bindCallback.ts 6.64 KB
/* @prettier */
import { SchedulerLike } from '../types';
import { Observable } from '../Observable';
import { bindCallbackInternals } from './bindCallbackInternals';

export function bindCallback(
  callbackFunc: (...args: any[]) => void,
  resultSelector: (...args: any[]) => any,
  scheduler?: SchedulerLike
): (...args: any[]) => Observable<any>;

// args is the arguments array and we push the callback on the rest tuple since the rest parameter must be last (only item) in a parameter list
export function bindCallback<A extends readonly unknown[], R extends readonly unknown[]>(
  callbackFunc: (...args: [...A, (...res: R) => void]) => void,
  schedulerLike?: SchedulerLike
): (...arg: A) => Observable<R extends [] ? void : R extends [any] ? R[0] : R>;

/**
 * Converts a callback API to a function that returns an Observable.
 *
 * <span class="informal">Give it a function `f` of type `f(x, callback)` and
 * it will return a function `g` that when called as `g(x)` will output an
 * Observable.</span>
 *
 * `bindCallback` is not an operator because its input and output are not
 * Observables. The input is a function `func` with some parameters. The
 * last parameter must be a callback function that `func` calls when it is
 * done.
 *
 * The output of `bindCallback` is a function that takes the same parameters
 * as `func`, except the last one (the callback). When the output function
 * is called with arguments it will return an Observable. If function `func`
 * calls its callback with one argument, the Observable will emit that value.
 * If on the other hand the callback is called with multiple values the resulting
 * Observable will emit an array with said values as arguments.
 *
 * It is **very important** to remember that input function `func` is not called
 * when the output function is, but rather when the Observable returned by the output
 * function is subscribed. This means if `func` makes an AJAX request, that request
 * will be made every time someone subscribes to the resulting Observable, but not before.
 *
 * The last optional parameter - `scheduler` - can be used to control when the call
 * to `func` happens after someone subscribes to Observable, as well as when results
 * passed to callback will be emitted. By default, the subscription to an Observable calls `func`
 * synchronously, but using {@link asyncScheduler} as the last parameter will defer the call to `func`,
 * just like wrapping the call in `setTimeout` with a timeout of `0` would. If you were to use the async Scheduler
 * and call `subscribe` on the output Observable, all function calls that are currently executing
 * will end before `func` is invoked.
 *
 * By default, results passed to the callback are emitted immediately after `func` invokes the callback.
 * In particular, if the callback is called synchronously, then the subscription of the resulting Observable
 * will call the `next` function synchronously as well.  If you want to defer that call,
 * you may use {@link asyncScheduler} just as before.  This means that by using `Scheduler.async` you can
 * ensure that `func` always calls its callback asynchronously, thus avoiding terrifying Zalgo.
 *
 * Note that the Observable created by the output function will always emit a single value
 * and then complete immediately. If `func` calls the callback multiple times, values from subsequent
 * calls will not appear in the stream. If you need to listen for multiple calls,
 *  you probably want to use {@link fromEvent} or {@link fromEventPattern} instead.
 *
 * If `func` depends on some context (`this` property) and is not already bound, the context of `func`
 * will be the context that the output function has at call time. In particular, if `func`
 * is called as a method of some object and if `func` is not already bound, in order to preserve the context
 * it is recommended that the context of the output function is set to that object as well.
 *
 * If the input function calls its callback in the "node style" (i.e. first argument to callback is
 * optional error parameter signaling whether the call failed or not), {@link bindNodeCallback}
 * provides convenient error handling and probably is a better choice.
 * `bindCallback` will treat such functions the same as any other and error parameters
 * (whether passed or not) will always be interpreted as regular callback argument.
 *
 * ## Examples
 *
 * ### Convert jQuery's getJSON to an Observable API
 * ```ts
 * import { bindCallback } from 'rxjs';
 * import * as jQuery from 'jquery';
 *
 * // Suppose we have jQuery.getJSON('/my/url', callback)
 * const getJSONAsObservable = bindCallback(jQuery.getJSON);
 * const result = getJSONAsObservable('/my/url');
 * result.subscribe(x => console.log(x), e => console.error(e));
 * ```
 *
 * ### Receive an array of arguments passed to a callback
 * ```ts
 * import { bindCallback } from 'rxjs';
 *
 * const someFunction = (n, s, cb) => {
 *   cb(n, s, { someProperty: 'someValue' });
 * };
 *
 * const boundSomeFunction = bindCallback(someFunction);
 * boundSomeFunction(5, 'some string').subscribe((values) => {
 *   console.log(values); // [5, 'some string', {someProperty: 'someValue'}]
 * });
 * ```
 *
 * ### Compare behaviour with and without async Scheduler
 * ```ts
 * import { bindCallback, asyncScheduler } from 'rxjs';
 *
 * function iCallMyCallbackSynchronously(cb) {
 *   cb();
 * }
 *
 * const boundSyncFn = bindCallback(iCallMyCallbackSynchronously);
 * const boundAsyncFn = bindCallback(iCallMyCallbackSynchronously, null, asyncScheduler);
 *
 * boundSyncFn().subscribe(() => console.log('I was sync!'));
 * boundAsyncFn().subscribe(() => console.log('I was async!'));
 * console.log('This happened...');
 *
 * // Logs:
 * // I was sync!
 * // This happened...
 * // I was async!
 * ```
 *
 * ### Use bindCallback on an object method
 * ```ts
 * import { bindCallback } from 'rxjs';
 *
 * const boundMethod = bindCallback(someObject.methodWithCallback);
 * boundMethod
 *   .call(someObject) // make sure methodWithCallback has access to someObject
 *   .subscribe(subscriber);
 * ```
 *
 * @see {@link bindNodeCallback}
 * @see {@link from}
 *
 * @param {function} func A function with a callback as the last parameter.
 * @param {SchedulerLike} [scheduler] The scheduler on which to schedule the
 * callbacks.
 * @return {function(...params: *): Observable} A function which returns the
 * Observable that delivers the same values the callback would deliver.
 */
export function bindCallback(
  callbackFunc: (...args: [...any[], (...res: any) => void]) => void,
  resultSelector?: ((...args: any[]) => any) | SchedulerLike,
  scheduler?: SchedulerLike
): (...args: any[]) => Observable<unknown> {
  return bindCallbackInternals(false, callbackFunc, resultSelector, scheduler);
}