esoteric-python

[low lynx]

that's just scoping

[lapis root]

:v

What is esoteric-python

[plush halo]

@lapis root, well...

it's the silly side of python xD

In [1]: def main(): 
   ...:     try: 
   ...:         main() 
   ...:     except RecursionError: 
   ...:         main() 
   ...:                                                                

i ran this and got ```
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
File "<ipython-input-1-396208332caf>", line 3 in main
...
Aborted (core dumped)

any more volunteers?

[lapis root]

Aborted mission failed

[plush halo]

xD

[low lynx]

core dumped lmao

[lapis root]

def main():
    try:
        try:
            int(input())
        except ValueError:
            main()
    except RecursionError:
        main()

It would cause another Recursion error right after it won't it

[lapis root]

Bruh

[plush halo]

the problem with linux -.-

[lapis root]

Why is your language set to alien tho

[plush halo]

wdym?

[lapis root]

Oh nothing〜

[versed eagle]

heres something fun to play around with

!e

__import__("ctypes").py_object.from_address(id(globals())+8).value=type("",(dict,),{"__getitem__":(lambda s,i:{}.__class__.__getitem__(s,i)if i in s else ({}.__class__.__getitem__(s,"__builtins__").__getattribute__(i)if {}.__class__.__getitem__(s,"__builtins__").hasattr({}.__class__.__getitem__(s,"__builtins__"),i)else i))})

print(undefined, variables)

[night quarry]

@versed eagle :x: Your 3.11 eval job has completed with return code 139 (SIGSEGV).

undefined variables

[fleet bridge]

I hate you 🖤

[rugged sparrow]

!e ```py
import("ctypes").py_object.from_address(id(globals())+8).value=type("",(dict,),{"slots":(),"getitem":(lambda s,i:{}.class.getitem(s,i)if i in s else ({}.class.getitem(s,"builtins").getattribute(i)if {}.class.getitem(s,"builtins").hasattr({}.class.getitem(s,"builtins"),i)else i))})

print(undefined, variables)```

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 eval job has completed with return code 0.

undefined variables

[rugged sparrow]

@versed eagle add __slots__ and it won't crash on close

[versed eagle]

yeah ik

but i was golfing

[plush halo]

@versed eagle ```ipython
In [4]: import("ctypes").py_object.from_address(id(globals())+8).value=type("",(dict,
...: ),{"getitem":(lambda s,i:{}.class.getitem(s,i)if i in s else ({}.cl
...: ass.getitem(s,"builtins").getattribute(i)if {}.class._getitem
...: _(s,"builtins").hasattr({}.class.getitem(s,"builtins"),i)else i))
...: })

SystemError Traceback (most recent call last)
/usr/lib/python3/dist-packages/IPython/core/async_helpers.py in _pseudo_sync_runner(coro)
66 """
67 try:
---> 68 coro.send(None)
69 except StopIteration as exc:
70 return exc.value

/usr/lib/python3/dist-packages/IPython/core/interactiveshell.py in run_cell_async(self, raw_cell, store_history, silent, shell_futures)
2985 # Store raw and processed history
2986 if store_history:
-> 2987 self.history_manager.store_inputs(self.execution_count,
2988 cell, raw_cell)
2989 if not silent:

/usr/lib/python3/dist-packages/IPython/core/history.py in store_inputs(self, line_num, source, source_raw)
737
738 if self.shell is not None:
--> 739 self.shell.push(to_main, interactive=False)
740
741 def store_output(self, line_num):

/usr/lib/python3/dist-packages/IPython/core/interactiveshell.py in push(self, variables, interactive)
1569
1570 # Propagate variables to user namespace
-> 1571 self.user_ns.update(vdict)
1572
1573 # And configure interactive visibility

SystemError: ../Objects/dictobject.c:1438: bad argument to internal function

In [4]:

[lapis root]

You guys tryna break python or do you hate your pc or what :v

[versed eagle]

yeah it wont work with ipython

[plush halo]

In [4]: iSegmentation fault (core dumped)

[versed eagle]

!e
shorter :D

__import__("ctypes").py_object.from_address(id(globals())+8).value=type("",(dict,),{"__getitem__":(lambda s,i:((g:={}.__class__.__getitem__,b:=g(s,"__builtins__")),g(s,i)if i in s else (b.getattr(b,i)if b.hasattr(b,i)else i))[1])})

print(undefined, variables)

[night quarry]

@versed eagle :x: Your 3.11 eval job has completed with return code 139 (SIGSEGV).

undefined variables

[low lynx]

!e upgraded my cursed input thing to allow for different messages

from typing import Callable
from attrs import define
# Overriding input because python bot can't take inputs
inputs = map(str, [1, 3.1415, -12, 1729, 'hello world', 0x603b5, 1+0j, 3, 'END', 10])
def input(_): return print(str(_), w:=next(inputs)) or w
@define(str=False)
class Message:
    msg: str
    def __str__(self):
        return self.msg
@define(eq=False)
class Verify:
    msg: Message
    fn: Callable[[str, Message], bool]
    def __eq__(self, other: str):
        return self.fn(other, self.msg)
def check_end(inp: str, msg: Message) -> bool:
    msg.msg = ['Invalid number, please try again: ', 'Enter another one: '][inp.isnumeric()]
    return inp.lower() in ('', 'stop', 'exit', 'end')
m = Message('Enter numbers: ')
numbers = [int(w) for w in iter(lambda: input(m), Verify(m, check_end)) if w.isnumeric()]
print(numbers)

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | Enter numbers:  1
002 | Enter another one:  3.1415
003 | Invalid number, please try again:  -12
004 | Invalid number, please try again:  1729
005 | Enter another one:  hello world
006 | Invalid number, please try again:  394165
007 | Enter another one:  (1+0j)
008 | Invalid number, please try again:  3
009 | Enter another one:  END
010 | [1, 1729, 394165, 3]

[plush halo]

yeah some funny errors i get while using it

[rugged sparrow]

!e there are useful things you can do with this knowledge ```py
from fishhook import *

@hook(dict)
def setattr(self, attr, name):
self[attr] = name

@hook(dict)
def getattr(self, attr):
if attr in self:
return self[attr]
return orig(self, attr)

x = {}
x.a = 1
print(x)
print(x.a)```

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | {'a': 1}
002 | 1

[plush halo]

!e```py
a = []
a.append(a)
print(a)

[night quarry]

@plush halo :white_check_mark: Your 3.11 eval job has completed with return code 0.

[[...]]

[plush halo]

the fishhook module looks pretty cool

i'll try it out

[versed eagle]

fishhook is very cool indeed

[lapis root]

Ayo what the fuck

[low lynx]

hmm

!e

from fishhook import hook
@hook(int)
def __iter__(self):
    return map(int, iter(str(self)))

print(sum(1729))

[rugged sparrow]

fishhook not fishook

[versed eagle]

two hs

[rugged sparrow]

sum has a default start arg of 0

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

19

[low lynx]

it was because the iterator returned strs

that's actually pretty cool

need digit sum? just override int.__iter__ and use normal sum()

[rugged sparrow]

!e ```py
from fishhook import *

@hook(int)
def iter(self):
yield from range(self)

print([*10])```

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 eval job has completed with return code 0.

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

[rugged sparrow]

!e ```py
from fishhook import *

@hook(int)
def getitem(self, idx):
return list(map(int, str(self)))[idx]

print(12345[0:2])```

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | <string>:7: SyntaxWarning: 'int' object is not subscriptable; perhaps you missed a comma?
002 | [1, 2]

[low lynx]

!e

from itertools import product
from fishhook import *

@hook(int)
def __iter__(self):
  yield from range(self)

for x, y, z in product(*[3]*3):
    print(x, y, z, end=' ')

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

0 0 0 0 0 1 0 0 2 0 1 0 0 1 1 0 1 2 0 2 0 0 2 1 0 2 2 1 0 0 1 0 1 1 0 2 1 1 0 1 1 1 1 1 2 1 2 0 1 2 1 1 2 2 2 0 0 2 0 1 2 0 2 2 1 0 2 1 1 2 1 2 2 2 0 2 2 1 2 2 2 

[rugged sparrow]

!e fishhook also supports nested hooks ```py
from fishhook import *

@hook(int)
def add(self, other):
print('hook 1', self, other)
return orig(self, other)

@hook(int)
def add(self, other):
print('hook 2', self, other)
return orig(self, other)

x = 5
print(x + 5)```

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | hook 2 5 5
002 | hook 1 5 5
003 | 10

[low lynx]

oh you can refer to the original function?

damn

[fleet bridge]

Wow, nice

[rugged sparrow]

orig will refer to the next level up function, so it will chain hooks if you always call orig

[plush halo]

@rugged sparrow ```ipython
In [1]: from fishhook import hook

In [2]: bak = dict.setitem

In [3]: @hook(dict)
...: def setitem(self, name, value):
...: if name == 'foo':
...: value = not value
...: return bak(self, name, value)
...:

Unhandled exception in event loop:
File "/usr/lib/python3/dist-packages/prompt_toolkit/eventloop/coroutine.py", line 92, in step_next
new_f = coroutine.throw(exc)
File "/usr/lib/python3/dist-packages/prompt_toolkit/buffer.py", line 1654, in new_coroutine
yield From(coroutine(*a, **kw))
File "/usr/lib/python3/dist-packages/prompt_toolkit/eventloop/coroutine.py", line 92, in step_next
new_f = coroutine.throw(exc)
File "/usr/lib/python3/dist-packages/prompt_toolkit/buffer.py", line 1611, in async_validator
yield From(self._validate_async())
File "/usr/lib/python3/dist-packages/prompt_toolkit/eventloop/coroutine.py", line 88, in step_next
new_f = coroutine.send(None)
File "/usr/lib/python3/dist-packages/prompt_toolkit/buffer.py", line 1202, in coroutine
document = self.document
File "/usr/lib/python3/dist-packages/prompt_toolkit/buffer.py", line 478, in document
return self._document_cache[
File "/usr/lib/python3/dist-packages/prompt_toolkit/cache.py", line 91, in missing
result = self.get_value(*key)
File "/usr/lib/python3/dist-packages/prompt_toolkit/document.py", line 111, in init
self._cache = _text_to_document_cache[self.text]
File "/usr/lib/python3.8/weakref.py", line 131, in getitem
o = self.datakey

Exception 'bool' object is not callable
Press ENTER to continue...

[rugged sparrow]

ipython crashed because of your hook, not because of fishhook

[plush halo]

i can't exit my terminal

xD

any help?

[rugged sparrow]

Ctrl-Z + Enter on windows, Ctrl-D on unix

[low lynx]

!e

from fishhook import *
@hook(int)
def __eq__(self, _):
    return False
print(*iter(int, 0))

heh

[plush halo]

doesn't work

[night quarry]

@low lynx :warning: Your 3.11 eval job timed out or ran out of memory.

[No output]

[low lynx]

nice

[rugged sparrow]

use task manager to kill the process then

[plush halo]

ipython does strange things

[versed eagle]

!e

from fishhook import *
@hook(tuple)
def __call__(self, *other):
    return tuple(map(lambda t:t[0]*t[1], zip(self, other)))
print((1, 2, 3)(1, 2, 3))

[night quarry]

@versed eagle :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | <string>:5: SyntaxWarning: 'tuple' object is not callable; perhaps you missed a comma?
002 | (1, 4, 9)

[plush halo]

i created an new tab and manually closed the former one.. thx btw

[rugged sparrow]

your hook changed everything into a bool in dictionaries if the name was foo so ipython broke

[plush halo]

i don't remember initializing foo at any point tho

[rugged sparrow]

it pulls the strings from your current scope

[low lynx]

!e

from fishhook import *
@hook(bool)
def __call__(self, *args, **kwargs):
    return [map, filter][self](*args, **kwargs)

print(*False(lambda x: x*x, True(lambda x: x%2, range(10))))

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | <string>:6: SyntaxWarning: 'bool' object is not callable; perhaps you missed a comma?
002 | <string>:6: SyntaxWarning: 'bool' object is not callable; perhaps you missed a comma?
003 | 1 9 25 49 81

[low lynx]

bool object is not callable?

look again python

[plush halo]

xd

@rugged sparrow, can you use fishhook to bypass setting read only attributes?

