Skip to content

Dart library for running asynchronous functions on time. For load balancing, rate limiting, lazy execution

License

Notifications You must be signed in to change notification settings

rtmigo/schedulers_dart

Repository files navigation

Dart library for running asynchronous functions on time. Useful for load balancing, rate limiting, lazy execution.

In the examples below, all the run calls are performed right after object creation. In fact all the schedulers can handle runs at random moments.

IntervalScheduler

Runs tasks asynchronously, maintaining a fixed time interval between starts.

final scheduler = IntervalScheduler(delay: Duration(seconds: 1));

scheduler.run(()=>download('pageA')); // starts download immediately
scheduler.run(()=>download('pageB')); // will start one second later
scheduler.run(()=>download('pageC')); // will start two seconds later

RateScheduler

Runs no more than N tasks in a certain period of time.

final scheduler = RateScheduler(3, Duration(seconds: 1)); // 3 per second

// the following tasks are executed immediately
scheduler.run(()=>download('pageA'));
scheduler.run(()=>download('pageB'));
scheduler.run(()=>download('pageC'));

// the following tasks are executed one second later
scheduler.run(()=>download('pageD'));
scheduler.run(()=>download('pageE'));
scheduler.run(()=>download('pageF'));
 
// the following tasks are executed two seconds later
scheduler.run(()=>download('pageG'));
scheduler.run(()=>download('pageH'));
scheduler.run(()=>download('pageI'));

ParallelScheduler

Limits the number of tasks running at the same time. This is somewhat similar to using a thread pool or process pool. But it just runs async functions.

// we will run a maximum of three tasks at the same time
final scheduler = ParallelScheduler(3); 

scheduler.run(()=>asyncDownload('pageA')); // executed immediately
scheduler.run(()=>asyncDownload('pageB')); // executed immediately
scheduler.run(()=>asyncDownload('pageC')); // executed immediately

scheduler.run(()=>asyncDownload('pageD'));
// task for pageD will execute when some of the previous tasks finish 

TimeScheduler

Runs tasks asynchronously at the specified time.

final scheduler = TimeScheduler();

// run the function on January 1st at 17:75
scheduler.run(() { ... }, DateTime(2020, 1, 1, 17, 45));

LazyScheduler

Runs only the last added task and only if no new tasks have been added during the time interval.

final scheduler = LazyScheduler(latency: Duration(seconds: 1));

scheduler.run(() => pushUpdate('1')); // maybe we will push 1
scheduler.run(() => pushUpdate('1+1')); // no we will push 1+1
scheduler.run(() => pushUpdate('1+1-1')); // no we will push 1+1-1
scheduler.run(() => pushUpdate('1')); // it's good we're so lazy
scheduler.run(() => pushUpdate('777')); // maybe we will push this

And one second later the scheduler runs pushUpdate('777'). Other tasks are ignored.

We can continue with the same scheduler:

scheduler.run(()=>pushUpdate('13')); // we pushed 777, now we maybe push 13
scheduler.run(()=>pushUpdate('10')); // no, we will not push 13...

Awaiting results

Each of the schedulers allows you to wait for the result of the function as a regular Future. You just have to await for the .result.

final a = await download('pageA');
final b = await scheduler.run(()=>download('pageB')).result;

These two calls to the download function behave in much the same way. The only difference is that the scheduler can delay the execution of the function for suitable times.

License

Copyright © 2021 Artsiom iG. Released under the MIT License.