esoteric-python

[severe canyon]

it just terminates after 1 iteration, i need the number of iterations before the condition is met

[grave rover]

oh I see

[severe canyon]

also i wanna avoid some recursion if possible

else you cant get more than 1000 iterations without raising the recursion limit

[grave rover]

well yeah that's why I didn't use recursion :)

recursion could probably make it shorter but be a pain in the ass

wait can't you just do open('o.ppm','w').write(f'P2\n{h} {w}\n{m}\n'+'\n'.join(str(r((x%w)/w*2.67-2.1+(x//h)/h*2.5j-1.25j))for x in range(w*h)]) for the last line

[severe canyon]

yes, indeed i can

i can also move the lambda function to that line, for -2 chars, but im not too worried about that

found out i can replace: z.real**2+z.imag**2<=4 with abs(z)<=2

:D

because apparently, taking the absolute value of a complex number, is like doing pythagoras

Also gave a massive speedup

[sick hound]

What the f'{string} is #esoteric-python?

@severe canyon

[severe canyon]

?

[unborn charm]

uesy

988uiui

[sleek sphinx]

just python weirdness and "lambdascores" i guess

[severe canyon]

w,h,m=256,256,64
open('o.ppm','w').write(f'P2\n{w} {h}\n{m}\n'+' '.join(str(len((z:=0j)or[(z:=z*z+x%w/w*2.67-2.1+x//w/h*2.5j-1.25j)for _ in[0]*m if abs(z)<=2]))for x in range(w*h)))

got it down to 181

[unique rose]

you could do: open(str(w:=256)+'.ppm', 'w') and just use a w variable, and replace m with 64

or m with w ** .75

open(str(w:=256) + '.ppm','w').write(f'P2\n{w} {w}\n64\n'+' '.join(str(len((z:=0j)or[(z:=z*z+x%w/w*2.67-2.1+x//w/w*2.5j-1.25j)for _ in[0]*64 if abs(z)<=2]))for x in range(w*w)))

is 174 chars

[severe canyon]

i kinda want to have both image dimensions available

[unique rose]

i was thinking you could just record powers of 2, but i dont think that serves any use

you could make w both the width and height, using the decimal part as the height

[severe canyon]

width and height can be any number

[unique rose]

w//100, w//1000 etc

[severe canyon]

but they dont have to be the same

[severe canyon]

hmmm

[unique rose]

depends what the game is -- lots of ideas, not many which actually shrink char count

[severe canyon]

i guess

[fleet bridge]

!e

X=(T:=type)('',(T,),{'__init__':lambda x,*_:x.__setattr__(x,'__class__',x)},)('X',(T,),{},)

assert X is X.__class__ is type(X) and X.__name__ == 'X'

[night quarry]

@fleet bridge :warning: Your eval job has completed with return code 0.

[No output]

[fleet bridge]

!e

class X:
    pass

class Y:
    __class__ = X

print(f'{Y().__class__ = }')
print(f'{type(Y()) = }')

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | Y().__class__ = <class '__main__.X'>
002 | type(Y()) = <class '__main__.Y'>

[fleet bridge]

weird

!e

X=(T:=type)('',(T,),{'__init__':lambda x,*_:x.__setattr__(x,'__class__',x)},)('X',(T,),{'__new__':lambda x,*_:x,'__init__':lambda *_:None},)

assert X() is X is X.__class__ is type(X) and X.__name__ == 'X'
print('OK')

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

OK

[severe canyon]

w,h,m=1024,1024,64
open('o.ppm','w').write(f'P3\n{w} {h}\n255\n'+'\n'.join(['122 108',f'{len((z:=0j)or[(z:=z*z+x%w/w*2.67-2.1+x//w/h*2.5j-1.25j)for _ in[0]*m if abs(z)<=2])*62%255} 173'][(abs(z)>2)]+' 227'for x in range(w*h))+'\n')

with colors ¯_(ツ)_/¯

@bright eagle

[fleet bridge]

it isnt circle 🧐

[severe canyon]

ranges from top to bottom arent exactly the same

the mandelbrot set ranges from -2 .. 0.47 , -1.12i .. 1.12i

and im plotting: -2.1 .. 0.57 , -1.25i .. 1.25i

[severe canyon]

@bright eagle final one:
julia sets this time:

s,m=2048,128;c=.285+.01j  
open('o.ppm','w').write(f'P2\n{s} {s}\n{m}\n{" ".join([z:=x%s/s*(R:=(4*abs(c)+1)**.5+1)-R/2+x//s/s*R*1j-R/2*1j,str(len([(z:=z*z+c)for _ in[0]*m if abs(z)<=R]))][1]for x in range(s*s))}')

[bright eagle]

Neato!

[patent heron]

!e

t,\
    s,r,C,E,      =tuple,\
   set, range,   "*"," ";f=r\
 (31);p=( t(s(r(4,12))|s(r(18,
26))),t(s(r(3,14))|s(r(17,28))),
t(s(r(1,15))|s(r(16,30))),f,f,f,
r(1,30),r(3,28),r(5,26),r(7,24),
 r(9,22),r(11,20),r(13,18),(15
   ,));print(*[(e:="").join(
     [(C if j in p[i] else
       E)for j in f] ) +
         "\n" for i in
           r(14) ] ,
             sep=e
               )

[night quarry]

@patent heron :white_check_mark: Your eval job has completed with return code 0.

001 |     ********      ********     
002 |    ***********   ***********   
003 |  ************** ************** 
004 | *******************************
005 | *******************************
006 | *******************************
007 |  ***************************** 
008 |    *************************   
009 |      *********************     
010 |        *****************       
011 |          *************         
... (truncated - too many lines)

Full output: https://paste.pythondiscord.com/nufipuhalu.txt?noredirect

[patent heron]

:)

[severe canyon]

<3

coooooool!!!!

[patent heron]

That took a lot longer than I thought

and I do kinda have to cheat at the start but it is python after all

[severe canyon]

worth <3

[patent heron]

[night quarry]

Bad argument

Converting to "int" failed for parameter "pep_number".

[patent heron]

still slightly off center

!e ```py

t,
s,r,C,E, =tuple,
set, range, ""," ";f=r
(31);p=( t(s(r(5,13))|s(r(18,
26))),t(s(r(3,14))|s(r(17,28))),
t(s(r(1,15))|s(r(16,30))),f,f,f,
r(1,30),r(3,28),r(5,26),r(7,24),
r(9,22),r(11,20),r(13,18),(15
,));print([(e:="").join(
[(C if j in p[i] else
E)for j in f] ) +
"\n" for i in
r(14) ] ,
sep=e
)

[night quarry]

@patent heron :white_check_mark: Your eval job has completed with return code 0.

001 |      ********     ********     
002 |    ***********   ***********   
003 |  ************** ************** 
004 | *******************************
005 | *******************************
006 | *******************************
007 |  ***************************** 
008 |    *************************   
009 |      *********************     
010 |        *****************       
011 |          *************         
... (truncated - too many lines)

Full output: https://paste.pythondiscord.com/uqevunulom.txt?noredirect

[patent heron]

Now it's symmetrical

[severe canyon]

:O

[patent heron]

kinda want the top line to be a bit smaller but that is like impossible to reflect in the code

It's fine

[severe canyon]

:( i still think its beautiful

[grave rover]

that's so much better than when I tried it lol ```py
heart =
(lambda n:(lambda o=int(.5*n//1):
[([( lambda m=(x/n *3),l=(
y / n *3): print ([" ","#"
][( m2 +l2
-1) 3
-m2 *l
**3 <0
],end =""
)) ()
for x in
range (-o,
o)] ,print
()) for
y in
range (o,
-o, -1
)] )(
))

[severe canyon]

on my small discord window hehe 😁

[patent heron]

rip

[severe canyon]

time to go wiiiide

:)

[grave rover]

but hey at least my heart() function takes a target size

I could probably make it work if I un-golf it

that said I can't believe it's almost 3 years since I made that code

[thorny tapir]

is that like that one donut shaped code

!e

print("#####")
print("#####")
print("#####")
print("#####")

I can make rectangle shaped code that prints rectangle B)

[night quarry]

@thorny tapir :white_check_mark: Your eval job has completed with return code 0.

001 | #####
002 | #####
003 | #####
004 | #####

[restive void]

Reminds me of the "calculate Pi by changing the size of a circle in the code" IOCCC thing: http://www.cise.ufl.edu/~manuel/obfuscate/pi.c

[patent heron]

Thats cool

[radiant shale]

Hi, when I use this code (I'm using PyQt5)

widget.resize(496, 600)
self.line.move(-10, 540)

The line isn't showing when it is in a part of window which was resized, anyone know how to fix that?

[severe canyon]

@patent heron

class v:
 def __init__(s,x=0,y=0,z=0): s.x,s.y,s.z=x,y,z
v.__str__=lambda s:f'{s.x},{s.y},{s.z}' #print values
v.__add__=lambda s,r:v(s.x+r.x,s.y+r.y,s.z+r.z) #addition
v.__mul__=lambda s,r:v(s.x*r,s.y*r,s.z*r) #scaling
v.__mod__=lambda s,r:s.x*r.x+s.y*r.y+s.z*r.z #dot-product
v.__xor__=lambda s,r:v(s.y*r.z-s.z*r.y,s.z*r.x-s.x*r.z,s.x*r.y-s.y*r.x) #cross-product
v.__invert__=lambda s:s*(1/(s%s)**0.5) #normalize

🤔

[patent heron]

nice

[severe canyon]

not nice xD its so ugly

[patent heron]

I mean, yea

[unique rose]

perhaps a small tidy,

!e ```py

class v:
init = lambda s, kw: s.dict.update(kw)
getitem = lambda s, i: s.dict[i]
iter = lambda s: iter(s.dict.items())
str = repr = lambda s: str(dict(s))
add=lambda s,r: v({k:v + r[k] for k,v in s})
mul=lambda s,r: v({k:v * r for k,v in s})
matmul=lambda s,r: sum(v * r[k] for k,v in s)
invert=lambda s: s * ((s@s)-0.5)

v(x=10, y=20) @ v(x=5, y=16)

[rich hound]

hey, not sure if i used it wrong but doing * causes this TypeError: unsupported operand type(s) for *: 'int' and 'v'

Fixed, you forgot a [k]

[unique rose]

whereabouts?

[rich hound]

__mul__

@unique rose ^^

[unique rose]

yeah, it's correct -- 3 * v(...)

but it does highlight that v... * v... fails

[rich hound]

ah

my mistake then ^^

[unique rose]

!e```py
class v:
init = lambda s, *ps, kw: s.dict.update(kw or zip('xyz', ps))
getitem = lambda s, i: s.dict[i]
iter = lambda s: iter(s.dict.items())
str = repr = lambda s: str(dict(s))
add=lambda s,r: v({k:v + r[k] for k,v in s})
_cmul = lambda s,r: v({k: v * r[k] for k,v in s})
_smul = lambda s,r: v({k: v * r for k,v in s})
mul = lambda s,r: [v._smul, v._cmul][isinstance(r, v)](s, r)
matmul=lambda s,r: sum(v * r[k] for k,v in s)
invert=lambda s: s * ((s@s)**(-1/len(s)))
len=lambda s: len(s.dict.values())

print( v(2, 2) + v(3, 4))
print( v(2, 2) * 3 )
print( v(2, 2) * v(3,4) )
print( ~v(x=2, y=2) )
print( v(x=10, y=10) @ ~v(x=10, y=10) )

[rich hound]

:0

[severe canyon]

Nice!

Although, it's a little too much for what I need

[earnest wing]

could go the shader language route and make a heckload of overloads and definitions for vec2, vec3, ..., vecn

[unique rose]

v above is already overloaded

[earnest wing]

yeah

[sick hound]

I'm gonna reverse engineer that heart thats cool

its so cool!

[thorny tapir]

i cant even print a heart with normal python ;-;

[floral meteor]

I could probably print a heart in python, but Java style of code writing

!e ```py
void=None;
class System:
out=type("stdout",(),{'println':print})();
class Main:
def main(argv) -> void: {
System.out.println(" *** *** "),
System.out.println(" ***** ***** "),
System.out.println("**************"),
System.out.println(" ************** "),
System.out.println(" ********** "),
System.out.println(" ****** "),
System.out.println(" ** ")
}
name=='main'==Main.main(import('sys').argv)

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 |    ***    ***   
002 |   *****  *****  
003 | ****************
004 |  ************** 
005 |    **********   
006 |      ******     
007 |        **       

[floral meteor]

!e Now that yall have been tortured here's the pythonic way

print("""    ***    ***   
   *****  *****  
 ****************
  ************** 
    **********   
      ******     
        **""")

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 |     ***    ***   
002 |    *****  *****  
003 |  ****************
004 |   ************** 
005 |     **********   
006 |       ******     
007 |         **

[thorny tapir]

how tf does it not give syntax error for System.out.println

[floral meteor]

Because I snuck a comma in at the end

[sick hound]

wait a sec

[dense nova]

wait

python != java

tf

[sick hound]

I realized that you could like split the heart in half and like oh okay

[restive void]

!e

(        "hi",i:=
     __file__,C:=chr,
     s:=open(i,"rt").
   read(),sc:=s.count(C
   (10)),sn:=sc+s.count
   (C(32)),a:=len(s)-sn
   ,print("π","is",":",
     .5*a/((.5*sc)**2
     )),"#eso- teric"
         ,"pthn") 

[night quarry]

@restive void :x: Your eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 2, in <module>
003 | NameError: name '__file__' is not defined. Did you mean: '__name__'?

[restive void]

Un-golf for higher precision

[vague cairn]

The exact same thing happens in tkinter, I think it's happening at a deeper level than you can control, the workaround I use is to destroy the circle and re-create it at the new coords.

[coarse shuttle]

print(8)

[floral meteor]

I have a structure for an object oriented code for a grid game (e.g. minesweeper, sudoku, 2048) run in python

class Cell: # contains a value
  ...

class Engine: # contains a matrix
  ...

class Interface: # contains an event loop
  ...

