My situation is as follows: I am performing sequential HTTP requests, where one HTTP request depends on the response of the previous. I would like to combine the response data of all these HTTP requests into one observable. I have implemented this before using an async generator. The code for this was relatively simple:
async function* AsyncGeneratorVersion() {
let moreItems = true; // whether there is a next page
let lastAssetId: string | undefined = undefined; // used for pagination
while (moreItems) {
// fetch current batch (this performs the HTTP request)
const batch = await this.getBatch(/* arguments */, lastAssetId);
moreItems = batch.more_items;
lastAssetId = batch.last_assetid;
yield* batch.getSteamItemsWithDescription();
}
}
I am trying to move away from async generators, and towards RxJs Observables. My best (and working) attempt is as follows:
const observerVersion = new Observable<SteamItem>((subscriber) => {
(async () => {
let moreItems = true;
let lastAssetId: string | undefined = undefined;
while (moreItems) {
// fetch current batch (this performs the HTTP request)
const batch = await this.getBatch(/* arguments */, lastAssetId);
moreItems = batch.more_items;
lastAssetId = batch.last_assetid;
const items = batch.getSteamItemsWithDescription();
for (const item of items) subscriber.next(item);
}
subscriber.complete();
})();
});
Now, I believe that there must be some way of improving this Observer variant - this code does not seem very reactive to me. I have tried several things using pipe, however unfortunately these were all unsuccessful.
I found concatMap to come close to a solution. This allowed me to concat the next HTTP request as an observable (done with the this.getBatch method), however I could not find a good way to also not abandon the response of the current HTTP request.
How can this be achieved? In short I believe this problem could be described as appending data to an observable inside the observable itself. (But perhaps this is not a good way of handling this situation)
TLDR;
Here's a working StackBlitz demo.
Explanation
Here would be my approach:
// Faking an actual request
const makeReq = (prevArg, response) =>
new Promise((r) => {
console.log(`Running promise with the prev arg as: ${prevArg}!`);
setTimeout(r, 1000, { prevArg, response });
});
// Preparing the sequential requests.
const args = [1, 2, 3, 4, 5];
from(args)
.pipe(
// Running the reuqests sequantially.
mergeScan(
(acc, crtVal) => {
// `acc?.response` will refer to the previous response
// and we're using it for the next request.
return makeReq(acc?.response, crtVal);
},
// The seed(works the same as `reduce`).
null,
// Making sure that only one request is run at a time.
1
),
// Combining all the responses into one object
// and emitting it after all the requests are done.
reduce((acc, val, idx) => ({ ...acc, [`request${idx + 1}`]: val }), {})
)
.subscribe(console.warn);
Firstly, from(array) will emit each item from the array, synchronously and one by one.
Then, there is mergeScan. It is exactly the result of combining scan and merge. With scan, we can accumulate values(in this case we're using it to have access to the response of the previous request) and what merge does is to allow us to use observables.
To make things a bit easier to understand, think of the Array.prototype.reduce function. It looks something like this:
[].reduce((acc, value) => { return { ...acc }}, /* Seed value */{});
What merge does in mergeScan is to allow us to use the accumulator something like (acc, value) => new Observable(...) instead of return { ...acc }. The latter indicates a synchronous behavior, whereas with the former we can have asynchronous behavior.
Let's go a bit step by step:
when 1 is emitted, makeReq(undefined, 1) will be invoked
after the first makeReq(from above) resolves, makeReq(1, 2) will be invoked
after makeReq(1, 2) resolves, makeReq(2, 3) will be invoked and so on...
