Do not assign to the exception parameter

When do I get this error?

The "Do not assign to the exception parameter" error is thrown when JSLint, JSHint or ESLint encounters an assignment inside a catch block to the identifer associated with that block. In the following example we attempt to assign a new value to the exception parameter e:

try {
    // Some code that might throw an exception
    throw new Error();
} catch (e) {
    e = 10;
}

In April 2013 the message given by JSLint in this situation changed from "Do not assign to the exception parameter" to the more generic "Read only", which is given in numerous other scenarios. As of August 2013 this functionality is no longer present in JSLint. It remains in JSHint and ESLint as explained above.

Why do I get this error?

This error is raised to highlight a lack of convention. The obvious issue with assigning something to an exception parameter is that you lose access to the value passed to the catch block completely. There is nothing like the arguments object that will allow you to get back a reference to it. This could be confusing to other developers maintaining your code in the future.

In JSHint 1.0.0 and above you have the ability to ignore any warning with a special option syntax. The identifier of this warning is W022. This means you can tell JSHint to not issue this warning with the /*jshint -W022 */ directive.

A note regarding the exception parameter

According to the ECMAScript specification, a catch block has its own lexical environment, which is created upon entering the block. The lexical environment maintains a reference to the lexical environment from which it was entered, which means you have access to variables defined in outer scopes. The lexical environment created when entering the catch block gets a binding between the identifier and the exception value passed to the block (ES5 §12.14):

The production Catch : catch ( Identifier ) Block is evaluated as follows:

  1. Let C be the parameter that has been passed to this production.
  2. Let oldEnv be the running execution context’s LexicalEnvironment.
  3. Let catchEnv be the result of calling NewDeclarativeEnvironment passing oldEnv as the argument.
  4. Call the CreateMutableBinding concrete method of catchEnv passing the Identifier String value as the argument.
  5. Call the SetMutableBinding concrete method of catchEnv passing the Identifier, C, and false as arguments...
  6. Set the running execution context’s LexicalEnvironment to catchEnv.
  7. Let B be the result of evaluating Block.
  8. Set the running execution context’s LexicalEnvironment to oldEnv.
  9. Return B.

NOTE No matter how control leaves the Block the LexicalEnvironment is always restored to its former state.

That note at the end is important. When the catch block exits, we return to the original lexical environment. This means that the binding for the identifier in the catch block will cease to exist (since the lexical environment that contained it no longer exists).

However, in Internet Explorer 8 and below, the catch block identifier leaks into the outer scope. In those browsers, you could accidently be assigning a value to the exception parameter, rather than a variable with the same identifier in an outer scope. For more information, have a read of this excellent article by Ben Alman.


About the author

James Allardice

This article was written by James Allardice, Software engineer at Tesco and orangejellyfish in London. Passionate about React, Node and writing clean and maintainable JavaScript. Uses linters (currently ESLint) every day to help achieve this.