I try to black-box Engine with Interface, so as multiple options are available, i.e. gui or cli or AI solution

but I'm not always successful at implementing black box

[prisma coral]

I suspect you intended to send this question in #software-architecture, which is about designing beautiful software, instead of here ( #esoteric-python ), which is where beautiful software goes to die

[severe canyon]

'where beautiful software goes to die'

[prisma coral]

things can be masterpieces without being beautiful 🤷

[severe canyon]

spirits lifted

[severe canyon]

is a raytracer in 1200 chars impressive?

[vestal solstice]

doesn't sound impressive

[earnest wing]

they're a frequent of this channel

[severe canyon]

o fancy role

[sick hound]

hi

[last locust]

@sick hound let's keep this channel on topic please, it's not the please to be dumping yt videos

And given this is an English speaking discord server this probably isn't the right server

[sick hound]

excuse me I do not understand English well and I understood that esoteric I am only looking to expand my community on twiter

[last locust]

We don't allow advertisement here. I recommend you read our #rules before continuing

Although if you can't speak English, then I'm afraid this isn't the server for you

[maiden vine]

Well my dudes!

There's obviously a lot of infrastructure I'm not showing here, but

I think I've managed to implement private attributes in Python XD

    def getCallingPrototype(self) -> Prototype:
        """retrieve the class which owns the scope, if the scope is indeed owned at all"""
        
        callingFrame = inspect.currentframe()
        callingCode  = getattr(callingFrame, 'f_code')

        while callingCode.co_flags & inspect.CO_NESTED:
            # climb the stack up to the first not-nested frame

            callingFrame = callingFrame.f_back
            callingCode  = getattr(callingFrame, 'f_code')


        # get every object which refers to the code object in question, then
        # filter out any objects which are not functions
        callingFunctions = gc.get_referrers(callingCode)
        callingFunctions = filter(inspect.isfunction, callingFunctions)
        callingFunctions = tuple(callingFunctions)


        if len(callingFunctions) > 1: # the code object has been used to manufacture more
                                      # then one function — throw an error
                                      #
                                      # NOTE: in theory, one could check ALL of these
                                      #       for privacy authorization

            error =  f"cannot access {self.privacy} member '{self.name}' of positronic "
            error += f"object with type '{instance.classname}' from within "
            error += f"recycled code object"

            raise AssertionError(error)

        if len(callingFunctions) < 1: # the code object is not attached to any function
                                      # (likely compiled manually) as is therefore
                                      # not authorized

            error =  f"cannot access {self.privacy} member '{self.name}' of positronic "
            error += f"object with type '{instance.classname}' from within "
            error += f"orphaned code object"

            raise AssertionError(error)


        # from the computed function, get its decorator (set as one of its attributes) and
        # from that, get the class the decorator is embedded on
        callingFunction  = operator.getitem(callingFunctions, 0)
        callingDecorator = getattr(callingFunction,  'decorator', None)
        callingPrototype = getattr(callingDecorator, 'prototype', None)

        # return the computed class
        return callingPrototype

Here is a snippet, the algorithm for finding which class owns a scope

[floral meteor]

"positronic"

Your variables are more than one word each

My variables are usually single letter

[maiden vine]

https://tenor.com/view/surprise-italianspiderman-gif-8081202

Its still very raw, I'll admit. I just wanted to get you guys' opinion on the approach TBH

Its the first half-sane means of determining the owner of a scope I've seen

But I'm sure there are edge cases I'm not thinking of. One improvement I've made already is that I've attached the code object directly to the decorator. For context, this is part of a group of classes that all work together — in this case, private variables are only eligible for access within code blocks decorated by a custom decorator. I realized I can just attach the code block as an attribute directly to the decorator and as such, I don't need to filter referents twice. I can directly detect any of my decorators which reference the code block in question

And that seems sane enough. Look at the current frame's code block, find any custom decorators which reference it, and whatever class that decorator is attached to (if any) is the owner of the code block

[earnest wing]

How does this respond to object.__getattribute__(x, "blah")?

[fading lion]

!e
import numpy as np

data = [i for i in np.arange(1,7)]
hori = vert = np.array(data)
arr = np.array([hori, vert])

def sample(n):
samp = []
for i in range(1,n+1):
for j in range(1,n+1):
num = (i,j)
samp.append(num)
return samp

sample = sample(6)
samp = np.array(sample)
print(samp)

[night quarry]

@fading lion :white_check_mark: Your eval job has completed with return code 0.

001 | [[1 1]
002 |  [1 2]
003 |  [1 3]
004 |  [1 4]
005 |  [1 5]
006 |  [1 6]
007 |  [2 1]
008 |  [2 2]
009 |  [2 3]
010 |  [2 4]
011 |  [2 5]
... (truncated - too many lines)

Full output: https://paste.pythondiscord.com/haqegazora.txt?noredirect

[fading lion]

!p
import numpy as np

data = [i for i in np.arange(1,7)]
hori = vert = np.array(data)
arr = np.array([hori, vert])

def sample(n):
samp = []
for i in range(1,n+1):
for j in range(1,n+1):
num = (i,j)
samp.append(num)
return samp

sample = sample(6)
samp = np.array(sample)
print(samp)

[night quarry]

Bad argument

Converting to "int" failed for parameter "pep_number".

[fleet bridge]

!e

from pprint import pprint

def is_constructible(cls: type) -> bool:
    try:
        cls()
    except TypeError as exc:
        s = exc.args[0]
        if not isinstance(s, str):
            return True
        if 'cannot create ' not in s:
            return True
        if ' instances' not in s:
            return True
        return False
    except:
        return True
    else:
        return True

x = [c for c in object.__subclasses__()[:100] if not is_constructible(c)]
x = sorted(x, key=lambda c: c.__name__[::-1])
pprint(x)

list of classes which raises TypeError: cannot create '<classname>' instances when called

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | [<class 'Token.MISSING'>,
002 |  <class 'async_generator_asend'>,
003 |  <class 'builtin_function_or_method'>,
004 |  <class 'hamt_collision_node'>,
005 |  <class 'hamt_bitmap_node'>,
006 |  <class 'hamt_array_node'>,
007 |  <class 'anext_awaitable'>,
008 |  <class 'PyCapsule'>,
009 |  <class 'frame'>,
010 |  <class 'coroutine'>,
011 |  <class 'types.UnionType'>,
... (truncated - too many lines)

Full output: https://paste.pythondiscord.com/elageloqig.txt?noredirect

[floral meteor]

It would be fun to make a net-oriented programming framework in python,

it'd be like neurons in a brain

possibly even turing complete

As long as you specified input nodes, output nodes, and a way to store memory

but you might be able to make memory with a logic loop

It would really just be an object that stores either 1 or 0 each clock cycle, and makes its next decision based on what's input to its "dendrites"

maybe even a glorified search algorithm

It would certainly be very esoteric to make programs with

[severe canyon]

so... like a neural network?

[maiden vine]

Yeah that sounds like a neural network to me XD

[severe canyon]

hehe

[marsh void]

what's the hamt_name_node thingy?

[rugged sparrow]

its part of from _testcapi import hamt

!e py from _testcapi import hamt a = hamt() b = a.set('x', 1) c = b.set('y', 1) print(*a.items()) print(*b.items()) print(*c.items())

[night quarry]

@rugged sparrow :white_check_mark: Your eval job has completed with return code 0.

001 | 
002 | ('x', 1)
003 | ('x', 1) ('y', 1)

[rugged sparrow]

https://github.com/python/cpython/blob/main/Python/hamt.c

[rugged sparrow]

i believe it was part of pep 567

!pep 567

[night quarry]

**PEP 567 - Context Variables**

Link

Status

Final

Python-Version

3.7

Created

12-Dec-2017

Type

Standards Track

[sick hound]

!e

for n in range(2, 100): print(n, end=', ') if len([d for d in range(2, n) if n % d == 0]) == 0 else print(end='')

[night quarry]

@sick hound :white_check_mark: Your eval job has completed with return code 0.

2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 

[sick hound]

One-liner that generates primes. Not the best, but it works.

[night quarry]

@sick hound :white_check_mark: Your eval job has completed with return code 0.

[2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97]

[restive void]

!e

print([n for n in range(2,100)if not __import__("re").match(r'^1?$|^(11+)\1+$','1'*n)])

[night quarry]

@restive void :white_check_mark: Your eval job has completed with return code 0.

[2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97]

[restive void]

cursed version

[ornate mist]

a little challenge in this genre is to make a one-liner that prints a list of length n that contains the integers

[1,2,3,4,5,6,7,8,9,10,20,30,40,50,60,70,80,90,100,200,300,400,500 ...]

[fleet bridge]

!e

n = 37
print([((i-1)%9+1)*10**((i-1)//9) for i in range(1,n+1)])

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000]

[restive void]

!e

print([n for i,n in zip(range(37),(n for n in __import__("itertools").count()if len(str(n).strip("0"))==1))])

[night quarry]

@restive void :white_check_mark: Your eval job has completed with return code 0.

[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000]

[dense nova]

imagine my confusion when i first joined this server and read 'Topical Chat/Help' as 'Tropical Chat/Help', the fact that theres a palm tree next to the channel name didn't help either.

I just found out its 'Topical Chat/Help' by the survey and I've been in this server for like a month.

[earnest wing]

fantastic

[severe canyon]

I feel like there should be a tropical chat anyway

🌴

[severe canyon]

!e

print([(i%9+1)*10**(i//9)for i in range(37)])

[night quarry]

@severe canyon :white_check_mark: Your eval job has completed with return code 0.

[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000]

[severe canyon]

cleaned up your solution a bit

[fleet bridge]

ty

[torn grotto]

!e
print("test")

[rugged sparrow]

!e ```py
def make_locals():
import sys
to_update = {}
def locals():
frame = sys._getframe(1)
return to_update.setdefault(frame, {**frame.f_locals})

def tracefunc(frame, what, arg):
    if frame in to_update:
        for key in frame.f_locals.copy():
            if key not in to_update[frame]:
                del frame.f_locals[key]
        for key, val in to_update[frame].items():
            if key not in frame.f_locals or frame.f_locals[key] != val:
                frame.f_locals[key] = val
        del to_update[frame]
    return tracefunc
sys.settrace(tracefunc)
return locals

builtins.locals = make_locals()

def foo():
locals()['abc'] = 'New Value'
print(abc)
abc = None # forces abc to be a local

foo()```

[night quarry]

@rugged sparrow :white_check_mark: Your eval job has completed with return code 0.

New Value

[rugged sparrow]

makes writing to the locals() dict actually stick using sys.settrace

not quite as good as my ctypes version, as the variables arent written until before the next line instead of immediatly

[golden finch]

oh my god

[floral meteor]

Beautiful

Now I can throw this in a file in python path and import mutable_locals to get locals that work like globals.
Now you just gotta figure out how to make local annotations work XD

[ornate mist]

this just makes me feel uncomfortable XD

[radiant cloak]

I'm guessing that tracefunc is meant to be invoked at least every time a variable is set. It's worth noting that sys.settrace which sets tracefunc to the trace-function, is used by debugger tools, so that could cause conflicts. sys._getframe is not guaranteed to be implemented outside of CPython. All in all, horrible code. 8/10

[torn grotto]

!e
print("test")

[night quarry]

@torn grotto :white_check_mark: Your eval job has completed with return code 0.

test

[floral meteor]

debugger tools: I don't have bugs in the first place. I'm too good at programming :P
CPython: why would I use any other version of python?

I jest of course, but I still think that doesn't warrant a number as low as 8

[rugged sparrow]

If sys._getframe doesn't exist I can get the frame from a trace back (and if that doesn't work then the implementation prob doesn't have user facing frame objects and thus has no settrace)

[floral meteor]

I think ok boomer would have sufficed.

[fiery hare]

#bot-commands is a place we all can go.

[floral meteor]

I think !e print("test") is a daily occurrence in this channel.

and there's nothing we can do about it

except program the bot to detect that exact message content in this channel

since there's rarely any deviation and they don't use backticks

I could find it eventually, but where's the link to do that?

and what would be an appropriate action, just delete and warn?

it seems !source gets me where I want to go

so I'll call it esoteric python print("test") prevention

I'll do it a little later though

[vague cairn]

Sometimes I love the errors you get when you muck around with ctypes on slot classes...

    getattr(cls, attr)(item)
TypeError: descriptor '__iter__' requires a 'dict' object but received a 'dict'

[prisma coral]

lol

[floral meteor]

XD

[ornate mist]

those people who can't make the difference between dict and dict smh my head

[severe canyon]

must have been an english and a german dictionary

[maiden vine]

How does one fool type(something)?

As best I can tell, if I implement an __instance_check__ and __subclass_check__, as well as somehow disguise its __mro__ and __class__ attributes then class A can effectively be disguised as class B. Are there any edge cases I'm missing?

[astral rover]

why do you want to do this?

[shut trail]

i think you can make a metaclass that implements instance_check right?

[maiden vine]

Why do bears poop in the woods?

[maiden vine]

Definitely

[shut trail]

!e

class MyMeta(type):
  def __instancecheck__(self, other):
    print(other)
    pass

class Test(metaclass=MyMeta): ...

isinstance(2, Test)

[maiden vine]

But its Python. Its very resistant to introspection manipulation like this XD I just want a full solution

[night quarry]

@shut trail :white_check_mark: Your eval job has completed with return code 0.

2

[shut trail]

ah

[astral rover]

you cant object.__x__ always exists

[maiden vine]

Hmmmm

[astral rover]

hence the question of why are you doing this, do you actually plan on having people use this?

[maiden vine]

Nah

I'm doing some work with metaclasses in general, and so I'm sorta in that zone

But I'm on break, so, I'm feeling a bit experimental

Just thought I'd take a look

[earnest wing]

!e

@type
@type.__call__
@type
@type.__call__
class C: pass
print(C)

[night quarry]

@earnest wing :white_check_mark: Your eval job has completed with return code 0.

<class '__main__.C'>

[desert flame]

h

[languid hare]

interesting discrepency:

__builtins__ = {}
print(__builtins__)

running the above as a python file results in {}
running with IDLE or a REPL results in NameError: name 'print' is not defined

anyone might have an idea why this happens

[restive void]

Because the two lines are compiled together when run as a file, but one by one when run from a REPL. You can also make it work in a REPL if you somehow put them in one compilation step:

class block:
    def __enter__(*whatever): pass
    def __exit__(*whatever): pass

with block():
    __builtins__ = {}
    print(__builtins__)

works in the REPL, but afterwards the builtins are gone.