Somebody I consulted regarding this matter came up with this solution, I think it's quite elegant:
defer(() => this.getBatch(options)).pipe(
expand(({ more_items, last_assetid }) =>
more_items
? this.getBatch({ ...options, startAssetId: last_assetid })
: EMPTY,
),
concatMap((batch) => batch.getSteamItemsWithDescription()),
);
From my understanding the use of expand here is very similar to the use of mergeScan in #Andrei's answer
I have the following scenario: There is a service which is called "ContextProvider" that holds information regarding the context of the applicaiton (Logged In User, Things he can acess, etc). Right now I am observing this as following:
this.contextProvider.Context.subscribe(context => {
//Do Something
})
Now I have a service that will also be observable. I want this service to observe the context and return an observable. This would be easy with the map function:
let observable = this.contextProvider.Context.pipe(map(context => {
let aux: number = somevar + someothervar;
return aux;
})) //observable variable now holds the type Observable<number>
My scenario is a little bit more complex, because in order to fetch the result, I have to make an Http call, which is also an observable/promise:
let observable = this.contextProvider.Context.pipe(map(context => {
return this.httpClient.get<number>("Some URL").pipe(take(1));
})); //observable var now holds Obsevable<Observable<number>>
How can I make the "observable" var hold Observable?
EDIT: The URL value depends on the some values of the "context" variable
If I understand right your problem, you need to use concatMap for this case, like this
this.contextProvider.Context.pipe(
concatMap(context => {
return this.httpClient.get<number>("Some URL" + context.someData);
}));
You can find more patterns around the use of Observables with http calls in this article
I have an Angular 2 service:
import {Storage} from './storage';
import {Injectable} from 'angular2/core';
import {Subject} from 'rxjs/Subject';
#Injectable()
export class SessionStorage extends Storage {
private _isLoggedInSource = new Subject<boolean>();
isLoggedIn = this._isLoggedInSource.asObservable();
constructor() {
super('session');
}
setIsLoggedIn(value: boolean) {
this.setItem('_isLoggedIn', value, () => {
this._isLoggedInSource.next(value);
});
}
}
Everything works great. But I have another component which doesn't need to subscribe, it just needs to get the current value of isLoggedIn at a certain point in time. How can I do this?
A Subject or Observable doesn't have a current value. When a value is emitted, it is passed to subscribers and the Observable is done with it.
If you want to have a current value, use BehaviorSubject which is designed for exactly that purpose. BehaviorSubject keeps the last emitted value and emits it immediately to new subscribers.
It also has a method getValue() to get the current value.
The only way you should be getting values "out of" an Observable/Subject is with subscribe!
If you're using getValue() you're doing something imperative in declarative paradigm. It's there as an escape hatch, but 99.9% of the time you should NOT use getValue(). There are a few interesting things that getValue() will do: It will throw an error if the subject has been unsubscribed, it will prevent you from getting a value if the subject is dead because it's errored, etc. But, again, it's there as an escape hatch for rare circumstances.
There are several ways of getting the latest value from a Subject or Observable in a "Rx-y" way:
Using BehaviorSubject: But actually subscribing to it. When you first subscribe to BehaviorSubject it will synchronously send the previous value it received or was initialized with.
Using a ReplaySubject(N): This will cache N values and replay them to new subscribers.
A.withLatestFrom(B): Use this operator to get the most recent value from observable B when observable A emits. Will give you both values in an array [a, b].
A.combineLatest(B): Use this operator to get the most recent values from A and B every time either A or B emits. Will give you both values in an array.
shareReplay(): Makes an Observable multicast through a ReplaySubject, but allows you to retry the observable on error. (Basically it gives you that promise-y caching behavior).
publishReplay(), publishBehavior(initialValue), multicast(subject: BehaviorSubject | ReplaySubject), etc: Other operators that leverage BehaviorSubject and ReplaySubject. Different flavors of the same thing, they basically multicast the source observable by funneling all notifications through a subject. You need to call connect() to subscribe to the source with the subject.
I had similar situation where late subscribers subscribe to the Subject after its value arrived.
I found ReplaySubject which is similar to BehaviorSubject works like a charm in this case.
And here is a link to better explanation: http://reactivex.io/rxjs/manual/overview.html#replaysubject
const observable = of('response')
function hasValue(value: any) {
return value !== null && value !== undefined;
}
function getValue<T>(observable: Observable<T>): Promise<T> {
return observable
.pipe(
filter(hasValue),
first()
)
.toPromise();
}
const result = await getValue(observable)
// Do the logic with the result
// .................