[rugged sparrow]

probably, it provides force_setattr

[plush halo]

i'll check it out

thx

or maybe not xD

[rugged sparrow]

ah that one crashes because it tries to free the original string

[plush halo]

>>> from fishhook import hook
>>> bak = globals().__setitem__
>>> @hook(dict)
... def __setitem__(self, name, value):
...     if name == 'foo':
...             value = not value
...     return bak(name, value)
... 
>>> foo  = True
>>> foo
True
>>> globals()
{'__name__': '__main__', '__doc__': None, '__package__': None, '__loader__': <class '_frozen_importlib.BuiltinImporter'>, '__spec__': None, '__annotations__': {}, '__builtins__': <module 'builtins' (built-in)>, 'hook': <function hook at 0x7fb6b1b8c1f0>, 'bak': <method-wrapper '__setitem__' of dict object at 0x7fb6b1dc0680>, 'Foo': <class '__main__.Foo'>, 'f': <__main__.Foo object at 0x7fb6b138ac10>, '__setitem__': None, 'foo': True, 'cStringIO': <weakref at 0x7fb6b1359360; to '_ModuleLock' at 0x7fb6b1d83d00>, 140422651197248: _ModuleLock('cStringIO') at 140422644514048}
>>> globals()['foo']
True
>>> globals()['foo'] = True
>>> foo
False

i'm trying to make fishhook automatically change the value of foo in the global namespace

any ideas?

@rugged sparrow @versed eagle @lapis root @low lynx

[versed eagle]

i mean it seemed to work already

>>> globals()['foo'] = True
>>> foo
False

[plush halo]

without calling globals :/

[versed eagle]

you shouldnt need to

oh actually nvm it wouldnt call the method on the python object

in that case, change the type of globals

[rugged sparrow]

globals is handled weird by bytecode, it bypasses the regular code path

[plush halo]

is there any hope left without threading/multiprocessing?

[versed eagle]

!e

from ctypes import py_object

class globals_overload(dict):
    __slots__ = ()
    def __setitem__(self, item, value):
        {}.__class__.__setitem__(self, item, not value if item == "foo" else value)

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

foo = True
print(foo)

[night quarry]

@versed eagle :white_check_mark: Your 3.11 eval job has completed with return code 0.

False

[versed eagle]

change the type of globals

and you can control it pretty nicely

[plush halo]

i guess there is xD

[versed eagle]

fun
i haven't really messed about with fishhook.asm

[rugged sparrow]

that asm.hook doesnt seem to be working anymore, it might have been inlined by the compiler

it works for other stuff tho

[rugged sparrow]

it lets you hook functions defined in C by patching assembly at runtime

[versed eagle]

ooh
thats a fun idea

[rugged sparrow]

*currently it only explicitly supports 64bit intel

[pure dew]

since when does fishhook support dunders

[astral rover]

since forever?

[arctic kettle]

!e py from ctypes import c_byte import sys n, v = 68, 69 (c_byte * sys.getsizeof(n)).from_address(id(n))[-4] = v print(68, 67 + 1, 69 - 1)

[night quarry]

@arctic kettle :white_check_mark: Your 3.11 eval job has completed with return code 0.

69 69 69

[arctic kettle]

😎

[versed eagle]

I think it's always supported dunders

[meager zinc]

!e ```py
import fishhook
print(dir(fishhook))
print(fishhook.doc)

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | ['BYTES_HEADER', 'NOT_FOUND', 'NULL', 'Py_TPFLAGS_HEAPTYPE', 'Py_TPFLAGS_IMMUTABLE', 'Py_TPFLAGS_READY', 'TYPE_BASICSIZE', '__all__', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'add_cache', 'alloc', 'allocate_structs', 'assert_cls', 'build_unlock_lock', 'c_char', 'classproperty', 'dis', 'find_offset', 'fishhook', 'force_delattr', 'force_setattr', 'get_cache', 'get_structs', 'getdict', 'getmem', 'hook', 'hook_cls', 'hook_property', 'hook_var', 'lock', 'newref', 'orig', 'patch_object', 'py_object', 'pythonapi', 'sizeof', 'sys', 'unhook', 'unlock', 'version']
002 | 
003 | This module allows for swapping out the slot pointers contained in static
004 | classes with the `generic` slot pointers used by python for heap classes.
005 | This allows for assigning arbitrary python functions to static class dunders
006 | using `hook` and `hook_cls` and for applying new functionality to previously
007 | un
... (truncated - too long)

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

[lapis root]

!e

from fishhook import *

@hook(list)
def __setattr__(self, attr, name):
    self+=[(attr, name)]

@hook(list)
def __getattr__(self, attr):
    for tup in self:
        if attr in tup:
            return tup[1]
    return orig(self, attr)

x = []
x.a = 1
print(x)
print(x.a)

[night quarry]

@lapis root :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | [('a', 1)]
002 | 1

[lapis root]

I decided to make a list version

But actually it's still error prone lol.

[rugged sparrow]

cool

[lapis root]

I just fixed it

[rugged sparrow]

fishhook has supported dunders since the first version

[lapis root]

In that example you can only have tuples in the list

But now:

!e

from fishhook import *

@hook(list)
def __setattr__(self, attr, name):
    self+=[(attr, name)]

@hook(list)
def __getattr__(self, attr):
    for i in self:
        if isinstance(i, tuple) and attr in i:
            return i[1]
    return orig(self, attr)

x = [10]
x.a = 1
x.append(6)
print(x)
print(x.a)

[night quarry]

@lapis root :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | [10, ('a', 1), 6]
002 | a

[lapis root]

Do idea why anyone would wanna do this but I decided to be silly-

[rugged sparrow]

i really need to update the docstring for fishhook

[lapis root]

Oh crap I forgot I was messing around and forgot to set it to the second value

!e

from fishhook import *

@hook(list)
def __setattr__(self, attr, name):
    self+=[(attr, name)]

@hook(list)
def __getattr__(self, attr):
    for i in self:
        if isinstance(i, tuple) and attr in i:
            return i[1]
    return orig(self, attr)

x = [10]
x.a = 1
x.append(6)
print(x)
print(x.a)

[night quarry]

@lapis root :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | [10, ('a', 1), 6]
002 | 1

[lapis root]

There we goo

[pure dew]

maybe i'm thinking of forbiddenfruit

[lapis root]

I'm trying to think of how to do it with a string

Seems kinda impossible but eh

[rugged sparrow]

forbiddenfruit also had dunders, but only specific ones

(it also did not support exceptions inside hooks)

fishhook's original purpose was for me to find a way around those limitations

[lapis root]

Strings are immutable.. so it really is impossible. How could I forget

[old socket]

laughs in ctypes

[rugged sparrow]

look at einspect by @dry mirage it could be used to do what you want

it has structure definitions for most builting types

[lapis root]

!e

from fishhook import *

@hook(str)
def __setattr__(self, attr, name):
     global x
     x = self+"{}:{}-".format(attr, name)

@hook(str)
def __getattr__(self, attr):
    for i in x.split("-"):
        if ":" in i:
            atr, e = i.split(":")
            if attr == atr:
                return e

x = ""
x.a = 1
x.b = 2
print(x.a)
print(x.b)

[night quarry]

@lapis root :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | 1
002 | 2

[lapis root]

I was able to get this far

Only 1 problem

How tf do I reference the variable name without hard coding it in

[versed eagle]

also, suppose you set one of the attributes to "-"

then the .split won't do what you want it to

or set it to ":"

[lapis root]

Then split by space

[versed eagle]

so this string will mess with you then "- :"

[lapis root]

Split by some fucking unicode character then 💀

[versed eagle]

"".join(map(chr, range(149_186)))

thats a string containing every unicode character

whatcha gonna do now :P

[lapis root]

That plus my social security number

[versed eagle]

hm
well, you've got a problem then
bc if i look at the code object for str.__getattr__ and str.__setattr__, i could get your number

[lapis root]

XD

[versed eagle]

from .__code__.co_consts

[lapis root]

That's the point-

Jk

[versed eagle]

i thought it was a bad thing if other people had your number

[lapis root]

Only if they do bad things ┐('～`;)┌

[versed eagle]

hm

well

regardless

it'd be breakable if someone obtained your fancy number

[lapis root]

:'

Can't let a guy dream〜

[versed eagle]

you can use ctypes to do it instead

in a way that's less breakable

[lapis root]

What are ctypes

[versed eagle]

import ctypes

https://docs.python.org/3/library/ctypes.html

gives a way to access c stuff directly from python

[versed eagle]

lets you do interesting stuff like control globals().__getitem__

[lapis root]

Bruh I'm too illiterate to read all dat-

[dry mirage]

!e

from fishhook import hook
from einspect import view

@hook(str)
def append(self, value):
    size = len(enc := value.encode())
    with view(self).unsafe() as v:
        v.length += size
        v.buffer[-size:] = enc

@hook(str)
def __setitem__(self, index, value):
    size = len(enc := value.encode())
    view(self).buffer[index:index+size] = enc

s = "hello"
s.append("!")
print(s)

s[1] = "3"
s[4] = "0"
print(s)
print("hello")

[night quarry]

@dry mirage :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | hello!
002 | h3ll0!
003 | h3ll0!

[versed eagle]

basically, it exposes a whole bunch of python wrappers to types and functions from C

[lapis root]

Excuse me but wutdafuk

Time to study python from scratch then cuz my life issa lie

[dry mirage]

!e or mutate ints as well 🥴

from einspect import view

t = 100
view(t).value = 500

print(100)
print(100 == 500)
print(list(range(98, 103)))

[night quarry]

@dry mirage :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | 500
002 | True
003 | [98, 99, 500, 101, 102]

[lapis root]

You're banned

[versed eagle]

that's sad

ionite is nice
and einspect is cool

[meager zinc]

!e ```py
print("".join(map(chr, range(55296))))

too long to upload?

never seen that

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | �
	
002 | 
003 |  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~€‚ƒ„…†‡ˆ‰Š‹ŒŽ‘’“”•–—˜™š›œžŸ ¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿĀāĂ㥹ĆćĈĉĊċČčĎďĐđĒēĔĕĖėĘęĚěĜĝĞğĠġĢģĤĥĦħĨĩĪīĬĭĮįİıIJijĴĵĶķĸĹĺĻļĽľĿŀŁłŃńŅņŇňʼnŊŋŌōŎŏŐőŒœŔŕŖŗŘřŚśŜŝŞşŠšŢţŤťŦŧŨũŪūŬŭŮůŰűŲųŴŵŶŷŸŹźŻżŽžſƀƁƂƃƄƅƆƇƈƉƊƋƌƍƎƏƐƑƒƓƔƕƖƗƘƙƚƛƜƝƞƟƠơƢƣƤƥƦƧƨƩƪƫƬƭƮƯưƱƲƳƴƵƶƷƸƹƺƻƼƽƾƿǀǁǂǃDŽDždžLJLjljNJNjnjǍǎǏǐǑǒǓǔǕǖǗǘǙǚǛǜǝǞǟǠǡǢǣǤǥǦǧǨǩǪǫǬǭǮǯǰDZDzdzǴǵǶǷǸǹǺǻǼǽǾǿȀȁȂȃȄȅȆȇȈȉȊȋȌȍȎȏȐȑȒȓȔȕȖȗȘșȚțȜȝȞȟȠȡȢȣȤȥȦȧȨȩȪȫȬȭȮȯȰȱȲȳȴȵȶȷȸȹȺȻȼȽȾȿɀɁɂɃɄɅɆɇɈɉɊɋɌɍɎɏɐɑɒɓɔɕɖɗɘəɚɛɜɝɞɟɠɡɢɣɤɥɦɧɨɩɪɫɬɭɮɯɰɱɲɳɴɵɶɷɸɹɺɻɼɽɾɿʀʁʂʃʄʅʆʇʈʉʊʋʌʍʎʏʐʑʒʓʔʕʖʗʘʙʚʛʜʝʞʟʠʡʢʣʤʥʦʧʨʩʪʫʬʭʮʯʰʱʲʳʴʵʶʷʸʹʺʻʼʽʾʿˀˁ˂˃˄˅ˆˇˈˉˊˋˌˍˎˏːˑ˒˓˔˕˖˗˘˙˚˛˜˝˞˟ˠˡˢˣˤ˥˦˧˨˩˪˫ˬ˭ˮ˯˰˱˲˳˴˵˶˷˸˹˺˻˼˽˾˿̀́̂ͰͱͲͳʹ͵Ͷͷ͸͹ͺͻͼͽ;Ϳ΀΁΂΃΄΅Ά·ΈΉΊ΋Ό΍ΎΏΐΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡ΢ΣΤΥΦΧΨΩΪΫάέήίΰαβγδεζηθικλμνξοπρςστυφχψωϊϋόύώϏϐϑϒϓϔϕ
... (truncated - too long)

Full output: too long to upload

[flint hollow]

!e

print(len("".join(map(chr, range(55296)))))

[night quarry]

@flint hollow :white_check_mark: Your 3.11 eval job has completed with return code 0.

