Flask

Thoughts on Flask?

Attached: flask.png (3000x1174, 208K)

Other urls found in this thread:

techempower.com/benchmarks/#section=test&runid=fcd8886c-e0ff-407c-979c-d93f1f8100ab&hw=ph&test=fortune
twitter.com/NSFWRedditGif

flask is extremely nice and easy to get set up with. I like flask.

slow as fuck, kind of kludgey but comfy and easy to use.

fine for small apps or simple APIs but a nightmare for larger applications. that goes for Python in general though.

Currently using it for microservices. Since Flask is most useful for small, simple things, it's been a joy to work with.

>python
>2018

Python got me laid and I am regularly fucking now thanks to Python 3. 11/10 would recommend.

>2018
>2019

Same here. Before Python, women wouldn't give me the time of day. I was alone, and a loser. But now with Python 3.7+ I am picking up beautiful women and getting laid on a regular basis.

story

Please expand on this. I would like to know your thoughts.

javascript won. python is still dog ass slow due its fundamentally broken design which was never fixed.

Please explain kind sir....

javascript is literally 100-1000x faster and has better syntax, no forced whitespace

I'm assuming you're referring to things like the global interpreter lock. Will it ever get fixed? Python 4.0 at 10x speed would kill all new languages like crystal and nim.

This argument is pointless. If you need server-side speed for heavy compute, you don't use javascript. Doing a high-performance backend in node that does heavy computation is like doing a high performance engine with duct tape and super glue. Sure you can slap it together 100-1000x faster, but man when that first rod gets thrown nobody is going to have a fucking clue how the thing was put together.

generally, GIL only applies to compute. the GIL locks are mostly avoided when a thread is awaiting IO resources, it's only compute that causes the bottleneck.

if you are doing heavy compute with python you are doing heavy compute wrong.

also, threads, asyncio, and multiprocessing all have different modes of parallelism, so the conversation is more nuanced than just "muh GIL" ... it depends on your parallelism needs and what your workload is (IO vs compute)

Eh. Realistically more like 3-10x

built around being used with WSGI like bullshit like gunicorn. it's 2019, web frameworks should be self-standing on the scripting language they run on.

If it's good enough for Google, it's good enough for me.

Things like encoding/decoding json and ORM are still all "compute" though. I realize in most real-world web applications IO will be 80%+, it'd still be nice to optimize the remaining 20.

techempower.com/benchmarks/#section=test&runid=fcd8886c-e0ff-407c-979c-d93f1f8100ab&hw=ph&test=fortune

now that we are at it, post the best resources to learn python please

learncpp.com

It's literally pseudocode. Coming from C/C++ docs.python.org was enough for me for casual coding (scripting and API prototypes)

In all honesty? is there a reason to use flask over bottle? for everything flask is useful for, bottle basically fits the same use and doesn't need a templating engine and werkzueg for wsgi.

It works. Microframeworks are my go-to

Jetty for JVM Languages
NodeJS for Javascript
Flask for Python

>doesn't contain alcohol
>contains snakes instead

Attached: you_have_a_new_item.jpg (600x500, 37K)