// .................
// .................
You can check the full article on how to implement it from here.
https://www.imkrish.com/blog/development/simple-way-get-value-from-observable
I encountered the same problem in child components where initially it would have to have the current value of the Subject, then subscribe to the Subject to listen to changes. I just maintain the current value in the Service so it is available for components to access, e.g. :
import {Storage} from './storage';
import {Injectable} from 'angular2/core';
import {Subject} from 'rxjs/Subject';
#Injectable()
export class SessionStorage extends Storage {
isLoggedIn: boolean;
private _isLoggedInSource = new Subject<boolean>();
isLoggedIn = this._isLoggedInSource.asObservable();
constructor() {
super('session');
this.currIsLoggedIn = false;
}
setIsLoggedIn(value: boolean) {
this.setItem('_isLoggedIn', value, () => {
this._isLoggedInSource.next(value);
});
this.isLoggedIn = value;
}
}
A component that needs the current value could just then access it from the service, i.e,:
sessionStorage.isLoggedIn
Not sure if this is the right practice :)
A similar looking answer was downvoted. But I think I can justify what I'm suggesting here for limited cases.
While it's true that an observable doesn't have a current value, very often it will have an immediately available value. For example with redux / flux / akita stores you may request data from a central store, based on a number of observables and that value will generally be immediately available.
If this is the case then when you subscribe, the value will come back immediately.
So let's say you had a call to a service, and on completion you want to get the latest value of something from your store, that potentially might not emit:
You might try to do this (and you should as much as possible keep things 'inside pipes'):
serviceCallResponse$.pipe(withLatestFrom(store$.select(x => x.customer)))
.subscribe(([ serviceCallResponse, customer] => {
// we have serviceCallResponse and customer
});
The problem with this is that it will block until the secondary observable emits a value, which potentially could be never.
I found myself recently needing to evaluate an observable only if a value was immediately available, and more importantly I needed to be able to detect if it wasn't. I ended up doing this:
serviceCallResponse$.pipe()
.subscribe(serviceCallResponse => {
// immediately try to subscribe to get the 'available' value
// note: immediately unsubscribe afterward to 'cancel' if needed
let customer = undefined;
// whatever the secondary observable is
const secondary$ = store$.select(x => x.customer);
// subscribe to it, and assign to closure scope
sub = secondary$.pipe(take(1)).subscribe(_customer => customer = _customer);
sub.unsubscribe();
// if there's a delay or customer isn't available the value won't have been set before we get here
if (customer === undefined)
{
// handle, or ignore as needed
return throwError('Customer was not immediately available');
}
});
Note that for all of the above I'm using subscribe to get the value (as #Ben discusses). Not using a .value property, even if I had a BehaviorSubject.
Although it may sound overkill, this is just another "possible" solution to keep Observable type and reduce boilerplate...
You could always create an extension getter to get the current value of an Observable.
To do this you would need to extend the Observable<T> interface in a global.d.ts typings declaration file. Then implement the extension getter in a observable.extension.ts file and finally include both typings and extension file to your application.
You can refer to this StackOverflow Answer to know how to include the extensions into your Angular application.
// global.d.ts
declare module 'rxjs' {
interface Observable<T> {
/**
* _Extension Method_ - Returns current value of an Observable.
* Value is retrieved using _first()_ operator to avoid the need to unsubscribe.
*/
value: Observable<T>;
}
}
// observable.extension.ts
Object.defineProperty(Observable.prototype, 'value', {
get <T>(this: Observable<T>): Observable<T> {
return this.pipe(
filter(value => value !== null && value !== undefined),
first());
},
});
// using the extension getter example
this.myObservable$.value
.subscribe(value => {
// whatever code you need...
});
There are two ways you can achieve this.
BehaviorSubject has a method getValue() which you can get the value in a specific point of time.