55296

[flint hollow]

i suppose that's obvious

[flint hollow]

i've seen this a few times - it occurs over 10,000 characters

!e print(''.join('a' for _ in range(10001)))

[night quarry]

@flint hollow :white_check_mark: Your 3.11 eval job has completed with return code 0.

aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
... (truncated - too long)

Full output: too long to upload

[low lynx]

!e why not just

print('a'*10001)
```?

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
... (truncated - too long)

Full output: too long to upload

[meager zinc]

I wonder if it's possible to print hello world without using any numbers or letters

[flint hollow]

no letters is a problem

restricted letters can be done, though, with just aelrstv

https://github.com/wanqizhu/pyfuck#hello-world

i would have pybot run it, but at 6019 characters, it's too long to paste in

[meager zinc]

what about exclusively letters without parens or brackets or braces

or numbers

[flint hollow]

the trick to reduce the character space is using either exec or eval, and they both take arguments, which require parentheses

[flint hollow]

there's also a version that drops to just 9 characters: https://github.com/satoki/PyFuck

!e

exec(chr(111+1)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+11)+chr(111+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(111)+chr(11+11+1+1+1+1+1+1+1+1+1+1)+chr(111+1+1+1+1+1+1+1+1)+chr(111)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1+1))

[night quarry]

@flint hollow :white_check_mark: Your 3.11 eval job has completed with return code 0.

hello world

[sick hound]

What the

How’s that work

[flint hollow]

each of the chrs creates one of the letters of print('hello world'), they're concatenated together with +, and then sent to exec to be run

the very first, for example, is chr(111+1), or chr(112), which is just p

[quartz wave]

simplest version of obfuscation

[sick hound]

So who do I know what chr(x) means?

Like is there a list for chr [number] is what?

[quartz wave]

it uses the ASCII set of characters + the extension that Unicode provides

[sick hound]

Alriht I’ll try it later

[low lynx]

the inverse is ord()

iirc it's how python compares strings

so it's good to know

[karmic pumice]

!e

_ = __builtins__.int();  
while _ < (((x:=(o:=(-~__builtins__.int().__add__(-~__builtins__.int())))).__pow__(o.__add__(-~__builtins__.int()))).__add__(o)):
    getattr(__builtins__, dir(__builtins__)[(o.__pow__(o.__pow__((o.__add__(-~__builtins__.int()))).__sub__(-~__builtins__.int())).__add__(((o.__pow__(o)).__add__(-~__builtins__.int()))))])(_:=(_.__add__(-~__builtins__.int())))

[night quarry]

@karmic pumice :white_check_mark: Your 3.11 eval job has completed with return code 0.

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

[karmic pumice]

btw this code is unstable

there is a huge chance that it wont work on your machine

because for some reason dir(__builtins__) differs from person to person 🤷

!e

for e, i in enumerate(dir(__builtins__)):
  print(e, i)

[night quarry]

@karmic pumice :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | 0 ArithmeticError
002 | 1 AssertionError
003 | 2 AttributeError
004 | 3 BaseException
005 | 4 BaseExceptionGroup
006 | 5 BlockingIOError
007 | 6 BrokenPipeError
008 | 7 BufferError
009 | 8 BytesWarning
010 | 9 ChildProcessError
011 | 10 ConnectionAbortedError
... (truncated - too many lines)

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

[karmic pumice]

as you can see here, print is at the 133 index for the bot (in dir builtins)

[crystal goblet]

whats wroing with this

[karmic pumice]

for me its 139 tho

[main imp]

How does that even work?

[karmic pumice]

wdym how does that work?

it just assigns characters by their number in ascii table

chr(97) = 'a'

[main imp]

Explain

Like I'm 5

[main imp]

My brain hurts

[karmic pumice]

there is a lot more stuff tho

!e

print(chr(8364))

[night quarry]

@karmic pumice :white_check_mark: Your 3.11 eval job has completed with return code 0.

€

[versed eagle]

it's changes based on what version of python you're using

[main imp]

BRUHHHHHH

WAIT

Huh-

This is fucking with ma Brain

[fleet bridge]

Soke implementations can assign new items into builtins.
Also sitebuiltins is adding stuff like help, exit, quit, copyright...
I have custom sitecustomize where i am assigning several helper functions like dis (so i can use dis.dis without importing).

So, relying on dir(__builtins__) is a bad idea even in this channel :-)

[karmic pumice]

yeah ik

sometimes you just want to write funky stuff 🤷

[fleet bridge]

I once wrote code that was relying on object.__subclasses__(). Its behaviour differs in 3.10 and 3.11, and it also differs in REPL and non-REPL mode. And for some reason it behaved differently in snekbox, iirc

[arctic kettle]

bored so I made a script for this.. ```py
def fuck(line: str) -> str:
numbers: list[list[int]] = [[] for _ in range(len(line))]
for i, c in enumerate(line):
remainder = ord(c)
for j in range(len(str(remainder)), 0, -1):
n = int('1'*j)
amount, remainder = divmod(remainder, n)
numbers[i].extend(n for _ in range(amount))
return 'exec(' + '+'.join('chr(' + '+'.join(map(str, n))+ ')' for n in numbers) + ')'

line = '''print('Hello World!')'''
print(fuck(line=line))

exec(chr(111+1)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+11)+chr(111+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(111)+chr(11+11+1+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1+1)+chr(111)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1)+chr(11+11+11)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1+1))```

!e exec(chr(111+1)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+11)+chr(111+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(111)+chr(11+11+1+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1+1)+chr(111)+chr(111+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1+1+1+1+1+1+1+1+1)+chr(11+11+11+11+11+11+11+11+11+1)+chr(11+11+11)+chr(11+11+11+1+1+1+1+1+1)+chr(11+11+11+1+1+1+1+1+1+1+1))

[night quarry]

@arctic kettle :white_check_mark: Your 3.11 eval job has completed with return code 0.

Hello World!

[fleet bridge]

!e ```py
='=%r;exec(%%)';exec(%)

[night quarry]

@fleet bridge :x: Your 3.11 eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 1, in <module>
003 |   File "<string>", line 1, in <module>
004 |   File "<string>", line 1, in <module>
005 |   [Previous line repeated 497 more times]
006 | RecursionError: maximum recursion depth exceeded during compilation

[fleet bridge]

during compilation
why? compile(_%_,'','exec') works

!e s = 'exec(s)'; exec(s)

[night quarry]

@fleet bridge :x: Your 3.11 eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 1, in <module>
003 |   File "<string>", line 1, in <module>
004 |   File "<string>", line 1, in <module>
005 |   [Previous line repeated 497 more times]
006 | RecursionError: maximum recursion depth exceeded during compilation

[fleet bridge]

i got it: ```py

sys.setrecursionlimit(1000)
s = 'exec(s)'; exec(s)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "<string>", line 1, in <module>
File "<string>", line 1, in <module>
File "<string>", line 1, in <module>
[Previous line repeated 496 more times]
RecursionError: maximum recursion depth exceeded during compilation
sys.setrecursionlimit(1001)
s = 'exec(s)'; exec(s)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "<string>", line 1, in <module>
File "<string>", line 1, in <module>
File "<string>", line 1, in <module>
[Previous line repeated 497 more times]
RecursionError: maximum recursion depth exceeded while calling a Python object

[karmic pumice]

hi guys, want to share this thing with you. Its a function that allows you to convert an int into a sum of powers of 2 but only uses -~int() and dunder add and pow.

import pyperclip
def power_find(n: int) -> list[int]:
    """Convert an integer into a list of powers of 2.

    Args:
        n (int): Integer to convert

    Returns:
        list[int]: List of powers
    """
    result = []
    binary = bin(n)[:1:-1]
    for x in range(len(binary)):
        if int(binary[x]):
            result.append(x)
    return result

def int_to_eso(n: int):
    """Convert an integer into a esoteric implementation of itself

    Args:
        n (int): Positive only integer that is going to be converted to an esoteric monster

    Returns:
        str: Esoteric implementation of the number {n}
    """
    powers = power_find(n)
    buf = []
    for power in powers:
        if power == 0:
            buf.append(f"(-~int().__add__(-~int())).__pow__(int())")
        else:
            buf.append(f"(-~int().__add__(-~int())).__pow__(-~int(){''.join([f'.__add__(-~int())' for _ in range(power-1)])})")
            
    res = f"sum({str(buf).translate({39: None})})"
    pyperclip.copy(res)
    return res

Example input -> Output:

int_to_eso(123001)
>>> sum([(-~int().__add__(-~int())).__pow__(int()), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()))])

!e

print(sum([(-~int().__add__(-~int())).__pow__(int()), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())), (-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()))]))

[night quarry]

@karmic pumice :white_check_mark: Your 3.11 eval job has completed with return code 0.

123001

[versed eagle]

for x in range(len(binary)):

why not use enumerate

[karmic pumice]

🤷

[versed eagle]

lol

[fleet bridge]

#esoteric-python message
similar idea, but here i am costructing integers using only lists literals and math operations

[karmic pumice]

ooh thats interesting too

[karmic pumice]

lmao, the implementation of the number 1234567891234567890 takes 11.5K chars with my method

[night quarry]

Hey @karmic pumice!

You either uploaded a .txt file or entered a message that was too long. Please use our paste bin instead.

[karmic pumice]

https://paste.pythondiscord.com/ejurocazis

[versed eagle]

there are some optimisations you could make

[versed eagle]

long addition sections can be turned into powers

e.g.
-~int().__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int()).__add__(-~int())
->
(-~int().__add__(-~int())).__pow__(-~int().__add__(-~int()).__add__(-~int()))

that example is a bit contrived, since i picked the number of operations to be a power of 2, but the strategy extends past powers of 2

[karmic pumice]

yeah i would like to do that somehow

[versed eagle]

well, if you know the number of additions that you're going to do in a section

you can check if doing it with pows would be shorter

[karmic pumice]

i was actually thinking of making it with walruses originally so i would just add the difference between closest powers to the latest walrus assigned power

[versed eagle]

ah

doing it with walruses could lead to some sort of recursion

[karmic pumice]

this was my first thing that converted something to a code implementation so im still kinda proud that i made it work 🤷

[versed eagle]

really, thats all you need

something you're proud of

[karmic pumice]

this is beautiful haha

[versed eagle]

instead of using sum, why not add them all together manually

using .__add__

[karmic pumice]

yeah true

k i cant think of a way of how to actually implement this without some stupid ifs 🤷

[versed eagle]

want a hint?

[karmic pumice]

i think im on the right way right now, ill test it a bit

yeah i did it, lets go

maybe there are now some 1 extra () but 🤷

[night quarry]

Hey @karmic pumice!

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