Like print is in the co_names of the compiled module before any deletion happens. This is of course just a string, but I imagine that's how that happens.

If you replace print(__builtins__) with exec("print(__builtins__)") it stops working, in my example.

[languid hare]

hmm interesting

what about the reverse

what if I wanted behaviour similar to in idle

[restive void]

exec("print(...)") should always fail, also in a file

(Unless print is used already, I guess)

[sick hound]

is this channel for script help?

[dark wharf]

about as far from it as concievable

if you ask for help here your code will emerge a hideous abomination of its former self

[prisma coral]

try #❓｜how-to-get-help

[dark wharf]

yea that was probably the correct response but hey im not wrong either

[prisma coral]

!e This is kinda satisfying:

__import__("fishhook").hook(console := type, "log")(print)

console.log("woohoo")

[night quarry]

@prisma coral :white_check_mark: Your eval job has completed with return code 0.

woohoo

[fleet bridge]

!e

__import__("fishhook").hook(type, "log")(print)

int.log("woohoo")

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

woohoo

[fleet bridge]

😄

!e

from fishhook import hook
hook(object, 'fizz')(lambda *args: print('bazz'))
1 .fizz()
[].fizz()
{}.fizz()
None.fizz()
....fizz()
....fizz().fizz()

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | bazz
002 | bazz
003 | bazz
004 | bazz
005 | bazz
006 | bazz
007 | bazz

[rugged sparrow]

!e ```py
import sys, dis

def trace(tf):
def wp(f):
def inner(*a, **k):
try:
sys.settrace(tf)
return f(*a, **k)
finally:
sys.settrace(None)
return inner
return wp

def jump_trace(frame, line, arg):
frame.f_trace_opcodes = True
if line == 'opcode':
idx = frame.f_lasti
op, arg = frame.f_code.co_code[idx: idx + 2]
if dis.opname[op] in ['LOAD_GLOBAL', 'LOAD_NAME']:
sarg = frame.f_code.co_names[arg]
if isinstance(sarg, str) and sarg.startswith('JUMP_'):
, d, n = sarg.split('')
try:
mvs = {
'UP': frame.f_lineno - int(n),
'DOWN': frame.f_lineno + int(n),
'TO': int(n)
}
frame.f_lineno = mvs.get(d, frame.f_lineno)
return
except:
pass
return jump_trace

@trace(jump_trace)
def foo():
x = 0
print(x)
x += 1
if x > 10:
return
JUMP_UP_4

foo()

[night quarry]

@rugged sparrow :white_check_mark: Your eval job has completed with return code 0.

001 | 0
002 | 1
003 | 2
004 | 3
005 | 4
006 | 5
007 | 6
008 | 7
009 | 8
010 | 9
011 | 10

[rugged sparrow]

It's not at versatile as my opcode replacement version but it works

[winged barn]

!e

[night quarry]

Command Help

!eval [code]
Can also use: e

*Run Python code and get the results.

This command supports multiple lines of code, including code wrapped inside a formatted code block. Code can be re-evaluated by editing the original message within 10 seconds and clicking the reaction that subsequently appears.

We've done our best to make this sandboxed, but do let us know if you manage to find an issue with it!*

[winged barn]

noobie question ... can i post code here and get a question answered?

[astral rover]

you probably shouldnt be getting code help here if youre a noob see #❓｜how-to-get-help

[winged barn]

@astral rover thanks

[fleet bridge]

!e

def f():
    print('before change #1')
    f.__code__ = f.__code__.replace(co_consts=new_consts)
    print('before change #2')


new_consts = (None, 'after change #1', ('co_consts',), 'after change #2')
print(f.__code__.co_consts)
f()
print(f.__code__.co_consts)
f()

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | (None, 'before change #1', ('co_consts',), 'before change #2')
002 | before change #1
003 | before change #2
004 | (None, 'after change #1', ('co_consts',), 'after change #2')
005 | after change #1
006 | after change #2

[grand tree]

!e

arr = []
class test:
  @classmethod
  def retinst():
    return self
for k in range(10):
  arr.append(test.retinst)
for it in range(len(arr)):
  print(it)

[night quarry]

@grand tree :white_check_mark: Your eval job has completed with return code 0.

001 | 0
002 | 1
003 | 2
004 | 3
005 | 4
006 | 5
007 | 6
008 | 7
009 | 8
010 | 9

[novel meadow]

I was told to try asking here instead by Mustafa
Someone sent this in another server

import ctypes, inspect, gc
def main():
    frame = not print("trolled with voidpointer haha") and inspect.currentframe()
    ctypes.cast(ctypes.c_void_p(int([referer for referer in gc.get_referrers(frame.f_code) if getattr(referer, "__code__", None) is frame.f_code and set(inspect.getclosurevars(referer).nonlocals.items()) <= set(frame.f_locals.items())][0].__repr__()[-15:][:-1], 16)), ctypes.py_object).value.__call__()
main()

which just repeatedly prints the print line

What in the accursed world is this?

[grand tree]

why

i was just messing aroudn, didnt wanna create a file

[rugged sparrow]

Use a trace function to change the line number

[sick hound]

whats code gore?

[next flame]

very awful code

[terse mortar]

This isn't even esoteric, this is just kinda useless

I get a lot of esoteric code is useless, but this is literally just ```py
for i in range(10):
nums.append(i)

for i_2 in len(nums):
print(i_2)

[night quarry]

Hey @inland crystal!

Uh-oh! It looks like your message got zapped by our spam filter. We currently don't allow .txt attachments, so here are some tips to help you travel safely:

• If you attempted to send a message longer than 2000 characters, try shortening your message to fit within the character limit or use a pasting service (see below)

• If you tried to show someone your code, you can use codeblocks
(run !code-blocks in #bot-commands for more information) or use a pasting service like:

https://paste.pythondiscord.com

[inland crystal]

!codeblocks

[night quarry]

Here's how to format Python code on Discord:

```py
print('Hello world!')
```

These are backticks, not quotes. Check this out if you can't find the backtick key.

[grand tree]

That's crazy bro

[eager sphinx]

!e

from math import *
r=range
itc= (10,129297,128530,128542,32,33,128545,128548,128038,40,41,128683,44,128555,46,127790,128560,128575,63,65,66,127939,128580,69,68,*r(71,81),83,84,127829,85,89,127837,97,98,128483,100,101,103,104,105,*r(107,113),114,115,116,117,119,121)
def decode(e,b=1<<3):
    l=ceil(floor((log2(e))+1) / b)
    return"".join(map(lambda v:chr(itc[v]),reversed((*map(lambda i:(n:=e>>((l-i-1)*b))-(n>>b<<b)-1,r(l)),))))
print(decode(3244649123016202304891620531740982762891576544349418611903359070093996148368271375664004833158782419693562368836100015717309746308332940478214165459771461696755758941698312170665438354835055834831744615294239453342422211676208604486492899923465219708682188660541207378354834529120737764067488908516785255937372371826538196424288440513734017759507973354255076505610764439310038685057105380756911991480116903918691598975498051255492241347215219269210770330616723537702123696223194741388965316956049794151445622869939476096941182589287463344214230469776438876833152703640187393087978507269978387253970430932734015582132700799781747041561406164252203015589507175416946093177848124798931251421662043923210397855460937219260958556753825036738197210775503124127439473622168605484957252044318980335590468794557627590658425490077211094575885174664166519538163711538014654477691416995561377842095391962921377395420086534902478441806398390468998664407161668998259836607346093060894177235495713257626700960105270542207746133468723296932517388083783230442828998535786514706960160148499916637788255417695785324346500982590476571292941003412146388465011509827394655562950181687950396815699310393422661751300725350352875585493140797753767221528333636979638462863703881880397526185202577677344133834472196407668967934862043413785868761398687053999727462972065072563600212346884760118092965616800572377416888571625464970966174085783681256507818947953372411031341568724531967057894693591683032679322585434294326037397807531219870135510123849601,b=6))

[night quarry]

@eager sphinx :white_check_mark: Your eval job has completed with return code 0.

001 | 
002 | Kala mo di masakit 😫 yung pinagsasabi mo sakin??? 🗣🗣🗣Na patay gutom akohh??? 😤😤😤Bakit?Hindi bayon masakit? 😞😡😡E ikaw nga hinuhuli mo kalapati e.🐦🐦🐦Sabi niya tara Joela ku,hulihin natin yung kalapati 🏃🐦kuhain natin 🚫🚫🚫(MASAMA YON)Oh siya yun,siya yung pagsabihan mo😫🤑😤😡.Siya pa naninisi😿,hati nga kami sa pagkain🍝🌮🍕,hindi naman siya ang bumili🏃.Hindi niya nalang hingiin😡😡😡para hindi ma...🤑🤑🤑O,wag mo na siyang anohin 😡😡😡!!!wag mo na siyang anohin(HULI TAYONG KALAPATI🐦)Bakittt??!!! 😫😡😤Nakikisali ka dito kahit di ka naman kasali.O bakit? 😒🙄🙄Bat mo inaano pinsan ko 😤😡.Iiyak na nga diba😰😰😰,tapos yun na nga,wag mo ng anuhin 😡😡😡.Iiyak na nga eyeyeyye.O bakit??? 🙄🙄🙄O bakit din??!!! 🙄😒😒😒Umalis kana 🗣🗣🗣Aalis ako di naman ako dito eh.😒😒🙄🙄.Bb ka pala eh😒😡Bakit ka nag sasabing bb??? 🙄😤😤Bat ka namamangga??😰😫😿OYY DONDUNGGN INGAY INGAY NIYO!!!HOY🗣🗣🗣.

[eager sphinx]

this is so satisfying

the original encoded number was 4,000 chars long i couldn't fit it in a discord message

i had to do this crap to compress it to 2,000

"you think it doesn't hurt, what you were saying about me? that i was eating a ton? why? is it not hurtful? ..." etc. etc.

it's funnier in tagalog

[hot hazel]

dont know a single code
im alone
not catched up with anything

[sick hound]

b = lambda __92_ : [[__92_ for n in range(2**63)]]==[[__92_ for n in range(2**63)]]
_____ = [(___-8)*-b("Eror 404")
 for 
  ___ in range(
      len(f"+{'_'*2}"))]
s = sum(_____)//(b(9223372036854775808)+b("""Letting the days go by, let the water hold me down"""))
print(_____)

mistakes have been made.

if anyone with a fast computer and an entire night free wants to try this, go ahead.

i believe this is what the industry calls a mistake

[prisma coral]

Ah yes. for n in range(2**63)

[sick hound]

mhm.

you see

it turns out that time.sleeps limit is 2**63 nanoseconds

my plan was to have a program that counts upwards, each time testing whether time.sleep() works on it

when it eventually reaches 2**63, that is used as the value

that way it would take a very long time

[rapid sparrow]

what do you mean?

we can help you if you're confused about python

[nova ginkgo]

can someone transform this code into some cursed one-liner

a = [100, 400, 500, 300, 200]
b = 345

def closest_to_int(x, itr):
    return sorted(itr, key=lambda j: abs(j - x))[0]

print(closest_to_int(b, a)) # -> 300

[shut trail]

!e ```py
print((lambda x, itr: sorted(itr, key=lambda j: abs(j - x))[0])(345, [100, 400, 500, 300, 200]))

[night quarry]

@shut trail :white_check_mark: Your eval job has completed with return code 0.

300

[shut trail]

@nova ginkgo ^

if you need me to explain anything in there LMK

[astral rover]

fwiw you can use min here

[terse mortar]

You don't need all the spaces after the commas

[nova ginkgo]

great, thanks guys

[shut trail]

ah true, that will probably up the byte count if this is a smallest amount of bytes solution and that kinda stuff

[terse mortar]

You can also get rid of the space after the :s

!e ```py
x = lambda test:print(test)

x("test")

[night quarry]

@terse mortar :white_check_mark: Your eval job has completed with return code 0.

test

[terse mortar]

!e ```py
print((lambda x,itr:sorted(itr,key=lambda j:abs(j-x))[0])(345,[100,400,500,300,200]))

[night quarry]

@terse mortar :white_check_mark: Your eval job has completed with return code 0.

300

[arctic pine]

Can someone help with this script?

[night quarry]

Hey @arctic pine!

Uh-oh! It looks like your message got zapped by our spam filter. We currently don't allow .txt attachments, so here are some tips to help you travel safely:

• If you attempted to send a message longer than 2000 characters, try shortening your message to fit within the character limit or use a pasting service (see below)

• If you tried to show someone your code, you can use codeblocks
(run !code-blocks in #bot-commands for more information) or use a pasting service like:

https://paste.pythondiscord.com

Hey @arctic pine!

It looks like you tried to attach a Python file - please use a code-pasting service such as https://paste.pythondiscord.com

[arctic pine]

https://paste.pythondiscord.com/ubozuyiraz.py

Please help me with this code I get, xsrf update error: list index out of range <class 'IndexError'>

Dm me if you figured it out!

