A functional reactive programming lib for TypeScript JavaScript, written in TypeScript.
Turns your event spaghetti into clean and declarative feng shui bacon, by switching
from imperative to functional. It's like replacing nested for-loops with functional programming
concepts like map
and filter
. Stop working on individual events and work with event streams instead.
Combine your data with merge
and combine
.
Then switch to the heavier weapons and wield flatMap
and combineTemplate
like a boss.
Here's the stuff.
Bacon.js starting from version 3.0 is a Typescript library so you won't need any external types. Just
Install using npm
.
npm install baconjs
Then you can
import { EventStream, once } from "baconjs"
let s: EventStream<string> = once("hello")
s.log()
As you can see, the global methods, such as once
are imported separately.
Check out the new API Documentation, that's now generated using Typedoc from the Typescript source code.
You can directly import Bacon.js as single aggregated ES6 module.
import * as Bacon from 'node_modules/baconjs/dist/Bacon.mjs';
Bacon.once("hello").log();
If you're on to CommonJS (node.js, webpack or similar) you can install Bacon using npm.
npm install baconjs
Try it like this:
node
Bacon=require("baconjs")
Bacon.once("hello").log()
The global methods, such as once
are available in the Bacon
object.
For bower users:
bower install bacon
Both minified and unminified versions available on cdnjs.
So you can also include Bacon.js using
<script src="https://cdnjs.cloudflare.com/ajax/libs/bacon.js/2.0.9/Bacon.js"></script>
<script>
Bacon.once("hello").log()
</script>
Bacon.js is an UMD module so it should work with AMD/require.js too. Not tested lately though.
Prefer to drink from the firehose? Download from Github master.
The idea of Functional Reactive Programming is quite well described by Conal Elliot at Stack Overflow.
Bacon.js is a library for functional reactive programming. Or let's say it's a library for working with events in EventStreams and dynamic values (which are called Properties in Bacon.js).
You can wrap an event source, say "mouse clicks on a DOM element" into an EventStream by saying
let $ = (selector) => document.querySelector(selector)
var clickE = Bacon.fromEvent($("h1"), "click")
The $
helper function above could be replaced with, for instance, jQuery or Zepto.
Each EventStream represents a stream of events. It is an Observable, meaning
that you can listen to events in the stream using, for instance, the onValue
method
with a callback. Like this:
clickE.onValue(() => alert("you clicked the h1 element") )
But you can do neater stuff too. The Bacon of Bacon.js is that you can transform,
filter and combine these streams in a multitude of ways (see EventStream API). The methods map
,
filter
, for example, are similar to same functions in functional list programming
(like Underscore). So, if you say
let plusE = Bacon.fromEvent($("#plus"), "click").map(1)
let minusE = Bacon.fromEvent($("#minus"), "click").map(-1)
let bothE = plusE.merge(minusE)
.. you'll have a stream that will output the number 1 when the "plus" button is clicked
and another stream outputting -1 when the "minus" button is clicked. The bothE
stream will
be a merged stream containing events from both the plus and minus streams. This allows
you to subscribe to both streams with one handler:
bothE.onValue(val => { /* val will be 1 or -1 */ console.log(val) })
Note that you can also use the log
method to log stream values to console
:
bothE.log()
In addition to EventStreams, bacon.js has a thing called Property
, that is almost like an
EventStream, but has a "current value". So things that change and have a current state are
Properties, while things that consist of discrete events are EventStreams. You could think
mouse clicks as an EventStream and mouse cursor position as a Property. You can create Properties from
an EventStream with scan
or toProperty
methods. So, let's say
let add = (x, y) => x + y
let counterP = bothE.scan(0, add)
counterP.onValue(sum => $("#sum").textContent = sum )
The counterP
property will contain the sum of the values in the bothE
stream, so it's practically
a counter that can be increased and decreased using the plus and minus buttons. The scan
method
was used here to calculate the "current sum" of events in the bothE
stream, by giving a "seed value"
0
and an "accumulator function" add
. The scan method creates a property that starts with the given
seed value and on each event in the source stream applies the accumulator function to the current
property value and the new value from the stream.
