Don't go sleeping on taskGroups: they're probably my favourite mechanic in all of Swift Concurrency. Infinite parallelism; basically quite easy.

We shipped structured data collection for voice agents. Use data collection mode in Agent Builder, or Tasks and TaskGroups in our Agents SDKs. Every session ends with a clean JSON payload for your CRM, form, or database. Great for lead qualification, patient intake, and surveys.

Is #LiteStar faster than #FastAPI? Benchmarks say yes, but look at Dependency Injection. 📉 Adding just 2 flat dependencies significantly degrades the throughput. I dug into the source to find why. Thread: TaskGroups, Kahn's algorithm & performance trade-offs. 🧵

While taskGroups is good for parallelisation, less typing and simpler code, I still do prefer subclassing Operation and have an OperationQueue execute it for me. As taskGroups is still bound to the thread the current Actor is running on. OperationQueue will execute those multiple Operations Concurrently. Making the users device spent less total time on getting those operations done.

Swift interview tips: Advanced parallelisation Sometimes, we need to make a bunch of similar network requests. Swift Concurrency has an app for that: taskGroups allow us to wait for any number of functions, provided they return the same type.

هي الswift concurrency ايه غير async و await وشويه ايرور هاندلينج في حاجه تانيه ؟ ← Async/await Tasks Tasks management/Cancellation AsyncStreams TaskGroups Actors Actors reentrancy Sendables Continuations Async let Supporting pre concu... onvo.me/AmrHossam96/msg/9178…

Advanced Swift Concurrency - Limit taskGroup concurrency TaskGroups are great for running lots of work in parallel. While tasks are cheap to spin up, you might not want to lock up your entire system processing hundreds of expensive image processing or ML requests. You can use this trick to limit the number of requests done simultaneously, reducing CPU load, peak memory, and freeing other threads to perform work (if you have enough cores). While this example is a little verbose, the end result works just like the classic OperationQueue’s maxConcurrentOperationCount. Gone but not forgotten 🙏. This example comes from WWDC 2024 - Discover Swift enhancements in the Vision framework. (Sorry, dark mode bros, but I am super lazy and screenshotted it directly)

The importance of Dispatch Queue in iOS development What if @vinayjn7 started writing Swift iOS articles in my style? maybe I can be his ghostwriter wdyt? ========================= let me tell you about dispatch queues the backbone of iOS concurrency they're like task managers FIFO queues, first in first out keeping your app running smooth two ways they work: serial - one task at a time concurrent - multiple tasks at once the main queue that's where UI lives don't mess with it keep it responsive global queues system gives you six different priorities from urgent to "whenever" custom queues? make your own serial or concurrent your choice quality of service that's how urgent it is userInteractive - needs to happen now userInitiated - user's waiting default - standard stuff utility - longer tasks background - no rush unspecified - legacy things ways to use them: sync - wait till it's done async - fire and forget asyncAfter - do it later asyncAndWait - async but wait concurrentPerform - parallel loops patterns to remember: UI updates? main queue heavy math? background network calls? background image processing? utility database stuff? background animations? main queue best practices, listen up: never sync on main pick the right priority avoid nested sync calls keep critical sections small use barriers when needed groups for coordination watch out for: too many queues slow you down high priority means high power use batch small tasks together monitor your queue depth profile with instruments common mistakes? blocking the main thread deadlocks from bad sync race conditions memory leaks in closures priority inversions too many serial queues modern swift gives you more: async/await for cleaner code actors for thread safety taskgroups for structure @ MainActor for UI safety AsyncSequence for iterations Sendable for thread-safe types advanced stuff: DispatchWorkItem wraps your code DispatchGroup coordinates tasks semaphores control access DispatchSource handles events target queues inherit priority context stores queue data master these concepts your apps will fly smooth and responsive that's what we want dispatch queues they're not just code they're the heartbeat of your iOS app learn them well use them right your users will thank you even if they don't know why

反馈与交流 | 肘子的 Swift 周报 #070 - Apple Vision Pro 一年使用体会 - SwiftUI ControlGroup - iOS 的声音引擎 - 用 Swift Charts 绘制地图 - 掌握 TaskGroups - 参数化测试 fatbobman.com/zh/weekly/issu…

Feedback and Communication | Fatbobman's Swift Weekly #070 - Apple Vision Pro 1 Year Later - ControlGroup - Sound Engineering on iOS - Drawing Maps with Swift Charts - TaskGroups - Parameterized Tests weekly.fatbobman.com/p/fatbo…

Swift interview tips: Advanced parallelisation Sometimes, we need to make a bunch of similar network requests. Swift Concurrency has an app for that: taskGroups allow us to wait for any number of functions, provided they return the same type.

I've written a short article on using TaskGroups to speedup routes, pgjones.dev/blog/how-to-task…. This includes a Quart-DB example where each task needs its own connection for the speedup to work.

Do you know how to use TaskGroups in Swift? I just applied some minor updates to my blog post that explains exactly how and when you should be using TaskGroups in your project 👍 donnywals.com/swift-concurre…

🚀 The ICPDR Expert and Task Groups Season is here! 🌊 1️⃣ First up: #TaskGroup Placeholder Meeting 2️⃣ Next in line: #ExpertGroup Placeholder Meeting Stay in the loop at: icpdr.org 💡 #ICPDR #DanubeRiver #TaskGroups #ExpertGroups

In a nutshell, #Airflow #TaskGroups organize tasks into groups within your #DAGs. ⚙️📁 📢 Airflow 2.7 Update: Now, manage task groups like individual tasks. Clear them or mark them as success/failure. 👀 Read our guide to learn more. bit.ly/3qKwXXN #DataEngineering

I refactored a 10 year old method today with TaskGroups thanks to this article ⁦from @twannl⁩. Here is why and how I did it. avanderlee.com/concurrency/t… 1/n

Unlock the secret to effortlessly organizing complex #Airflow DAG structures. Join our webinar tomorrow to delve into the world of #taskgroups. Learn how to organize your #DAGs, manage task state & modularize common patterns in your pipelines. Register👉 bit.ly/44nHwOS

This is why we have TaskGroups now.

The good news is that TaskGroups will save you. Assuming all your tasks are in a TG somewhere. docs.python.org/3/library/as…

I would have given a #fosdem presentation today in the Python devroom about asyncio's new features: TaskGroups and ExceptionGroups. Unfortunately I got a viral throat infection and won't be able to present. :'( Hopefully, I can present it another time. Have a good Fosdem!