Python Asynchronous Programming - AsyncIO & Async/Await
HTML-код
- Опубликовано: 28 сен 2024
- In today's video, I'll be talking to you about asynchronous programming in python. This Python Async tutorial will cover the 'async' and 'await' keyword, coroutines, futures and tasks, and some basic features from the asyncio module in Python. This video is for intermediate programmers, and it's recommended you have Python 3.7 or above.
💻 AlgoExpert is the coding interview prep platform that I used to ace my Microsoft and Shopify interviews. Check it out and get a discount on the platform using the code "techwithtim" algoexpert.io/...
📄 Documentation
AsyncIO: docs.python.or...
⭐️ Timestamps ⭐️
01:43 - Synchronous vs Asynchronous
04:30 - Coroutines
07:35 - Async Event-Loop
08:58 - Async/Await Keywords
12:12 - Tasks
17:00 - Async Example
◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️
💰 Courses & Merch 💰
💻 The Fundamentals of Programming w/ Python: tech-with-tim....
👕 Merchandise: teespring.com/...
🔗 Social Medias 🔗
📸 Instagram: / tech_with_tim
📱 Twitter: / techwithtimm
⭐ Discord: / discord
📝 LinkedIn: / tim-ruscica-82631b179
🌎 Website: techwithtim.net
📂 GitHub: github.com/tec...
🔊 Podcast: anchor.fm/tech...
🎬 My RUclips Gear 🎬
🎥 Main Camera (EOS Canon 90D): amzn.to/3cY23y9
🎥 Secondary Camera (Panasonic Lumix G7): amzn.to/3fl2iEV
📹 Main Lens (EFS 24mm f/2.8): amzn.to/2Yuol5r
🕹 Tripod: amzn.to/3hpSprv
🎤 Main Microphone (Rode NT1): amzn.to/2HrZxXc
🎤 Secondary Microphone (Synco Wireless Lapel System): amzn.to/3e07Swl
🎤 Third Microphone (Rode NTG4+): amzn.to/3oi0v8Z
☀️ Lights: amzn.to/2ApeiXr
⌨ Keyboard (Daskeyboard 4Q): amzn.to/2YpN5vm
🖱 Mouse (Logitech MX Master): amzn.to/2HsmRDN
📸 Webcam (Logitech 1080p Pro): amzn.to/2B2IXcQ
📢 Speaker (Beats Pill): amzn.to/2XYc5ef
🎧 Headphones (Bose Quiet Comfort 35): amzn.to/2MWbl3e
🌞 Lamp (BenQ E-reading Lamp): amzn.to/3e0UCr8
🌞 Secondary Lamp (BenQ Screenbar Plus): amzn.to/30Dtafi
💻 Monitor (BenQ EX2780Q): amzn.to/2HsmUPZ
💻 Monitor (LG Ultrawide 34WN750): amzn.to/3dSD7tS
🎙 Mic Boom Arm (Rode PSA 1): amzn.to/30EZw9m
🎚 Audio Interface (Focusrite Scarlet 4i4): amzn.to/2TjXsih
💸 Donations 💸
💵 One-Time Donations: www.paypal.com...
💰 Patreon: / techwithtim
◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️
⭐️ Tags ⭐️
- Asynchronous programming
- Coroutines
- Futures in Python
- Tasks in Python
- Python tutorial
- Tech With Tim
⭐️ Hashtags ⭐️
#Programming #Asynchronous #TechWithTim
A quick story. I started programming 4 years ago at Tim was the first instructor in this vast sea of programming. When I learned the basic I kinda move on and start my own journey. Today in my work I need to use asynchronous programming and of course I knew the place to learn. Tim you are amazing, thank you so much for the knowledge you have given us!
"a programmer had a problem and he thought to solve it with thread.
has problems. two Now he"
About talking you are what.
good one
good but irrelevant.
