promising-utils v0.2.2

promising-utils

A utility library for promise, supports both commonjs and ESM

npm install promising-utils --save
yarn add promising-utils

wait

Used when try to wait/sleep/delay for some time options

- duration: number
// miliseconds

timeout

- duration?: number
// milliseconds
- timeoutMsg?: string
// custom error message returns from reject

if duration is not defined, then it will never time out

retry

Used to retry a promise if rejected

- retries?: number
// Number of retry times

- timeout?: number
 // define timeout for each promsie

- retryIf?: (err: any, retriedTimes: number) => boolean;
// only retry when this return true

- retryUntil?: (err: any, count: number) => boolean;
// stop retry when this return true

-  delay?: number | ((count: number) => number);
// A number or function that returns how long it should wait before next retry. `count` is the number of times it's been retried

if 2 or more of retries, retryIf and retryUntil are defined at same time, retry will stop when first condition is met.

series

Used when to run a promise one at a one in sequence.

delay?: number | ((count: number) => number);
// delay between each promise execution
onEachStart?: (index: number) => void;
// called when each promise is executed
onEachEnd?: (index: number) => void;
// calle when each promise is resolved
stopOnReject?: boolean;
// If a promise is rejected, then stop the flow
waterFall?: boolean;
// If waterFall is true, then the current promise will receive resolved result from last promise as input

series is basically batch with size is 1

batch

Similar to series, but can run a batch of promises at one time, 2nd batch won't start until all promises in first batches finish

size?: number;
delay?: number | ((count: number) => number);
onBatchStart?: (index: number) => void;
onBatchEnd?: (index: number) => void;
stopOnReject?: boolean;
waterFall?: boolean;
waterFallInitialValue?: any;

parallel

Used when want to have fixed number of concurrent promises run at any time.

concurrency?: number;
// Number of concurrent promises running
timeout?: number;
// timeout for each promise
retries?: number;
// default is 0
stopOnReject?: boolean;

The difference between parallel and batch:

Assume we have 30 promises to run

if use parallel with concurrency to be 10. Then first 10 promises start to run, if 1 promise finished, the 11th promise will fill in to maintain total concurrency number as 10. and so on so forth.

if use batch, the first 10 promises start to run, if 1 promise finish, nothing happens, it continue to wait for rest 9 promises to finish, only after all 10 promises finish, the next batch of 10 will start to run.