I'm encounter myself writing a custom hook that exposes a socket connection, the fact is that the code is there and it is working but I have the following API.
const socket = useWebSocket(
companyId,
accessToken,
resourceUrl,
!settings.data.manualDispatch,
options
);
The hook encapsulate several things:
Authenticate the WebSocket connection by asking first for a ticket that is appended to the WebSocket url (so it actually does a first HTTP request that returns a ticket and then trigger other effect that create a connection and update the socket state that is being return)
As I mentioned previously it does a request GET to a ticket and create a WebSocket connection
Cleanup any existing WebSocket connection when a new one is created (through state management)
The problem is:
I am currently using a reference within the hook to avoid create new objects and ending up in an infinite loop BUT It will be that consider an anti-pattern?.
Use memo alongside the hook?
Yes sure is valid approach but I feel like I am forcing the developers using my hook to write the same code over and over (memoize the options before pass it to the hook).
In summary:
OPTION 1:
Don't memoize the options to use the hook and do that process within the hook (because options are made of things that are not mean to change, urls and callbacks).
const socket = useWebSocket(
companyId,
accessToken,
resourceUrl,
!settings.data.manualDispatch,
{
onMessageCallback: message => {
// do things with message
},
onUnexpectedCloseCallback: () => {
// do things on close
},
url: websocketUrl,
}
);
OPTION 2:
We memoize the options in order to use the hook because the option 1 is an anti-pattern.
const options = useMemo(
() => ({
onMessageCallback: message => {
// do things with message
},
onUnexpectedCloseCallback: () => {
// do things on close
},
url: websocketUrl,
}),
[websocketUrl]
);
const socket = useWebSocket(
companyId,
accessToken,
resourceUrl,
!settings.data.manualDispatch,
options
);
Related
I have a react query which writes the state variable- follower, and I want to access this variable in other component to find its .length can someone tell me how do I do it
const ModalFollower = ({profile}) => {
const [follower,setFollower] = useState([])
const {
data: followerName,
isLoading: followerLoading,
isFetching: followerFetching
} = useQuery(["invitations", profile?.id], () => {
getFollowers(profile?.id).then((response) => {
if (response) {
setFollower(response);
}
});
});
return(
{
!followerLoading && (
follower.map((e) => {
return(<>
<p>{e.requested_profile.Userlink}</p>
</>}
)
}
)
I want to access the length of follower in some other component
There is no need to copy data from react-query to local state, because react-query is a full-blown state manager for server state. As long as you use the same query key, you will get data from its cache. This is best abstracted away in custom hooks.
Please be aware that with the default values, you will get a "background refetch" if a new component mount, so you will see two network requests if you use it twice. That might look confusing at first, but it is intended, as it is not react-query's primary goal to reduce network requests, but to keep your data on the screen as up-to-date as possible. So when a new component mounts that uses a query, you'll get the stale data from the cache immediately, and then a background refetch will be done. This procedure is called stale-while-revalidate.
The best way to customize this behaviour is to set the staleTime property to tell react-query how long your resource is "valid". For that time, you will only get data from the cache if available. I've written about this topic in my blog here: React Query as a State Manager.
React Query also provides selectors, so if your second component is only interested in the length, this is what my code would look like:
const useInvitations = (profile, select) =>
useQuery(
["invitations", profile?.id],
() => getFollowers(profile?.id),
{
enabled: !!profile?.id
select
}
)
Note that I also added the enabled property because apparently, profile can be undefined and you likely wouldn't want to start fetching without that id.
Now we can call this in our main component:
const ModalFollower = ({profile}) => {
const { data } = useInvitations(profile)
}
and data will contain the result once the promise resolves.
In another component where we only want the length, we can do:
const { data } = useInvitations(profile, invitations => invitations.length)
and data will be of type number and you will only be subscribed to length changes. This works similar to redux selectors.
What is the best way to pass data from one app to another app.
I have a two extJS app lets called appA and appB.
In my appA I am using some of the views of appB. Therefore I need to pass some data to be passed.
what is the correct way to pass the data.