[next flame]

@arctic pine think you want a help channel instead lol #❓｜how-to-get-help

[dark wharf]

for the sorted[0] can't you just use min, it accepts key arg too

[astral rover]

#esoteric-python message

[floral meteor]

I find this brainfuck derivative useful in arithmetic encryption

def intfuck(code:list[int]):
  a=__import__("collections").defaultdict(int)
  i=p=t=0
  def add(i):
    a[i]=(a[i]+1)%256
    return i,0
  def sub(i):
    a[i]=(a[i]-1)%256
    return i,0
  def left(i):return i-1,0
  def rite(i):return i+1,0
  def dot(i):
    print(end=chr(a[i]))
    return i,0
  def com(i):
    a[i]=ord(__import__('sys').stdin.read(1))
    return i,0
  def loop(i):return i,not a[i]
  def end(i):return i,-bool(a[i])
  while 0<=p and p<len(c):
    if t:t+=(c[p]==7)-(c[p]==8)
    else:i,t=[add,sub,left,rite,dot,com,loop,end,lambda i:(i,0)][int(''.join([*range(8)]).find(str(c[p]-1))](i)
    p+=1-2*(t<0)
  return int(bool(t))

I had to write it from memory so it might have errors

this one prints the output and returns the exit code

[floral meteor]

I tried to reverse engineer intfuck, but it looks like it belongs where beautiful code goes to die... oh wait that's here!

def random_noop_insert(victim):
  from random import randint
  _0,_1=0,len(victim)
  while _0<_1:
    if randint(0,4):victim=''.join([*victim[:_0],str(9*randint(0,1)),*victim[_0:]])
    _0+=1
  return map(int,victim)

def unfuck(message:str):
  output=[ord(this)%256 for this in message]
  if randint(0,9):code=''
  else:code='9'
  _0=[*filter(lambda x:any([t//10==x for t in output]),range((max(output)+2)//10)),max(output)//10]
  _0=[*{*_0}]
  code+='1'*10+'74'+'4'.join([t*'1'for t in _0])+'3'*len(_0)+'284'
  j=0
  for c in output:
    a = c.__divmod__(10)
    while a[0]>d[j]:j+=1;code+='4'
    while a[0]<d[j]:j-=1;code+='3'
    code+='1'*a[1]+'5'+'2'*a[1]
  _={_ and(code:=code[:-1])for _ in iter(lambda:code.endswith('2'),False)},{_ and(code:=code.replace'12',''))for _ in iter(lambda:'12'in code,False)},{_ and(code:=code.replace('21',''))for _ in iter(lambda:'21'in code,False)}
  return[*random_noop_insert(code)]

[inland monolith]

prio = {                   1: 'priority', 2: 176027, 3: '',             4: 'priority'}
supp = {0: 'supplemental',                2: 160620, 3: 'supplemental', 4: 'supplemental'}
merged = supp | prio
merged = {k: supp[k] if not merged[k] and supp.get(k) else merged[k] for k in merged}
merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'}

is anyone able to golf this further?

[last locust]

