Exception Handling

When errors can occur, it is good practice to handle them to either change the program or debug.

However, one may also wish to throw user-defined errors for specific cases and handle them in a try...catch.

• An example of this is when checking for, say, the boundaries of an object.

Error handling is all about this. Predicting and managing errors, or throwing your own errors for possibly bad data.

---

try...catch

The try...catch statement is powerful for handling errors; it makes code robust, safe and error-proof.

info

Throwing exceptions isn't always the best error-handling method.

A try...catch lets you execute things with access to the error, thereby resolving it.

• A try...catch statement has a try block, then a catch block, and an optional finally block.

• The code in the try block is executed first. If the try throws an exception or error:

  • Then the code in the catch/finally block is executed.

• Note that when there is a finally block, it will always be executed.

  • It will execute before control flow exits the try...catch construct.

This is the syntax for try...catch blocks:

try {

    tryStatements
} catch (exceptionVar) {

    catchStatements
} [finally {

    finallyStatements
}]

• The exceptionVar is an optional identifier to hold the caught exception.

  • If you don't need this, ignore it and write catch { catchStatements }.

• An exception being thrown in the try block immediately moves control flow to catch.

• If finally is included, it will be executed no matter what.

  • A great example of finally is when trying to open a file, at the end no matter what close it

Let's see an example utilizing a try...catch with a finally too:

tip

A great use case of try...catch is to only catch (and silence) a small subset of expected errors, say RangeErrors.

As for other errors, just re-throw them if they're not relevant in your scenario.

For example: if (e instanceof RangeError) { //handle range error} else { throw e; }

The optional exception identifier can provide detailed information regarding the thrown Error/Exception.

Here's an example where the exception identifier isn't needed at all:

---

throw

The keyword throw is also used for exception handling, and using it throws a user-defined exception.

Similarly, control flow will be passed to the first catch block in the "stack".

• If there is no catch blocks, the program simply terminates after throwing the error.

The different between throw and try...catch is:

• throw is user-defined, so one can throw exceptions when a try block wouldn't

• Also, throw lets you throw objects and expressions

Here's an example of using throw:

 

tip

throw and try...catch can be combined. That is when exception handling is powerful.

---