Asyncio in Python - Full Tutorial

Asynchronous programming allows our code to be more efficient by doing multiple things at once without any unnecessary waiting. Asyncio is your choice for running multiple tasks concurrently such as network requests or reading files, without using much CPU power.
If you want to land a developer job click here and view my free introduction course with Course Careers for becoming a developer in 2024: techwithtim.net/dev
Skool community for free resources: www.skool.com/software-develo...
Discord server: / discord
🎞 Video Resources 🎞
Python Async IO Docs: docs.python.org/3/library/asy...
⏳ Timestamps ⏳
00:00 | What is Asyncio?
00:48 | When to use Async IO?
02:04 | #1 - The Event Loop
02:47 | #2 - Coroutines
09:52 | #3 - Tasks
16:50 | #4 - Futures
18:23 | #5 - Synchronization
Hashtags
#techwithtim
#asynchronousprogramming
#python
#pythonprogramming

[@TechWithTim]

If you want to become a developer in 2024 you can watch my free introduction course here: techwithtim.net/dev - yes, there is a paid course as well but we've provided a free 90 minute course you can go through that explains exactly what you need to do to land a developer job.

[@benjaminbaxter4103]

Tim, is your course good for software development or engineering?

[@TechWithTim]

@@benjaminbaxter4103 of course it is! That's what it is designed for

[@noticedcloud]

Sorry, but how can I do in Python that while an app is running, printing certain things like a server that prints every time someone enters, it can accept a command in the console without the print overlapping with the command?

[@garrettsmith315]

Great work on the visual and graphics here, you have stepped your game up from you early Python/KZbin days. Glad to be a part of the journey.

[@TechWithTim]

Thank you!

[@benjaminbaxter4103]

Tim, is your course good for software dev or engineering?

[@MrTaddelie]

Honestly the short animation what the difference between a thread, process and async task is was the most helpfull thing i've seen since i started hobby programming! Thanks, gonna try read some sensors async now^^

[@night23412]

this is an outstanding video, it might just be the single best explanation of asyncio on the internet! well done tim

[@soumodas782]

Thanks for the upload Tim. This was a concise and well rounded explanation of asynio usage in Python. Exactly what I was looking for

[@user-hy1oz6lk3k]

Hi Tim, thanks for this tutorial! Clear and on point with good examples!

[@surfzion]

Great tutorial ! I've been struggling with this subject for some time, but this videos has made things much more clear to me. Thank you sir.

[@joshuaarinaitwe8351]

Thanks tim. Glad to have followed you from way back. Great video

[@haimroizman6440]

Really I've heard couple of asyncio tutorials but I have to say, this is the best. Thanks a lot

[@user-ts9ks8in2n]

Clear concepts. Thanks Tim

[@nicolaspillot5789]

Great tutorial, simple, concise, clear. I like it.

[@kerrykreiter445]

Excellent work! Thank you!!

[@ygr4943]

Beautiful explanation. Thank youu ❤❤❤

[@idopshik]

Subscribed fast. Top quality.

[@oj43085]

Nice explanation! I use semaphores for co-current web request to a client that has some throttling. I like how you explained it and the others which i didn't know about.

[@petegranneman1623]

Great tutorial! Great depth while still being very accessible. Keep up the great work! Perhaps worth a mention for our Spyder brethren and sistren... the iPython console runs its own loop which can interfere with the asyncio code. If the iPython console throws errors when running asyncio code, (but the code runs just fine in terminal), the nest_asyncio package can be imported to improve compatibility in the iPython console.

[@rajeshseemakurthi1595]

Awesome Video and helpfulll for middle class learners..Great Contribution bro.

[@GRSG_tech]

Thank you Tim for the wonderful presentation of Asyncio. I'd like for a full python course which for an intermediate level to advanced like diving into machine learning, computer vision(OpenCV or YOLO) and other data science Anyways Great content, appreciate it

[@Dexter4o4]

🎯 Key Takeaways for quick navigation: 00:00 *🏞️ Asynchronous programming allows tasks to start at different points and proceed independently, improving efficiency by handling multiple tasks simultaneously.* 00:59 *🛠️ Async IO is ideal for tasks with waiting times like network requests or file reading, excelling in handling many tasks concurrently with low CPU usage.* 01:56 *🚀 Async IO for managing many waiting tasks efficiently, threads for parallel tasks with minimal CPU use, and processes for CPU-intensive tasks maximize performance.* 02:25 *🔄 The event loop in Python's Async IO manages and distributes tasks efficiently, ensuring smooth and responsive program flow.* 03:53 *⚙️ Starting Async IO involves importing the module, defining asynchronous functions, and using `asyncio.run()` with a coroutine function to initiate the event loop.* 05:17 *🔍 Calling an async function returns a coroutine object that must be awaited to execute, distinguishing async functions from normal ones.* 06:45 *🕰️ The `await` keyword in asynchronous functions allows waiting for routines to execute and fetch results, crucial for managing IO-bound operations.* 09:11 *🔄 Tasks enable running multiple coroutines concurrently, optimizing efficiency by switching between tasks when one is idle or waiting.* 11:37 *🌐 Using `asyncio.create_task()` optimizes code execution by scheduling coroutines to run concurrently without waiting for each to finish.* 14:04 *🚀 The `asyncio.gather()` function concurrently runs multiple coroutines, simplifying code for concurrent execution and result collection.* 16:00 *🛡️ Utilizing `asyncio.TaskGroup` with an asynchronous context manager organizes and executes multiple tasks with built-in error handling and synchronization.* 17:57 *🔄 Futures represent a promise of a future result, useful in lower-level libraries for asynchronous programming.* 19:23 *🔒 Synchronization primitives like locks help manage shared resources by ensuring only one coroutine accesses them at a time, preventing issues like race conditions.* 20:50 *🛠️ Using a lock in asyncio synchronizes code routines, ensuring critical sections are accessed by one routine at a time, preventing concurrency issues.* 21:48 *🔒 Semaphores in asyncio limit simultaneous access to shared resources, controlling the number of routines that can access them concurrently.* 23:46 *🚥 Events act as boolean flags in asyncio, allowing code to wait until a specific condition is met before continuing execution.* Made with HARPA AI

[@yaswanthanumolu765]

clear info 👍

[@MrDerterminator1992]

@TechWithTim: Awesome tutorial. I used asyncio already in a couple of projects and never really understood the concept of coroutines. What I also came across is Rate Limiting. So it would be awesome if u could add Semaphores in a future video to have a usecase for async API Requests and Concurrency Limiting with Semaphores.