p='priority'
prio={1:p,2:176027,3:'',4:p}
s='supplemental'
supp={0:s,2:160620,3:s,4:s}
merged={k:x if not (x:=merged[k]) and (y:=sup.get(k))else x for k in supp|prio}
```this?

[inland monolith]

!e

p='priority'
prio={1:p,2:176027,3:'',4:p}
s='supplemental'
supp={0:'supplemental',2:160620,3:s,4:s}
merged={k:x if not (x:=merged[k]) and (y:=sup.get(k)) else x for k in merged}
print(merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'})

can't walrus in an inline if

[last locust]

You need() around the x:= probably

edited

[inland monolith]

how do i tell it to try again

[last locust]

Press the reaction

nvm, you took too long to edit

Just resend

[inland monolith]

!e

p='priority'
prio={1:p,2:176027,3:'',4:p}
s='supplemental'
supp={0:'supplemental',2:160620,3:s,4:s}
merged = supp | prio
merged={k:x if not (x:=merged[k]) and (y:=supp.get(k)) else x for k in merged}
print(merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'})

[last locust]

oh wait

fixed

[night quarry]

@inland monolith :white_check_mark: Your eval job has completed with return code 0.

False

[inland monolith]

aaaan it doesn't even work

[last locust]

And obviously you can shorten further by using shorter variable names

sec

!e ```py
p='priority'
prio={1:p,2:176027,3:'',4:p}
s='supplemental'
supp={0:s,2:160620,3:s,4:s}
merged={k:x if not (x:=(supp|prio)[k]) and (y:=sup.get(k))else x for k in supp|prio}

[inland monolith]

i don't really care about variable names, i just want to reduce the number of valid lines

[night quarry]

@last locust :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 5
002 | SyntaxError: assignment expression cannot be used in a comprehension iterable expression

[inland monolith]

lol

[last locust]

!e py p='priority' prio={1:p,2:176027,3:'',4:p} s='supplemental' supp={0:s,2:160620,3:s,4:s} merged={k:x if not (x:=(supp|prio)[k]) and (y:=supp.get(k))else x for k in supp|prio} print(merged)

[night quarry]

@last locust :white_check_mark: Your eval job has completed with return code 0.

{0: 'supplemental', 2: 176027, 3: '', 4: 'priority', 1: 'priority'}

[inland monolith]

3 is wrong

[last locust]

What are you actually trying to do?

[inland monolith]

merge supplemental into priority such that anything missing from priority is supplemented

missing can be defined as both the absence of a key and a value that would equate to false, like '' or None

!e

prio = {                   1: 'priority', 2: 176027, 3: '',             4: 'priority'}
supp = {0: 'supplemental',                2: 160620, 3: 'supplemental', 4: 'supplemental'}
merged = {k: supp[k] if not prio.get(k) and supp.get(k) else prio[k] for k in supp|prio}
merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'}

[last locust]

So take prio if it exists, otherwise supp?

[inland monolith]

yes

!e

prio = {                   1: 'priority', 2: 176027, 3: '',             4: 'priority'}
supp = {0: 'supplemental',                2: 160620, 3: 'supplemental', 4: 'supplemental'}
merged = {k: supp[k] if not prio.get(k) and supp.get(k) else prio[k] for k in supp|prio}
print(merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'})

[night quarry]

@inland monolith :white_check_mark: Your eval job has completed with return code 0.

True

[inland monolith]

ah nice, now i don't have to declare merged twice

[last locust]

Is that not just prio | supp?

[inland monolith]

!e

prio = {                   1: 'priority', 2: 176027, 3: '',             4: 'priority'}
supp = {0: 'supplemental',                2: 160620, 3: 'supplemental', 4: 'supplemental'}
print(prio | supp == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'})

[night quarry]

@inland monolith :white_check_mark: Your eval job has completed with return code 0.

False

[last locust]

!e py prio = { 1: 'priority', 2: 176027, 3: '', 4: 'priority'} supp = {0: 'supplemental', 2: 160620, 3: 'supplemental', 4: 'supplemental'} print(prio | supp) print(supp | prio)

[inland monolith]

lol

[night quarry]

@last locust :white_check_mark: Your eval job has completed with return code 0.

001 | {1: 'priority', 2: 160620, 3: 'supplemental', 4: 'supplemental', 0: 'supplemental'}
002 | {0: 'supplemental', 2: 176027, 3: '', 4: 'priority', 1: 'priority'}

[last locust]

What's wrong with the first one?

[inland monolith]

neither are what i want (i did look before)

[last locust]

Oh I see

[inland monolith]

all of them are supplemental

and the 2nd one is missing the value i needed for 3

[last locust]

!e ```py
prio = { 1: 'priority', 2: 176027, 3: '', 4: 'priority'}
supp = {0: 'supplemental', 2: 160620, 3: 'supplemental', 4: 'supplemental'}

d = {k: x if (x:=prio.get(k)) else supp[k] for k in prio | supp}
print(d)```

[night quarry]

@last locust :white_check_mark: Your eval job has completed with return code 0.

{1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority', 0: 'supplemental'}

[last locust]

Is that good?

[inland monolith]

i think so

!e

prio = {                   1: 'priority', 2: 176027, 3: '',             4: 'priority'}
supp = {0: 'supplemental',                2: 160620, 3: 'supplemental', 4: 'supplemental'}
merged = {k: x if (x:=prio.get(k)) else supp[k] for k in prio | supp}
print(merged == {0: 'supplemental', 1: 'priority', 2: 176027, 3: 'supplemental', 4: 'priority'})

[night quarry]

@inland monolith :white_check_mark: Your eval job has completed with return code 0.

True

[inland monolith]

hm what did you change

i see the walrus

{k: supp[k] if not prio.get(k) and supp.get(k) else prio[k] for k in supp|prio}```
vs```py
{k: x if (x:=prio.get(k)) else supp[k] for k in prio | supp}```

[last locust]

I also tidied your conditional

[inland monolith]

why remove supp.get(k)

[last locust]

Note that mine assumes it'll always be in prio or supp

If that's not guaranteed then change to supp.get(k)

[inland monolith]

it may be in neither

[last locust]

^

[inland monolith]

alrighty that seems to be the simplest form, thank you

[fleet bridge]

!e

import _testcapi as t
print(t.the_number_three)

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

3

[floral meteor]

I think I get what whitespace is for now

it's to "water down" the code

so you can retain your sanity whilst reading it

[restive void]

Ithinkit'sthesamereasonwehavewhitespaceinEnglish:Yes,youcanalmostalwaysstilldecipherit,butit'sunnecessarilyhard.

[floral meteor]

what's so hard about my event loop?

english isn't colour coded ;)

now that i think about it i should've used the acknowledge character instead of \1

for some reason this line

char=='�':lambda:(c:=msvcrt.getch().decode('ANSI'))not in'KHMP'and(self.reset()or 1)or self.move(**{{'K':'a','H':'w','M':'d','P':'s'}[c]:1})

works on an older laptop, but not this newer laptop

so I have to use this line

char in'wasdqezx':lambda:self.move(w=char in'qwe',a=char in'qaz',s=char in'zsx',d=char in'edx')

for movement

[floral meteor]

alright I fixed it, spot the difference, and someone extremely experienced please tell me why

char in'�\xe0':lambda:(c:=msvcrt.getch().decode('ANSI'))not in'KHMP'and(self.reset()or 1)or self.move(**{{'K':'a','H':'w','M':'d','P':'s'}[c]:1})

the invalid unicode character is actually á

[floral meteor]

in the previous device, it was a different character

so in order for the same code to work on two different machines, that line has to look for two different characters

although now that I have fixed that plus a small visual glitch, I believe I have the game perfectly working

except maybe the problem of too much processing power available

in this snippet, the last line is an aid for speed running to prevent mistakes when hitting zero, and clearing a spot the user was gonna clear next.

            if not self.numbers[j][i]:
                _={self[m,n]==1 or self.select(m,n)for k,l in product(range(-1,2),repeat=2)for m,n in[(i+k,j+l)]if m in range(x)and n in range(y)}
                while msvcrt.kbhit(): msvcrt.getch()

without it, the user would instead attempt to clear the next square over, which was not the spot they were planning.
The problem with too much processing power is the above might run quicker than the user's reaction time, thus finishing before the next keyboard hit.
Slow it down a bit and the user can react to a 0-square clearing other squares and recalculate

aight i'm done someone else show me some amazing python torture

[errant crescent]

A nice exercise -

a[:], b = a = [[]], []
print(a)

Tell me what you get after running this

🙃

[trim grove]

#esoteric-python message

any1 knows about that?

[thorny tapir]

how tf would that work

!e

a[:], b = a = [[]], []
print(a)

[night quarry]

@thorny tapir :x: Your eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 1, in <module>
003 | NameError: name 'a' is not defined

[thorny tapir]

ok then

[frigid widget]

!e

try:
    a = a[:], b = [[]], []
except: pass
print(a)

[night quarry]

@frigid widget :white_check_mark: Your eval job has completed with return code 0.

([[]], [])

[frigid widget]

assignments in Python are left-to-right, a = b = c does a = c first and b = c second

[floral meteor]

Interestingly, the "intuitive order" is less intuitive than the "actual order"

[earnest wing]

a, b = a[b] = [{},()]

[earnest wing]

!e ```py
a, b = a[b] = [{},()]
print(a, b)

[night quarry]

@earnest wing :white_check_mark: Your eval job has completed with return code 0.

{(): [{...}, ()]} ()

[fleet bridge]

empty tuple is singleton

!e

x = ()
y = (1,2,3)[:0]
z = tuple()
print(x is y is z)

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

True

[floral meteor]

!e someone asked this question in a help channel, so I'm just gonna post here what I shouldn't post there #help-corn message

from __future__ import annotations
from ctypes import*
flogbals = py_object.from_address(id(globals())+8).value
__name__ = 'Throwable'

class TypeException(TypeError):
  def __init__(self, message):
    super().__init__(message)
    py_object.from_address(id(globals())+8).value=flogbals

class __annotations__(metaclass=lambda*a:type(*a)()):
  queue = []
  type_hints = {}
  def __setitem__(self, cls_name, name):
    if name in self.type_hints:
      self.queue[:]=()
      raise TypeException(f'Variable \'{name}\' already declared as \'{self.type_hints[name]}\'.')
    if not self.queue:
      self.type_hints[name] = globals()[cls_name]
      return
    cls, value = self.queue.pop(0)
    self.type_hints.update({name:cls})
    if cls==cls_name:raise NameError(f'Type {cls} not found')
    dict.update(globals(),{name:cls(value)})

class __globals__(dict):
  def __missing__(self, key):
    return __builtins__.__dict__.get(key,key)
  def __setitem__(self, key, value):
    if key in __annotations__.type_hints:
      dict.update(globals(), {key:__annotations__.type_hints[key](value)})
      return
    __annotations__.queue+=[(globals()[key], value)]


py_object.from_address(id(globals())+8).value=__globals__

str: x
str: y
int: z
x = 11
y = 10
z = x + y
print(z)

[night quarry]

@floral meteor :x: Your eval job has completed with return code 139 (SIGSEGV).

1110

[errant crescent]

Whoops, placed the slice in the wrong place... 👀

[floral meteor]

👀

[fleet bridge]

!e

class _ProxyGetter:
    def __ror__(self, other):
        return other

def _get_mappingproxy_dict(mp, *, __proxy_getter=_ProxyGetter()):
    return mp | __proxy_getter

del _ProxyGetter

_get_mappingproxy_dict(str.__dict__)['trim'] = str.strip

print("   hello w orld   ".trim())

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

hello w orld

[astral rover]

Hmm

Why does that work

[fleet bridge]

https://bugs.python.org/issue43838

[rugged sparrow]

!e ```py
def get_cls_dict(cls, E=type('',(),{'eq':lambda s,o:o})()):
return cls.dict == E

get_cls_dict(int)['inc'] = lambda self: self + 1
print(1 .inc())```

[night quarry]

@rugged sparrow :white_check_mark: Your eval job has completed with return code 0.

2

[rugged sparrow]

__eq__ works too

[trim grove]

!e

def get_cls_dict(cls, E=type('',(),{'__eq__':lambda s,o:o})()): return cls.__dict__ == E
get_cls_dict(int)['__ne__'] = get_cls_dict(int)['__eq__']
print((1).__eq__(1))
print((1).__ne__(1))
print(1==1)
print(1!=1)

[night quarry]

@trim grove :white_check_mark: Your eval job has completed with return code 0.

001 | True
002 | True
003 | True
004 | False

[trim grove]

any way to override the c func

[rugged sparrow]

use fishhook

!e ```py
from fishhook import *

@hook(int)
def add(self, other):
print(self, other)
return orig(self, other)

1 + 2```

[night quarry]

@rugged sparrow :x: Your eval job has completed with return code 1.

001 | 8 48
002 | 0 48
003 | 0 48
004 | 0 48
005 | 0 48
006 | 0 48
007 | 0 48
008 | 0 48
009 | 0 48
010 | 0 48
011 | 0 48
... (truncated - too many lines)

Full output: too long to upload

[rugged sparrow]

yea i need to fix that __eq__ bug (but its hard cause I wrote the hooker in python so it has to use the hooks)

v2 will fix it because that introduces a with no_hooks: context manager

[trim grove]

is this calculating the offsets at runtime

[rugged sparrow]

yea

[floral meteor]

Because error handling requires a reset state otherwise it segfaults before the improvised atexit kicks in

[floral meteor]

Error handling and exiting both crash "corrupted" python unless safeguards like these reset it

[rugged sparrow]

*you can also use some more complex hooks that don't break pythons built in garbage collector (thats what segfaults in the end)

[floral meteor]

My prevention is really a gamble that the garbage collector would come for my user created object before whatever builtin object got corrupted

[rugged sparrow]

@floral meteor this should show why changing a dict instances class to a basic dict subclass causes issues. The basic subclass thinks it larger, so it frees more memory than the original dict instance. By adding __slots__ you can help to nullify this as it makes the dict subclass have the same internal structure as the original dict class

[shut trail]

Quick question - why wouldn't #bot-commands message

change the value of x from 1 to 2?

or is that not how it works

[shut trail]

Ohh

Is an integer not a void pointer like most things? Or are most things not void pointers

[fleet bridge]

TIL except: catches BaseException, not Exception:

>>> try:
...   raise BaseException()
... except:
...   print('hi')
...
hi

[sick hound]

saw that on yt or someplace saying how to confuse a python dev

[thorny tapir]

oh so this is just
len((True, True, True, True == True, True, True, True))

so it finds True == True which is true so the length i s7
right?

[thorny tapir]

and this it only looks at the last 1 and checks if itss equal to the tuuple which its not

[rapid sparrow]

what sorcery is this? your python is possessed

[last locust]

It essentially evaluates as (1, 1, (1 == (1, 1, 1)))

[rapid sparrow]

oh wow, I wouldn't have guessed so

[last locust]

Yeah, it's funky lol

Down to precedence ig

[rapid sparrow]

so 1, 1, 1 isnt the same as (1, 1, 1) in this case?

[last locust]

It's not checking what you expect it to check

The point is that it does (1, 1, x) where x is 1 == (1, 1, 1) (it evaluates the == before "building" the left tuple)

And an int isn't equal to a tuple, so it evaluates to False

Hence (1, 1, False)

You could equally do 1, 1, 1 == 1 and you'd get (1, 1, True) for the same reason

The == is done before the tuple constructor

[last locust]

It's literally just precedence. It's like 10 - 3 * 2 is 4 (since multiplication is evaluated before substraction), even though reading left-to-right you'd expect it be 14

[gleaming timber]

!e

class Foo:
  def __init__(self):
    x = 42
    def get_x():
      return x
    def set_x(val):
      nonlocal x
      x = val
    self.get_x = get_x
    self.set_x = set_x

foo = Foo()
print(foo.get_x())
foo.set_x(3)
print(foo.get_x())
print(foo.x)

[night quarry]

@gleaming timber :x: Your eval job has completed with return code 1.

001 | 42
002 | 3
003 | Traceback (most recent call last):
004 |   File "<string>", line 16, in <module>
005 | AttributeError: 'Foo' object has no attribute 'x'

[gleaming timber]

private variables in python

[snow beacon]

!e ```py
class Foo:
def init(self):
x = 42
def get_x():
return x
def set_x(val):
nonlocal x
x = val
self.get_x = get_x
self.set_x = set_x

foo = Foo()
foo.set_x.closure[0].cell_contents = 10
print(foo.get_x())

[night quarry]

@snow beacon :white_check_mark: Your eval job has completed with return code 0.

10

[gleaming timber]

😮

[restive void]

Ew

[floral meteor]

You broke your own private variables system

[rapid sparrow]

lol

privacy breach

needs friend keyword

[floral meteor]

I have recently used |{...} to fix a syntax error

literally those 6 characters

in that order

[floral meteor]

I could use kivy, and then make one Image object, then the event loop would only take raw input from clicking, and draw everything onto an image to display that. 2D rendering maybe.
Then with just that as io i could make a touchscreen game

I'm gonna make another minesweeper, this time with an actual gui

but first I gotta black box the existing minesweeper so I can use it

anyone have esoteric tips on black boxing an interface the most cursed way possible?

I wanna get creative

[light peak]

!rank

[night quarry]

Iterating over range(len(...)) is a common approach to accessing each item in an ordered collection.

for i in range(len(my_list)):
    do_something(my_list[i])

The pythonic syntax is much simpler, and is guaranteed to produce elements in the same order:

for item in my_list:
    do_something(item)

Python has other solutions for cases when the index itself might be needed. To get the element at the same index from two or more lists, use zip. To get both the index and the element at that index, use enumerate.

[floral meteor]

You teaching people how to hack? :P

[sick hound]

lol is this the brainfuck python?

[fleet bridge]

https://media.discordapp.net/attachments/835020265301606420/929972462622482494/forest_1641793101.gif

[faint dirge]

huh

[snow beacon]

I'm emphasising to people that their security measures could use some work.

I'd hate for someone to get hacked, after all.

[upbeat sonnet]

this is why python is nice

!e```python
import sys, os

def restart():
os.execv(sys.executable, ['python'] + sys.argv)
return False

while True:
print(restart())
restart()```

it wont let me break it 😢

[floral meteor]

!e you probably don't need the brackets, but you are right

try:raise BaseException
except:print(':(')

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

:(

[floral meteor]

!e ```py
try:raise SystemExit
except:print(SystemExit)
try:raise KeyboardInterrupt
except:print(KeyboardInterrupt)

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 | <class 'SystemExit'>
002 | <class 'KeyboardInterrupt'>

[floral meteor]

bruh is this recent? I always remember having to explicitly reference BaseException to cover these two exceptions

[floral meteor]

I started making a proper gui version of minesweeper. I made the skeleton in Cell-Engine-Interface style but my brain is already fried :)

[sick hound]

!e !e
import sys, os

def restart():
os.execv(sys.executable, ['python'] + sys.argv)
return False

while True:
print(restart())
restart()

[night quarry]

@sick hound :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 1
002 |     !e
003 |     ^
004 | SyntaxError: invalid syntax

[sick hound]

!e
import sys, os

def restart():
os.execv(sys.executable, ['python'] + sys.argv)
return False

while True:
print(restart())
restart()

[night quarry]

@sick hound :x: Your eval job has completed with return code 2.

001 | Argument expected for the -c option
002 | usage: python [option] ... [-c cmd | -m mod | file | -] [arg] ...
003 | Try `python -h' for more information.

[sick hound]

!e @zinc aurora

[night quarry]

@sick hound :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 1
002 |     <@​683001325440860340>
003 |     ^
004 | SyntaxError: invalid syntax

[restive void]

Posted this in #advanced-discussion, but it probably fits better here:

import os
import sys
import inspect
from contextlib import contextmanager

@contextmanager
def atomic():
    ppid = os.getpid()
    cpid = os.fork()
    if cpid:
        os.waitpid(cpid, 0)
        # we didn't get killed: error occured, skip body:
        sys.settrace(lambda *a, **k:None)
        frame = inspect.currentframe().f_back.f_back
        def raiser(*a): raise RuntimeError
        frame.f_trace = raiser
        try:
            yield
        except RuntimeError:
            return
    try:
        yield
    except Exception:
        # rollback: we exit
        os._exit(0)
    else:
        # all good: kill state-preserved parent
        os.kill(ppid, 9)

x = 23
with atomic():
    x += 1
    1/0

print(x) 

[tiny lodge]

@restive void why can’t I access #advanced-discussion? Is it invite only?

[prisma coral]

It got renamed to #internals-and-peps recently, and L3v was using the old name

[tiny lodge]

Thanks @prisma coral

[restive void]

@tiny lodge so how it works is it first forks: this splits the execution in half across two "clones" of the same Python interpreter.
There parent process waits for the child process to exit. The child process enters the context manager's body and tries to execute it. If it succeeds, it just kills the parent process and continues normally.
If it fails, it kills itself.
In that case, the parent knows that the child died and is in the situation you desire a solution for: we are still "back in time", before the body of the context manager was entered, and we want to skip it. To do so, we set a trace function (the mechanism used by debuggers, coverage tools, etc.) to raise an exception inside the body immediately on entering it. We then catch that exception, and carry on.

[rugged sparrow]

ah raising an exception in the frame with a trace is clever, i was thinking about using a trace to jump lines

[restive void]

I stole it from StackOverflow

https://stackoverflow.com/questions/12594148/skipping-execution-of-with-block

[steel knot]

!e
print("haha xd")

[night quarry]

@steel knot :white_check_mark: Your eval job has completed with return code 0.

haha xd

[steel knot]

i dont python

[last locust]

#bot-commands please @steel knot

[floral meteor]

But beautifully esoteric way of going back in time in case of a failure. Just like saving your game, and going back to your last save if you die or accidentally kill an essential npc.

It's like code that can control time itself

Keep it up and you might even be able to code a Terminator that goes back in time to assassinate Hitler

[last locust]

Let's not be making suicide jokes please.

[floral meteor]

fair enough

I thought that might've been to far hence the /s

but now i know for sure, thank you

the assassinate hitler joke is good though?

!e okay how about this one? ```py
for()in(),{},[],'',"",range(0),iter(lambda:...,...),set():print(end='.')

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

........

[floral meteor]

instead of doing the same thing 8 times like

for i in range(8):print(end='.')

you can

for()in[()]*8:print(end='.')

it's shorter and confusinger

[fleet bridge]

!e

for {}[()] in range(5): print(end='.')

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

.....

[sick hound]

!e

await ctx.send("test?")

[night quarry]

@sick hound :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 1
002 | SyntaxError: 'await' outside function

[floral meteor]

!e ```py
for{}[()]in[{}]*8:print('something in between?')

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 | something in between?
002 | something in between?
003 | something in between?
004 | something in between?
005 | something in between?
006 | something in between?
007 | something in between?
008 | something in between?

[floral meteor]

probably #bot-commands for generic trying to hack the bot?

[sick hound]

No, that's not hacking the bot buddy.

[floral meteor]

this channel is more for torturing python itself

[sick hound]

That's a message to be sent.

Lul

[floral meteor]

yeah, that's obviously not gonna work

[sick hound]

It does on most python bots.

Thanks.

[floral meteor]

and this still isn't the right channel

[sick hound]

Please refer to the #argument channel then.

😉

[floral meteor]

what wrong with the #bot-commands channel?

[fleet bridge]

!e ```py
for[]in[[]]*8:print(end='.')

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

........

[fleet bridge]

i think it is shortest

[floral meteor]

indeed

well, assuming an abstract code block in place of print(end='.')

!e otherwise ```py
print('.'*8)

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

........

[fleet bridge]

😄

i meant for[]in[[]]*n: is shortest for-cycle over n elements

[floral meteor]

yes, that's another way of saying it

[fleet bridge]

yea)

[floral meteor]

but for{():[]}[()]in[()]*n: makes it deliciously more confusing for the beginner coder

[fleet bridge]

exactly

[sick hound]

[fleet bridge]

!e

a = a[0] = [None]
print(a)
assert a is a[0]

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

[[...]]

[floral meteor]

!e ```py
for{():[]}[()]in[()]*8:print(import('time').time())

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 | 1641988848.2122383
002 | 1641988848.2122772
003 | 1641988848.2122831
004 | 1641988848.2122874
005 | 1641988848.2122915
006 | 1641988848.2122958
007 | 1641988848.2122998
008 | 1641988848.2123039

@sick hound :white_check_mark: Your eval job has completed with return code 0.

Fatal  Failure

[floral meteor]

two spaces?
maybe make it a \b

[fleet bridge]

print(a, b) add one space

[sick hound]

!e

a = "Fatal"
b = "Failure"

print(a, b)

[night quarry]

@sick hound :white_check_mark: Your eval job has completed with return code 0.

Fatal Failure

[floral meteor]

!e ```py
print("Fatal",'\b',"Failure")

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

Fatal  Failure

[sick hound]



[floral meteor]

it works in a "real" terminal

[fleet bridge]

!e {}[()]={():[{()}]}

[night quarry]

@fleet bridge :warning: Your eval job has completed with return code 0.

[No output]

[proper vault]

ye, the message doesn't interpret control characters other than a newline

[sick hound]

\n

hm

[fleet bridge]

you can edit message

[sick hound]

You can?

[fleet bridge]

you can

[sick hound]

huh

[fleet bridge]

too late

also pls delete messages with errors

[sick hound]

bruh

[fleet bridge]

what are you doing??? go to #bot-commands

[sick hound]

Sorry, sir. This is not a general chat.

Thank you.

[night quarry]

@sick hound :warning: Your eval job has completed with return code 0.

[No output]

[fleet bridge]

cool

>>> [*[a,[*([b,[f],d],c)],*(*[(),e],)],[]] = 1,[(2,[3],4),5],(),7,()
>>> a,b,f,d,c,e
(1, 2, 3, 4, 5, 7)

>>> class PrintSetAttr:
...  def __setattr__(self, attr, value):
...   print(f'{attr}={value!r}')
...
>>> _ = PrintSetAttr()
>>> [*[_.a,[*([_.b,[_.f],_.d],_.c)],*(*[(),_.e],)],[]] = 1,[(2,[3],4),5],(),7,()
a=1
b=2
f=3
d=4
c=5
e=7

order of assignments is left-to-right

is there a way to convert any code to one-liner?

especially with and try-except

import -> __import__
for -> [list comprehension]
if -> a if c else b
class -> type(...)
def -> lambda
with -> type('',(contextlib.ContextDecorator,),{'__enter__': ..., '__exit__': ...})()(func)()
try -> ???
while -> [list comprehension with callavle iterator]

[proper vault]

the only thing I have yet to see done algorithmically is nonlocal

with is easy with ContextDecorator, try is just a context manager

while is just iter

while a: -> [_ for _ in iter(lambda: a,False)]

[fleet bridge]

can i do something like:

(cm := SomeContextManager(), x := cm.__enter__(), some_code, cm.__exit__())

i think it wont work

[proper vault]

no

you need the ContextDecorator thing

!d contextlib.ContextDecorator

[night quarry]

contextlib.ContextDecorator

class contextlib.ContextDecorator```
A base class that enables a context manager to also be used as a decorator.

Context managers inheriting from `ContextDecorator` have to implement `__enter__` and `__exit__` as normal. `__exit__` retains its optional exception handling even when used as a decorator.

`ContextDecorator` is used by [`contextmanager()`](https://docs.python.org/3/library/contextlib.html#contextlib.contextmanager "contextlib.contextmanager"), so you get this functionality automatically.

Example of `ContextDecorator`:

[fleet bridge]

type('',(contextlib.ContextDecorator,),{'__enter__': ..., '__exit__': ...})()(func)()

!e

from __future__ import annotations
from types import TracebackType
from typing import Any, Callable

from contextlib import ContextDecorator


class TryExceptContext(ContextDecorator):
    def __init__(
        self,
        *exceptions: tuple[
            type[BaseException],
            Callable[[type[BaseException] | BaseException], Any],
        ]
    ) -> None:
        self.exceptions = exceptions

    def __enter__(self) -> None:
        pass

    def __exit__(
        self,
        exc_type: type[BaseException],
        exc_value: BaseException,
        traceback: TracebackType,
    ) -> bool:
        if exc_type is None:
            exc_type = type(exc_value)
        for exc, fun in self.exceptions:
            if issubclass(exc_type, exc):
                fun(exc_value if exc_value is not None else exc_type)
                return True
        return False


catch_typeerror = TryExceptContext((TypeError, lambda exc: print(exc)))

catch_typeerror(lambda: 1 + 2)()
catch_typeerror(lambda: 1 + '')()
try:
    catch_typeerror(lambda: 1 / 0)()
except ZeroDivisionError:
    pass

catch_attr_and_name_error = TryExceptContext(
    (AttributeError, lambda exc: print(exc)),
    (NameError, lambda exc: print(exc)),
)
catch_attr_and_name_error(lambda: [].a)()
catch_attr_and_name_error(lambda: a)()
try:
    catch_attr_and_name_error(lambda: [][0])()
except IndexError:
    pass

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | unsupported operand type(s) for +: 'int' and 'str'
002 | 'list' object has no attribute 'a'
003 | name 'a' is not defined

[fleet bridge]

it works

def f():
  for x in y:
    if z(x):
      return a(x)
  return b

how to convert it to lambda?

(without messing with bytecode, of course)

[last locust]

lambda: a if any(z(x) for x in y) else b```

[fleet bridge]

oh

a is expression

it may contain x

[last locust]

Then a(x) instead of a 🤷

[fleet bridge]

you cant

lambda: a(x) if any(z(x) for x in y) else b is invalid

x is local to gen-expression inside any()

[proper vault]

f=lambda: next((a(x) for x in y if z(x)),b)

[fleet bridge]

nice!

[last locust]

lambda: c[0] if all(z==c[0] for z in (c:=[a(x) if z(x) else b for x in y])) else b
```I think this also works