[karmic pumice]

https://paste.pythondiscord.com/oyalocuyeb

well yeah with large numbers the amount of adds it has to do to get the powers is.. something

i actually want to make that package now lmao

never created one, think that this is a good one to start with

[kindred stone]

💀

[meager zinc]

!e this is a compressed version: ```py
exec(eval("'print%@$#@$&!$)#%@$#@$&!@$~??~..~..?~..??~;~;?~;.~,?~,;~,;.??~,,?~,,??~,,.??~,,..~,,;?~,,;.~,,;.?~,,;.??~,,,?~,,,.~,,,.?~,,,..~,,,..??~,,,;~,,,;??~,,,,~,,,,..``.?.??....?..??;?;.?;.?:..?:;,;:;?:,,,:,?:,.:,.?:,;,,;:,;?:,;.,,;.:,,#@$&&)'" + '''z;y...xz:y?,xz,y.....xz.y???xz`y~"+"?"*69)z~y#@$&&#%@$#@$&!@$xz?y#@$)xz&y))xz%y((xz#y.add(xz$yint()xz@y-~xz!y.pow(x'''.replace("z",".replace("").replace("y","","").replace("x","")")))

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

123456789123456789123456789123456789

[little ledge]

If you don't segfault an interpreted language, are you even trying?

[old socket]

!e import("ctypes").py_object.from_address(0)

[night quarry]

@old socket :warning: Your 3.11 eval job has completed with return code 0.

[No output]

[old socket]

what

[quartz wave]

!e ```py
print(import("ctypes").py_object.from_address(0))

[night quarry]

@quartz wave :warning: Your 3.11 eval job has completed with return code 139 (SIGSEGV).

[No output]

[quartz wave]

!e ```py
import("ctypes").py_object.from_address(0)
print("it's fine")

[night quarry]

@quartz wave :white_check_mark: Your 3.11 eval job has completed with return code 0.

it's fine

[old socket]

Oh I was so confused to why it wasn't segfaulting

So I need to use the object

[meager zinc]

Well if it's just a pointer to an out-of-bounds address then it doesn't matter - you have to dereference the pointer

Classic python memory and pointer management

[quartz wave]

!e ```py
import("ctypes").py_object.from_address(214912748963731)
print("it's fine")

[night quarry]

@quartz wave :white_check_mark: Your 3.11 eval job has completed with return code 0.

it's fine

[quartz wave]

!e ```py
import("ctypes").py_object.from_address(214912748963731)
print("it's fine")

[night quarry]

@quartz wave :white_check_mark: Your 3.11 eval job has completed with return code 0.

it's fine

[quartz wave]

ye so

as long as you don't use it it's fine

[rough moat]

bro used a corruptor on the code

[rough moat]

oh so that’s how you check adresses

[meager zinc]

!e ```py
exec(bytes('汣獡\u2073㩟彟瑳彲㵟慬扭慤张∺效汬Ɐ眠牯摬∡眊瑩⡨㩳弽楟灭牯彴⡟攢湩灳捥≴⸩楶睥猨牴⤩甮獮晡⡥㨩\u2e73祴数弽','u16')[2:])
print(type('Foo bar.'))

[night quarry]

@meager zinc :x: Your 3.11 eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 1, in <module>
003 |   File "<string>", line 2, in <module>
004 |   File "/snekbox/user_base/lib/python3.11/site-packages/einspect/views/view_type.py", line 111, in __setattr__
005 |     super().__setattr__(key, value)
006 |   File "/snekbox/user_base/lib/python3.11/site-packages/einspect/views/view_base.py", line 107, in type
007 |     self._pyobject.ob_type = value
008 |     ^^^^^^^^^^^^^^^^^^^^^^
009 | TypeError: expected LP_PyTypeObject instance, got type

[meager zinc]

wait it doesn't work in snekbox

works on my machine

[rough moat]

you guys know that percentage bar meme with the if else

lambda n:("⚪"*(10-int(n*10))).rjust(10,"🔵")

1 liner

[meager zinc]

Here's the actual output:

[rough moat]

oh i swapped the colors

oops

nope nvm lmao

it was fine at the start

[meager zinc]

what about

!e ```py
print((lambda m:[("🔵"*n).ljust(m,"⚪")for n in range(-~m)])(10))

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

['⚪⚪⚪⚪⚪⚪⚪⚪⚪⚪', '🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪', '🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪', '🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪', '🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪', '🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪', '🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪', '🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪', '🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪', '🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪', '🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵']

[meager zinc]

the actual function:

lambda m:[("🔵"*n).ljust(m,"⚪")for n in range(-~m)]

[rough moat]

mine is shorter 😎

[meager zinc]

yours makes a single bar

[rough moat]

well yeah

that's the point

it's not supposed to make the whole thing

cool using ~ tho

[meager zinc]

mine is shorter and is a single bar 😎

lambda n:("🔵"*(n//1)).ljust(10,"⚪")

[rough moat]

that uses an int value

[meager zinc]

round it

[rough moat]

i mean yours does

it should take a float from 0-1

[meager zinc]

ah I see

lambda n:("🔵"*(n*//.1)).ljust(10,"⚪")

-1 char by //.1

[arctic skiff]

!epy print('''⚪⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪ 🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵''')

[night quarry]

@arctic skiff :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | ⚪⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪
002 | 🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵

[meager zinc]

best progress bar

[rough moat]

using //1 is the same amount of chars as int()

[meager zinc]

well //.1 is less

we need to multiply it by 10, no?

[arctic skiff]

!epy print('''⚪⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪ 🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪ 🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪ 🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵''')

[night quarry]

@arctic skiff :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | ⚪⚪⚪⚪⚪⚪⚪⚪⚪⚪
002 | 🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪
003 | 🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪
004 | 🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪
005 | 🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪
006 | 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪
007 | 🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪
008 | 🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪
009 | 🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪
010 | 🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪
011 | 🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵

[arctic skiff]

Now thats ok

[rough moat]

wait you can use float index on string

nope

[meager zinc]

no

that's why it's floordiv

[rough moat]

//.1 returns 1.0

[meager zinc]

what?

[rough moat]

    print("abcdefg"[0.1//.1])
          ~~~~~~~~~^^^^^^^^^
TypeError: string indices must be integers, not 'float'

hm

[arctic skiff]

!epy print('abc'[1.0])

[night quarry]

@arctic skiff :x: Your 3.11 eval job has completed with return code 1.

001 | <string>:1: SyntaxWarning: str indices must be integers or slices, not float; perhaps you missed a comma?
002 | Traceback (most recent call last):
003 |   File "<string>", line 1, in <module>
004 | TypeError: string indices must be integers, not 'float'

[rough moat]

maybe its different for multiplication

[meager zinc]

wait it doesn't floor it?

[rough moat]

nope

well it does

[meager zinc]

wat

[rough moat]

but it doesnt return int

[meager zinc]

but it's a float....

[rough moat]

    print("abcdefg"*(0.1//.1))
          ~~~~~~~~~^^~~~~~~~~
TypeError: can't multiply sequence by non-int of type 'float'

[arctic skiff]

!e print(0.1//0.1)

[night quarry]

@arctic skiff :white_check_mark: Your 3.11 eval job has completed with return code 0.

1.0

[meager zinc]

boo

I would just pull a

from fishhook import hook

@hook(float)
def __div__(self, other):
  return __import__("math").floor(self / other)

[arctic skiff]

cheating!!

[meager zinc]

nothing is cheating in #esoteric-python

in a way

shortest so far ig

lambda n:("🔵"*int(n*10)).ljust(10,"⚪")

[arctic skiff]

!e```py
(lambda n:("🔵"int(n10)).ljust(10,"⚪") )()

[night quarry]

@arctic skiff :x: Your 3.11 eval job has completed with return code 1.

001 |   File "<string>", line 1
002 |     (lambda n:("🔵"*int(n*10)).ljust(10,"⚪") )()
003 |                                            ^
004 | SyntaxError: invalid non-printable character U+200A

[arctic skiff]

https://cdn.discordapp.com/emojis/854310158070448150.gif?quality=lossless&size=48

bonk @night quarry

[versed eagle]

that happens to me sometimes when i copy and paste stuff

it inserts extra characters

[meager zinc]

wait this is actually shorter by a bit:

#prev
lambda n:("🔵"*int(n*10)).ljust(10,"⚪")
#new
lambda n:("🔵"*(a:=int(n))*10)+"⚪"*a

[arctic skiff]

bonk discord

[meager zinc]

@rough moat

wait nevermind

you have to subtract from 10

so its longer

[rough moat]

i made one thats exactly the same length

lambda n:(("🔵"*int(n*10))+"⚪"*10)[:10]

[meager zinc]

smart

[versed eagle]

ooh thats fun

[rough moat]

oh i shaved off two chars

[meager zinc]

wait

can we := the 10

[versed eagle]

it'd be too long

[rough moat]

never used walrus

[meager zinc]

darn

[versed eagle]

cause it'd only save 1 char per use of 10

[rough moat]

lambda n:("🔵"*int(n*10)+"⚪"*10)[:10]

36 chars

[meager zinc]

I always miss this

lambda n:(("🔵"*int(n*10))+"⚪"*10)[:10]
#        ^               ^
lambda n:("🔵"*int(n*10)+"⚪"*10)[:10]

[pulsar tangle]

Hey

[meager zinc]

what

[arctic skiff]

def progress_bar(n):
  match n:
    case 1:return '🔵⚪⚪⚪⚪⚪⚪⚪⚪⚪'
    case 2:return '🔵🔵⚪⚪⚪⚪⚪⚪⚪⚪'
    case 3:return '🔵🔵🔵⚪⚪⚪⚪⚪⚪⚪'
    case 4:return '🔵🔵🔵🔵⚪⚪⚪⚪⚪⚪'
    case 5:return '🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪'
    case 6:return '🔵🔵🔵🔵🔵🔵⚪⚪⚪⚪'
    case 7:return '🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪'
    case 8:return '🔵🔵🔵🔵🔵🔵🔵🔵⚪⚪'
    case 9:return '🔵🔵🔵🔵🔵🔵🔵🔵🔵⚪'
  return '🔵🔵🔵🔵🔵🔵🔵🔵🔵🔵'```

[meager zinc]

💀

[rough moat]

but that uses an int!!!!!

[arctic skiff]

Doesn't matter

[meager zinc]

!e yoo 2 chars (())

#ignore this line, doesn't do anything important
__import__('fishhook').hook(float,'__call__',lambda n:("🔵"*int(n*10)+"⚪"*10)[:10])

print(.5())

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | <string>:4: SyntaxWarning: 'float' object is not callable; perhaps you missed a comma?
002 | 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪

[rough moat]

grrr non stdlib

[versed eagle]

!e
stdlib :D

__import__("ctypes").py_object.from_address(id(f:=.5)+8).value=type("",(float,),{"__slots__":(),"__call__":(lambda n:("🔵"*int(n*10)+"⚪"*10)[:10])})

print(.5())

[night quarry]

@versed eagle :x: Your 3.11 eval job has completed with return code 139 (SIGSEGV).

001 | <string>:3: SyntaxWarning: 'float' object is not callable; perhaps you missed a comma?
002 | 🔵🔵🔵🔵🔵⚪⚪⚪⚪⚪

[quartz wave]

trying to remove segfault
pathetic

[versed eagle]

the red x in the result is peer pressuring me 😭

also its like midnight im not at my best here
cut me some slack lmao

[quartz wave]

lambda n:f"{'🔵'*int(n*10):⚪<10.10}"
``` 35 chars

this is genius damn

[meager zinc]

I feel like have a m (maximum) argument is shorter and better for actual use

but I think that won't work with the fstring

unless <m.m} would format by the variable m instead of string m

[quartz wave]

earlier would be shorter ```py
lambda n,m:("🔵"int(nm)+"⚪"*m)[:m]

OR

lambda n,m=10:("🔵"int(nm)+"⚪"*m)[:m]

[sick hound]

Hi

Guys I need help in solving a coding task

[tough willow]

#❓｜how-to-get-help @sick hound

[sick hound]

that should probs be considered whitespace

[royal brook]

best

[dense nova]

!e

print((lambda n: ("🔵"*n).ljust(10, "⚪"))(7))

[night quarry]

@dense nova :white_check_mark: Your 3.11 eval job has completed with return code 0.