What
terrible joke
It would be great to learn about the differences between acyncio, multiprocessing and threading
asyncio and threading are almost identical, but threading is more intuitive. They are mainly used for I/O bound tasks.
Mulltiprocessing is very different. It bypasses the GIL (look it up if you don't know what it is). It is mainly used to improve performance with parallel programming.
Both asyncio and threading are concurrent.
Threading is being handled by the OS and AsyncIO is being handled within python ( the async event loop ).
Threads can execute small operations to speeds things up ( e.g. adding more robots to handle the work, difficult to oversee and manage when program grows big. Also race conditions where: comes out weird the outcome )
asyncio is done by using an event loop where task are being awaited ( e.g. non-blocking! tasks that are fired up and deliver some result while the program is doing other things )
At the end of the event loop ( yes it is a loop, all the results are gathered from the loop and it starts over again )
If tasks are blocking, asyncio isn't the right tool at all. Luckily you can also "throw" blocking tasks into a different executor so that they dont slow down your async stuff.
Multiprocessing is using multiple processes to do the grunt work. It kind of bypasses the GIL by having more processes with another GIL.
Drawback can be copied memory and such.
Basically every technique has an advantage and a drawback. I do like asyncio because it is a simple solution to a complex problem.
The thing is that when using an Executor you can test different scenario's within Python because you can swap out multiprocessing, threading and asyncio easily.
Test and see what does what.
@@sleepymarauder4178 thank you.
i see so many people spreading knowledge about threading vs. async which is wrong. good to see at least there's people out there explaining things completely and correctly.
Threading and multiprocessing are preemptive, coroutines are not. Coroutines and threading are default-share-everything, multiprocessing is default-share-nothing. Coroutines are easier to debug than threads.
@@sleepymarauder4178 do you know when is it better to use threading over async?
Just started with asynchronous programming in python, and this video of yours helps me to get comfortable with the asynchronous concept.
Thank you @tim
(well, I must include, I have a good grasp in the primary concepts of python, then I headed to the asynchronous concept in python)
😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊
😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊😊
asyncio is pretty useful when you're coding server's responses etc. While one request is being processed, the other one begins.
Hi do you know maybe some good tutorials how to use asyncio to handle multiple tcp sockets on a server? What would be the advantage to handle multiple sockets with asyncio compared to handle multiple server sockets with threads?
Can you compare the asyncio vs thread pool executor for me. I'm creating a script about sending and receving the request but when using TPE, the thread is locked for no reason
I wanted this for so long thank you so much! :)
Great stuff! I've been reading up on this for a few days now and your breakdown has been the clearest explanation. Thank you.
Oh, man. I needed this. I watched lots of videos but none of them were as good and simple-explained as yours.
This was suggested by Dovid S. on this same video, but I would really like a series where you go through the code of popular Python packages and/or modules .
I'm about a month into learning Python. That was a great explanation and gives me a foundation to dig deeper. It shows me how I can have multiple things happening at the same time, and the concept of futures really is exciting; in that I can see how the idea in my mind can actually be built.
I learned asynchronous programming on hard mode, means that when I started my first asyncio project (discord bot) I knew nothing about it, and barely a beginner in programming world and python In general i struggled days even for simple stuff, though this made me really learn not just scrolling through videos and remember nothing after over one year of working at my project ( which is alive even now, because I'm always performing maintenance checks on my bot) I still feel like I have a lot to learn even if I'm able to write a generous amount of code without "cheating" (searching for answers or something)
Great video for people who want to dive more deeper about asynchronous code
Man what the actual F, I was learning to make discord bot and now I get this video, you must be reading my head, Thanks man learning a new topic fromfav youtuber is really Awesome!
One of the best videos to get started with asyncio!
Thanks Tim, this was a really helpful and useful and awesome tutorial, I really enjoyed it. I 'await' fot the next tutorial!!!
This is an absolutely spendid demonstration of asyncio, I needed to use it today and had no clue how to go about it, now I'm off to testing my understanding on my product. Thanks Tim.
