Frontend utilities - Retry operation

Retry Operation

Retry Operation is a handler around any async operation (not just fetch requests) that will automatically retry every X ms, N times when specific errors are thrown

Team

Activity Platform / Recent Work

Slack: #rw-team

Instructions

import { DEFAULT_RETRIES, FailedFetchError, retryOnException } from "@atlaskit/frontend-utilities/retry-operation";

FailedFetchError / GraphQLFragmentError

These are custom error classes that extends Error. Feel free to add your own!

retryOnException

retryOnException allows you to retry an async operation any number of times, at any interval, when specifically defined exceptions are thrown. You can define your own exceptions to be retried by creating a new class and extending Error or by passing through a function that performs a test on the error thrown and returning a boolean.

We also provider handlers so that you can gain a better understanding of what is happening underneath.

Usage

const result = await retryOnException<TResult>(
    asyncOperation,
    {
      intervalsMS: DEFAULT_RETRIES,
      retryOn: [FailedFetchError],
      captureException: this.captureException,
      onRetry: this.logNumRetries
    },
);

Constants

We also export predefined retry intervals for convenience.

NO_RETRIES                  = []
UP_TO_TWO_INSTANT_RETRIES   = [0, 0]
DEFAULT_RETRIES             = [0, 50, 100]
LAZY_LOAD_RETRIES           = [100, 500, 1000]