🔵🔵🔵🔵🔵🔵🔵⚪⚪⚪

[meager zinc]

!e fishhook and einspect and ctypes are simply perfect for this channel (this uses fishhook)

exec(bytes('彟浩潰瑲彟✨楦桳潨歯⤧栮潯⡫畴汰ⱥ弧损污彬❟氬浡摢\u2061㩟牰湩⡴䠧汥潬眠牯摬✡⤩','u16')[2:])

()()

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

001 | <string>:3: SyntaxWarning: 'tuple' object is not callable; perhaps you missed a comma?
002 | Hello world!

[half oak]

ops='+-*/%rq'
while 1:
 for i in ops[:6]:print(f"Enter '{i}' for {'square root' if i=='r'else i}")
 i=input()
 if i in ops:
  a,b=float(input()),float(input())if i!='r'else None
  exec(f'print({a}{i}{b if b else "**0.5"})')
  if i=='q':break

Can this be more short

[versed eagle]

yes

[golden finch]

for starters there's whitespace that can be stripped

[versed eagle]

heres some pretty easy expression shortenings
'square root' if i=='r'else i -> i=='r'and'square root'or i
b if b else "**0.5" -> b or"**0.5"
also you should probably walrus input, and maybe float

and im not sure why you're defining a,b instead of using them inside the fstring directly

[versed eagle]

(i:='',o:='+-*/%rq',I:=input,{(print("\n".join(map(lambda c:f"Enter '{c}' for {c=='r'and'square root'or c}",o))),exec(f"print({float(I())}{i}{(i!='r')*float(I())or"**0.5"})")if(i:=I())in o else i:='q')for _ in iter(lambda:i=='q',1)})

[meager zinc]

F:=float is an improvement

[versed eagle]

alr

[quartz wave]

it's broken for square root so i fixed it while golfing, 204 ```py
ops=i='+-/%rq'
I=input
while'q'!=i:[print(f"Enter {i!r} for {['square root',i][i!='r']}")for i in ops[:6]];exec(((i:=I())in ops)"exec(f'print({float(I())}{i*(d:=i!="r")}{d and float(I())or"**.5"})')")

wait

fixed

[meager zinc]

ops=i='+-*/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in ops[:6]];exec(((i:=I())in ops)*"exec(f'print({F(I())}{i*(d:=i!="r")}{d and F(I())or\"**.5\"})')")

[quartz wave]

it's ```py
ops=i='+-/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in ops[:6]];exec(((i:=I())in ops)"exec(f'print({F(I())}{i*(d:=i!="r")}{d and F(I())or"**.5"})')")

have to escape the quotes in the latter part

[meager zinc]

ah

[quartz wave]

so that's 199

[meager zinc]

why is the varname ops still here

o=i='+-*/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in o[:6]];exec(((i:=I())in ops)*"exec(f'print({F(I())}{i*(d:=i!=\"r\")}{d and F(I())or\"**.5\"})')")

[quartz wave]

oh yeah

195 ```py
o=i='+-/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in ops[:6]];i=I();exec((i in o)"exec(f'print({F(I())}{i*(d:=i!="r")}{d and F(I())or"**.5"})')")

[meager zinc]

you didn't replace the second one

[quartz wave]

uh huh sure that was just me for sure

[quartz wave]

193 ```py
o=i='+-/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in o[:6]];i=I();exec((i in o)"exec(f'print({F(I())}{i*(d:=i!="r")}{d and F(I())or"**.5"})')")

[quartz wave]

sneaky little ops hiding in the corner

[meager zinc]

Oh we both removed one

*forgot one

[quartz wave]

187 ```py
o=i='+-/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in o[:6]];i=I();exec(i in o and f"print({F(I())}{i(d:=i!='r')}{d and F(I())or'**.5'})")

186 ```py
o=i='+-/%rq'
I=input
F=float
while'q'!=i:[print(f"Enter {i!r} for {['sqrt',i][i!='r']}")for i in o[:6]];i=I();d=i!='r';exec(i in o and f"print({F(I())}{id}{d and F(I())or'**.5'})"or"")

183 ```py
o,I=i='+-/%rq',input
while'q'!=i:[print("Enter %r for "%i+['sqrt',i][i!='r'])for i in o[:6]];i=I();d=i!='r';exec(i in o and f"print({float(I())}{id}{d and float(I())or'**.5'})"or"")

169 disregarding unsafe input ```py
o,I=i='+-/%rq',input
while'q'!=i:[print("Enter %r for "%i+['sqrt',i][i!='r'])for i in o[:6]];i=I();d=i!='r';exec(i in o and"print("+I()+id+(d and I()or'**.5')+")"or"")

[split kestrel]

167 with the original interfacepy o=f="+-*/%r" i=input while'q'!=f:[print("Enter %r for "%n+["square root",n][n<'r'])for n in o];f=i();f in o and print(eval("float(i())"+["**.5",f+"eval(i())"][f<'r']))

[quartz wave]

160 with the shorter sqrt ```py
o=f="+-*/%r"
i=input
while'q'!=f:[print("Enter %r for "%n+["sqrt",n][n<'r'])for n in o];f=i();f in o and print(eval("float(i())"+["**.5",f+"eval(i())"][f<'r']))

159 unsafe input ```py
o=f="+-*/%r"
i=input
while'q'!=f:[print("Enter %r for "%n+["sqrt",n][n<'r'])for n in o];f=i();f in o and print(eval("eval(i())"+["**.5",f+"eval(i())"][f<'r']))

[earnest wing]

f in o==print(...)

-2

[quartz wave]

good one

156 ```py
o=f="+-*/%r"
i=input
while'q'!=f:[print("Enter %r for "%n+["sqrt",n][n<'r'])for n in o];f=i();f in o==print(eval("eval(i())"+["**.5",f+"eval(i())"][f<'r']))

150 ```py
o=f="+-*/%r"
c="eval(input())"
while'q'!=f:[print("Enter %r for "%n+["sqrt",n][n<'r'])for n in o];f=input();f in o==print(eval(c+["**.5",f+c][f<'r']))

[split kestrel]

aaa comparison chaining

[magic wraith]

b if b else "**0.5"
can be
b or'**.5'
but i'm probably late

[half oak]

Ok

[fleet bridge]

it still works weird for square root
your code: sqrt(-2) = -2**0.5 = -(2**0.5) = -1.41
how it should work: sqrt(-2) = sqrt(-1) * sqrt(2) = 1.41*i

>>> -2**0.5
-1.4142135623730951
>>> (-2)**0.5
(8.659560562354934e-17+1.4142135623730951j)

[eager ore]

👋

[glass stone]

this seems like the right place - WTF is Jython?

[earnest wing]

JVM implementation of python

aka pava

[quartz wave]

8.659560562354934e-17 is FPE noise

ye

[quartz wave]

so it still needs to handle those properly

[glass stone]

lol pava. Is it python syntax or just the interpreter class in Java?

[old socket]

Interpreter class in Java + Module in python that interprets java

[glass stone]

So can I write pure python code and run it on the JVM with few changes?

[old socket]

import org.python.util.PythonInterpreter;