As always a great video!
There is just one point that was missing for me (a python-noob):
Global variables are shared between the async routines, so it looks to me like is multithreading and not multitasking - that makes it even better!
Before i was confused with threading and async. Thanks to tim, i now see the difference between threading and async. Threading suspends the thread/task, and runs the other thread, whereas, asycnio doesnt suspend anything it just gives execution
Absolutely fantastic and best video explanation of the asyncio and the keywords. Actually , very clear and to the point and thanks for not using ridiculous analogies.
12:58 There is a chance that could fail. asyncio only holds weak references to the tasks it creates; it relies on the caller (you) to maintain strong references, to stop those tasks from disappearing randomly. Here it executes after the “task” variable holding the reference has gone.
Thanks man, I have toi say one of the best tutorial over the whole asyncio python package ;)
ikr
Hi Tim, thanks a bunch for this video. I was struggling to understand asyncio and you made it simple to grasp. God bless you!
This was a good video to get a basic understanding of it but not full enough to actually know what applications to use it for. It reminded me of threads or background jobs.
Awesome!! Was waiting for this from soo long
This is the first time I really understood the basic mechanics of asyncio. Thank you!
Best explanation on RUclips. Thanks TIM !!
Man you make things so simple. Why are you not famous yet! Thank you so much for this! :D
thanks bro.... shortest tutorial with the most beneficial concept!
Since I found out that Python supports asyncio, I always avoid it, I prefer multi threading. But thanks to this tutorial I will start my project using asyncio. Multi threading feat asyncio is the best duet. Great Job Tim, greeting from Indonesia 🍻
this is the 5th video I watched today about asyncio and finally got it. Thanks for this, it was really clear :)
Finally, someone explained this so I could understand. Liked, subscribed 👍
Here we go another video from 'not a prodigy' kid! Boy I am 39 and you are my idol and inspiration, finally after 20 yrs of lingering with coding - you taught me something productive ;) Keep doing this great job. Thanks.
After 6 months I actually could apply some of Python in electrical engineering (electrical!). Soon my paperwork will be gone...
i would just say "PERFECT" for this explanation .
Thanks Tim !!
Really great video! One of the best ways of explaining I have seen.
I paused the video and tried the code.... and finally I figured it ! thx !！
you gave the crux of the async
now i can finally learn FASTAPI
thanks a ton buddy
This guy is amazing. Respect
thanks dude, i know you might not see this but thanks for your help.
You are welcome!!
exellent bro! you are the only one❤
Thanks a lot for your detailed explanation. It really helped me solving one problem related to the sending mail notification.
7:35 Worth mentioning at this point the difference between “stackless” and “stackful” coroutines. Python has the “stackless” kind, which is why the await keyword can only be used in the top level of an async-function, not anywhere else.
best explanation ever - including example and explanation. good work
I've been waiting for this video for months ❤️❤️❤️❤️
AT FIRST I DIDN'T UNDERSTOOD BUT THE 2ND TIME IT WAS SO CLEAR! THANK YOU!!!!
So clear about all informatiion. Tnx!
Awesome work dude
Thank you
very simply explained! Very well done Tim.
Yess! I was waiting for this for a long time
ive been trying to learn this for so long lmfao u made it so easy
THIS VIDEO IS AWESOME. THANK YOU SO MUCH FOR SHARING THIS WITH US.
Wow great introduction into asyncio. Thanks!!
No problem !
I am building my own spider Project and hope this tourial might work for me,you teached so detailed,i love it
Nicely explained such a complicated topic master tim. It's a good tutorial.
Thank you!! It was hard for me to grasp all this just from the asyncio documentation
This was very interesting, thank you. But it looks like you could get into all kinds of timing issues. A follow up to this might be how to identify time issues and overcome or prevent them?