You can subscribe directly with the BehaviorSubject and you may pass the subscribed value to a class member, field or property.
I wouldn't recommend both approaches.
In the first approach, it's a convenient method you can get the value anytime, you may refer to this as the current snapshot at that point of time. Problem with this is you can introduce race conditions in your code, you may invoke this method in many different places and in different timing which is hard to debug.
The second approach is what most developers employ when they want a raw value upon subscription, you can track the subscription and when do you exactly unsubscribe to avoid further memory leak, you may use this if you're really desperate to bind it to a variable and there's no other ways to interface it.
I would recommend, looking again at your use cases, where do you use it? For example you want to determine if the user is logged in or not when you call any API, you can combine it other observables:
const data$ = apiRequestCall$().pipe(
// Latest snapshot from BehaviorSubject.
withLatestFrom(isLoggedIn),
// Allow call only if logged in.
filter(([request, loggedIn]) => loggedIn)
// Do something else..
);
With this, you may use it directly to the UI by piping data$ | async in case of angular.
A subscription can be created, then after taking the first emitted item, destroyed. In the example below, pipe() is a function that uses an Observable as its input and returns another Observable as its output, while not modifying the first observable.
Sample created with Angular 8.1.0 packages "rxjs": "6.5.3", "rxjs-observable": "0.0.7"
ngOnInit() {
...
// If loading with previously saved value
if (this.controlValue) {
// Take says once you have 1, then close the subscription
this.selectList.pipe(take(1)).subscribe(x => {
let opt = x.find(y => y.value === this.controlValue);
this.updateValue(opt);
});
}
}
You could store the last emitted value separately from the Observable. Then read it when needed.
let lastValue: number;
const subscription = new Service().start();
subscription
.subscribe((data) => {
lastValue = data;
}
);
The best way to do this is using Behaviur Subject, here is an example:
var sub = new rxjs.BehaviorSubject([0, 1])
sub.next([2, 3])
setTimeout(() => {sub.next([4, 5])}, 1500)
sub.subscribe(a => console.log(a)) //2, 3 (current value) -> wait 2 sec -> 4, 5
Another approach, If you want / can to use async await (has to be inside of an async functions) you can do this with modern Rxjs:
async myFunction () {
const currentValue = await firstValueFrom(
of(0).pipe(
withLatestFrom(this.yourObservable$),
map((tuple) => tuple[1]),
take(1)
)
);
// do stuff with current value
}
This will emit a value "Right away" because of withLatestFrom, and then will resolve the promise.
The goal is to iterate through a collection of IDs, making an HTTP call for each ID. For each ID, I'm using a service with a get() method that returns an Observable. Each time the get() method is called, I'm subscribing to the returning Observable and trying to push the result into an array, which will eventually get passed on to a different method for a new operation.
Relevant service method:
public get(departmentId: number): Observable<IDepartmentModel> {
return super.get<IDepartmentModel>(departmentId);
}
note: the super class is leveraging Angular Http, which is well tested and confirmed to be working correctly. The problem with the logic isn't here...
Relevant component methods:
note the departmentService.get() call that's being called several times within the forEach.
setInitialDepartmentsAssignedGridData(): void {
this.settingsForDropdownSelectedCompanyId = this.userForm.get('defaultCompany').get('defaultCompanyId').value;
let departments: IDepartmentModel[] = [];
this.userService.user.getValue() //confirmed: valid user is being pulled back from the userService (logic is fine here..)
.userCompanies.find(comp => comp.companyId === this.settingsForDropdownSelectedCompanyId) // getting a valid match here (logic is fine here..)