Currently I am using
var myOb= {"key1" : "value1","key2" : "value2"}
sessionStorage.setItem('obJectData', muObj);
After use I am removing this like
sessionStorage.removeItem("obJectData");
Can anyone help me how and where to store the data in correct way.
I also have a though to be global vaiable.
If you need to have this across several different microapps all loaded on the same page, you might consider a combo approach of:
A singleton that holds the current shared state items
An events bus that can be used for pubsub (publish-subscribe) so that components/apps can subscribe to events and get updates when changes occur.
A rudimentary example could be seen here:
const appState = {
count: 0,
};
const eventBus = {
// just using the built-in DOM handlers, but you could create a custom one
subscribe: (eventName, handler) => document.addEventListener(eventName, handler),
publish: (eventName, payload) => document.dispatchEvent(new CustomEvent(eventName, payload)),
};
const incrementCount = () => {
appState.count = appState.count + 1;
eventBus.publish('newCount', { detail: { count: appState.count } });
}
const addCountDisplay = () => {
const newDisplay = document.createElement('div');
newDisplay.classList.add('countDisplay');
newDisplay.textContent = appState.count;
document.body.append(newDisplay);
}
// updateAllCountDisplays
eventBus.subscribe('newCount', (e) => {
const newCount = e.detail.count;
const displays = document.querySelectorAll('.countDisplay');
([...displays]).forEach((display) => display.textContent = newCount.toString());
});
<button onclick="incrementCount()">increment count</button>
<button onclick="addCountDisplay()">add count display</button>
In this example, you can see that:
Anytime you create a new count display, it is able to fetch the current count from the singleton appState
Anytime you update the count, a custom event is triggered that is able to be subscribed to to update the UI in reaction to the change in state
This could obviously be improved in many ways, but it serves as an example of how you can combine a globally available singleton in memory serving as a state cache with custom events to give all your components access to the current state as well as a way to subscribe and react to changes in state. This is not the only pattern you could use; it's just one option to consider.
I have the following scenario in redux-observable. I have a component which detects which backend to use and should set the backend URL used by the api-client. Both the client and URL are held in the global state object.
The order of execution should be:
1. check backend
2. on error replace backend URL held in state
3. trigger 3 actions to load resources using new backend state URL
What i did so far is, in step 1. access the state$ object from within my epic and modify the backed URL. This seems to only half work. The state is updated by actions triggered in 3. still see the old state and use the wrong backend.
What is the standard way to update state in between actions if you depend on the order of execution?
My API-Epic looks like this:
export const authenticate = (action$, state$) => action$.pipe(
ofType(actions.API_AUTHENTICATE),
mergeMap(action =>
from(state$.value.apiState.apiClient.authenticate(state$.value.apiState.bearer)).pipe(
map(bearer => apiActions.authenticatedSuccess(bearer))
)
)
)
export const authenticatedSuccess = (action$, state$) => action$.pipe(
ofType(actions.API_AUTHENTICATED_SUCCESS),
concatMap(action => concat(
of(resourceActions.doLoadAResource()),
of(resourceActions.doLoadOtherResource()),
of(resourceActions.doLoadSomethingElse()))
)
)
A common approach I've found users discussing on GitHub & StackOverflow is chaining multiple epics, much like what I believe your example tries to demonstrate. The first epic dispatches an action when it's "done". A reducer listens for this action and updates the store's state. A second epic (or many additional epics if you want concurrent operations) listen for this same action and kick off the next sequence of the workflow. The secondary epics run after the reducers and thus see the updated state. From the docs:
Epics run alongside the normal Redux dispatch channel, after the reducers have already received them...
I have found the chaining approach works well to decouple phases of a larger workflow. You may want the decoupling for design reasons (such as separation of concerns), to reuse smaller portions of the larger workflow, or to make smaller units for easier testing. It's an easy approach to implement when your epic is dispatching actions in between the different phases of the larger workflow.
However, keep in mind that state$ is an observable. You can use it to get the current value at any point in time -- including between dispatching different actions inside a single epic. For example, consider the following and assume our store keeps a simple counter:
export const workflow = (action$, state$) => action$.pipe(
ofType(constants.START),
withLatestFrom(state$),
mergeMap(([action, state]) => // "state" is the value when the START action was dispatched
concat(
of(actions.increment()),
state$.pipe(
first(),
map(state => // this new "state" is the _incremented_ value!
actions.decrement()),
),
defer(() => {
const state = state$.value // this new "state" is now the _decremented_ value!
return empty()
}),
),
),
)
There are lots of ways to get the current state from the observable!
