Process promise-based tasks in series and parallel, controlling concurrency and throttling.
npm install porch
tasks
(Array
): An array oftasks
, where each task is a function that expects no arguments and will return aPromise
.concurrency
(Number
): Default:1
. Maximum number of tasks to run concurrently (in parallel).interval
(Number
): Default:0
. Interval between each batch of concurrent tasks.failFast
(Boolean
): Default:true
. Whether to bail out when one of the promises fails. If set tofalse
errors will be included in the results passed tothen()
instead of being passed independently via thecatch()
method.
A Promise
that will resolve to an array with the results for each task.
Results will be in the same order as in the input tasks array.
Process each task after the other, sequentially. Each task will wait for the
previous one to complete. Concurrency set to 1
(one task at a time).
import porch from 'porch';
const tasks = users.map(user => () => auth.deleteUser(user.localId);
porch(tasks)
.then(console.log)
.catch(console.error);
Process tasks in batches based on a given concurrency. In this example tasks will be processed in batches of 5 tasks each. Each batch waits for the previous one to complete and then performs its tasks in parallel.
porch(tasks, 5)
.then(console.log)
.catch(console.error);
Same as example above but adding a 1000ms delay between batches.
porch(tasks, 5, 1000)
.then(console.log)
.catch(console.error);
Same as above, but in this case if a promise fails, processing will continue
instead of stopping the whole thing. When failFast
is set to false
, errors
will appear as the value/result for the relevant element in the results array
(failed tasks/promises won't end up in the catch()
method).
porch(tasks, 5, 1000, false)
.then(console.log)
Same as porch()
.
A readable stream (stream.Readable
) instead of a Promise
. Each result will
be emitted as a data event and the stream will operate in objectMode
.
import { createStrean } from 'porch';
createStream(tasks, 5, 1000, false)
.on('error', err => console.error('error', err))
.on('data', data => console.log('data', data))
.on('end', _ => console.log('ended!'));
// This example assumes that tasks will resolve to string values so that the
// resulting stream can be directly piped to stdout.
createStream(tasks, 5, 1000, false).pipe(process.stdout);