But next() is definitely the nicer option

[fleet bridge]

no, x is local to this generator: (a(x) if z(x) else b for x in y)

[last locust]

Wait I sent the wrong one

Edited

Nvm sec

[fleet bridge]

>>> (lambda: x:=0)()
  File "<stdin>", line 1
    (lambda: x:=0)()
     ^^^^^^^^^
SyntaxError: cannot use assignment expressions with lambda

>>> lambda: c[0] if all(z==a(x) for z in (c:=[a(x) if z(x) else b for x in y])) else b
  File "<stdin>", line 1
SyntaxError: assignment expression cannot be used in a comprehension iterable expression

[last locust]

>>> (lambda: (x:=0))()
0```you just need parentheses

[fleet bridge]

interesting 🧐

[last locust]

Can I have a, b, x and y? @fleet bridge

[last locust]

This is close but might not quite be right

[fleet bridge]

y = [1,2,3,4,5]
z = lambda x: x % 2 == 0
# z = lambda x: False # for returning b
a = lambda x: x*x
b = 0

[last locust]

Thanks

[fleet bridge]

(lambda:t[0]if(t:=[a(x)for x in y if z(x)])else b)()

[last locust]

Yeah, that could be what I was going for

And then you can remove unnecessary spaces etc.

(lambda:t[0]if(t:=[a(x)for x in y if z(x)])else b)()

[fleet bridge]

how to convert match-case into one-liner? 😄

I have seen a way to dynamically generate patterns and match them in early drafts, iirc

[floral meteor]

not sure about all the dynamiccy stuff but...

{caseA:lambda:...,caseB:lambda:...,caseC:lambda:...}.get(True,lambda:...)()

that's how I do it, even in preference to the match case suite

[fleet bridge]

it is switch-case, not pattern matching actually

[floral meteor]

for example if i want multiple keybinds to do certain functions, just observe the core of my minesweeper event loop

            (lambda i,j:{
char in'\t/?':lambda:self.right_click(i, j),
char in'\b\xff':self.reset,
char in' \r':lambda:self.react_select(i, j),
char in'wasdqezx':lambda:self.move(w=char in'qwe',a=char in'qaz',s=char in'zsx',d=char in'edx'),
char in'[\x1b':lambda:self.DIE(True),
char in'�\xe0':lambda:(c:=msvcrt.getch().decode('ANSI'))not in'KHMP'and(self.reset()or 1)or self.move(**{{'K':'a','H':'w','M':'d','P':'s'}[c]:1}),
char in'h;:':self.hint,
char in'rl':self.random_move,
char.isdigit():lambda:self.digit_move(int(char))
            }.get(1,lambda:0)()or self.update_board())(*self.here)

[fleet bridge]

hmm

!e print({1:2,1:3})

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

{1: 3}

[floral meteor]

exactly

[fleet bridge]

yes, it implements some flavor of pattern-matching

[floral meteor]

the only "flavor" I really would ever need to use

the python version of ```matlab
switch s
case 1:
do_this()
case 2:
do_that()
case 3:
print('wheeee')
otherwise:
0
end

