Rise In Logo

Rust Programming for Smart Contract Development

Creating and using custom error types

Alrighty then, let's dive into the world of custom error types! Buckle up because this is going to be an enlightening ride!

First things first, why would anyone want to create custom error types? After all, isn't it easier to just use the built-in error types? Well, sure, but remember that with great power comes great... customizability! (And a tad bit more complexity, but who's counting?)

Improving Error Clarity

Imagine you're a detective trying to solve a mystery (just go with me on this one). You've got a bunch of clues, but they're all vague and generic. How much harder would it be to solve the mystery? Now, imagine if you had specific, clear, and detailed clues. Much easier, right?

The same principle applies to error handling. Using generic error types is like saying, "Something went wrong. I dunno, you figure it out!" But with custom error types, you can pinpoint exactly what went wrong, which makes debugging your code much less of a Sherlock-Holmes-style mystery.

Consider this piece of code:

fn divide(numerator: f64, denominator: f64) -> Result<f64, &'static str> {
    if denominator == 0.0 {
        Err("You can't divide by zero, silly!")
    } else {
        Ok(numerator / denominator)

If a division by zero happens, we'll get a custom error message that's a lot more specific and helpful than a generic panic!

Creating Specific Responses

Sometimes, we want to do more than just spit out an error message. We might want to trigger a specific response when certain errors occur. With custom error types, we can categorize our errors in a way that allows us to write code to handle specific kinds of errors differently.

Let's say we're building a rocket ship in our backyard (as one does). There are a lot of things that can go wrong, and we want to handle each of them differently.

enum RocketError {

// Now we can write code to handle each of these errors specifically!

fn handle_error(error: RocketError) {
    match error {
        RocketError::OutOfFuel => {
            println!("Houston, we have a problem. We're out of fuel!");
            // Trigger fuel replenishment procedures
        RocketError::NavigationSystemFailure => {
            println!("Houston, we're lost. I knew we should've taken that left turn at Albuquerque!");
            // Trigger navigation system diagnostics
        RocketError::AlienInvasion => {
            println!("Houston, we're not alone. There are aliens. This is not a drill!");
            // Trigger intergalactic diplomacy protocols

In this example, we're not only providing detailed error messages, but also triggering specific responses for each error. How cool is that?

In conclusion, custom error types are like your favorite superhero costume. Sure, you could go to the party in your regular clothes, but wouldn't it be so much more fun to show up as Captain Error-Handler? (Okay, maybe we need to work on that name...)

Defining a Custom Error Type

So, you're probably wondering, "Why can't I just use the pre-existing error types in Rust?" Well, you can, but sometimes you want to add a bit of your own flavor to your error messages, right? It's like cooking; sure, you can order takeout, but sometimes you want to add your own secret sauce to the mix. That's where defining a custom error type comes in handy.

Creating a Custom Error Type

Let's start with creating our own custom error type. In Rust, we can use enums and structs to create our own types. It's like building your own custom action figure. You can give it whatever attributes and behaviors you want. Here's an example:

enum MyCustomError {

In this example, we have a custom error type named MyCustomError, which is an enum that can be one of three variants: Io, Parse, or Other. Io and Parse are wrapping standard error types, while Other is a general-purpose variant that contains a String.

Implementing the Error Trait for a Custom Error Type

Now that we have our custom error type, we need to make it behave like an error. It's like when Clark Kent puts on his glasses and everyone suddenly believes he's not Superman (go figure). In Rust, we do this by implementing the Error trait for our custom type.

use std::fmt;

impl fmt::Display for MyCustomError {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        match self {
            MyCustomError::Io(err) => write!(f, "I/O error: {}", err),
            MyCustomError::Parse(err) => write!(f, "Parse error: {}", err),
            MyCustomError::Other(message) => write!(f, "Other error: {}", message),

impl std::error::Error for MyCustomError {}

In the above code, we first implement the Display trait for MyCustomError. This allows us to print out our error messages in a human-readable format. Each variant of MyCustomError has its own custom message.