Thanks Tim! Great video
Terrific job! Finally, I start understanding this topic :)
Starts at 4:30
Thank you, Tim. This video saved me a lot of time..
SUper clear. Thanks!!
damnn ... i have never seen a teacher like TIM
Excellent demo & explainations!
I noticed that, In the cosole, Task1 or Task2 and alike will only be said that they are pending If the value returning from them is being printed
I learnt from this
Thanks a lot
I love your videos, you have high quality content, Tim
Best simple explain, thanks
Process of Querying the database should also be a coroutine when using await?
Too good Tim.. 🎉
great video tim, I am a Typescript developer but I use python too. I was wondering if handling errors with Futures is like handler errors with promises in Typescript
I love this man.... exactly what i needed
Thank you for making it so easy to follow.
Awesome video, using this to help me understand some IoT applications in Azure
At 16:00, why doesn't the execution take at least 10.5 seconds to fully complete? I get that it's smart to hand off while you await, but why doesn't the "main" object/function finish running the task before being considered completely done? Or is it that the task runs outside of the main function and when you pass back off the memory or CPU to the function it is technically complete once it runs the last line in that function which is print(finished)?
Does the task ever get finished? I get the function completed, but does the task ever get compete itself? Like say it wasn't time.wait, but a different method that still caused a delay would it complete? That part is foggy.
Ya that part confused me as well. Ask chat gpt!
Thank you very well explained
THANK YOU!!!!!
Thank you so much! I could understand this clearly 🤯
you are the best:)
Really interesting! Good Explanation!!
very helpful intro, cheers
Thank you for making the awesome video
clear explanation
It was very useful thanks dude
Cool, the video I needed to really understand this!
at 12:57 i got what i needed!
Great video. Thanks so much, very helpful.
wonderful class! thank you so much!
If we were to add the "print('done')" part in my code below (exactly Tim's code at 16:56, but with the extra print), why would the 'done' message not be printed? The program prints 'finished' then exits. Is it because the execution of 'main' finished, therefore the execution of 'foo' is "prematurely" must be finished too, because "foo" was called from within "main"? (similar to how daemon threads work..?)
import asyncio
async def main():
print('tim')
task = asyncio.create_task(foo('text'))
await asyncio.sleep(0.5)
print('finished')
async def foo(text):
print(text)
await asyncio.sleep(10)
print('done')
asyncio.run(main())
Thank you, very nice video
thanks man you're very helpfull
kind of new to programming, one thing that is somewhat confusing me is, howcome he is able to execute foo function within the main, before defining it.
Interesting topic. As a beginner, it seems to me, that the artificial 'sleep command' hands tasks over nicely, but, what about a nasty huge for-loop that bogges down a program? Then it seems "busy", and can NOT hand over a task (to, for example, the exit button) and the program seems to freeze.
This is for I/O stuff. Since a huge for loop is a data processing task, it needs to run it somehow.
Great video. Is that the same as Celery does?
Nice explanation
Шикарно! Спасибо!
Weird! When I ran your code @08:44 , I got the error:
RuntimeError: asyncio.run() cannot be called from a running event loop
Just figured out why: It is because I am using the spyder IDE which uses an event loop to run its python interpreter/console. The solution is to insert the following at teh top of your code:
import nest_asyncio
nest_asyncio.apply()
Greate Content! Thanks
so only one task is actively running at a point in time ? (the coroutine waiting for something is pushed to the back?)
In every other use of the word "synchronous" in the human language, it means things happening together, synchronized like "synchronized swimming." I get that having all routines using the same thread and not being unhinged from each other could be considered as synchronous in some way, but even a dictionary definition of synchronous is "occurring or existing at the same time." I wish the people who named these things had a better command of the English language.
thank you for great content
Thanks, it helps me a lot
Hi, when I use this code in python, it is ok but when I run the same code in anaconda v5 (spyder), the error will rise :
"RuntimeError: asyncio.run() cannot be called from a running event loop"
what should I do?