would be
```py
{1:do_this,2:do_that,3:lambda:print('wheeee')}.get(s,lambda:0)()

switching True can be useful for substituting extended if statements

keep in mind if all the middle ifs are elifs, your dict would actually be backwards in the one-line version

another example is where I sum a list excluding elements between and including 6 and 9

def summer_69(a):
  state = [0]
  output = [0]
  for e in a:
    {
state[0]:lambda:...,
state[0]and e==9:lambda:state.__setitem__(0,0),
not state[0]:lambda:output.__setitem__(0,output[0]+e),
not state[0]and e==6:lambda:state.__setitem__(0,1)
    }[1]()
  return output[0]

[fleet bridge]

it is sadly that you cant do a[b]:=c or a.b:=c

[floral meteor]

you can a.__setitem__(b,c) and a.__setattr__(b,c)

[fleet bridge]

yes, but it is ugly imo

[floral meteor]

or even juicier

[0 for a[b]in[c]]
[0 for a.b in[c]]

[fleet bridge]

lol

[floral meteor]

anticlimactic loops can modify mutable objects

lambda a:((state:=[0]),(output:=[0],[{state[0]:lambda:0,state[0]and e==9:lambda:[0 for state[0]in[0]],not state[0]:lambda:[0 for output[0]in output[0]+e],not state[0]and e==6:lambda:[0 for state[0]in[1]]}[1]()for e in a]))[1][0]

that's the oneliner version of summer_69

using a state machine to get a binary state and the input to decide on modifying the output state and the binary state

and using anticlimactic loops for variable modification

and that's how I do switch suites

[fleet bridge]

import -> __import__
for -> [list or set comprehension]
if -> a if c else b
class -> type(...)
def -> lambda
with -> type('',(contextlib.ContextDecorator,),{'__enter__': ..., '__exit__': ...})()(func)()
try -> custom context manager based on ContextDecorator
while -> [list or set comprehension with callable iterator]
assignment -> := or [0 for a[b]in[c]] or [0 for a.b in[c]]
match-case -> {cond3:f3,cond2:f2,...}.get(True,fdefault)()
switch-case -> {val1:f1,val2:f2,...}.get(val,fdefault)()

[floral meteor]

don't forget
switch-case-default -> {casen:fn, case3:f3,case2:f2,case1:f1}.get(switched_variable,f_otherwise)()

[fleet bridge]

this is interesting

[floral meteor]

and you got your conds backwards if you want it to execute f1 in priority to f2 if cond1 and cond2

also for the while, it's better on memory if you use set comprehension otherwise you cannot truly make an infinite while loop

[floral meteor]

I could make this more cursed

lambda a:((state:=[0]),(output:=[0]),[{state[0]:lambda:0,state[0]and e==0b1001:lambda:[0 for state[0]in[0]],not state[0]:lambda:[0 for output[0]in output[0]+e],not state[0]and e==0b0110:lambda:[0 for state[0]in[1]]}[1]()for e in a])[1][0]

[fleet bridge]

you cant make it more cursed than this

[floral meteor]

yeah you should see one of the cursed hello worlds way back a few years ago in this channel

trying to understand it requires immense knowledge, and even then that's just to write a deobfuscator to make it a little bit more readable, then you need advanced python knowledge and algorithmic understanding just to read that

[floral meteor]

compared to that, this is the most elegant solution to the given problem

woops forgot a bracket, fixed

bracket matching is easy to stuff up in this line of onelinerisation

[fleet bridge]

!e import __hello__

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

Hello world!

[floral meteor]

obfuscation is an entirely different field which is fun but tiresome for me

onelinerisation with just a touch of cursed is my comfort zone

but just a touch of cursed is present in all of my code

!e ```py
print((lambda a:((state:=[0]),(output:=[0]),[{state[0]:lambda:0,state[0]and e==0b1001:lambda:[0 for state[0]in[0]],not state[0]:lambda:[0 for output[0]in[output[0]+e]],not state[0]and e==0b0110:lambda:[0 for state[0]in[1]]}1for e in a])[1][0])([1,2,3,4,5,6,7,8,9,1,1,1,1]))

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

19

[dense nova]

i saw this video 'calculate electricity bill in c' and im like: 'I'll do that in 1 line in python'
Behold: ```py
print(f"Electricity bill: {(add := 0 if (units := int(input('Enter number of units you used: '))) < 50 else 130 if units <= 100 else 292.5 if units <= 200 else 818.5) + (((units - (0 if units <= 50 else 50 if units <= 100 else 100 if units <= 200 else 200))) * (2.6 if units < 50 else 3.25 if units <= 100 else 5.26 if units <= 200 else 7.75)) + (25 if add == 0 else 35 if add == 130 else 45 if add == 292.5 else 55)}")

la code

[floral meteor]

!e print(1+2+3+4+5+4)

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

19

[floral meteor]

okay this code actually works the way i intend it to

[dense nova]

what is this supposed to do???

[dense nova]

.....

[floral meteor]

the last two evaluations i did both added up to 19

the second one summed the list, but without the elements between and including 6 and 9

the first one did that algorithmically, taking the whole list as input

!e ```py
á, 🤔, 🪟 = 0, False, []
print(dir())

[night quarry]

@floral meteor :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 1
002 |     á, 🤔, 🪟 = 0, False, []
003 |        ^
004 | SyntaxError: invalid character '🤔' (U+1F914)

[floral meteor]

dammit

[fleet bridge]

!e print('🤔')

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

🤔

[dense nova]

and

[floral meteor]

thonk

you made an AI that does nothing but think

[fleet bridge]

not all emoji are allowed to be in names

[dense nova]

the python variable naming rules do state that the variable name cannot have any special characters except _ but any ascii character can be in a sting

[floral meteor]

or does it?

[fleet bridge]

unicode lambda can be in names

!e λ = ''; print(dir())

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

['__annotations__', '__builtins__', '__doc__', '__loader__', '__name__', '__package__', '__spec__', 'λ']

[next flame]

any Unicode letters and numbers can be in a variable name

see https://docs.python.org/3/reference/lexical_analysis.html#identifiers

and certain punctuation too I think

[sick hound]

.topic

[glass drum]

**What are the pros and cons of messing with __magic__()?**

Suggest more topics here!

[unique rose]

.topic

[glass drum]

**What's a common part of programming we can make harder?**

Suggest more topics here!

[severe canyon]

esoteric in a nutshell

[dense nova]

what exactly is magic?? is it the dunder methods?

[restive void]

Yes, the bot tried to say __magic__, but forgot the backticks

[fleet bridge]

>>> True, False = False, True
>>> True, False
(False, True)

[restive void]

Only in deprecated Python 2

[languid hare]

gross

in a good way

[gaunt ferry]

how to handle null and none values in pyspark dataframe, glue dynamic frame after reading CSV files

[fleet bridge]

>>> id([]) == id([])
True
>>> [] is []
False

[fleet bridge]

>>> id([[]]) == id([[]])
False

[proper vault]

try id([1]) == id([2])

[dawn stone]

a = b = c = d = "hi"

[floral meteor]

try mixing assignment chains with implicit tuple construction and unpacking

that's how I easily make code look initimidating

[floral meteor]

this only has one builtin: print
the rest is literals, variables, keywords and syntax-sugar

I only just noticed that and i wrote it 🤔

which means it's a universal algorithm, it can port to another language without looking too different

[night quarry]

Hey @sick hound!

It looks like you tried to attach a Python file - please use a code-pasting service such as https://paste.pythondiscord.com

[sick hound]

This here, https://paste.pythondiscord.com/tedeciluho.apache

Prints "hello".

[restive void]

What do you need math for?

[arctic elm]

I was about to start decoding the base64 and dive down that rabbit hole...but then I came to my senses

[night quarry]

@sick hound :x: Your eval job has completed with return code 1.

001 |   File "<string>", line 22
002 | SyntaxError: 'named expression' can not be used within an annotation

[rich hound]

!e

print((len("RATIO")**(int([]==[])/int(int("OwO"!="XwX")*len(":3")))+int(len("Furry")-4))/int(len("uwu")-1))

[night quarry]

@rich hound :white_check_mark: Your eval job has completed with return code 0.

1.618033988749895

[rich hound]

math

[night quarry]

@sick hound :x: Your eval job has completed with return code 139 (SIGSEGV).

5:3.300000

[simple crystal]

can you make type(a) is int without messing with pointers?

when a isn't an int

[restive void]

type = lambda*_:int

[simple crystal]

lol good point

[earnest wing]

Incredible

I really like the reversed type annotations

[solemn jungle]

!e

n,l=[1000,900,500,400,100,90,50,40,10,9,5,4,1],'M CM D CD C XC L XL X IX V IV I'.split()
for N in range(1,4000):
 r='';p=N
 for x,y in zip(n,l):
  while p>=x:r+=y;p-=x
 exec('%s=%d'%(r,N))

print(CDXX+MCCCVII-MDCLXXXVIII)

[night quarry]

@solemn jungle :white_check_mark: Your eval job has completed with return code 0.

39

[solemn jungle]

!e

n,l=[1000,900,500,400,100,90,50,40,10,9,5,4,1],'M CM D CD C XC L XL X IX V IV I'.split()
for N in range(1,4000):
 r='';p=N
 for x,y in zip(n,l):
  while p>=x:r+=y;p-=x
 exec('%s=%d'%(r,N))

print(CDXX+MCCCVII-MDCLVIII)

[night quarry]

@solemn jungle :white_check_mark: Your eval job has completed with return code 0.

69

[restive void]

Now change the __repr__ of int accordingly

[floral meteor]

bugger that, don't use int, use an int subclass like booleans do

[marsh void]

roman type

haha

[floral meteor]

obviously, the __str__ method would be the algorithm to convert an integer to roman numerals

class Numeral(int):
  def __repr__(self):
    ...
  def __init__(self, *args):
    super().__init__(*args)
    self.__str__ = self.__repr__

[floral meteor]

alright I thought of a challenge. shorten and expand the functionality of this oneliner: ```py
class SirPrintalot:add=sub=mul=floordiv=truediv=mod=getattribute=getitem=setattr=setitem=pow=and=or=radd=rsub=rmul=rfloordiv=rtruediv=rmod=rpow=rand=ror=rxor=xor=iadd=isub=imul=ifloordiv=itruediv=imod=ipow=iand=ior=ixor=call=lt=eq=neq=gt=ge=le=req=rneq=rlt=ilt=rle=ige=rge=ieq=ineq=ile=rgt=igt=print;pos=lambda s:print(end=' ');invert=lambda s:print();neg=lambda s:print(' ');str=lambda s:print("print('...')")or'print(...)'

ideally, anything you can do to an instance prints something

[fleet bridge]

__repr__, not __str__

object.__str__ uses object.__repr__

[restive void]

Maybe use a metaclass' getattr?

[fleet bridge]

no, it wont initialise method's slots

x.__getitem__(...) will work, but x[...] wont

[cloud fossil]

Why does this work

[solemn jungle]

you mean "how does it work?"

it converts every number from 1 to 4000 to roman numerals

then the exec function execute like MMXXII=2022

[weak shuttle]

@solemn junglehelp me in #discord-bots

[solemn jungle]

sorry i dont code bots

[cloud fossil]

Oh I thought just the digits

[sick hound]

To convert some binary into text

[sly root]

genius, f-string that has only one item

[fleet bridge]

>>> dis('f"{x}"')
  1           0 LOAD_NAME                0 (x)
              2 FORMAT_VALUE             0
              4 RETURN_VALUE

[solemn jungle]

!e

exec("""π=print;ρ=range

φ=λξ:1ηξ<2εφ(ξ-1)+φ(ξ-2)
ζνορ(20):π(φ(ν))

""".replace('λ','lambda ').replace('η',' if ').replace('ε',' else ').replace('ζ','for ').replace('ο',' in '))

[night quarry]

@solemn jungle :white_check_mark: Your eval job has completed with return code 0.

001 | 1
002 | 1
003 | 2
004 | 3
005 | 5
006 | 8
007 | 13
008 | 21
009 | 34
010 | 55
011 | 89
... (truncated - too many lines)

Full output: https://paste.pythondiscord.com/duvuyajeya.txt?noredirect

[languid hare]

i mean

i guess

[rich hound]

!e

print((len("RATIO")**(int([]==[])/int(int("what the fuck"!="am i doing")*len("with")-2))+int(len("my life")-6))/int(len("...")-1))

[night quarry]

@rich hound :white_check_mark: Your eval job has completed with return code 0.

1.618033988749895

[sly root]

>>> dis("x")
1           0 LOAD_NAME                0 (x)
            2 RETURN_VALUE

[fleet bridge]

f'{x}' is equal to format(x) (str(x) in most cases), while x just gets reference

[errant crescent]

print(
   (5
   ** (1 / int(1 * 4 - 2))
   + 7 - 6 )
   / int(3 - 1)
)

==

print(
( sqrt(5) + 1 ) / 2
)

==

1.618033...

[rich hound]

yes

well done

i knew what it did

just wanted to over complicate it

[errant crescent]

And this was all using the phrase "RATIO what the fuck am i doing with my life" 🙃

[rich hound]

yes

[errant crescent]

[rich hound]

but its true

[errant crescent]

Having an exclamation mark(!) would have made it even more hilarious.

[rich hound]

where.

[torn oak]

# Create a program that will allow a user to enter in ten numeric test grades. Then it drops the lowest two test grades and finds the average. 
# 
# Requirements:
# 
# Prompt the user for ten test scores from 0 - 100
# Do not allow numbers out of that range
# Drop the lowest two scores and find the average
# Output the average to the user.
# i need to do that

print(sum(sorted([(a if ((a := int(input(f"Enter your {i + 1}{'th' if i + 1 > 3 else ('rd' if i + 1 == 3 else ('nd' if i + 1 == 2 else 'st'))} number > ")))) <= 100 else exit("Numbers over 100 are not allowed.")) for i in range(10)])[2:])/8)```

uhhh clean, I never really used .get tbh

I love it

[sick hound]

I love this channel

[errant crescent]

Use a dictionary instead of multiple if-statments -

input( f"Enter your {i}{{1: 'st', 2:'nd', 3:'rd'}.get(i, 'th')}" ... for i in range(1, 11) )

Oh, lol. You already answered that Q.Q

[fleet bridge]

ive implemented stdout coloring

[solemn knoll]

dangg thats cool, how did u do that?

[fleet bridge]

1. wrapped sys.stdout
2. used pygments, my style and lexer to color text
3. changed sys.displayhook to pprint

i can share code, but it is really bad

[finite blaze]

What's interesting going on in the esoteric python world currently?

Tldr, what are you guys working on?

[indigo dock]

any alternatives to OS and Subprocess libraries?

[proper vault]

for what purpose?

[rapid sparrow]

well I successfully screwed myself over

>>> import main
KeyError called: args=('main',) kwargs={}
Calling my_excepthook on KeyError
Traceback (most recent call last):
  File "<frozen importlib._bootstrap>", line 175, in _get_module_lock
OSError called: args=(2, 'No such file or directory', '<frozen importlib._bootstrap>') kwargs={}
Calling my_excepthook on OSError
Traceback (most recent call last):
  File "/data/media/0/pythonrc.py", line 433, in my_excepthook
    return old_excepthook(exctype, value, traceback)
  File "<stdin>", line 1, in on_exception
OSError called: args=(2, 'No such file or directory', '<stdin>') kwargs={}
Calling my_excepthook on OSError
...
File "<stdin>", line 1, in on_exception
Fatal Python error: _Py_CheckRecursiveCall: Cannot recover from stack overflow.
Python runtime state: initialized

Current thread 0x0000007ff7ce7010 (most recent call first):
  File "<stdin>", line 1 in <lambda>
  File "/data/media/0/pythonrc.py", line 433 in my_excepthook
  File "<stdin>", line 1 in on_exception
  File "<stdin>", line 1 in <lambda>
  File "/data/media/0/pythonrc.py", line 433 in my_excepthook
  File "<stdin>", line 1 in on_exception
  File "<stdin>", line 1 in <lambda>
File "/data/media/0/pythonrc.py", line 433 in my_excepthook
  File "<stdin>", line 1 in on_exception
  File "<stdin>", line 1 in <lambda>
  ...
zsh: abort (core dumped)  python3

that was my attempt to hook the __init__ method of all exception types

so I won't lose exceptions that get thrown in async code

🤦‍♂️

using fishhook

now it's not saved my python histfile.. serves me right for trying to do something with exceptions, I suppose

[languid hare]

oof

[rapid sparrow]

whenever I try to do something esoteric with exceptions or reprs, this is inevitably what awaits

lol

[languid hare]

awaits

ba dum tss

[fleet bridge]

!e

z,n,o=' \n1';[(print('#'*a+' '*b,end=c))for a,b,c in zip(map(int,'115113113414'+o*16+'54'+o*9+'4'+o*22+'55531131154'),map(int,'2004552111402044424202020420014424000201420204442400020204202000160212000'),z*11+n+z*15+n+z*14+n+z*16+n+z*12+n,)]

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | #   # ##### #     #      ###      #   #  ###  ####  #     ####
002 | #   # #     #     #     #   #     #   # #   # #   # #     #   #
003 | ##### ####  #     #     #   #     # # # #   # ####  #     #   #
004 | #   # #     #     #     #   #     # # # #   # #   # #     #   #
005 | #   # ##### ##### #####  ###       # #   ###  #   # ##### ####

[fleet bridge]

🥳

[floral meteor]

then this is the right channel for you

[floral meteor]

it took me a while but i got it.

!e ```py
{
print(f"assigned {x} to x") for x in [69]
}

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

assigned 69 to x

[floral meteor]

actually I think that has it's own local namespace, so...

!e ```py
for x in [69]:{
print(f"assigned {x} to x")
}
print(x, 'is still assigned to x')

[night quarry]

@floral meteor :white_check_mark: Your eval job has completed with return code 0.

001 | assigned 69 to x
002 | 69 is still assigned to x

[floral meteor]

this is a subtle way to avoid =

hol' up

I guess that's what happens when i import externally located code

i didn't know onedrive could be a module

this is almost comical

[languid hare]

lmao

[languid hare]

I had to do a double take on those braces

nice

[vague cairn]

I don't care how bad it looks, I want to see your code as a reference implementation, even if afterwards I try to golf it or make it pythonic or anything. 😉

[night quarry]

Hey @fleet bridge!

It looks like you tried to attach a Python file - please use a code-pasting service such as https://paste.pythondiscord.com

