npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

cntxt

v5.7.0

Published

Pipeline context

Downloads

43

Readme

cntxt

Build Status NPM version

Context for executing and accumulating data through a function pipeline.

Index

Latest Version

5.6.0

Installation

npm install --save cntxt
yarn add --exact cntxt

Usage

Constructor

A context instance can be created via its constructor and then run by passing an array of pipeline steps to the .run method which returns a promise that resolves or rejects appropriately after all the steps are either successfully run, an error is thrown, or a given step short-circuits execution via .succeed, .throw, or .fail.

let context = new Context({
  overwrite: false,
  data: {
    wow: 9
  }
});

function a (context) {
  // can call context.next explicitly
  context.next({hello: 10});
}

function b (context) {
  // or implicity by return defined value
  return {barf: 11};
}

function c (context) {
  context.next({honk: 12});
}

context.run([
  a,
  [b, c] // these steps will be run in parallel
]).then((context)=> {
  let {wow, hello} = context.data;
  Assert.deepEqual(data, {
    wow: 9,
    hello: 10,
    barf: 11,
    honk: 12
  });
});

Context.series, Context.parallel

The shorthand method of using nested arrays above for parallel steps is limited to single level nesting so as to avoid readability issues. These factory methods can be used for building more complicated pipelines.

Context.series([
  stepA,
  stepB,
  Context.parallel([
    stepC1,
    stepC2,
    stepC2
  ]),
  Context.parallel([
    Context.series([
      stepD1,
      stepD2
    ]),
    stepE
  ]),
  stepF
])

Context.runSeries, Context.runParallel

If configuration via the constructor isn't needed, the static .runSeries and .runParallel methods can be used directly to instantiate a context and run it in series or parallel respectively. A completion callback can be passed as an optional second argument. It can either accept a single context argument, or node-style (error, data) arguments. If the callback is omitted, a promise is returned.

// A. callback: (context)
Context.runSeries([
  function (context) {
    context.next({hello: 10});
  }
], function (context)=> {
  Assert(!context.errored());
  Assert.equal(context.data.hello, 10);
});

// B. callback: (error, data)
Context.runSeries([
  function (context) {
    context.next({hello: 10});
  }
], function (error, data)=> {
  Assert(!error);
  Assert.equal(data.hello, 10);
});

// C. promise
Context.runParallel([
  function (context) {
    context.next({hello: 10});
  },
  function (context) {
    context.next({goodbye: 11});
  }
]).then((context)=> {
  Assert.equal(context.data.hello, 10);
  Assert.equal(context.data.goodbye, 11);
});

Steps

Individual pipeline steps should be in one of the following forms

// A. A single argument function that takes a context argument.
//    It has a data attribute consisting of accumulated data from
//    previous steps and is also used for flow control via the
//    following methods: .next, .fail, .error, .succeed
//
//    Instead of calling .next directly, steps may alternatively
//    return an object which .next will be called with.

function contextStep (context) {
  context.next({more: 'data'});
}

function contextStepImplicitNext (context) {
  return {more: 'data'};
}

async function asyncContextStep (context) {
  const data = await somethingAsync();
  return {more: data};
}

// B. A two argument Function that accepts data and callback as arguments.
//    The data argument holds the context's accumulated data and the callback
//    is node style, and accepts two arguments: (error, data)
function callbackStep (data, callback) {
  callback(null, {more: 'data'});
}

// C. A Promise. When the promise resolves, the resulting value should be an
//    object with data to add to the context's accumulated data.
let promiseStep = new Promise((reject, resolve)=> {
  resolve({more: 'data'});
});

Context control flow

The context object has four methods used for control flow

.next(data)

Called within a step in order to advance to the next step and add data, which should be an object, to the context's accumulated data.

Context.runSeries([
  function (context) {
    context.next({more: 'filling'});
  }
]).then((context)=> {
  Assert.equal(context.data.more, 'filling');
});

Promises passed as data values to context.next will be set to their resolved value. If the promise rejects context.error will be called with that error. Note that promises will be evaluated in parallel.

Context.runSeries([
  function (context) {
    context.next({
      more: 'data'
      wow: new Promise((resolve, reject)=> {
        resolve('extra');
      })
    });
  }
]).then((context)=> {
  Assert.deepEqual(context.data, {
    more: 'filling',
    wow: 'extra'
  });
});

.throw(error)

Should be called when a step encounters an unexpected error (for example, a database isn't reachable), and the pipeline should stop processing in an errored state.