Regarding the following line of code in your example:
state$.value.apiState.apiClient.authenticate(state$.value.apiState.bearer)
First, passing an API client around using the state is not a common/recommended pattern. You may want to look at injecting the API client as a dependency to your epics (this makes unit testing much easier!). Second, it's not clear how the API client is getting the current backend URL from the state. Is it possible the API client is using a cached version of the state? If yes, you may want to refactor your authenticate method and pass in the current backend URL.
Here's an example that handles errors and incorporates the above:
/**
* Let's assume the state looks like the following:
* state: {
* apiState: {
* backend: "URL",
* bearer: "token"
* }
*/
// Note how the API client is injected as a dependency
export const authenticate = (action$, state$, { apiClient }) => action$.pipe(
ofType(actions.API_AUTHENTICATE),
withLatestFrom(state$),
mergeMap(([action, state]) =>
// Try to authenticate against the current backend URL
from(apiClient.authenticate(state.apiState.backend, state.apiState.bearer)).pipe(
// On success, dispatch an action to kick off the chained epic(s)
map(bearer => apiActions.authenticatedSuccess(bearer)),
// On failure, dispatch two actions:
// 1) an action that replaces the backend URL in the state
// 2) an action that restarts _this_ epic using the new/replaced backend URL
catchError(error$ => of(apiActions.authenticatedFailed(), apiActions.authenticate()),
),
),
)
export const authenticatedSuccess = (action$, state$) => action$.pipe(
ofType(actions.API_AUTHENTICATED_SUCCESS),
...
)
Additionally, keep in mind when chaining epics that constructs like concat will not wait for the chained epics to "finish". For example:
concat(
of(resourceActions.doLoadAResource()),
of(resourceActions.doLoadOtherResource()),
of(resourceActions.doLoadSomethingElse()))
)
If each of these doLoadXXX actions "starts" an epic, all three will likely run concurrently. Each action will be dispatched one after another, and each epic will "start" running one after another without waiting for the previous one to "finish". This is because epics never really complete. They're long-lived, never ending streams. You will need to explicitly wait on some signal that identifies when doLoadAResource completes if you want to doLoadOtherResource to run after doLoadAResource.
I have a React Native app I'm going to be building that uses WebSockets. I have a WebSocket library written in JavaScript and I'm simply re-using it for this project, which is fantastic.
My question is, being new to React/React Native, what is the best practice for setting up and maintaining all of the traffic going through the WebSocket?
Initially my idea was to create the websocket in the main App component, something like this:
export default class App extends Component {
constructor(props) {
super(props);
this.ws = new WebSocket;
}
componentWillMount() {
console.log(this.ws);
}
render() {
console.log("We are rendering the App component.....");
return (
<View style={styles.container}>
<Text style={styles.welcome}>Hello, world</Text>
</View>
);
}
}
The actual WebSocket class would contain all of the respective connection handling:
ws.onopen = () => {
// connection opened
ws.send('something'); // send a message
};
ws.onmessage = (e) => {
// a message was received
console.log(e.data);
};
ws.onerror = (e) => {
// an error occurred
console.log(e.message);
};
ws.onclose = (e) => {
// connection closed
console.log(e.code, e.reason);
};
My question is, since the data coming through WebSocket will be applicable for state through many components in the React Native app, but it is not a class that will extend React.Component, do I not interact with Redux in the WebSocket class? Do I move all of the WebSocket connection handling to the App component and dispatch actions there to Redux?
What's the common pattern here to instantiate my WebSocket class and ensure that all traffic in it is properly getting passed to Redux so all component's state will funnel correctly?
Great answers here so far. Just wanted to add that where you keep your data should really be a decision based on what type of data it is. James Nelson has an excellent article on this topic that I refer to regularly.