After that, we implement the Error trait for MyCustomError. This is what really makes our custom type an "error" in the eyes of Rust. Notice that we don't have to provide any method implementations inside the impl block because the Error trait doesn't require any.

And there you have it! You've just created your very own custom error type. You're now officially a Rust Error Chef, cooking up your own special blend of errors. Enjoy!

Returning Custom Errors

Alright, now that we've introduced our own superstar error type into the world, it's time to let it shine in the spotlight! Brace yourself as we now dive into the realm of returning custom errors from functions.

Imagine you are working on a project about a library. You have a function borrow_book() that lets people borrow books from your collection. However, if a book is already borrowed, your function should return an error.

First, let's define our custom error type. For this library, we have a very specific kind of error: a BookError. We can define it as follows:

pub struct BookError {
    pub message: String,

impl fmt::Display for BookError {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        write!(f, "{}", self.message)

impl Error for BookError {}

We've created a struct BookError with a single field message to hold our error message. We then implemented the Display trait for our BookError so it can be printed in a user-friendly format. Lastly, we marked BookError as implementing the Error trait, which is empty but essential for interoperability with other error types.

Now, let's use our BookError in our borrow_book() function:

fn borrow_book(book_id: u32, library: &mut Library) -> Result<(), BookError> {
    if library.is_book_borrowed(book_id) {
        Err(BookError {
            message: String::from("Sorry, this book is already borrowed!"),
    } else {

In this function, we're returning a Result with the Err variant being our newly minted BookError. If the book is already borrowed, we return our error with a sassy message: "Sorry, this book is already borrowed!". If not, we let the borrower take the book and return Ok(()) to indicate that all went smoothly.

So, there you have it! You've just learned how to make your functions return your very own, home-brewed error type. Now go forth and sprinkle your custom errors everywhere (but, you know, only where appropriate)!

Handling Custom Errors

Alright, so you've created your custom error type. Now, we need to handle these quirky little troublemakers. Don't fret though, it's just like handling any other errors but with a personal touch.

Let's start by talking about strategies.

Using match or ?

Remember our good friends match and ?. Just like they've been there for us with regular errors, they're not leaving us hanging with custom ones.

For instance, imagine we have a function that returns our custom MyError type in a Result. Handling it with a match would look something like this:

match might_return_error() {
    Ok(value) => println!("Got value: {}", value),
    Err(e) => match e {
        MyError::Type1 => println!("Error of Type 1 occurred"),
        MyError::Type2 => println!("Error of Type 2 occurred"),

Look familiar? That's because it's the same old match we've been using, but now it has some custom error action. We're matching on the specific variants of our MyError enum to provide tailored responses.

But what about the ? operator? Well, it's business as usual:

let value = might_return_error()?;

This will return the value if our function is Ok, or return the error and exit the function if it's Err. The custom error will propagate up to be handled elsewhere, just like it's used to.

Helpful Error Messages

Having custom error types is great, but what makes them even better is if they can tell us what went wrong. A nice, helpful error message can turn a debugging nightmare into a walk in the park.

When we define our custom error type, we can include a description method that returns a detailed message:

impl fmt::Display for MyError {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        match self {
            MyError::Type1 => write!(f, "Type 1 error happened! Oh no!"),
            MyError::Type2 => write!(f, "Type 2 error happened! Not again!"),

Now, when we print our error, we get a detailed message:

if let Err(e) = might_return_error() {
    println!("Error: {}", e);

In this case, our program will print "Error: Type 1 error happened! Oh no!" or "Error: Type 2 error happened! Not again!" depending on the error variant. Now that's an error message that's hard to ignore!

And there you have it, handling custom errors like a pro! Now go out there and handle those errors with style and a touch of personal flair!

Rise In Logo

Rise together in web3