Skip to content

Zyclotrop-j/iterate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

52 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

iterate-async

Iteration of iterables using a async function with concurrency

Example

ES6 Module

import ProcessConcurrently from 'iterate-async';


const result = await ProcessConcurrently(async (item) => {
  await Promise.resolve(); // do something async
  return item * 2
}, [1, 2, 3, 4, 5, 6]);
// result = [2, 4, 6, 8, 10, 12]

Common JS

const ProcessConcurrently = require('iterate-async').default; // don't forget the 'default'!

ProcessConcurrently(async (item) => {
    await Promise.resolve(); // do something async
    return item * 2;
}, [1, 2, 3, 4, 5, 6]).then(
    result => console.log(result)
    // result = [2, 4, 6, 8, 10, 12]
);

Browser

Normally you'd use this with your own bundler with one of the two methods above. You can also compile this from source; the full source is ES6 and in the src folder. The npm bundle includes all builds in the build folder. It contains .mjs (ES6), .cjs (common js) and .js ('classic' oldschool js).

When importing into the browser directly, preferably, use the ES6 export - all modern browsers support this!

<script type="module">
  import ProcessConcurrently from '<your path>/iterate-async/build/index.mjs';
  ... as abve ...
</script>

If you need this in the global namespace for some reason, you can use the 'classic' oldschool export.

<script type="javascript" src="<path>/iterate-async/build/index.js"></script>
// window.ProcessConcurrently is now globally available holding all exports

Options

You can configure the behaviour of ProcessConcurrently using options (third argument to ProcessConcurrently). All options are optional. Below you find the defaults.

const options = {
  commonArgs: undefined,  // type: any; anything additional you'd like to pass to the processing-function.
  concurrency: 4,  // default: 4; type = positive integer; how many items are processed at the same time
  log: (...args) => console.log(...args),  //  type = function(...args: any[]): void; logs to global console by default; attach your logger here!
};
const fn = async (item, commonArgs) => {
  // commonArgs is whatever you pass in options
  // e.g. with options.commonArgs = { foo: 'bar' }, this will be { foo: 'bar' }
  ...
};
const result = await ProcessConcurrently(fn, my_iterable, options);

Meta-information and context

The current progress-information is passed to the process-function as a third argument. The following context information is available:

type Meta = {
  idx: number // index of the current item being processed
  done: number // count of items already processed
  active: number // count of items currently processing
  idxArg: T // the current item in raw being processed
  results: R[] // whatever you have returned from the processing function
  signal: AbortSignal // if the current processing job was cancelled, this Abort-signal will go off; usefull to pass to network requests, e.g. fetch
  worker: number // index of current worker - same as count of jobs that have been spawned to handle items when this worker was instantiated
  waiting?: number // count of items waiting to be processed - not present if not available, e.g. because arr doesn't have a length
  total?: number // total count of items (in any state) - not present if not available, e.g. because arr doesn't have a length
}
const fn = async (item, commonArgs, metaInformation) => {
  // metaInformation has the form of Meta above
  console.log(metaInformation.idxArg); // 1, 2, 3
  console.log(metaInformation.results); // [], [10], [10, 20]
  ...
  return Promise.resolve(item * 10)
};
const arr: T[] = [1, 2, 3];
const result = await ProcessConcurrently(fn, arr, { concurrency: 1 });

If you are doing multiple async operations within your function and one task might "overtake" another, you can always get the latest status by just reading from metaInformation after each step - the stats returned by metaInformation are always the latest as of the time of when you're reading them!

Processing things other than arrays

You can process any iterable, not just arrays!

const result = await ProcessConcurrently(async (item) => {
  await Promise.resolve(); // do something async
  return item * 10
}, new Set([1, 2, 3, 4, 5, 6]));
// result = [10, 20, 30, 40, 50, 60]
const arr = function* () {
  yield 1;
  yield 2;
  yield 3;
  yield 4;
  yield 5;
  yield 6;
};
const result = await ProcessConcurrently(async (item) => {
  await Promise.resolve(); // do something async
  return item * 10
}, arr());
// result = [10, 20, 30, 40, 50, 60]

Examples

Call an api in the browser.

const fn = (item, { url }) => {
  return fetch(`${url}${item}`)
    .then(response => response.json());
};
const result = await ProcessConcurrently(fn, [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14], {
  commonArgs: {
    url: 'https://jsonplaceholder.typicode.com/todos/'
  }
});
console.log(result);

File-system in nodejs.

const fs = require('node:fs/promises');

const fn = (item, { folder }) => {
  return fs.readFile(`${folder}${item}`);
};
const result = await ProcessConcurrently(fn, ['foo', 'bar'], {
  commonArgs: {
    folder: './'
  }
});
console.log(result);

Use an iterator.

const fs = require('node:fs');
const fsp = require('node:fs/promise');

const upload = async (file, fielname, target) => {
  ....
}
const fn = async (filename, { target, source }) => {
  const file = await fsp.readFile(`${source}${filename}`);
  const isOK = await upload(file, filename, target);
  return isOK;
};

const iterateDir = function*(dir) {
  for(const fileOrDir of fs.readdirSync(dir)) {
    if (fs.statSync(fileOrDir).isDirectory()) {
      return yield* iterateDir(fileOrDir);
    } else {
      yield fileOrDir;
    }    
  }
};

const result = await ProcessConcurrently(fn, iterateDir('./source'), {
  commonArgs: {
    target: 'https://myapi/upload',
    source: './source'
  }
});
console.log(result);

Use an async iterator.

const fs = require('node:fs');
const fsp = require('node:fs/promise');

const upload = async (file, fielname, target) => {
  ....
}
const fn = async (filename, { target, source }) => {
  const file = await fsp.readFile(`${source}${filename}`);
  const isOK = await upload(file, filename, target);
  return isOK;
};

const iterateDir = async function*(dir) {
  for(const fileOrDir of await fsp.readdir(dir)) {
    if (fs.statSync(fileOrDir).isDirectory()) {
      return yield* iterateDir(fileOrDir);
    } else {
      yield fileOrDir;
    }    
  }
};

const result = await ProcessConcurrently(fn, iterateDir('./source'), {
  commonArgs: {
    target: 'https://myapi/upload'
    source: './source'
  }
});
console.log(result);

Using meta-information.

const fn = (item, { progressIndicator }, meta) => {
  progressIndicator.innerHTML = `Progress: ${(meta.done / meta.total * 100)}%`;
  .... // actually progress your item here
};
const result = await ProcessConcurrently(fn, [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14], {
  commonArgs: { progressIndicator: document.querySelector('#progressindicator') }
});
console.log(result);

About

Iterate of iterables using a given concurrency

Resources

License

Stars

Watchers

Forks

Packages

No packages published