public class JythonHelloWorld {
  public static void main(String[] args) {
    try(PythonInterpreter pyInterp = new PythonInterpreter()) {
      pyInterp.exec("print('Hello Python World!')");
    }
  }
}
``` from Jython docs

[eager ore]

https://gist.github.com/dzshn/1b8853023f364dc559c2b11c6a5f38b7

can someone explain stuff like this?

[glass stone]

@old socket that looks especially brutal

[old socket]

Yea lol

Certified java moment

[glass stone]

I reckon you could encapsulate it and make it as easy as saying py.do(stuff);

[glass stone]

wtf is this

[eager ore]

...esoteric python?

do you know what channel you're in?

[glass stone]

well before just a moment ago I considered Jython in that realm. This is next level crazy and it kind of scares me I must know how it works.

[eager ore]

lol

[glass stone]

I am ready for my transformation to the dark side

naw, I make python challenges for my friends and that seems like an awesome way to blow their minds.

[eager ore]

esoteric python and code golfing are great challenges

[glass stone]

I will have to check them out, buried in school work rn. Fate brought me here. Soon young skywalker will be one of us

[versed eagle]

if you're looking for something fun to play around with try checking out fishhook

is a rather brilliant module made by chilaxan

[glass stone]

yeah reading the description it seems over my head

[languid hare]

!e

from fishhook import *

@hook(str)
def __mul__(self, other):
    if not isinstance(other, float):
        return orig(self, other)

    integer = int(other)
    decimal = int((other - integer) * len(self))

    return self * integer + self[:decimal]

print("0123456789" * 1.5)

[night quarry]

@languid hare :white_check_mark: Your 3.11 eval job has completed with return code 0.

012345678901234

[languid hare]

@glass stone it lets you replace / add methods onto existing classes

[glass stone]

I have been cramming Java for the last week because we got a new professor at school and the class changed from the intro class I took. My brain is at capacity for new ideas at the moment. It looks really neat. Is mul a standard dunder? I haven't delved too deep into the dunders

[meager zinc]

__mul__ is *

[glass stone]

ya that makes sense, like I said my poor brain hurts. CCNA, Java OOP and Ethical hacking I'm hit and it's only monday of week 2.

So its a dunder of the type int eh

[languid hare]

mhm, so this code replaces the int __mul__ with the one i wrote, allowing it to work with floats

orig points to the original __mul__ through some inspection magic

[eager ore]

dunder is a weird word

[versed eagle]

means double underscore

[eager ore]

i'm aware of what it means

i use the term magic method

[versed eagle]

ah

[burnt pasture]

i don't like "magic" method, because it implies you can't understand the mechanism.

[versed eagle]

also dunder is more than just magic methods-

__import__ is a dunder

same with __builtins__

and so on

whereas they aren't methods

[fleet bridge]

__class__, __name__, __dict__ are dunder attributes

Types of dunders:

1. dunder methods
   2) dunder attributes on instance level (__dict__, __name__, __class__)
   3) dunder attributes on class level (__slots__, __bases__)
2. dunder attributes (__dict__, __name__, __class__, __slots__, __bases__, ...)
3. dunder global variables (__name__, __builtins__, __file__, __loader__, __all__, __path__...) + global methods (__getattr__, __dir__)
4. dunder builtin variables (__import__)
5. __debug__ - it is replaced with corresponding bool value during comppilation, and you cant assign to it. You also cant assign to x.__debug__
6. dunder local variables (like __class__ in methods)
7. dunder string values like in __name__ == '__main__'
8. dunder filenames (__init__.py, __main__.py, __pycache__/)
9. dunder modules (__future__, __hello__)

[versed eagle]

there's also module level __getattr__

and there's __future__

[meager zinc]

how do I invert bool

why doesn't this work:

__import__("ctypes").py_object.from_address(id(False)+8).value = True

[sharp moon]

!e dunder

[night quarry]

@sharp moon :x: Your 3.11 eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 1, in <module>
003 | NameError: name '__dunder__' is not defined. Did you mean: '__loader__'?

[next flame]

>>> import ctypes
>>> ctypes.c_uint32.from_address(id(False)+16).value=1 # ob_size
>>> ctypes.c_uint32.from_address(id(False)+24).value=1 # ob_digit
>>> int(False)
1

[tough willow]

Where did you get the 16 and 24 From?

[true mortar]

Is anybody familiar with Polygon API or where might I be able to go to find an algorithmic trading focused server?

[next flame]

// from https://github.com/python/cpython/blob/main/Include/object.h
#define PyObject_VAR_HEAD      PyVarObject ob_base;
// ...
typedef struct {
    PyObject ob_base;
    Py_ssize_t ob_size; /* Number of items in variable part */
} PyVarObject;
// from https://github.com/python/cpython/blob/main/Include/cpython/longintrepr.h
struct _longobject {
    PyObject_VAR_HEAD
    digit ob_digit[1];
};

(bools are ints under the hood)

I can't find the source of PyObject right now, but it has 2 fields: ob_refcnt (a ssize_t = 8 bytes on x64) and ob_type (a pointer = 8 bytes)
so ob_size is 16 bytes offset, and the first element of ob_digit is 24 bytes offset

@tough willow

[dry mirage]

you can, it's just that the repr of bools use the id of the singletons instead of their value

!e

from einspect import view, impl, orig

@impl(bool)
def __repr__(self):
    if self is False and int(self):
        return repr(True)
    return orig(bool).__repr__(self)
 
with view(False).unsafe() as v:   
    v <<= True
    
print(False)
print(True)
print(False == True)
print(False is True)

[night quarry]

@dry mirage :white_check_mark: Your 3.11 eval job has completed with return code 0.

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

[languid hare]

oh nice i didnt know einspect was added to !e

[quartz wave]

who pinged

[low lynx]

why do you need view().unsafe()?

[dry mirage]

!e cuz the memory move not safe 😔

from einspect import view

v = view(True)
v <<= False

[night quarry]

@dry mirage :x: Your 3.11 eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 4, in <module>
003 |   File "/snekbox/user_base/lib/python3.11/site-packages/einspect/views/view_base.py", line 243, in __lshift__
004 |     return self.move_from(other)
005 |            ^^^^^^^^^^^^^^^^^^^^^
006 |   File "/snekbox/user_base/lib/python3.11/site-packages/einspect/views/unsafe.py", line 55, in unsafe_call
007 |     raise UnsafeError(f"Call to {func.__qualname__} requires unsafe context")
008 | einspect.errors.UnsafeError: Call to View.move_from requires unsafe context

[earnest wing]

8. is just 10) in disguise

[low lynx]

fair enough

[sick hound]

guessing_game = (lambda number: print("Guess the number!") or (list((lambda itertools = __import__("itertools"): itertools.takewhile(lambda guess: ((not ((guess[1:] if guess.startswith("-") else guess).isnumeric() or print("Enter an integer!"))) or int(guess) != number and ((int(guess) > number and print("Guess lower!")) or (int(guess) < number and print("Guess higher!")) or True)), (input("--> ") for _ in itertools.count())))()) or print("First try! Damn!")) and print("Well done!"))

[fleet bridge]

i believe there exists some other string values that are special to interpreter, but are not variable/method/module/file names, but i cant recall any

[finite blaze]

Hey, what would be the shortest way to check if a numer is prime?

Numbers from 0 to 10000

Speed doesnt matter

[last locust]

Probs something like all(n%x!=0 for x in range(n))
Or for a fair bit more efficiency, range(int(n**.5)+1)

[low lynx]

just all(n%p for p in range(n)) is enough

since 0 is false

[finite blaze]

Why the second one is more efficient?

[astral rover]

the first prime factor cannot be greater than the numbers sqrt

[low lynx]

they mean all(n%x!=0 for x in range(int(n**.5)+1))

[astral rover]

hence you dont need to check anything after that

[last locust]

So it doesn't work

You'd need not all

Or I guess any() rather

[low lynx]

!e

is_prime = lambda n:all(n%p for p in range(2,n))
print(is_prime(2), is_prime(17), is_prime(81))

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

False False False

[low lynx]

ah we need a 1

[last locust]

Yeah

[low lynx]

no a 2

x % 1 is

!e

is_prime = lambda n:all(n%p for p in range(2,n))
print(is_prime(2), is_prime(17), is_prime(81))

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

True True False

[low lynx]

yeah

[last locust]

Right you meant that

[low lynx]

you want to check if it isn't divisible and if it isn't divisible then it leaves a nonzero remainder

[quartz wave]

!e ```py
f=lambda x:all(x%i for i in range(2,x))
print(f(0), f(1), f(5), f(15))

[night quarry]

@quartz wave :white_check_mark: Your 3.11 eval job has completed with return code 0.

True True True False

[low lynx]

which is truthy

[quartz wave]

um

0 and 1?

[low lynx]

we ignore 0 and 1

those are annoying

[quartz wave]

no we don't
it's part of the golf now

[low lynx]

fine

[quartz wave]

!e ```py
f=lambda x:x>1==all(x%i for i in range(2,x))
print(f(0), f(1), f(5), f(15))

ok

[low lynx]

5 isn't prime damn

[night quarry]

@quartz wave :white_check_mark: Your 3.11 eval job has completed with return code 0.

False False True False

[low lynx]

if you already have a factorial defined then

!e

g=lambda n:0**n or n*g(n-1)
f=lambda x:x>1>~g(x-1)%x
print(*map(f, [0, 1, 2, 5, 13, 15, 19, 53]))

[night quarry]

@low lynx :white_check_mark: Your 3.11 eval job has completed with return code 0.

False False True True True False True True

[restive void]

!e

p=lambda n:not __import__("re").match("1?$|(11+)\\1+$","1"*n)
print(*map(p, [0, 1, 2, 5, 13, 15, 19, 53]))

[night quarry]

@restive void :white_check_mark: Your 3.11 eval job has completed with return code 0.

False False True True True False True True

[low lynx]

wow that's clever

[finite blaze]

What's going on there?

What's this regex doing?

[low lynx]

it basically checks against a string of n 1s

(11+) matches against some number of 1s greater than or equal to 2

and \1+ matches against some number of the above pattern

so if n is divisible by some integer p

then (11+) would be able to match against p and \1+ would match (n/p)-1 times, and you would have a match

but if it's prime and no such p exists, then you don't get a match

[fleet bridge]

i love this trick

[plush halo]

need help with this

!e```py
from fishhook import hook

@hook(int)
def add(self, other):
return 11

print(1 + 1)

[night quarry]

@plush halo :white_check_mark: Your 3.11 eval job has completed with return code 0.

2

[restive void]

Constant folding, not much you can do about it..

[plush halo]

maybe i can override __new__?

[restive void]

!e

from fishhook import hook

@hook(int)
def __add__(self, other):
    return 11

x = 1
print(x+1)

[night quarry]

@restive void :white_check_mark: Your 3.11 eval job has completed with return code 0.

11

[plush halo]

oof

[restive void]

Won't help. The compiler sees 1 + 1 in the code and pretty much replaces it with 2.

[plush halo]

yeah you make a point

thx for the help @restive void

oh and one more thing @restive void sry for the ping

!e```py
from future import barry_as_FLUFL
from fishhook import hook

@hook(bool)
def ne(self, _): return False

print(True <> False)```

[night quarry]

@plush halo :x: Your 3.11 eval job has completed with return code 1.

001 |   File "<string>", line 7
002 |     print(True <> False)
003 |                ^^
004 | SyntaxError: invalid syntax

[plush halo]

barry_as_FLUFL doesn't work here :/

[rugged sparrow]

weird

[dry mirage]

I think it's REPL only now

[plush halo]

i guess... thx for the tip

and btw @dry mirage , you're the creator of einspect?

[versed eagle]

yep

[dry mirage]

mhm

[versed eagle]

im not sure why, though

[rugged owl]

They usually do that with "april-fools" features

afaik

[plush halo]

oh cool.. i just heard of it today

i'll add it to my forbidden modules(along with fishhook)

[dry mirage]

!e you could do this tho

from __future__ import barry_as_FLUFL
from fishhook import hook

@hook(bool)
def __ne__(self, _):
    return False

print(eval("True <> False"))

[night quarry]

@dry mirage :white_check_mark: Your 3.11 eval job has completed with return code 0.

False

[dry mirage]

since compile inherits futures

[plush halo]

i thought eval was cheating 😅

but i guess anything goes in this twisted world of esoteric python xD

[void ivy]

from inspect import getouterframes as ___,currentframe as ____
__=setattr
_=lambda _:(__(_,'_',_),print(len(___(____()))-4),_(_._));_(_)

I don't think this is how you are supposed to do a python for loop

[rugged owl]

Looks right

[void ivy]

You sure? Maybe I should rewrite some old programs.

[rugged owl]

!e

from inspect import getouterframes as ___,currentframe as ____
__=setattr
_=lambda _:(__(_,'_',_),print(len(___(____()))-4),_(_._));_(_)

[night quarry]

@rugged owl :x: Your 3.11 eval job timed out or ran out of memory.

001 | -2
002 | -1
003 | 0
004 | 1
005 | 2
006 | 3
007 | 4
008 | 5
009 | 6
010 | 7
011 | 8
... (truncated - too many lines)

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

[plush halo]

!e```py
i = -3
while True:
i += 1
print(i)

[void ivy]

Weird how that starts at -2

[plush halo]

xD yeah weird

[void ivy]

Starts at 0 for me.

When I run on my machine

[earnest wing]

[void ivy]

Basically what the function is doing is calling itself and counting how deep it is in recursion.

[rugged sparrow]

!e ```py
{*iter(lambda n=[]:(print(len(n)),n.append(1))if len(n) < 10 else None, None)}

[night quarry]

@rugged sparrow :white_check_mark: Your 3.11 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

[void ivy]

Fun fact, a python scrip can import itself. Like if you have a script called foo.py, you can do import foo. Using that, I came up with this loop

import importlib.util
import sys
from inspect import getouterframes, currentframe
print(len(getouterframes(currentframe()))//3)
spec = importlib.util.spec_from_file_location( __file__[-__file__[::-1].find(r'/'):], __file__)
foo = importlib.util.module_from_spec(spec)
sys.modules[__file__[-__file__[::-1].find(r'/'):]] = foo
spec.loader.exec_module(foo)

[void ivy]

If you run into a maximum recursion depth error in this first script I posted, or the one I just posted, you can add something like this to increase the recursion depth ```py
sys.setrecursionlimit(1500)

[long fulcrum]

any way to golf this any further:

Q=range
K=enumerate
G='#'
E='.'
C=len
import sys
for P in sys.argv[1:]:A=P.split('\n')
for H,V in K(A):A[H]=A[H].center(C(A[H])+2,E)
A.insert(0,E*C(A[0]))
A.append(E*C(A[0]))
L=[]
for B,I in K(A):
    D=''
    for F,W in K(I):
        M=[]
        if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:J=0
        else:[[M.append(A[B+N][F+O])for O in Q(-1,2)if N or O]for N in Q(-1,2)]
        J=M.count(G)
        if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:continue
        elif A[B][F]==G:D+=G if 1<J<4else E
        elif J==3:D+=G
        else:D+=E
    if B>=1:L.append(D)
for I in L:print(I)```

its incredibly ugly and also long

[gleaming linden]

Indent with 1 space for one

[long fulcrum]

i am

its tabs discord just converts

[low lynx]

firstly

C,*L=len,

ternary looks to be replacable with [] indexing

[long fulcrum]

i have no idea what that is tbh

[gleaming linden]

print(*L,sep='\n')

[low lynx]

1 if x else 2
[2,1][x]

if x is a condition

that returns a bool

so

D+=[E,G][1<J<4]

[gleaming linden]

A.append(x)
``` can also be shortened to ```py
A+=x,

[low lynx]

.insert(0) as well

probably combine the appending on both sides or smthing

[long fulcrum]

how can the insert be shortened?