Hiding and showing the result div depending on the content of the property value is equally straightforward
let hiddenIfZero = value => value == 0 ? "hidden" : "visible"
counterP.map(hiddenIfZero)
.onValue(visibility => { $("#sum").style.visibility = visibility })
For an actual (though a bit outdated) tutorial, please check out my blog posts
There's a multitude of methods for creating an EventStream from different sources, including the DOM, node callbacks and promises for example. See EventStream documentation.
Properties are usually created based on EventStreams. Some common ways are introduced in Property documentation.
You can combine the latest value from multple sources using combine, combineAsArray, combineWith or combineTemplate.
You can merge multiple streams into one using merge or mergeAll.
You can concat streams using concat or concatAll.
If you want to get the value of an observable but emit only when another stream emits an event, you might want to use sampledBy or its cousin withLatestFrom.
One of the common first questions people ask is "how do I get the latest value of a stream or a property". There is no getLatestValue method available and will not be either. You get the value by subscribing to the stream/property and handling the values in your callback. If you need the value of more than one source, use one of the combine methods.
Bus
is an EventStream
that allows you to push
values into the stream.
It also allows plugging other streams into the Bus.
There are essentially three kinds of Events that are emitted by EventStreams and Properties:
map
, filter
and most of the other operators
also deal with values only.If you want to subscribe to all events from an Observable, you can use the subscribe method.
Error
events are always passed through all stream operators. So, even
if you filter all values out, the error events will pass through. If you
use flatMap, the result stream will contain Error events from the source
as well as all the spawned stream.
You can take action on errors by using onError
.
See also mapError
, errors
, skipErrors
,
Bacon.retry
and flatMapError
.
In case you want to convert (some) value events into Error
events, you may use flatMap
like this:
stream = Bacon.fromArray([1,2,3,4]).flatMap(function(x) {
if (x > 2)
return new Bacon.Error("too big")
else
return x
})
Conversely, if you want to convert some Error
events into value events, you may use flatMapError
:
myStream.flatMapError(function(error) {
return isNonCriticalError(error) ? handleNonCriticalError(error) : new Bacon.Error(error)
})
Note also that Bacon.js operators do not catch errors that are thrown.
Especially map
doesn't do so. If you want to map things
and wrap caught errors into Error events, you can do the following:
wrapped = source.flatMap(Bacon.try(dangerousOperation))
For example, you can use Bacon.try
to handle JSON parse errors:
var jsonStream = Bacon
.once('{"this is invalid json"')
.flatMap(Bacon.try(JSON.parse))
jsonStream.onError(function(err) {
console.error("Failed to parse JSON", err)
})
An Error does not terminate the stream. The method endOnError
returns a stream/property that ends immediately after the first error.
Bacon.js doesn't currently generate any Error
events itself (except when
converting errors using fromPromise
). Error
events definitely would be generated by streams derived from IO sources
such as AJAX calls.
See retry for retrying on error.
Bacon.js provides ways to get some descriptive metadata about all Observables.
See toString
, deps
, desc
,
spy
.
Function construction rules, which allowed you to use string shorthands for properties and methods, were removed in version 3.0, as they are not as useful as they used to be, due to the moderd, short lambda syntax in ES6 and Typescript, as well as libraries like Ramda and partial.lenses.
Lazy evaluation of event values has been removed in version 2.0
As described above, a subscriber can signal the loss of interest in new events in any of these two ways:
noMore
from the handler functiondispose()
function that was returned by the subscribe
or onValue
call.Based on my experience, an actual side-effect subscriber in application-code almost never does this. Instead you'll use methods like takeUntil to stop listening to a source when something happens.
