Package Exports
- ext-channel-queue
This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (ext-channel-queue) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.
Readme
channel-queue
A queue management library with channels.
About
This library provides a queue system, organised by channels (or "topics"), that allows for finer-grain control over asynchronous method execution. It allows for queuing tasks by channels (string-based names) - tasks executing sequentially when the previous task is done. It also allows for 3 priority levels : normal, high and tail.
Check out the API documentation.
Installation
Simply run:
npm install @buttercup/channel-queue --save
Usage
The simplest usage is to just queue some tasks:
const ChannelQueue = require("@buttercup/channel-queue");
const queue = new ChannelQueue();
const work = queue.channel("myChannel").enqueue(() => 123);
// 'work' will resolve with 123
const someChannel = queue.channel("someChannel");
someChannel.enqueue(job1);
someChannel.enqueue(job2);
someChannel.enqueue(job3);
You can push tasks above others by setting them as high-priority:
const ChannelQueue = require("@buttercup/channel-queue");
const { TASK_TYPE_HIGH_PRIORITY } = ChannelQueue.Task;
const queue = new ChannelQueue();
const workChannel = queue.channel("work");
// some other tasks added to the queue...
workChannel.enqueue(importantJobMethod, TASK_TYPE_HIGH_PRIORITY).then(result => {
// 'result' is the resolved result from the 'importantJobMethod' function
});
Tasks can also be run as "low" priority or at the tail end of the queue:
const ChannelQueue = require("@buttercup/channel-queue");
const { TASK_TYPE_TAIL } = ChannelQueue.Task;
const queue = new ChannelQueue();
const workChannel = queue.channel("work");
// some other tasks added to the queue...
workChannel.enqueue(runNearEnd, TASK_TYPE_TAIL);
Stacking
Items can be "stacked", meaning that if specified, items can be limited to only 1 pending item in queue. All items of the same stack name would simply queue on the same item and not create more tasks. The stack can be specified when enqueuing:
const ChannelQueue = require("@buttercup/channel-queue");
const queue = new ChannelQueue();
const workChannel = queue.channel("work");
const promise1 = workChannel.enqueue(saveWorkFn, undefined, /* Stack ID */ "save");
// work start
const promise2 = workChannel.enqueue(saveWorkFn, undefined, "save");
const promise3 = workChannel.enqueue(saveWorkFn, undefined, "save");
// promise2 and promise3 will be equal, as promise2 was still in the queue when promise3
Parallel Execution
Tasks can be run in parallel using the ParallelChannel
class. You can create a parallel channel, in place of a regular channel, by calling ChannelQueue#createParallelChannel
:
const ChannelQueue = require("@buttercup/channel-queue");
const queue = new ChannelQueue();
const workChannel = queue.createParallelChannel("work");
workChannel.enqueue(someTask);
// The same channel can be fetched later using the familiar channel() method:
queue.channel("work"); // The parallel channel
Parallel channels, like their name implies, can run tasks in parallel. Instead of running them head-to-tail like regular channels, parallel channels can execute many tasks side-by-side. You can also limit them to a certain number of threads (default is 2) by calling queue.createParallelChannel("name", 5)
(where 5
is the maximum number of simultaneous tasks).
Parallel channels by default do not run tasks of different priorities simultaneously. This means that if the current running tasks are high-priority, no normal priority tasks will be started. This feature can be disabled by running parallelChannel.canRunAcrossTaskTypes = true
.
Clearing
You can clear all enqueued items in a Channel
by calling clear()
:
queue.channel("myChannel").clear();
Development & Supported Node Versions
This library is intended to be used with NodeJS version 6 and later.
To contribute, clone this project and run npm install
before beginning development.
To test, simply run npm test
.