For your case, let's talk about the first 3 types of state:
Data
Communication State
Control State
Data
Your WebSocket connection is generic and could technically return anything, but it's likely that the messages you're receiving are data. For example, let's say you're building a chat app. Then, the log of all messages that have been sent and received would be the data. You should store this data in redux with a messages reducer:
export default function messages(state = [], action) {
switch (action.type) {
case 'SEND_MESSAGE':
case 'RECEIVE_MESSAGE': {
return [ ...state, action.message ];
}
default: return state;
}
}
We don't have to (and we shouldn't) have any WebSocket logic in our reducers, as they are generic and don't care where the data is coming from.
Also, note that this reducer is able to handle sending and receiving in exactly the same way. This is because the network communication is handled separately by our communication state reducer.
Communication State
Since you're using WebSockets, the types of communication state you want to track may differ from my example. In an app that uses a standard API, I would track when a request is loading, failed, or successful.
In our chat app example, you'll probably want to track these request states whenever you send a message, but there could be other things you want to classify as communication state as well.
Our network reducer can use the same actions as the messages reducer:
export default function network(state = {}, action) {
switch (action.type) {
case 'SEND_MESSAGE': {
// I'm using Id as a placeholder here. You'll want some way
// to tie your requests with success/failure receipt.
return {
...state,
[action.id]: { loading: true }
};
} case 'SEND_MESSAGE_SUCCESS': {
return {
...state,
[action.id]: { loading: false, success: true }
};
} case 'SEND_MESSAGE_FAILURE': {
return {
...state,
[action.id]: { loading: false, success: false }
};
}
default: return state;
}
}
This way, we can easily find the status of our requests, and we don't have to bother with loading/success/failure in our components.
However, you might not care about the success/failure of any given request since you're using WebSockets. In that case, your communication state might just be whether or not your socket is connected. If that sounds better to you, then just write a connection reducer that responds to actions on open/close.
Control State
We'll also need something to initiate the sending of messages. In the chat app example, this is probably a submit button that sends whatever text is in an input field. I won't demonstrate the whole component, as we'll use a controlled component.
The takeaway here is that the control state is the message before it's sent. The interesting bit of code in our case is what to do in handleSubmit:
class ChatForm extends Component {
// ...
handleSubmit() {
this.props.sendMessage(this.state.message);
// also clear the form input
}
// ...
}
const mapDispatchToProps = (dispatch) => ({
// here, the `sendMessage` that we're dispatching comes
// from our chat actions. We'll get to that next.
sendMessage: (message) => dispatch(sendMessage(message))
});
export default connect(null, mapDispatchToProps)(ChatForm);
So, that addresses where all of our state goes. We've created a generic app that could use actions to call fetch for a standard API, get data from a database, or any number of other sources. In your case, you want to use WebSockets. So, that logic should live in your actions.
Actions
Here, you'll create all of your handlers: onOpen, onMessage, onError, etc. These can still be fairly generic, as you've already got your WebSocket utility set up separately.
function onMessage(e) {
return dispatch => {
// you may want to use an action creator function
// instead of creating the object inline here
dispatch({
type: 'RECEIVE_MESSAGE',
message: e.data
});
};
}
I'm using thunk for the async action here. For this particular example, that might not be necessary, but you'll probably have cases where you want to send a message then handle success/failure and dispatch multiple actions to your reducers from within a single sendMessage action. Thunk is great for this case.
Wiring It All Together
Finally, we have everything set up. All we have to do now is initialize the WebSocket and set up the appropriate listeners. I like the pattern Vladimir suggested--setting up the socket in a constructor--but I would parameterize your callbacks so that you can hand in your actions. Then your WebSocket class can set up all the listeners.
By making the WebSocket class a singleton, you're able to send messages from inside your actions without needing to manage references to the active socket. You'll also avoid polluting the global namespace.
By using the singleton set up, whenever you call new WebSocket() for the first time, your connection will be established. So, if you need the connection to be opened as soon as the app starts, I would set it up in componentDidMount of App. If a lazy connection is okay, then you can just wait until your component tries to send a message. The action will create a new WebSocket and the connection will be established.
