Hi John 👋 Sanskar here, the author of Robyn. Firstly, thank you for covering the framework and for the great content. Documentation is one of our top most priorities and will improve substantially over the coming months. ✨ One suggestion, you can pass in the `--workers` flag to allow Robyn to serve a higher throughput. The default 1 worker is for prototyping or development mode. e.g. `python app.py --workers=5`. The number can vary depending on the power of your machine.
I will not be cheering Robyn on. I was looking through the issues and PRs and saw several red flags that led me to believe that the maintainer is largely incompetent. See issue #145. > I see a project with Rust and I am very excited thinking about type safety, but there is no Python typing in the current functions; have you considered adding them? "I am not a big fan of strongly typed languages(sorry but this is just me xD) BUT I found out that the community loves them. So, this is definitely a high priority issue. But, I don't have much experience with it so any help(either docs, resources or code contributions) would be highly appreciated." His reply has told me three things. 1) Rust is a strongly-typed language which the framework is based on and he hates such languages. This signals to me that the current or future Rust codebase in Robyn will be largely disappointing. 2) As a Python developer, he has no experience with typing in Python, which is arguably a core feature of Python 3 as of now. I can only expect the framework to have poor DX. 3) The maintainer himself is not dedicated to implementing types and the framework will probably see the same fate as Tensorflow, now losing popularity to the more typed PyTorch. Will not be wasting time on this framework. I'll stick with FastAPI until someone who actually cares about both Python and Rust technologies appear.
Hello sir I have watched your scrapy playlist very great content you have made There's a website i am trying to scrape the scrapy is reading only the head of the html code . The body is not getting read. I also tried to do it with splash but it is getting 502 bad gateway Could please help me 🙏
Hi John. Bit off topic but how would you normally distribute a python web scraping script to a client if they needed it to run once a month etc? Would you convert it to exe and send it to them or perhaps run it in the cloud and have it email them? Cheers
Hi - I would run it myself on my own server and either give them a simple interface to login and download/see data or email a complete file over each week
Thank you very much. I follow you and your videos are very helping. Could you please do a video to scape autosuggestion from any website please. Thank you in advance.
Great video!!🔥 Can you make a video where you use httpx? I have a project where I need to input stuff into a text field and then hit a submit button. How do I interact with a site using httpx? Thanks!
0:47 Polars is already there from quite some time and it’s blazing fast compared to our lovely Pandas 🐼. Unfortunately Pandas is the go to library for data manipulation and not enough people talking about Polars
@@JohnWatsonRooney yes, it is if you compared it to Pandas. The API is also simpler, more like PySpark API. And the performance comparison with Pandas is similar to what you've shown when comparing Robyn and FastAPI.
So guy who created new framework. - "Hmm, i have fastapi with 100.000 rps/sec, in my project i have 0.02 rps/sec. Its time to create new faster framework" =)
When I discovered Flask, I was happy because it was far more interesting than Django! And the rest are only variations (improved for sure)! Rust? I'm not interested, imagine thousands of programs to maintain and you are having a mix of API with different languages and more or less documentation? 3 months after the delivery, a modification and you are f...!
Interesting. But it looks to be too a little bit new still to use for production. I think though a wait and see approach. Good to know though, thanks again!
Hi John 👋
Sanskar here, the author of Robyn. Firstly, thank you for covering the framework and for the great content. Documentation is one of our top most priorities and will improve substantially over the coming months. ✨
One suggestion, you can pass in the `--workers` flag to allow Robyn to serve a higher throughput. The default 1 worker is for prototyping or development mode. e.g. `python app.py --workers=5`. The number can vary depending on the power of your machine.
Hey! Thanks for comment! I appreciate the efforts you've put into Robyn and I've enjoyed using it + yes you are right, I missed out the workers part
I will not be cheering Robyn on. I was looking through the issues and PRs and saw several red flags that led me to believe that the maintainer is largely incompetent. See issue #145.
> I see a project with Rust and I am very excited thinking about type safety, but there is no Python typing in the current functions; have you considered adding them?
"I am not a big fan of strongly typed languages(sorry but this is just me xD) BUT I found out that the community loves them. So, this is definitely a high priority issue. But, I don't have much experience with it so any help(either docs, resources or code contributions) would be highly appreciated."
