SwiftConcurrency Library

An open source library with utilities and extensions to support Swift async/await concurrency.

Developed as re-usable components for various projects at XII's iOS, macOS, and watchOS applications.

1. In Xcode, select File > Swift Packages > Add Package Dependency.
2. Follow the prompts using the URL for this repository
3. Select the SwiftConcurrency library to add to your project

See the LICENSE file.

extension Task where Failure == Error {
  init(
    delaySeconds: Double,
    priority: TaskPriority? = nil,
    operation: @escaping @Sendable () async throws -> Success
  )
}

extension Task where Failure == Never {
  init(
    delaySeconds: Double,
    priority: TaskPriority? = nil,
    operation: @escaping @Sendable () async -> Success
  )
}

Creates a task with an optional priority that executes the given operation after an initial delay (in seconds).

This is a shorthand for starting a task and then calling Task.sleep with a number of nanoseconds to wait.

extension Task {
  static func sleep(seconds duration: Double) async throws
}

Suspends the current task for at least the given duration in seconds, unless the task is cancelled. If the task is cancelled, throws CancellationError without waiting for the duration. This function does not block the underlying thread.

extension Task {
  static func poll(
    intervalSeconds: Double,
    timeoutSeconds: Double,
    action: @escaping () async throws -> Bool
  ) async throws
}

Performs a polling action at a specific interval, timing out if the action does not return true within the timeout interval.

Returns true if the polling action succeeds and false if the task has timed out or has been cancelled.

struct ParallelProcessing {
  static let DEFAULT_MAX_PARALLEL_TASKS = 4

  static func processItemsInParallel<Item, Result>(
    items: [Item],
    maxParallelTasks: Int = DEFAULT_MAX_PARALLEL_TASKS,
    processInRandomOrder: Bool = true,
    processItem: @escaping (Item) async -> Result?
  ) async -> [Result]
}

Applies a processing task (processItem) to each item in parallel. Limits the maximum parallelism of the processing to maxParallelTasks to limit the number of threads created.

All items provided are added to a work queue that is then drained by multiple parallel Tasks.