throw is, the better. Always keep in mind that a coworker who doesn’t know your code (or maybe you in a few months) may need to call your method and handle the exception.
Therefore make sure to provide them as much information as possible. That makes your API easier to understand.
And as a result, the caller of your method will be able to handle the exception better or avoid it with an additional check.
So, always try to find the class that fits best to your exceptional event, e.g. throw a NumberFormatException instead of an IllegalArgumentException. And avoid throwing an unspecific Exception.
public void doNotDoThis() throws Exception { ... } public void doThis() throws NumberFormatException { ... }
Whenever you specify an exception in your method signature, you should also document it in your Javadoc.
That has the same goal as the previous best practice: Provide the caller as many information as possible so that he can avoid or handle the exception.
So, make sure to add a @throws declaration to your Javadoc and to describe the situations that can cause the exception.
/** * This method does something extremely useful ... * * @param input * @throws MyBusinessException if ... happens */ public void doSomething(String input) throws MyBusinessException { ... }
The idea behind this best practice is similar to the two previous ones. But this time, you don’t provide the information to the caller of your method.
The exception’s message gets read by everyone who has to understand what had happened when the exception was reported in the log file or your monitoring tool.
It should, therefore, describe the problem as precisely as possible and provide the most relevant information to understand the exceptional event.
Don’t get me wrong; you shouldn’t write a paragraph of text. But you should explain the reason for the exception in 1-2 short sentences.
That helps your operations team to understand the severity of the problem, and it also makes it easier for you to analyze any service incidents.
If you throw a specific exception, its class name will most likely already describe the kind of error. So, you don’t need to provide a lot of additional information.
A good example for that is the NumberFormatException. It gets thrown by the constructor of the class java.lang.Long when you provide a String in a wrong format.
try { new Long("xyz"); } catch (NumberFormatException e) { log.error(e); }
The name of the NumberFormatException class already tells you the kind of problem. Its message only needs to provide the input string that caused the problem.
If the name of the exception class isn’t that expressive, you need to provide the required information in the message.
17:17:26,386 ERROR TestExceptionHandling:52 - java.lang.NumberFormatException: For input string: "xyz"
Most IDEs help you with this best practice. They report an unreachable code block when you try to catch the less specific exception first.
The problem is that only the first catch block that matches the exception gets executed.
So, if you catch an IllegalArgumentException first, you will never reach the catch block that should handle the more specific NumberFormatException because it’s a subclass of the IllegalArgumentException.
Always catch the most specific exception class first and add the less specific catch blocks to the end of your list.
You can see an example of such a try-catch statement in the following code snippet. The first catch block handles all NumberFormatExceptions and the second one all IllegalArgumentExceptions which are not a NumberFormatException.
public void catchMostSpecificExceptionFirst() { try { doSomething("A message"); } catch (NumberFormatException e) { log.error(e); } catch (IllegalArgumentException e) { log.error(e) } }
Throwable is the superclass of all exceptions and errors. You can use it in a catch clause, but you should never do it!
If you use Throwable in a catch clause, it will not only catch all exceptions; it will also catch all errors.
Errors are thrown by the JVM to indicate serious problems that are not intended to be handled by an application.
Typical examples for that are the OutOfMemoryError or the StackOverflowError. Both are caused by situations that are outside of the control of the application and can’t be handled.
So, better don’t catch a Throwable unless you’re absolutely sure that you’re in an exceptional situation in which you’re able or required to handle an error.
public void doNotCatchThrowable() { try { // do something } catch (Throwable t) { // don't do this! } }
Have you ever analyzed a bug report where only the first part of your use case got executed?
That’s often caused by an ignored exception. The developer was probably pretty sure that it would never be thrown and added a catch block that doesn’t handle or logs it.
And when you find this block, you most likely even find one of the famous “This will never happen” comments:
public void doNotIgnoreExceptions() { try { // do something } catch (NumberFormatException e) { // this will never happen } }
Well, you might be analyzing a problem in which the impossible happened.
So, please, never ignore an exception.
You don’t know how the code will change in the future. Someone might remove the validation that prevented the exceptional event without recognizing that this creates a problem. Or the code that throws the exception gets changed and now throws multiple exceptions of the same class, and the calling code doesn’t prevent all of them.
You should at least write a log message telling everyone that the unthinkable just had happened and that someone needs to check it.
public void logAnException() { try { // do something } catch (NumberFormatException e) { log.error("This should never happen: " + e); } }
Don’t log and throw is probably the most often ignored best practice in this list. You can find lots of code snippets and even libraries in which an exception gets caught, logged, and rethrown.
try { new Long("xyz"); } catch (NumberFormatException e) { log.error(e); throw e; }
It might feel intuitive to log an exception when it occurred and then rethrow it so that the caller can handle it appropriately. But it will write multiple error messages for the same exception.
17:44:28,945 ERROR TestExceptionHandling:65 - java.lang.NumberFormatException: For input string: "xyz" Exception in thread "main" java.lang.NumberFormatException: For input string: "xyz" at java.lang.NumberFormatException.forInputString(NumberFormatException.java:65) at java.lang.Long.parseLong(Long.java:589) at java.lang.Long.(Long.java:965) at com.stackify.example.TestExceptionHandling.logAndThrowException(TestExceptionHandling.java:63) at com.stackify.example.TestExceptionHandling.main(TestExceptionHandling.java:58)
The additional messages also don’t add any information.
As explained in best practice #4, the exception message should describe the exceptional event. And the stack trace tells you in which class, method, and line the exception was thrown.
If you need to add additional information, you should catch the exception and wrap it in a custom one. But make sure to follow best practice number 9.
public void wrapException(String input) throws MyBusinessException { try { // do something } catch (NumberFormatException e) { throw new MyBusinessException("A message that describes the error.", e); } }
So, only catch an exception if you want to handle it. Otherwise, specify it in the method signature and let the caller take care of it.
Try Stackify’s free code profiler, Prefix, to write better code on your workstation. Prefix works with .NET, Java, PHP, Node.js, Ruby, and Python.
It’s sometimes better to catch a standard exception and to wrap it into a custom one.
A typical example for such an exception is an application or framework specific business exception. That allows you to add additional information and you can also implement a special handling for your exception class.
When you do that, make sure to set the original exception as the cause. The Exception class provides specific constructor methods that accept a Throwable as a parameter.
Otherwise, you lose the stack trace and message of the original exception which will make it difficult to analyze the exceptional event that caused your exception.
public void wrapException(String input) throws MyBusinessException { try { // do something } catch (NumberFormatException e) { throw new MyBusinessException("A message that describes the error.", e); } }
As you’ve seen, there are lots of different things you should consider when you throw or catch an exception. Most of them have the goal to improve the readability of your code or the usability of your API.
Exceptions are most often an error handling mechanism and a communication medium at the same time.
You should, therefore, make sure to discuss the Java exception handling best practices and rules you want to apply with your coworkers so that everyone understands the general concepts and uses them in the same way.
When using Retrace APM with code profiling, you can collect exceptions directly from Java, without any code changes!
If you would like to be a guest contributor to the Stackify blog please reach out to [email protected]