[@pgrvloik]

it's difficult to me to comprehend that an async function is actually synchronous, since "awaited'

[@hughesadam87]

Agreed. I think of it like async keyword makes a function potentially asynchronous but we can still use the function synchronously if need be😊

[@aashayamballi]

Thanks

[@danielcrigan1301]

Decent course guide! I am working now with aiogram 3.0 framework for buildigt tg chatbots. Watching your video made asyncio easier to understand!

[@tarzanat3409]

Hey tim, have u ever did something with Microsoft Graph?

[@satyabharadwaj7779]

Hey Tim, just need a clarification on the usage of threads. Do they really work in 'parallel', especially in python? Cause I often here that GIL prevents threads from running threads in parallel, but they are efficient given the fact that a thread waiting for I/O tasks can be suspended and allow other thread to run.

[@adarmawan117]

Hi Tim. Please, create video that explaind how to connect Firebase with Python. And how to use general function. Thankss

[@chrisw1462]

@15:20 I don't disagree with how this was done, but for a demonstration I think I would have made sleep_time a list, and used a range() for 'id' in the for loop. And let 'id' be zero for the first one. A little clearer what's happening with the tasks, imo.

[@artemiasalina1860]

How well does asyncio play with Qt's (PySide) event system? Do they interfere with each other? I'm writing a chat client that uses PySide6 for its UI in order to learn python better and am having a hard time deciding how to handle network I/O with asynchronous socket reads and Qt's signal/slot event system. I know that Qt has it's own networking classes like QTCPSocket but I'm specifically trying to avoid using them in order to learn Python's own native async mechanism. Any advise would be appreciated.

[@ffzz007]

I just opened youtube to search for this topic and there you go😂

[@pjotrekse]

What is the theme called that you use in this tutorial?

[@weedfreer]

So, if I needed to call other functions from my async function in order to complete my async function, would the other functions also need to be created as async functions or can they just be regular functions?

[@TechWithTim]

they can be normal functions unless you want to run them concurrently

[@weedfreer]

@@TechWithTim if many instances of the same concurrent function were attempting to run their own instance of the same standard 'tertiary' function, would that cause a problem, or, would the tertiary function also need to be concurrent?

[@mihhailkapustin7886]

Coming from JS background asyncio has always been super confusing to me. Could you perhaps make a video explaining the difference between JavaScript promises model and Python’s asyncio?

[@AlexJaeger716]

What if i want the code to dynamically create tasks to run concurrently? Is this possible? It seems like the tasks need to be hard coded from everything i’ve seen

[@noticedcloud]

Sorry, but how can I do in Python that while an app is running, printing certain things like a server that prints every time someone enters, it can accept a command in the console without the print overlapping with the command?

[@ti6i]

Quick question: Lets assume I make a class with nothing but async functions, if I call the async functions inside another file, lets say I call hello_world() and then I call async_function() and then hello_world() again, would the program freeze while waiting for the hello_world function to execute? Will it execute the hello_world function the 2nd time as the async_function is running?

[@TechWithTim]

well it depends on how you call it, you can't call a coroutine without awaiting it or running it some kind of task or future. So in this case if you awaited it ya it would freeze, but if you ran your hello_world() functions and your async_function() concurrently with tasks then it would not wait, it would go and execute the other hello _world function as soon as it encounters any blocking code

[@chrisw1462]

@16:02 'Gives .. access to' is a bit misleading, as 'tg' doesn't exist before this line. The whole interface is a bit clunky, IMO, so stuff like that just makes it harder to understand.

[@WaldoTheWombat]

Not sure I understand the difference between Asyncio and multithreading. Aren't they both used for IO bound situations?

[@soumodas782]

Yes but Async IO gives you direct control of the event loop to control concurrency. In multi threading , the GIL (global Interpreter Lock) only emulates concurrency with kernel handling much of the scheduling side of the control flows

[@WaldoTheWombat]

@@soumodas782 thanks!

[@eugenmalatov5470]

I don't understand 😓

[@dekisugihidetoshi9397]

Can anyone explain to me the loop For i, sleep_time in enumerate([2,1,3], start=1): ? Thank you in advance

[@seelam.hemasasank]

Enumerate function return the two values (index, value in the list) by default start valve == 0 if you give the start value the index starts from there. Every time it creates task and run the async func and sleeps the fun for given secs in list.

[@dennisdegraphicsdynamo_999]

Tim I thought your course are free

[@TechWithTim]

Most of them are but my premium one is not. It would be impossible for me to help everyone the way I do in that course if it was free

[@dennisdegraphicsdynamo_999]

@@TechWithTim okay I get you

[@dennisdegraphicsdynamo_999]

@@TechWithTim But we need more complex but beginner friendly projects on python and also Ai so that we can also learn.....you are really helping us

[@ZelabsAdministration]

Sorry, but how can I do in Python that while an app is running, printing certain things like a server that prints every time someone enters, it can accept a command in the console without the print overlapping with the command?