His reply has told me three things. 1) Rust is a strongly-typed language which the framework is based on and he hates such languages. This signals to me that the current or future Rust codebase in Robyn will be largely disappointing. 2) As a Python developer, he has no experience with typing in Python, which is arguably a core feature of Python 3 as of now. I can only expect the framework to have poor DX. 3) The maintainer himself is not dedicated to implementing types and the framework will probably see the same fate as Tensorflow, now losing popularity to the more typed PyTorch.
Will not be wasting time on this framework. I'll stick with FastAPI until someone who actually cares about both Python and Rust technologies appear.
Thanks! I like such videos, trending libraries and frameworks thanks so much! and best of luck
Thanks i appreciate it
Thank you for releasing constantly great and helpful content.
Thanks John for sharing your amazing discoveries with us
Thank you for this video! Is good to be aware of new stuff!
4:32 subtle and classy. very nice
therefor 👍
haha. thanks
Thanks for the enlightenment John!!
I miss the swaggerui out of the box. And the dependency injection of fastapi.
And the harmony with pydantic is missing.
Thank you for this useful information!
Thanks for watching
Great stuff. I have been waiting for the dethroning of Pandas and I am excited about building a career based on Rust
Well, it's pretty much dethroned thanks to Polars ;-)
Thanks for sharing.
Thanks to inform us
would be great if theres a way to make a swagger url like fastapi does
I think you can you just have to set it up yourself, I’ll have to look into it
What does "hey" command do ??
“Hey” is a terminal app for benchmarking and testing web apps and APIs
Hello sir I have watched your scrapy playlist very great content you have made
There's a website i am trying to scrape the scrapy is reading only the head of the html code .
The body is not getting read.
I also tried to do it with splash but it is getting 502 bad gateway
Could please help me 🙏
Hi John. Bit off topic but how would you normally distribute a python web scraping script to a client if they needed it to run once a month etc? Would you convert it to exe and send it to them or perhaps run it in the cloud and have it email them? Cheers
Hi - I would run it myself on my own server and either give them a simple interface to login and download/see data or email a complete file over each week
Again a good video
Thank you
How does it compare to Django Ninja i wonder
I’ve not used django ninja before, only DRF. This is much newer but I was interested as it was built on a rust server for performance
Will watch later. Like and comment to feed algorithms 😉👍
Thank you!
It looks interesting...
I think it is. Not production ready I don’t think but good for side projects or personal stuff
@@JohnWatsonRooney YES!!!
Thank you very much.
I follow you and your videos are very helping.
Could you please do a video to scape autosuggestion from any website please.
Thank you in advance.
Great video!!🔥 Can you make a video where you use httpx? I have a project where I need to input stuff into a text field and then hit a submit button. How do I interact with a site using httpx? Thanks!
Bro didnt disable log output….
0:47 Polars is already there from quite some time and it’s blazing fast compared to our lovely Pandas 🐼. Unfortunately Pandas is the go to library for data manipulation and not enough people talking about Polars
Yeah I have actually heard of it until recently but it seems to be very good
@@JohnWatsonRooney yes, it is if you compared it to Pandas. The API is also simpler, more like PySpark API. And the performance comparison with Pandas is similar to what you've shown when comparing Robyn and FastAPI.
So guy who created new framework. - "Hmm, i have fastapi with 100.000 rps/sec, in my project i have 0.02 rps/sec. Its time to create new faster framework" =)
They are becoming like a node module devs.
When I discovered Flask, I was happy because it was far more interesting than Django!
And the rest are only variations (improved for sure)!
Rust? I'm not interested, imagine thousands of programs to maintain and you are having a mix of API with different languages and more or less documentation? 3 months after the delivery, a modification and you are f...!
Interesting. But it looks to be too a little bit new still to use for production. I think though a wait and see approach. Good to know though, thanks again!
Yes it’s still a young project but I like the idea and what it might mean going forward
I will watch it, but the clickbaity title makes me sad 😢
Is Robyn displaying the log output like the FastAPI version? This is a bottleneck here. Try to suppress log output in FastAPI and then compare.
Good point thanks