You can create dedicated class for WebSocket and use it everywhere. It's simple, concise and clear approach. Moreover you will have all stuff related to websockets encapsulated in one place! If you wish you can even create singleton out of this class, but the general idea is this:
class WS {
static init() {
this.ws = new WebSocket('ws://localhost:5432/wss1');
}
static onMessage(handler) {
this.ws.addEventListener('message', handler);
}
static sendMessage(message) {
// You can have some transformers here.
// Object to JSON or something else...
this.ws.send(message);
}
}
You have only run init somewhere in index.js or app.js:
WS.init();
And now you can loosely send message from any application layer, from any component, from any place:
WS.sendMessage('My message into WebSocket.');
And receive data back from WebSocket:
WS.onMessage((data) => {
console.log('GOT', data);
// or something else or use redux
dispatch({type: 'MyType', payload: data});
});
So you can use it everywhere even in redux in any action or somewhere else!
There are no official guidelines about that. I think using a component is confusing because it will not be rendered, and I guess if you use Redux you want to share the data from websocket anywhere in the application.
You can give the dispatch function to your Websocket manager.
const store = createStore(reducer);
const ws = new WebSocketManager(store.dispatch, store.getState);
And use this.dispatch inside your class methods.
// inside WebSocketManager class
constructor(dispatch, getState) {
this.dispatch = dispatch;
this.getState = getState;
}
You can also use middlewares to handle side effects, I think it is the recommended way. There are two great libraries that you can look :
redux-saga
redux-observable
I want to create a websocket on demand when certain components want to subscribe to data. How can I share the websocket instance in a redux fashion?
action.js
export function subscribeToWS(url) {
return dispatch => {
let websocket = new WebSocket(url)
websocket.on('connect', () => {
websocket.send("subscribe")
}
websocket.on('message', (message) => {
dispatch(storeNewData(message))
}
}
}
I could do something like this, but this would require a new instance for every new subscription.
The standard place to put things like persistent connection objects is inside middleware. And, in fact, there's literally dozens of existing middlewares that demonstrate that approach, with most of them listed over at https://github.com/markerikson/redux-ecosystem-links/blob/master/middleware.md#sockets-and-adapters . You should be able to use some of those as-is, or at least as examples.
You can look at redux-websocket-bridge. It unfold Web Socket messages into Redux action, and relay Redux action to Web Socket.
Upside of this approach: you can use Redux on your server as API endpoint, replacing standard REST API with less code.
Also, if your server do not send Flux Standard Action, you can still use redux-websocket-bridge for raw messages. It works with string, ArrayBuffer, and Blob. Of course, you can always write a small middleware to translate them into Flux Standard Action, e.g. messages from Slack RTM API.
Although this is quite an old question by now, it popped up several times when looking for an example. As #matthewatabet and #abguy mention, https://github.com/luskhq/redux-ws just mentions it has been deprecated and you can use Redux Thunk, without an example specific for web sockets.
For future reference, I found this article that outlines an example, that is implemented in a Github repo, starting on this file. This is for socket.io, but using web sockets directly should be similar.
Summarizing, in the Component call dispatch with addNewItemSocket:
<RaisedButton
label="Click to add!" primary={true}
onTouchTap={ () => {
const newItem = ReactDOM.findDOMNode(this.refs.newTodo.input).value
newItem === "" ? alert("Item shouldn't be blank")
: dispatch(addNewItemSocket(socket,items.size,newItem))
{/*: dispatch(addNewItem(items.size,newItem))*/}
ReactDOM.findDOMNode(this.refs.newTodo.input).value = ""
}
}
/>
In the actions file, declare addNewItemSocket as:
export const addNewItemSocket = (socket,id,item) => {
return (dispatch) => {
let postData = {
id:id+1,
item:item,
completed:false
}
socket.emit('addItem',postData)
}
}
To handle incoming messages from the socket, in the constructor of the component:
socket.on('itemAdded',(res)=>{
console.dir(res)
dispatch(AddItem(res))
})
And in the actoins file, declare AddItem as:
export const AddItem = (data) => ({
type: "ADD_ITEM",
item: data.item,
itemId:data.id,
completed:data.completed
})
For me this is still new, so any feedback is appreciated. I will also file a PR with https://github.com/luskhq/redux-ws to have an example listed there.