Simple, scalable state management
- Installation:
npm install mobx --save
. React bindings:npm install mobx-react --save
. To enable ESNext decorators (optional), see below. - Ten minute, interactive MobX React tutorial
- Official documentation and API overview
- Videos:
- Egghead.io course: Manage Complex State in React Apps with MobX
- Practical React with MobX. In depth introduction and explanation to MobX and React by Matt Ruby on OpenSourceNorth.
- Screencast: intro to MobX
- State Management Is Easy, React Amsterdam 2016 conf (slides)
- Transparent Reactive Programming and Mutable Data, Reactive2015 conf (slides)
- More tutorials, blogs and videos can be found on the MobX homepage
- Boilerplates and related projects
MobX is a battle tested library that makes state management simple and scalable by transparently applying functional reactive programming (TFRP). The philosophy behind MobX is very simple:
Anything that can be derived from the application state, should be derived. Automatically.
which includes the UI, data serialization, server communication, etc.
React and MobX together are a powerful combination. React renders the application state by providing mechanisms to translate it into a tree of renderable components. MobX provides the mechanism to store and update the application state that React then uses.
Both React and MobX provide very optimal and unique solutions to common problems in application development. React provides mechanisms to optimally render UI by using a virtual DOM that reduces the number of costly DOM mutations. MobX provides mechanisms to optimally synchronize application state with your React components by using a reactive virtual dependency state graph that is only updated when strictly needed and is never stale.
MobX has only a few core concepts. The following snippets can be tried online using JSFiddle (or without ES6 and JSX).
MobX adds observable capabilities to existing data structures like objects, arrays and class instances. This can simply be done by annotating your class properties with the @observable decorator (ES.Next), or by invoking the observable
or extendObservable
functions (ES5). See Language support for language-specific examples.
// ESNext class example:
class Todo {
id = Math.random();
@observable title = "";
@observable finished = false;
}
// ES5 constructor function example:
function Todo() {
this.id = Math.random()
extendObservable(this, {
title: "",
finished: false
})
}
// ... or just create plain objects:
function createTodo() {
return observable({
id: Math.random(),
title: "",
finished: false
})
}
Using @observable
is like turning a value into a spreadsheet cell. But unlike spreadsheets, not only can these values be primitive values, but references, objects and arrays as well. You can even define your own observable data sources.
With MobX you can define values that will be derived automatically when relevant data is modified. By using the @computed
decorator or by using parameterless functions as property values in extendObservable
.
// ESNext class example:
class TodoList {
@observable todos = [];
@computed get unfinishedTodoCount() {
return this.todos.filter(todo => !todo.finished).length;
}
}
// ES5 constructor function example:
function TodoList() {
extendObservable(this, {
todos: [],
unfinishedTodoCount: function() {
return this.todos.filter(function (todo) {
return !todo.finished
}).length;
}
})
}
MobX will ensure that unfinishedTodoCount
is updated automatically when a todo is added or when one of the finished
properties is modified.
Computations like these can very well be compared with formulas in spreadsheet programs like MS Excel. They update automatically whenever, and only when, needed.
Reactions are similar to a computed value, but instead of producing a new value, a reaction produces a side effect for things like printing to the console, making network requests, incrementally updating the React component tree to patch the DOM, etc. In short, reactions bridge reactive and imperative programming.
If you are using React, you can turn your (stateless function) components into reactive components by simply adding the @observer
decorator from the mobx-react
package onto them.
import React, {Component} from 'react';
import ReactDOM from 'react-dom';
import {observer} from "mobx-react";
// ESNext decorator / JSX
@observer
class TodoListView extends Component {
render() {
return <div>
<ul>
{this.props.todoList.todos.map(todo =>
<TodoView todo={todo} key={todo.id} />
)}
</ul>
Tasks left: {this.props.todoList.unfinishedTodoCount}
</div>
}
}
const TodoView = observer(({todo}) =>
<li>
<input
type="checkbox"
checked={todo.finished}
onClick={() => todo.finished = !todo.finished}
/>{todo.title}
</li>
);
const store = new TodoList();
ReactDOM.render(<TodoListView todoList={store} />, document.getElementById('mount'));
In ES5 a component declaration looks like this:
var TodoListView = observer(React.createClass({ /* etc */ }))
observer
turns React (function) components into derivations of the data they render.
Also, reactions can be created using the autorun
, autorunAsync
or when
functions to fit your specific situations.
When using MobX there are no smart or dumb components.
All components render smartly but are defined in a dumb manner. MobX will simply make sure the components are always re-rendered whenever needed, but also no more than that. So the onClick
handler in the above example will force the proper TodoView
to render, and it will cause the TodoListView
to render if the number of unfinished tasks has changed.
However, if you would remove the Tasks left
line (or put it into a separate component), the TodoListView
will no longer re-render when ticking a box. You can verify this yourself by changing the JSFiddle.