[gleaming linden]

More unpacking

[low lynx]

well at the very least

A=[x]+A

but

A=x,*A,y

exists

[gleaming linden]

A.insert(0,x)
A.append(y)
# becomes
*A,=x,*A,y

[low lynx]

ah wait

we might need *A,=x,*A,y

yeah

[long fulcrum]

got this

j=E*C(A[0])
*A,=j,*A,j```

[gleaming linden]

What is this supposed to do though?

[long fulcrum]

its game of life

[gleaming linden]

Ah

[low lynx]

for P in sys.argv[1:]:A=P.split('\n')

why are you assigning A in a for loop

[long fulcrum]

oh also got this: (G,E)='#.'

I think its because theres an empty item in argv[0]?

let me check without it

[gleaming linden]

You don't need parens there

[long fulcrum]

yeah if I try A=sys.argv[1:].split('\n') I get:

Traceback (most recent call last):
  File "<stdin>", line 6, in <module>
AttributeError: 'list' object has no attribute 'split'
exit status 1```

[gleaming linden]

can't it just be sys.argv[-1]

[long fulcrum]

oh shit it can

im just so used to having to go through multiple things in argv

[gleaming linden]

Q can also be range(-1,2) instead of just range

[long fulcrum]

oh nice

[gleaming linden]

wait

what does .center() do in L8

oh it's a string

[low lynx]

it just puts . on either side I'm pretty sure

[long fulcrum]

changed this py if not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}): if A[B][F]==G:D+=[E,G][1<J<4] elif J==3:D+=G else:D+=E

[gleaming linden]

isn't E+A[H]+E shorter then

[long fulcrum]

yeah its just padding it

[low lynx]

honestly lines 7 to 10 could be shortened to one line

[long fulcrum]

yep

[gleaming linden]

A=[E+i+E for i in sys.argv[-1].split('\n')]

[long fulcrum]

this is lines 7-10 now

A=sys.argv[-1].split('\n')
for H,V in K(A):A[H]=E+A[H]+E
j=E*C(A[0])
*A,=j,*A,j```

[gleaming linden]

this can replace 7-8

[long fulcrum]

holy shit its down to 433 now

[gleaming linden]

we're not even onto the body yet lmao

[long fulcrum]

this is what it looks like now:

Q=range(-1,2)
K=enumerate
G,E='#.'
C,*L=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B,I in K(A):
    D=''
    for F,W in K(I):
        M=[]
        if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:J=0
        else:[[M.append(A[B+N][F+O])for O in Q if N or O]for N in Q]
        J=M.count(G)
        if not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}):
            if A[B][F]==G:D+=[E,G][1<J<4]
            elif J==3:D+=G
            else:D+=E
    if B>=1:L+=D,
print(*L,sep='\n')```

[gleaming linden]

this is definitely improvable

[low lynx]

fix the spaces

golf code looks so bad with just huge swathes of spaces in the middle

[gleaming linden]

Q=range(-1,2)
K=enumerate
G,E='#.'
L,*o=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*L(A[0])
*A,=j,*A,j
for B,I in K(A):
 D=''
 for F,W in K(I):
  M=[]
  if{B}&{0,L(A)-1}or{F}&{0,L(A[B])-1}:J=0
  else:[[M.append(A[B+N][F+O])for O in Q if N or O]for N in Q]
  J=M.count(G)
  if not({B}&{0,L(A)-1}or{F}&{0,L(A[B])-1}):
   if A[B][F]==G:D+=[E,G][1<J<4]
   elif J==3:D+=G
   else:D+=E
 if B>=1:o+=D,
print(*o,sep='\n')
```same thing with single space indent (also I renamed `len` to `L`)

[low lynx]

too many if else's

[gleaming linden]

that too

[meager zinc]

use ternary or [1,2,3][(math involving bools)] for the elifs

[gleaming linden]

you can convert first if/else into ternary

[meager zinc]

Instead of if N or O you can do if+N+O actually nevermind

[long fulcrum]

yeah I would probably have better variable names but I just run my readable code through a minifier to remove all unnecessary spaces and change variables to a single character

[low lynx]

the second condition seems unoptimal

I usually do capital letters for functions and lowercase for variables in golf

usually

[meager zinc]

same

[low lynx]

unless the case of variables matter

[meager zinc]

and usually I'd make it the first letter of the function (E=enumerate)

[low lynx]

E, R, L=enumerate,range,len

[icy sleet]

what if you're doing something like x.y.z(), same thing?

[meager zinc]

as in?

[icy sleet]

for renaming the function name

[versed eagle]

instead of if N or O, can you use if N|O

[low lynx]

depends on the situation

[meager zinc]

J=''.join?

[meager zinc]

typically it will be the name of the final function

Z=x.y.z

but again, it depends

[icy sleet]

Fair

[long fulcrum]

this fails with an error

[[[M.append(A[B+N][F+O])for O in Q if N or O]for N in Q],J=0][{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}]```

[low lynx]

if you're importing a module then it definitely depends

[versed eagle]

use walrus

[long fulcrum]

doesnt work

[versed eagle]

instead of J=0

use a walrus

[meager zinc]

Using walrus is helpful in other cases too:

[gleaming linden]

Q=range(-1,2)
K=enumerate
G,E='#.'
L,*o=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*L(A[0])
*A,=j,*A,j
for B,I in K(A):
 D=''
 for F,W in K(I):
  J=[sum(G==A[B+x][F+y]for x in Q for y in Q if x|y),0][L({B}&{0,L(A)-1}|{F}or{0,L(A[B])-1})]
  if not({B}&{0,L(A)-1}or{F}&{0,L(A[B])-1}):
   if A[B][F]==G:D+=[E,G][1<J<4]
   elif J==3:D+=G
   else:D+=E
 if B>=1:o+=D,
print(*o,sep='\n')
``` I think this should work

untested though

[long fulcrum]

doesnt work

[versed eagle]

then show what you did

[meager zinc]

# WALRUS:
A=[E+i+E for i in sys.argv[-1].split(n:='\n')]
print(*o,sep=n)

[versed eagle]

instead of saying "doesn't work"

and not providing any detail

whatsoever

[long fulcrum]

nope

[meager zinc]

n:= is one char less than '\n'

[meager zinc]

Traceback, logic error, output?

[long fulcrum]

[[[M.append(A[B+N][F+O])for O in Q if N or O]for N in Q],j:=0][{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}]

[gleaming linden]

nvm I made a mistake ```py
Q=range(-1,2)
K=enumerate
G,E='#.'
L,o=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=EL(A[0])
*A,=j,*A,j
for B,I in K(A):
D=''
for F,W in K(I):
J=[sum(G==A[B+x][F+y]for x in Q for y in Q if x|y),0][L({B}&{0,L(A)-1}or{F}&{0,L(A[B])-1})]
if not({B}&{0,L(A)-1}or{F}&{0,L(A[B])-1}):
if A[B][F]==G:D+=[E,G][1<J<4]
elif J==3:D+=G
else:D+=E
if B>=1:o+=D,
print(*o,sep='\n')

[long fulcrum]

list index error

[meager zinc]

Line, traceback?

[long fulcrum]

string index out of range

[long fulcrum]

TypeError: list indices must be integers or slices, not set

[meager zinc]

line #?

[gleaming linden]

same line?

[long fulcrum]

13

yeah same line

[meager zinc]

are you on 3.11 ?

[long fulcrum]

im pretty sure yeah

[meager zinc]

can you give the detailed traceback?

[gleaming linden]

^

[versed eagle]

the ... or{F}&{0,L(A[B])-1} is returning a set i think

[gleaming linden]

I put L() around it

[long fulcrum]

of this or this?

[meager zinc]

it would look like

x[1] / y / [2] / 0
         ^
ZeroDivisionError

[versed eagle]

ah nvm

i missed that

[meager zinc]

idk why it would be zerodivisonerror there but you get my point

[gleaming linden]

string index should only be the [F+O] in A[B+N][F+O]

[versed eagle]

this should work
if A[B][F]==G:D+=[E,G][1<J<4]
->
D+=[E,G][1<J<4]*(A[B][F]==G)

[long fulcrum]

this gives SyntaxError: assignment expression cannot be used in a comprehension iterable expression btw

[gleaming linden]

why can't you just loop from 1 to len-1

then you don't need to check if it's an edge cell

[meager zinc]

oh yeah I forgot about that

its dumb - it should be allowed

maybe I should write a PEP

[long fulcrum]

which one is this for?

[gleaming linden]

the for loops

B,I in K(A) and F,W in K(I)

you don't even use W so you can just use range here

[meager zinc]

but does that save anything?

[long fulcrum]

if I change it to for F in range(I): it doesnt work

[versed eagle]

you should be able to do this
if A[B][F]==G:D+=[E,G][1<J<4]
elif J==3:D+=G
else:D+=E
->
D+=[E,G][1<J<4]*(A[B][F]==G)or(((J==3)*G)or E)

[long fulcrum]

TypeError: 'str' object cannot be interpreted as an integer

[gleaming linden]

why not?

[long fulcrum]

string index out of range

[long fulcrum]

@gleaming lindenthis error

[gleaming linden]

oh

[versed eagle]

can you provide more context please?

[gleaming linden]

shouldn't it be range(len(i))

[versed eagle]

oh nvm ik why it broke

[gleaming linden]

yeah you forgot len()

[meager zinc]

well isn't that way longer

*longer

rangeint()()
enumerate()

[versed eagle]

this should work now
if A[B][F]==G:D+=[E,G][1<J<4]
elif J==3:D+=G
else:D+=E
->
D+=((A[B][F]==G)and[E,G][1<J<4])or(((J==3)*G)or E)

[long fulcrum]

yeah its longer

[long fulcrum]

  File "<stdin>", line 12, in <module>
  File "<stdin>", line 12, in <genexpr>
IndexError: string index out of range
exit status 1```

[versed eagle]

as long as neither E or G are empty

[gleaming linden]

we don't need enumerate anymore

[meager zinc]

I'm comparing their lengths

by using the monospace font

[versed eagle]

why are you telling me

when it breaks on something else

line 12 is not what i was messing with

[meager zinc]

rangeint()()
enumerate(),W # <-- longer

I didn't include the ,W

[gleaming linden]

we can do R=range as well

and L for len

[long fulcrum]

oh shit I thought you were trying to fix that code

[versed eagle]

ah

[long fulcrum]

yeah that works

[versed eagle]

im making the ifs below it shorter

can you send what the code currently is

[long fulcrum]

Q=range(-1,2)
K=enumerate
G,E='#.'
C,*L=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B,I in K(A):
    D=''
    for F,W in K(I):
        M=[]
        J=0if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}else[[M.append(A[B+N][F+O])for O in Q if N or O]for N in Q]
        J=M.count(G)
        if not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}):D+=[E,G][1<J<4]*(A[B][F]==G)or(((J==3)*G)or E)
    if B>=1:L+=D,
print(*L,sep='\n')```

[versed eagle]

im having trouble keeping all the changes sorted in my mind

[versed eagle]

at ty

[meager zinc]

ok is that one working

[long fulcrum]

yeah that works

[gleaming linden]

this probably doesn't work but just putting this out there ```py
R=range
Q=R(-1,2)
K=enumerate
G,E='#.'
L,*o=len,
import sys
A=[E+i+E for i in sys.argv[-1].split('\n')]
W=L(A[0])-1
H=L(A)-1
A,=EW,A,EW
for B in R(1,H):
D=''
for F in R(1,W):
J=sum(G==A[B+x][F+y]for x in Q for y in Q if x or y)
if A[B][F]==G:D+=[E,G][1<J<4]
elif J==3:D+=G
else:D+=E
o+=D,
print(*o,sep='\n')

[meager zinc]

there's unnecessary spaces I can tell

in the other one

[long fulcrum]

  File "<stdin>", line 14, in <module>
  File "<stdin>", line 14, in <genexpr>
IndexError: string index out of range
exit status 1```

[gleaming linden]

smh

[versed eagle]

enthusiastic person
you can get rid of the if-elif-else block

[gleaming linden]

yes ik

I just saw that

[meager zinc]

!e ```py
c = import('ctypes.util')
c_lib = c.CDLL(c.util.find_library('c'))
c_lib.puts(
c.cast(
b'Hello C!',
c.POINTER(c.c_char),
)
)

[night quarry]

@meager zinc :white_check_mark: Your 3.11 eval job has completed with return code 0.