Context.runSeries([
  function (context) {
    let error = new Error('OH NO!');
    context.throw(error);
  }
]).catch((error)=> {
  Assert.equal(error.message, 'OH NO!');
});

.succeed(data)

Should be called when a step wants to complete the context run successfully without running any other steps. data will be added to the context's accumulated data.

Context.runSeries([
  function (context) {
    context.succeed({ok: 'done'});
  },
  function (context) {
    context.next({not: 'run'});
  }
]).then((context)=> {
  Assert.deepEqual(context.data, {
    ok: 'done'
  });
});

.fail(failure)

Should be called when a step encounters an expected error (for example, a record isn't found or input validation fails), and the pipeline should stop processing, but instead of ending in an error state, it will ended in failed state. The failure will be added to the context's data under the key failure. The common use case is to return a user-facing error message.

Context.runSeries([
  function (context) {
    let error = new Error('Not Found!');
    context.fail(error);
  },
  function (context) {
    context.next({not: 'run'});
  }
]).then((context)=> {
  Assert(context.failed());
  Assert.equal(context.error, null);
  Assert.equal(context.failure.message, 'Not Found!');
});

.wrap(key)

Helper that returns a node style callback with (error, data) args which calls .error or .next appropriately. If the optional key is specified, data passed to .next will be namespaced under key.

function createMessage (context) {
  // The following:
  Message.create(messages, context.wrap('messages'))

  // is shorthand for this:
  Message.create(messages, function (error, data) {
    if (error) {
      context.throw(error);
    }
    else {
      context.next({
        messages: data
      });
    }
  }
}

Example

const Assert = require('assert');
const Context = require('cntxt');

function findUser(context) {
  let {user_id} = context.data;
  User.find({id: user_id }, (error, user)=> {
    if (error) {
      // .throw is used for unexpected errors
      context.throw(error);
      return;
    }

    if (!user) {
      // .fail is used for expected errors that are
      // passed to consumer via message
      context.fail(`User not found ${user_id}`);
      return;
    }

    // .next is used to pass data and invoke the next
    // function in pipeline. after this, the next step
    // will be called, and `data.user` will be `user`
    return {user};
  })
}

function findUserGames(context) {
  let {user_id} = context.data;

  // games will evaluate to the result of the findAll promise
  context.next({
    games: Game.findAll({user_id})
  });
}

function getOpponents(context) {
  let {
    user_id,
    games
  } = context.data;

  let opponents = games.map((game)=> {
    return games.players.filter((player)=> {
      return (player.user.id !== user_id);
    })[0];
  });

  context.next({opponents});
}

function taunt(context) {
  let {opponents} = context.data;
  let messages = opponents.map((user_id)=> {
    return {
      recipient: user_id,
      content: 'Heehaw!'
    };
  });
  // context.wrap returns a function with (error, data) args
  // which calls .error or .next depending on those values
  Message.create(messages, context.wrap('message'))
}

var params = {user_id : 'derrrrp'};

// .runSeries runs pipeline and returns a promise
// The pipeline is finished when
// A) any step calls `.succeed`, `.throw`, or `.fail`,
// B) any step throws unexpectedly
// C) final step calls `.next` (equivalent to `.succeed`)
Context.runSeries([
  params,
  [findUser, findUserGames], // these will be run in parallel
  getOpponents,
  taunt
]).then((context)=> {
  // this is called when the pipeline `failed` or `succeeded`
  Assert.equal(context.succeeded(), true);
  Assert.equal(context.error, null);

  // accumulated data is available in context.data
  Assert.deepEqual(context.data, {
    user_id   : 'derrrrp',
    user      : { /* the user      */ },
    games     : [ /* the games     */ ],
    opponents : [ /* the opponents */ ],
    messages  : [ /* the messages  */ ]
  });
}).catch((error)=> {
  // this is called when the pipeline errored
  Assert.equal(error, null);
});

Or instead of a promise, callbacks can be passed as the second argument to .run, .runSeries, .runParallel

Context.runSeries([
  params,
  [findUser, findUserGames],
  getOpponents,
  taunt
], function callback (context) {
  // A single argument callback receives the context
});

Context.runSeries([
  params,
  [findUser, findUserGames],
  getOpponents,
  taunt
], function callback (error, data) {
  // A two argument node style callback receives
  // error: context.error
  // data: context.data
});

This test also has some more usage examples.