The state of an EventStream can be defined as (t, os) where t
is time
and os
the list of current subscribers. This state should define the
behavior of the stream in the sense that
The rules are deliberately redundant, explaining the constraints from different perspectives. The contract between an EventStream and its subscriber is as follows:
Next
event.noMore
or
more
. The undefined
value is handled like more
.noMore
the source must never call the subscriber again.End
event. The return value of the subscribe function is
ignored in this case.A Property
behaves similarly to an EventStream
except that
subscribe
, it will deliver its current value
(if any) to the provided subscriber function wrapped into an Initial
event.x
to its subscribers and that is the latest value emitted, it will deliver
this value to the new subscriber.Bacon.js supports atomic updates to properties for solving a glitches problem.
Assume you have properties A and B and property C = A + B. Assume that both A and B depend on D, so that when D changes, both A and B will change too.
When D changes d1 -> d2
, the value of A a1 -> a2
and B changes b1 -> b2
simultaneously, you'd like C to update atomically so that it
would go directly a1+b1 -> a2+b2
. And, in fact, it does exactly that.
Prior to version 0.4.0, C would have an additional transitional
state like a1+b1 -> a2+b1 -> a2+b2
Earlier versions of Bacon.js automatically installed the asEventStream
method into jQuery.
Now, if you still want to use that method, initialize this integration by calling Bacon.$.init($)
.
Bacon.js is quite similar to RxJs, so it should be pretty easy to pick up. The major difference is that in bacon, there are two distinct kinds of Observables: the EventStream and the Property. The former is for discrete events while the latter is for observable properties that have the concept of "current value".
Also, there are no "cold observables", which means also that all EventStreams and Properties are consistent among subscribers: when an event occurs, all subscribers will observe the same event. If you're experienced with RxJs, you've probably bumped into some wtf's related to cold observables and inconsistent output from streams constructed using scan and startWith. None of that will happen with bacon.js.
Error handling is also a bit different: the Error event does not
terminate a stream. So, a stream may contain multiple errors. To me,
this makes more sense than always terminating the stream on error; this
way the application developer has more direct control over error
handling. You can always use endOnError
to get a stream
that ends on the first error!
See Examples
See Specs
First check out the Bacon.js repository and run npm install
.
Then build the Typescript sources into a javascript bundle (plus typescript type definitions):
npm run dist
Result javascript files will be generated in dist
directory. If your planning
to develop Bacon.js yourself, you'll want to run [tests] too using npm test
.
Run all unit tests:
npm test
The tests are run against the javascript bundle in the dist
directory. You can build the bundle using npm run dist
.
This will loop thru all files under spec
and build the library with the
single feature and run the test.
Run browser tests locally:
npm install
npm run browsertest-bundle
npm rum browsertest-open
Run performance tests:
performance/PerformanceTest.coffee
performance/PerformanceTest.coffee flatmap
Run memory usage tests:
coffee --nodejs '--expose-gc' performance/MemoryTest.coffee
Runtime: none Build/test: see [package.json].
Bacon.js doesn't mess with prototypes or the global object, except that it exports the Bacon object as window.Bacon
when installed using the <script>
tag.
So, it should be pretty much compatible and a nice citizen.
I'm not sure how it works in case some other lib adds stuff to, say, Array prototype, though. Maybe add test for this later?
TLDR: good.
Bacon.js is not browser dependent, because it is not a UI library. It should work on all ES5-ish runtimes.
Automatically tested on each commit on modern browsers in Browserstack.
Bacon.js exists largely because I got frustrated with RxJs, which is a good library, but at that time didn't have very good documentation and wasn't open-source. Things have improved a lot in the Rx world since that. Yet, there are still compelling reasons to use Bacon.js instead. Like, for instance,
If you're more into performance and less into atomic updates, you might want to check out Kefir.js!
Use GitHub issues and Pull Requests.
Note:
dist/Bacon*.js
files are assembled from files in src/
. After updating source files, run npm install
to update the generated files. Then commit and create your Pull Request.Thanks to BrowserStack for kindly providing me with free of charge automatic testing time.
Thanks also to Reaktor for supporting Bacon.js development and letting me use some of my working hours on open-source development.
Generated using TypeDoc