.departmentIds.forEach(deptId => this.departmentService.get(deptId).first().subscribe(dept => { // getting a valid department back here (logic is fine here...)
departments.push(dept); // HERE LIES THE PROBLEM
}));
this.setDepartmentsAssignedRowData(departments);
}
setDepartmentsAssignedRowData(departments: IDepartmentModel[]): void {
console.log('setDeptAssignedRowData called'); // confirmed: method is getting called...
console.log(departments); // confirmed: fully-composed collection of departments logged to the console...
departments.forEach(dept => {
console.log(dept);
}); // Y U NO WORK!?
departments.map((department) => {
console.log(department); // Y U NO WORK?
this.departmentAssignedRowData.push({
departmentId: department.id,
departmentName: department.name
});
});
this.departmentAssignedGridOptions.api.setRowData(this.departmentAssignedRowData);
}
The problem is, although what's getting logged to the console is a fully-composed department-objects array, it's not TRULY "there"; what's getting passed to setDepartmentsAssignedRowData is an empty array.
I'm sure what's happening is that the async operations are not complete before the departments array gets passed to the second method. Some of what I've read online says to use forkJoin, but I can't see how that will look in this context. I've also read concatMap may work, but again, in this context, I'm not sure how to make that work...
In this context, how do I leverage RxJS to make sure the intended, fully-composed departments array is truly ready to be passed?
thanks for any insight you can provide. help is much appreciated!
You are correct, you need forkJoin
let observableArray = this.userService.user.getValue()
.userCompanies.find(comp => comp.companyId === this.settingsForDropdownSelectedCompanyId)
.departmentIds.map(deptId => this.departmentService.get(deptId)) // map is building out an array of observables
This will be an array of http request observables that you want to make in parallel. Now you can pass this array to forkJoin.
Observable.forkJoin(...observableArray)
The return of forkJoin will be an array of results from observableArray. forkJoin will not emit to the next operator in the sequence until all of the observables in observableArray have completed (so when all of the http requests have finished)
So altogether the code will be
let observableArray = this.userService.user.getValue()
.userCompanies.find(comp => comp.companyId === this.settingsForDropdownSelectedCompanyId)
.departmentIds.map(deptId => this.departmentService.get(deptId));
Observable.forkJoin(...observableArray).subscribe(res => {
// res = [resId0, resId1, resId2, ..., resIdX];
});
You mentioned passing the result to another operator. If that operator is another http request where you pass an array of data (from forkJoin), then you can use the flatMap operator.
Observable.forkJoin(...observableArray)
.flatMap(res => {
return this.otherApi(res);
})
.subscribe(res => {
// res is the result of the otherApi call
});
flatMap will chain your api requests together. So altogether what is happening is
run array of observables in parallel
once complete, run second api (otherApi)
In my application i am using chain of of web request call for fetching data from the net. Ie from the result of one request i will send other request and so on. But when i am disposing the web request, only the parent request is disposing. The two other request are still running. How i can cancel all these request in Rx
For your subscription to terminate everything, you either cannot break the monad or you need to make sure that you work into the IDisposable model.
To keep the monad (ie. stick with IObservables):
var subscription = initialRequest.GetObservableResponse()
.SelectMany(initialResponse =>
{
// Feel free to use ForkJoin or Zip (intead of Merge) to
// end up with a single value
return secondRequest.GetObservableResponse()
.Merge(thirdRequest.GetObservableResponse());
})
.Subscribe(subsequentResponses => { });
To make use of the IDisposable model:
var subscription = initialRequest.GetObservableResponse()
.SelectMany(initialResponse =>
{
return Observable.CreateWithDisposable(observer =>
{
var secondSubscription = new SerialDisposable();
var thirdSubscription = new SerialDisposable();
secondSubscription.Disposable = secondRequest.GetObservableResponse()
.Subscribe(secondResponse =>
{
// Be careful of race conditions here!
observer.OnNext(value);
observer.OnComplete();
});
thirdSubscription.Disposable = thirdRequest.GetObservableResponse()
.Subscribe(thirdResponse =>
{
// Be careful of race conditions here!
});
return new CompositeDisposable(secondSubscription, thirdSubscription);
});
})
.Subscribe(subsequentResponses => { });
One approah is by using TakeUntil extnsion method as described here. In your case, the event that takes this method as parameter could be some event thrown by the parent request.
If you could show us some code we can face the problem more specifically.
regards,