[fleet bridge]

https://paste.pythondiscord.com/owawupexip.py @vague cairn

it is sitecustomize.py

also it slows python startup time from ~25ms to ~100ms

[vague cairn]

If I import it, from the interpreter it changes the color of the prompt, and pretty much blocks output. what am I doing wrong? you want me to drop it in /site_packages under the name sitecustomize.py?

[fleet bridge]

you want me to drop it in /site_packages under the name sitecustomize.py?
yes

because it is code for sitecustomize module

[vague cairn]

Cool, that got it.

I removed TypeGuard and underscore_numbers=True, to get it to work on my version of python, but anyway. Neat stuff.

[vague cairn]

And I added: ```py
if not 'idlelib' in sys.modules:
print('sitecustomize.py: Not in idle, console highlight enabled.')

on line 6 and indented the rest...

[arctic elm]

I hope this is the right kind of crowd for this old thing of mine
https://codeforces.com/contest/1154/submission/52853290

[devout quest]

If anyone here is knowledgeable in obfuscating Python, please DM me, I have a few questions I would like to ask c:

[last locust]

You should just ask here

Then everyone can see and people will know if they're able to help

[earnest wing]

Perfect

[torn oak]

I would just use pyarmor

[sick hound]

hello guys

i'm new to Python and i am assigned a task to extract time from a given time range this is as follows:

lets say a person is working in a company he started a work at start_time and end_time between these duration i have to extract how much hours does this man has worked during morining hours, evening hours and night hours.

[glad spade]

Hi Guys

[fleet bridge]

@zinc aurora crossposting here 👆

[snow beacon]

This isn't really the right channel for getting help. (Trust me: you don't want the kind of 'help' this channel would provide.) Instead, you should write code until you get stuck, then look in #❓｜how-to-get-help to see where you can get help with your task.

[floral meteor]

yeah this probably is the worst place for someone new to python to be. This place can give you nightmares if you're not equipped with advanced knowledge of the inner workings of python

[errant crescent]

This isn't the perfect channel for asking questions. Because there are many "esoteric" codes here, which rarely follow the standards for python.

Anyhow, what you're probably looking for are
• timedelta and datetime from the module datetime,
• along with maybe strftime from the time module.

These would help with managing and parsing the time frames.

[restive void]

s/deltatime/timedelta/

[errant crescent]

Yes. I always seem to get confused with them...

also, a quick esoteric code for validating inputs -

while not (k := input("...")).isnumeric():
   print("Invalid Input")

print(f"{k=:^8}is a number!")

[torn oak]

!e
k = 2
print(f"{k=:^8}is a number!")

[night quarry]

@torn oak :white_check_mark: Your eval job has completed with return code 0.

k=   2    is a number!

[torn oak]

!e
print(f"{'test'=:^13}")

[night quarry]

@torn oak :white_check_mark: Your eval job has completed with return code 0.

'test'=    test     

[torn oak]

@errant crescent what does that do , can u maybe explain? D:

[prisma coral]

@torn oak The k= writes the literal text of k= into the string, and then put the value of k after. The 8 says how wide the section after that = sign should be (so the 2 bit is 8 chars long). Finally, the ^ says to center that value of k within that 8 char section

[torn oak]

!e

print([a := '', [(a := f'{a}{i}{__import__("random").choice(["+", "-", "/", "*"])}') for i in [__import__("random").randint(1, 9) for _ in range(5)]], a[:-1]][2])```

[night quarry]

@torn oak :white_check_mark: Your eval job has completed with return code 0.

9*2*7+3/6

[torn oak]

!e

print([a := '', [(a := f'{a}{i}{__import__("random").choice(["+", "-", "/", "*"])}') for i in [__import__("random").randint(1, 9) for _ in range(10)]], a[:-1]][2])```

[night quarry]

@torn oak :white_check_mark: Your eval job has completed with return code 0.

9-7*3/3-2+9-5/4-8*8

[torn oak]

it works yay

[wanton flame]

Any tips for how I can further golf this? (not counting variable names)

    def line(x0: int, y0: int, x1: int, y1: int) -> Generator[int]:
        """
        Naive solution
        Bresenham's Line Algorithm, used for diagonals
        """
        deltax = x1 - x0
        dxsign = int(abs(deltax) / deltax)
        deltay = y1 - y0
        dysign = int(abs(deltay) / deltay)
        deltaerr = abs(deltay / deltax)
        error = 0
        y = y0
        for x in range(x0, x1, dxsign):
            yield x, y
            error = error + deltaerr
            while error >= 0.5:
                y += dysign
                error -= 1
        yield x1, y1

which I've cut down to:

r_,i_,a_=range,int,abs # used elsewhere in not shown code, so is a reduction
def _L(x0,y0,x1,y1,e=0):
  d=[x1-x0,y1-y0]
  de=a_(d[1]/d[0])
  d=[i_(a_(_)/_)for _ in d]
  for x in r_(x0,x1,d[0]):
    yield x,y0
    e+=de
    while e>=0.5:
      y0+=d[1]
      e-=1
  yield x1,y1

[errant crescent]

pretty much. although I was wondering whether to use < or ^ for the padding.

[errant crescent]

Use magic math to shrink things down -

def line(x0: int, y0: int, x1: int, y1: int) -> Generator[int]:
        """
        Naive solution
        Bresenham's Line Algorithm, used for diagonals
        """
        deltax = x1 - x0
        dxsign = int(abs(deltax) / deltax)
        
        deltay = y1 - y0
        dysign = int(abs(deltay) / deltay)
        
        deltaerr = abs(deltay / deltax)
        
        const_err = floor(deltaerr)
        deltaerr %= 1
        
        y = y0
        for x in range(x0, x1, dxsign):
            yield x, y
            
            y += dysign * (const_err + floor(deltaerr * i))
        yield x1, y1 

def line_compact(x0: int, y0: int, x1: int, y1: int) -> Generator[int]:
        """
        Naive solution
        Bresenham's Line Algorithm, used for diagonals
        """
        
        delta: float = abs((y1 - y0) / (x1 - x0)))
        const_err: int = floor(deltaerr)
        delta %= 1
        
        # from numpy import arrange
        
        cnv = (1, -1)
        x_axis = range(x0, x1, cnv[x0 > x1])
        y_axis = [floor(n + const_err + delta * i) for i, n in enumerate(range(y0, y1, cnv[y0 > y1]))]
        
        for x, y in zip(x_axis, y_axis):
            yield x, y     
        yield x1, y1 

[errant crescent]

Although when I look at it now ; I can see that there is no need for const_err, just use the entire delta instead.

You could also do x1+1 and y1+1 to include them in the range. However this may offset the final point.

@wanton flame

from math import floor

def line_compact(x0: int, y0: int, x1: int, y1: int) -> Generator[int]:
        """
        Naive solution
        Bresenham's Line Algorithm, used for diagonals
        """
        
        delta: float = abs((y1 - y0) / (x1 - x0)))
        # from numpy import arrange
        
        cnv = (1, -1)
        x_axis = list( range(x0, x1, cnv[x0 > x1]) )
        y_axis = [floor(cnv[y0 > y1] + delta * i) for i range(len(x_axis))]
        
        yield from zip(x_axis, y_axis)
        yield x1, y1 

oh right! I was so focused on the math, I didn't notice O.o

nvm

discord is wrapping my code, making things look funky

for a sec I though I didn't remove the old loop.

what elements?

Hmm...

Oh, I know why.
I accidentally though the constraints of error were between 0 - 1
after checking again, I see that it can go between -0.5 - 1.

Edit: It was also using separate values for both lists. e.g. x axis could be 6 elements long, while y axis might only be 4. Hence, the difference in the final output length.

give me a sec. imma fix this.

[wanton flame]

the algo is known to not work with 'steep' lines, for this data we can assert that all input will have a slope of 1 or -1 aka pure diagonal

[torn oak]

working hopefully soon™️

[torn oak]

print([a := int(input('Amount of numbers > ')), nums := [str(__import__("random").randint(1, 9)) for _ in range(a)], combo := list(__import__('itertools').product(["/", "*", "+", "-"], repeat=a)), nums :=  list(__import__('itertools').permutations(nums)), f := [([]) for _ in range(len(combo))], [[[(f[i].append([(l,n) for l,n in zip(n, c)])) for c in combo] for n in nums] for i in range(len(combo))], f := [''.join([list(__import__('itertools').chain.from_iterable(b))][0])[:-1] for b in list(__import__('itertools').chain.from_iterable(f))], g := [], r := [], generated := [], [([g.append(s := eval(b)), r.append([s, b])] if type(eval(b)) == int else '') for b in set(f)], [((generated.append(h)) if (h > 0 and r.count(h) == 1) else '') for h in set(g)], [sorted(set(g)), sorted(r), generated]][-1])```

not finished D:

[errant crescent]

!e
k = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.8]
n = [(int(i), i) for i in k]
print(n)

[errant crescent]

Well, Its sort of fixed for now... it should work for ±1 slopes, and it does get pretty close to the original logic.

I think this is as far as I could get writing on the phone, without even a pen or a piece of paper

In order to get the exact logic,
You might need to round the initial deviation then shift everything by -0.5 (to fix the original -1 offset, changing the difference from -0.5 - 0.5 to -1 - 0).

Afterwards either divide by 0.5 or something, to get the usual offset that would happen if you had used the original logic.

Finally substract that from the calculations in the loop, or do some other operations.
(Some bitshifts operations and floor division might be of use).

[golden finch]

how can I get the current frame in one expression w/o imports?

[fleet bridge]

__import__('sys')._getframe(0)

[sly root]

def l(a): #x:a[0], y:a[1], x1:a[2], y1:a[3], e:a[4]
 for x in range(a[0],a[2],int(abs((b:=a[2]-a[0]))/b)):
  yield x,a[1];a[4]+=abs((d:=a[3]-a[1])/b)
  while a[4]>=0.5:a[1]+=int(abs(d)/d);a[4]-=1
 yield a[2],a[3] 

for i,b in l([7,6,1,2,0]):
...   print(i,b)
...
7 6
6 5
5 5
4 4
3 4
2 4
1 2```

[arctic elm]

is this even so esoteric? seems like a reasonable use of walrus

[torn oak]

print([a := int(input('Enter amount of nums > ')), nums := [str(__import__("random").randint(1, 9)) for _ in range(a)], combo := list(__import__('itertools').product(["/", "*", "+", "-"], repeat=a)), '''nums :=  list(__import__('itertools').permutations(nums))''', f := [([]) for _ in range(len(combo))], [[(f[i].append([(l,n) for l,n in zip(nums, c)])) for c in combo] for i in range(len(combo))], f := [''.join([list(__import__('itertools').chain.from_iterable(b))][0])[:-1] for b in list(__import__('itertools').chain.from_iterable(f))], g := [], r := [], generated := [], [([g.append(s := eval(b)), r.append([s, b])] if type(eval(b)) == int and eval(b) > 0 else '') for b in set(f)], [((generated.append(h)) if (h > 0 and r.count(h) == 1) else '') for h in set(g)], [sorted(set(g)), sorted(r), generated]][-1])```

[golden finch]

literally uses imports

without imports

[languid hare]

!e

class MagicNamespace(dict):
    _getFailure = object()
    MagicType = object()
    
    def __init__(self, initial=None):
        self._dict = initial or {
            "__name__": "Magic Namespace"
        }

    def __setitem__(self, key, value):
        self._dict[key] = value
            
    def __getitem__(self, key):
        get_res = self._dict.get(key, MagicNamespace._getFailure)
        if (get_res == MagicNamespace._getFailure):
            self._dict[key] = self.MagicType
            
        return self._dict[key]
    
class MagicEnumMeta(type):
    @classmethod
    def __prepare__(mcs, name, bases):
        return MagicNamespace()
        
    def __new__(mcs, name, bases, namespace, **kwargs):
        
        enum = super().__new__(mcs, name, bases, namespace, **kwargs)
        enum.__repr__ = lambda self: self._key
        
        if not isinstance(namespace, MagicNamespace):
            return enum
        
        for key, value in namespace._dict.items():
            if value is MagicNamespace.MagicType:
                inst = enum()
                inst._key = key
                setattr(enum, key, inst)
        return enum
    
class MagicEnum(metaclass=MagicEnumMeta):
    def __init__(self):
        raise NotImplementedError("Enum Instantiation")
    
class Colors(MagicEnum):
    RED
    BLUE
    GREEN
    
print(Colors.RED == Colors.BLUE) #False
print(Colors.GREEN == Colors.GREEN)  #True

[night quarry]

@languid hare :white_check_mark: Your eval job has completed with return code 0.

001 | False
002 | True

[languid hare]

imagine using enum auto()

[golden finch]

this is very cheaty but I'll allow it

what's the 100th subclass of object on your machine?

I have never even encountered that before

nice loophole though

[torn oak]

I love oneliners

its unreal this is so hot

[golden finch]

love how they added whitespace to this abominations

for research purposes I need an expression containing no lambdas which gives me functionType

[fleet bridge]

it returns wrong frame

it returns frame of generator expr, not outer frame

!e

import sys
print(next(getattr(c, f).__globals__['sys']._getframe(0) for c in ().__class__.__base__.__subclasses__()for f in dir(c)if isinstance(getattr(c, f, None), type(lambda: None)) and'sys' in getattr(c, f).__globals__))
print(next(getattr(c, f).__globals__['sys']._getframe(1) for c in ().__class__.__base__.__subclasses__()for f in dir(c)if isinstance(getattr(c, f, None), type(lambda: None)) and'sys' in getattr(c, f).__globals__))
print(sys._getframe(0))

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

001 | <frame at 0x7efe1e5b09f0, file '<string>', line 2, code <genexpr>>
002 | <frame at 0x7efe1e5b9000, file '<string>', line 3, code <module>>
003 | <frame at 0x7efe1e5b9000, file '<string>', line 4, code <module>>

[fleet bridge]

=> you should call ._getframe(1)

[languid hare]

i imagine evaling an obfuscated string is cheating?

[golden finch]

I have no idea if that will make my code segfault

I'll try it

[languid hare]

hmm

what's this for

[golden finch]

truly anonymous lambda

with recursion

[languid hare]

that's

cursed

[golden finch]

yeah

so far

(lambda n:eval('type(lambda:0)')([i for i in __import__('sys')._getframe().f_back.f_code.co_consts if isinstance(i,type(__import__('sys')._getframe().f_back.f_code))][0],globals()))(3)

it can actually be golfed now with your hint though

(lambda n:eval('type(lambda:0)')([i for i in __import__('sys')._getframe().f_back.f_code.co_consts if isinstance(i,eval('type((lambda:0).__code__)'))][0],globals()))(3)

almost there

[golden finch]

also welcome to #esoteric-python

[languid hare]

heh

[restive void]

__import__("types").FunctionType?

[golden finch]

trying to use as few imports as possible

!e

(lambda n:eval('type(lambda:0)')([i for i in __import__('sys')._getframe().f_back.f_code.co_consts if isinstance(i,eval('type((lambda:0).__code__)'))][0],globals())(n-1)+eval('type(lambda:0)')([i for i in __import__('sys')._getframe().f_back.f_code.co_consts if isinstance(i,eval('type((lambda:0).__code__)'))][0],globals())(n-2)if n>1 else n)(3)

tried to make fibonacci, not sure where I messed up

[night quarry]

@golden finch :warning: Your eval job has completed with return code 139 (SIGSEGV).

[No output]

[golden finch]

yeah and there's that too

[sly root]

__builtins__.__loader__.load_module("_frozen_"+__builtins__.__loader__.__module__[8:])._ModuleLock.__init__.__globals__["sys"]._getframe(0)```
My attempt.

[golden finch]

smart, still a loophole

[sly root]

wdym?

[sly root]

__builtins__.__loader__.load_module(().__class__.__bases__[0].__subclasses__()[141].__module__).FunctionType```

but idk if load_module counts as import too

[golden finch]

uses external modules

[fleet bridge]

!e

function = ().__class__.__base__.__subclasses__()[32]
print(function)
assert type(lambda: ...) is function

[night quarry]

@fleet bridge :white_check_mark: Your eval job has completed with return code 0.

<class 'function'>

[fleet bridge]

@golden finch

[golden finch]

nice

[fleet bridge]

or object.__subclasses__()[32]

order of initialization of builtin types changes very rarely

so this code should work almost always

[golden finch]

thanks

[floral meteor]

delicious!

I usually try to run a search by subclass name to get the index rather than hardcoding it, just in case

[rapid sparrow]

😄

[sly root]

Now that's a good way to isolate function from main globals.

>>> ().__class__.__bases__[0].__subclasses__()[32]((lambda:globals()).__code__,{})()
{}```

[torn oak]

Okay, so I ended up with this:

print([a := int(input("How many numbers > ")), nums := [str(__import__("random").randint(1, 9)) for _ in range(a)], combo := list(__import__('itertools').product(["/", "*", "+", "-"], repeat=a)), f := [([]) for _ in range(len(combo))], [[(f[i].append([(l,n) for l,n in zip(nums, c)])) for c in combo] for i in range(len(combo))], f := [''.join([list(__import__('itertools').chain.from_iterable(b))][0])[:-1] for b in list(__import__('itertools').chain.from_iterable(f))], g := [], r := [], generated := [], [([g.append(s := eval(b)), r.append([s, b])] if type(eval(b)) == int and eval(b) > 0 else '') for b in set(f)], [((generated.append(i)) if [it[0] for it in sorted(r)].count(i[0]) == 1 else '') for i in sorted(r)], generated][-1])```

how could I optimize this

[fleet bridge]

it is same code, as globals()
your lambda globals and actual globals are the same dictionary

[ripe narwhal]

Hi I need some help with a python project