co

What is co?

The co npm package is a generator based flow-control utility for Node.js and the browser, making it easier to work with asynchronous JavaScript operations. It allows you to use generators to yield any function that returns a Promise. It can be used to simplify callback or promise-based code, especially in the context of async/await patterns.

What are co's main functionalities?

Sequential Execution

This feature allows for sequential execution of asynchronous tasks. The code sample demonstrates how you can use co to run promises in sequence using a generator function, which yields a promise that resolves to true.

co(function* () {
  var result = yield Promise.resolve(true);
  return result;
}).then(function (value) {
  console.log(value);
}, function (err) {
  console.error(err.stack);
});

Error Handling

This feature demonstrates how co can be used for error handling in asynchronous operations. The code sample shows a generator function yielding a promise that gets rejected, and the error is caught and logged.

co(function* () {
  try {
    yield Promise.reject(new Error('Oops!'));
  } catch (err) {
    console.error(err.message);
  }
});

Parallel Execution

This feature showcases how co can handle parallel execution of promises. The code sample illustrates a generator function yielding an array of promises, which co runs in parallel, and then logs the array of results.

co(function* () {
  var res = yield [
    Promise.resolve(1),
    Promise.resolve(2),
  ];
  return res;
}).then(function (value) {
  console.log(value);
});

Other packages similar to co

bluebird

Bluebird is a fully featured promise library with focus on innovative features and performance. It provides utilities for working with promises including but not limited to parallel execution, which is similar to what co offers. However, Bluebird does not use generator functions.

async

Async is a utility module which provides straight-forward, powerful functions for working with asynchronous JavaScript. Although it does not use promises or generators in the same way as co, it offers similar functionalities in terms of controlling the flow of asynchronous operations.

q

Q is a promise library for JavaScript which provides a toolset for creating and composing asynchronous promises. It is similar to co in that it helps manage asynchronous operations, but it does not utilize generator functions for flow control.

README

Co

Generator based flow-control goodness for nodejs (and soon the browser), using thunks or promises, letting you write non-blocking code in a nice-ish way.

Currently you must use the --harmony-generators flag when running node 0.11.x to get access to generators.

Co is careful to relay any errors that occur back to the generator, including those within the thunk, or from the thunk's callback. "Uncaught" exceptions in the generator are passed to co()'s thunk.

Make sure to view the examples.

Installation

$ npm install co

Associated libraries

View the wiki for libraries that work well with Co.

Example

var co = require('co');

co(function *(){
  var a = yield get('http://google.com');
  var b = yield get('http://yahoo.com');
  var c = yield get('http://cloudup.com');
  console.log(a.status);
  console.log(b.status);
  console.log(c.status);
})()

co(function *(){
  var a = get('http://google.com');
  var b = get('http://yahoo.com');
  var c = get('http://cloudup.com');
  var res = yield [a, b, c];
  console.log(res);
})()

Yieldables

The "yieldable" objects currently supported are:

• promises
• thunks (functions)
• array (parallel execution)
• objects (parallel execution)
• generators (delegation)
• generator functions (delegation)

To convert a regular node function that accepts a callback into one which returns a thunk you may want to use thunkify or similar.

Thunks vs promises

While co supports promises, you may return "thunks" from your functions, which otherwise behaves just like the traditional node-style callback with a signature of: (err, result).

For example take fs.readFile, we all know the signature is:

fs.readFile(path, encoding, function(err, result){

});

To work with Co we need a function to return another function of the same signature:

fs.readFile(path, encoding)(function(err, result){

});

Which basically looks like this:

function read(path, encoding) {
  return function(cb){
    fs.readFile(path, encoding, cb);
  }
}

Receiver propagation

When co is invoked with a receiver it will propagate to most yieldables, allowing you to alter this.

var ctx = {};

function foo() {
  assert(this == ctx);
}

co(function *(){
  assert(this == ctx);
  yield foo;
}).call(ctx)

You also pass arguments through the generator:

co(function *(a){
  assert(this == ctx);
  assert('yay' == a);
  yield foo;
}).call(ctx, 'yay');

API

co(fn)

