Is there a practical difference between creating Promises by hand and using the async/await API?

Question

I've inherited a codebase which is full of functions like this:

const someFunc = async (): Promise<string> => {
    return new Promise(async (resolve, reject) => {
        try {
            const result = await doSomething();
            resolve(result);
        } catch (error) {
            reject(error);
        }
    });
};

It is my understanding that since the error is not handled in the catch this is essentially the same as doing this:

const someFunc = (): Promise<string> => {
    return doSomething();
};

Did I miss something?

Answer 1

This is horrible indeed. Never pass an async function as the executor to new Promise!

Did I miss something?

Synchronous exceptions thrown by doSomething. We assume it returns a promise, so this should never happen, but if they do, then your code is not strictly equivalent to the original which returns a rejected promise. You'd fix this by simply making it an async function:

// eslint-disable-next-line require-await -- async is used to catch the synchronous exceptions
const someFunc = async (): Promise<string> => {
//               ^^^^^
    return doSomething();
};

If this is not an issue, you could shorten the code even further:

const someFunc: () => Promise<string> = doSomething;