Update :
In angular 7, they are the same as 6
In angular 6
the complete answer found in live example
/** POST: add a new hero to the database */
addHero (hero: Hero): Observable<Hero> {
return this.http.post<Hero>(this.heroesUrl, hero, httpOptions)
.pipe(
catchError(this.handleError('addHero', hero))
);
}
/** GET heroes from the server */
getHeroes (): Observable<Hero[]> {
return this.http.get<Hero[]>(this.heroesUrl)
.pipe(
catchError(this.handleError('getHeroes', []))
);
}
it's because of pipeable/lettable operators
which now angular is able to use tree-shakable
and remove unused imports and optimize the app
some rxjs functions are changed
do -> tap
catch -> catchError
switch -> switchAll
finally -> finalize
more in MIGRATION
and Import paths
For JavaScript developers, the general rule is as follows:
rxjs: Creation methods, types, schedulers and utilities
import { Observable, Subject, asapScheduler, pipe, of, from, interval, merge, fromEvent } from 'rxjs';
rxjs/operators: All pipeable operators:
import { map, filter, scan } from 'rxjs/operators';
rxjs/webSocket: The web socket subject implementation
import { webSocket } from 'rxjs/webSocket';
rxjs/ajax: The Rx ajax implementation
import { ajax } from 'rxjs/ajax';
rxjs/testing: The testing utilities
import { TestScheduler } from 'rxjs/testing';
and for backward compatability you can use rxjs-compat