Pass a generator fn and return a thunk. The thunk's signature is (err, result), where result is the value passed to the return statement.

var co = require('co');
var fs = require('fs');

function read(file) {
  return function(fn){
    fs.readFile(file, 'utf8', fn);
  }
}

co(function *(){
  var a = yield read('.gitignore');
  var b = yield read('Makefile');
  var c = yield read('package.json');
  return [a, b, c];
})()

You may also yield Generator objects to support nesting:

var co = require('co');
var fs = require('fs');

function size(file) {
  return function(fn){
    fs.stat(file, function(err, stat){
      if (err) return fn(err);
      fn(null, stat.size);
    });
  }
}

function *foo(){
  var a = yield size('.gitignore');
  var b = yield size('Makefile');
  var c = yield size('package.json');
  return [a, b, c];
}

function *bar(){
  var a = yield size('examples/parallel.js');
  var b = yield size('examples/nested.js');
  var c = yield size('examples/simple.js');
  return [a, b, c];
}

co(function *(){
  var a = yield foo();
  var b = yield bar();
  console.log(a);
  console.log(b);
})()

Or if the generator functions do not require arguments, simply yield the function:

var thunkify = require('thunkify');
var request = require('superagent');

var get = thunkify(request.get);

function *results() {
  var a = yield get('http://google.com')
  var b = yield get('http://yahoo.com')
  var c = yield get('http://ign.com')
  return [a.status, b.status, c.status]
}

co(function *(){
  // 3 concurrent requests at a time
  var a = yield results;
  var b = yield results;
  var c = yield results;
  console.log(a, b, c);

  // 9 concurrent requests
  console.log(yield [results, results, results]);
})()

If a thunk is written to execute immediately you may acheive parallelism by simply yield-ing after the call. The following are equivalent since each call kicks off execution immediately:

co(function *(){
  var a = size('package.json');
  var b = size('Readme.md');
  var c = size('Makefile');

  return [yield a, yield b, yield c];
})()

Or using join:

co(function *(){
  var a = size('package.json');
  var b = size('Readme.md');
  var c = size('Makefile');

  return yield [a, b, c];
})()

You can also pass arguments into the generator. The last argument, done, is the callback function. Here's an example:

var exec = require('co-exec');
co(function *(cmd) {
  var res = yield exec(cmd);
  return res;
})('pwd', done);

yield array

By yielding an array of thunks you may "join" them all into a single thunk which executes them all concurrently, instead of in sequence. Note that the resulting array ordering is retained.

var co = require('co');
var join = co.join;
var fs = require('fs');

function size(file) {
  return function(fn){
    fs.stat(file, function(err, stat){
      if (err) return fn(err);
      fn(null, stat.size);
    });
  }
}

co(function *(){
  var a = size('.gitignore');
  var b = size('index.js');
  var c = size('Makefile');
  var res = yield join(a, b, c);
  console.log(res);
  // => [ 13, 1687, 129 ]
})()

As an alias of join(array) you may simply yield an array:

co(function *(){
  var a = size('.gitignore');
  var b = size('index.js');
  var c = size('Makefile');
  var res = yield [a, b, c];
  console.log(res);
  // => [ 13, 1687, 129 ]
})()

Nested joins may also be expressed as simple nested arrays:

var a = [
  get('http://google.com'),
  get('http://yahoo.com'),
  get('http://ign.com')
];

var b = [
  get('http://google.com'),
  get('http://yahoo.com'),
  get('http://ign.com')
];

console.log(yield [a, b]);

yield object

Yielding an object behaves much like yielding an array, however recursion is supported:

co(function *(){
  var user = yield {
    name: {
      first: get('name.first'),
      last: get('name.last')
    }
  };
})()

Here is the sequential equivalent without yielding an object:

co(function *(){
  var user = {
    name: {
      first: yield get('name.first'),
      last: yield get('name.last')
    }
  };
})()

Performance

On my machine 30,000 sequential stat()s takes an avg of 570ms, while the same number of sequential stat()s with co() takes 610ms, aka the overhead introduced by generators is extremely negligable.

License

MIT