For an in-depth explanation about how MobX determines to which observables needs to be reacted, check out: Understanding what MobX reacts to
Unlike many flux frameworks, MobX is unopinionated about how user events should be handled.
- This can be done in a Flux like manner.
- Or by processing events using RxJS.
- Or by simply handling events in the most straightforward way possible, as demonstrated in the above
onClick
handler.
In the end it all boils down to: Somehow the state should be updated.
After updating the state MobX
will take care of the rest in an efficient, glitch-free manner. So simple statements, like below, are enough to automatically update the user interface.
There is no technical need for firing events, calling dispatcher or what more. A React component is in the end nothing more than a fancy representation of your state. A derivation that will be managed by MobX.
store.todos.push(
new Todo("Get Coffee"),
new Todo("Write simpler code")
);
store.todos[0].finished = true;
Nonetheless, MobX has an optional built-in concept of actions
.
Use them to your advantage; they will help you to structure your code better and make wise decisions about when and where state should be modified.
MobX is one of the least obtrusive libraries you can use for state management. That makes the MobX
approach not just simple, but very scalable as well:
With MobX you don't need to normalize your data. This makes the library very suitable for very complex domain models (At Mendix for example ~500 different domain classes in a single application).
Since data doesn't need to be normalized, and MobX automatically tracks the relations between state and derivations, you get referential integrity for free. Rendering something that is accessed through three levels of indirection?
No problem, MobX will track them and re-render whenever one of the references changes. As a result staleness bugs are a thing of the past. As a programmer you might forget that changing some data might influence a seemingly unrelated component in a corner case. MobX won't forget.
As demonstrated above, modifying state when using MobX is very straightforward. You simply write down your intentions. MobX will take care of the rest.
MobX builds a graph of all the derivations in your application to find the least number of re-computations that is needed to prevent staleness. "Derive everything" might sound expensive, MobX builds a virtual derivation graph to minimize the number of recomputations needed to keep derivations in sync with the state.
In fact, when testing MobX at Mendix we found out that using this library to track the relations in our code is often a lot more efficient then pushing changes through our application by using handwritten events or "smart" selector based container components.
The simple reason is that MobX will establish far more fine grained 'listeners' on your data then you would do as a programmer.
Secondly MobX sees the causality between derivations so it can order them in such a way that no derivation has to run twice or introduces a glitch.
How that works? See this in-depth explanation of MobX.
MobX works plain javascript structures. Due to it's unobtrusiveness it works with most javascript libraries out of the box, without needing MobX specific library flavors.
So you can simple keep using your existing router-, data fetching and utility libraries like react-router
, director
, superagent
, lodash
etc.
For the same reason you can use it out of the box both server- and client side, in isomorphic applications and with react-native.
The result of this is that you often need to learn less new concepts when using MobX in comparison to other state management solutions.
MobX is proudly used in mission critical systems at Mendix
MobX is inspired by reactive programming principles as found in spreadsheets. It is inspired by MVVM frameworks like in MeteorJS tracker, knockout and Vue.js. But MobX brings Transparent Functional Reactive Programming to the next level and provides a stand alone implementation. It implements TFRP in a glitch-free, synchronous, predictable and efficient manner.
A ton of credits for Mendix, for providing the flexibility and support to maintain MobX and the chance to proof the philosophy of MobX in a real, complex, performance critical applications.
And finally kudo's for all the people that believed in, tried and validated MobX.
After using #mobx for lone projects for a few weeks, it feels awesome to introduce it to the team. Time: 1/2, Fun: 2X
Working with #mobx is basically a continuous loop of me going “this is way too simple, it definitely won’t work” only to be proven wrong
Try react-mobx with es6 and you will love it so much that you will hug someone.
I have built big apps with MobX already and comparing to the one before that which was using Redux, it is simpler to read and much easier to reason about.
The #mobx is the way I always want things to be! It's really surprising simple and fast! Totally awesome! Don't miss it!
- Feel free to send small pull requests. Please discuss new features or big changes in a GitHub issue first.
- Use
npm test
to run the basic test suite,npm run coverage
for the test suite with coverage andnpm run perf
for the performance tests.
TypeScript
Enable the compiler option experimentalDecorators
in tsconfig.json
or pass it as flag --experimentalDecorators
to the compiler.
Babel:
Install support for decorators: npm i --save-dev babel-plugin-transform-decorators-legacy
. And enable it in your babelrc
file:
{
"presets": [
"es2015",
"stage-1"
],
"plugins": ["transform-decorators-legacy"]
}
Probably you have more plugins and presets in your .babelrc
already, note that the order is important and transform-decorators-legacy
should come as first.
Bower support is available through the infamous npmcdn.com:
bower install https://npmcdn.com/mobx/bower.zip
Then use lib/mobx.umd.js
or lib/mobx.umd.min.js
See the changelog for all the details about mobservable
to mobx
.
Was MobX key in making your project a success? Share the victory by using the donate button! MobX is developed largely in free time, so any ROI is appreciated :-). If you leave a name it will be added to the sponsors list.