Hello C!

[meager zinc]

only works on certain pythons

[versed eagle]

won't work on windows

since windows users dont have libc

[gleaming linden]

@long fulcrum can you give us a test input + expected output

so we can test it ourselves

[long fulcrum]

wait no hold on

[meager zinc]

Traceback (most recent call last):
  File "C:\Users\Desktop\main.py", line 2, in <module>
    c_lib = c.CDLL(c.util.find_library('c'))
  File "C:\Users\AppData\Local\Programs\Python\Python310\lib\ctypes\__init__.py", line 364, in __init__
    if '/' in name or '\\' in name:
TypeError: argument of type 'NoneType' is not iterable

[versed eagle]

mhm

[long fulcrum]

why does it keep messing up the first line

[meager zinc]

paste it in a code block

[long fulcrum]

it is the code block is deleting it for some reason?

ill use a hastebin

[meager zinc]

!paste

[night quarry]

Pasting large amounts of code

If your code is too long to fit in a codeblock in Discord, you can paste your code here:
https://paste.pythondiscord.com/

After pasting your code, save it by clicking the floppy disk icon in the top right, or by typing ctrl + S. After doing that, the URL should change. Copy the URL and post it here so others can see it.

[long fulcrum]

or that yeah

https://paste.pythondiscord.com/sojuzepido

[gleaming linden]

make sure the first line is on the line after the backticks

[long fulcrum]

yeah that was the problem

its clears up chat having it on the site anyways

[meager zinc]

.............

[versed eagle]

this should work

if not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}):D+=[E,G][1<J<4]*(A[B][F]==G)or(((J==3)*G)or E)
```->
```py
D+=not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1})and[E,G][1<J<4]*(A[B][F]==G)or(((J==3)*G)or E)

[long fulcrum]

<stdin>:13: SyntaxWarning: invalid decimal literal

also outputs wrong

[versed eagle]

thats from J=0if{B}&{0,C(...

the 0if

[long fulcrum]

ok no errors but its not removing the left, bottom, and right padding

[versed eagle]

try now

if not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}):D+=[E,G][1<J<4]*(A[B][F]==G)or(((J==3)*G)or E)
```->
```py
D+=(not({B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}))and[E,G][1<J<4]*(A[B][F]==G)or((J==3)*G)or E

i forgot that not is after and

im tired lol

[gleaming linden]

imo we should just get rid of edge checks altogether

by not looping over them entirely

[long fulcrum]

nope same issue with padding

[versed eagle]

hm

so what is this program supposed to do?

whats the intended behaviour

[long fulcrum]

game of life

[versed eagle]

ah

ok

[long fulcrum]

you should have seen the code before

individual checks for each of the 4 edges and 4 corners

[night quarry]

Hey @long fulcrum!

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

[versed eagle]

yeah that'd be better i think

[long fulcrum]

oh

well it looked like this

thats why I added the padding

to prevent that

[gleaming linden]

is using a dictset to store alive cells with complex numbers shorter

[versed eagle]

ooh thats a good idea

[long fulcrum]

i have never used complex numbers in python before

I didnt even think they had a use

[gleaming linden]

the only use we've had for them is 2d coordinates

[void ivy]

f=lambda :None
f.f=f
print(f is f.f.f)

#True
#False
#AttributeError

Without running the code, what do you think the answer is?

[versed eagle]

True

would be my guess

but im also tired as hell

so

[gleaming linden]

can you even assign attributes like that?

[versed eagle]

who knows

[versed eagle]

yea, you can assign function attributes

[gleaming linden]

then True probably

it's like that list self-nesting in #1051603408597024828

[void ivy]

It was indeed True. But still, wierd to me

Also you can only do f.f.f.f (and so on) until the recursion limit

[gleaming linden]

like .__call__

[versed eagle]

how is it weird?

[meager zinc]

you can only have a maximum of four nested f-strings without an eval or exec

[void ivy]

Never thought about assigning attributes to function like that, and didn't expect it to have the recursive effect, but it makes sense now

[versed eagle]

its the same as a class returning self for something

@type.__call__
class thing:
    def __call__(self):
        return self
thing()()()()()()()()()()()() is thing

[long fulcrum]

oh yeah this is what the golfed code was like before padding

M='.'
L=enumerate
G=len
E='#'
import sys
for I in sys.argv[1:]:B=I.split('\n')
H=[]
for A,J in L(B):
    F=''
    for C,N in L(J):
        D=0
        if A==0and C==0:D=[B[A][C+1],B[A+1][C],B[A+1][C+1]].count(E)
        elif A==0and C==G(B[A])-1:D=[B[A][C-1],B[A+1][C-1],B[A+1][C]].count(E)
        elif A==G(B)-1and C==0:D=[B[A-1][C],B[A-1][C+1],B[A][C+1]].count(E)
        elif A==G(B)-1and C==G(B[A])-1:D=[B[A-1][C-1],B[A-1][C],B[A][C-1]].count(E)
        elif A==0:D=[B[A][C-1],B[A+1][C-1],B[A+1][C],B[A+1][C+1],B[A][C+1]].count(E)
        elif A==G(B)-1:D=[B[A][C-1],B[A-1][C-1],B[A-1][C],B[A-1][C+1],B[A][C+1]].count(E)
        elif C==0:D=[B[A-1][C],B[A-1][C+1],B[A][C+1],B[A+1][C+1],B[A+1][C]].count(E)
        elif C==G(B[A])-1:D=[B[A-1][C],B[A-1][C-1],B[A][C-1],B[A+1][C-1],B[A+1][C]].count(E)
        else:D=[B[A-1][C-1],B[A-1][C],B[A-1][C+1],B[A][C-1],B[A][C+1],B[A+1][C-1],B[A+1][C],B[A+1][C+1]].count(E)
        if B[A][C]==E:
            if 2<=D<=3:F+=E
            else:F+=M
        elif D==3:F+=E
        else:F+=M
    H.append(F)
for K in H:print(K)```a complete fucking mess

[versed eagle]

cause there are 4 different types of string literals?

" ", ' ', """ """, ''' '''

[gleaming linden]

yes

you can't nest the same type

[flint hollow]

not yet, at least

[versed eagle]

time for a new PEP o.o

[meager zinc]

!e and this ```py
this=valid=False
print(f"{this * "is" * not valid}")

[night quarry]

@meager zinc :x: Your 3.11 eval job has completed with return code 1.

001 |   File "<string>", line 2
002 |     print(f"{this * "is" * not valid}")
003 |                      ^^
004 | SyntaxError: f-string: expecting '}'

[meager zinc]

discord syntax highlights incorrectly

[gleaming linden]

import sys
Q=range(-1,2)
K=enumerate
G,E='#.'
C,*L=len,
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B,I in K(A):
 D=''
 for F,W in K(I):
  if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:continue
  J=sum(G==A[B+N][F+O]for O in Q for N in Q if N or O)
  D+=[E,G][2-(A[B][F]==G)<J<4]
 L+=D,
print(*L,sep='\n')
```still optimizable

[meager zinc]

its actually more like

print("{this * "is" * not valid}")

[versed eagle]

f"{"something or other"}" should be possible imo
since it's inside the {, it's not ambiguous

[flint hollow]

!pep 701 why new?

[night quarry]

**PEP 701 - Syntactic formalization of f-strings**

Link

Status

Draft

Python-Version

3.12

Created

15-Nov-2022

Type

Standards Track

[meager zinc]

Ooh

[versed eagle]

cause im too lazy to read current PEPs until like 3 years after they're accepted and implemented

[long fulcrum]

almost works

[meager zinc]

do you know if there's a PEP on the usage of := in a comprehension?

[long fulcrum]

it prints a newline at the start

import sys
Q=range(-1,2)
K=enumerate
G,E='#.'
C,*L=len,
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B,I in K(A):
 D=''
 for F,W in K(I):
  if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:continue
  J=sum(G==A[B+N][F+O]for O in Q for N in Q if N or O)
  D+=[E,G][2-(A[B][F]==G)<J<4]
 if B!=0:L+=D,
print(*L,sep='\n')```

this works

[versed eagle]

know what i want?

i want continue and break

[gleaming linden]

import sys
R=range
Q=R(-1,2)
K=enumerate
G,E='#.'
C=len
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B in R(1,C(A)):
 D=''
 for F in R(C(A[0])):
  if{B}&{0,C(A)-1}or{F}&{0,C(A[B])-1}:continue
  J=sum(G==A[B+N][F+O]for O in Q for N in Q if N or O)
  D+=[E,G][2-(A[B][F]==G)<J<4]
 print(D)
``` even better

[versed eagle]

to be expressions

[gleaming linden]

return as well

[long fulcrum]

oh shit

thats a crazy reduction

[versed eagle]

for example:
[continue if somecondition else x**2 for x in range(whatever)]

[long fulcrum]

410 to 321

[versed eagle]

ooh yeah

[gleaming linden]

because you don't need to repeat {B}&{0,C(A)-1}or{F}&{0,C(A[B])-1} besides other things

[low lynx]

wouldn't this just be [x**2 for x in range(whatever) if somecondition]?

[versed eagle]

it would, but that was a toy example to demonstrate how im imagining the syntax would work

[gleaming linden]

this is one-linable now

[long fulcrum]

some people have done this in just over 100 chars which is crazy

[long fulcrum]

my code is?

[gleaming linden]

yes

[long fulcrum]

how wtf

[gleaming linden]

import sys
R=range
Q=R(-1,2)
K=enumerate
G,E='#.'
C=len
A=[E+i+E for i in sys.argv[-1].split('\n')]
j=E*C(A[0])
*A,=j,*A,j
for B in R(1,C(A)-1):print(''.join([E,G][2-(A[B][F]==G)<sum(G==A[B+N][F+O]for O in Q for N in Q if N or O)<4]for F in R(1,C(A[0])-1)))
``` just 1 more loop

[long fulcrum]

all i can see is a semicolon to get combine py J=sum(G==A[B+N][F+O]for O in Q for N in Q if N or O) D+=[E,G][2-(A[B][F]==G)<J<4]

[low lynx]

I mean the code was always onelineable

if you use exec

[meager zinc]

can we not do N|O

instead of N or O

[long fulcrum]

holy shit

[gleaming linden]

yeah N|O should work now

[versed eagle]

i suggested that a while ago 😢

[meager zinc]

It's also possible with deeply nested lambdas

[meager zinc]

you did?

whoops sorry

im working on homework in the background rn

[versed eagle]

great minds think alike ✨

[gleaming linden]

i did that here

x|y

idk where you did

[versed eagle]

here

[versed eagle]

i didnt even see when you did it lol

[gleaming linden]

yeah because that was on a nonfunctional version

[versed eagle]

ah

[gleaming linden]

and I renamed them to x and y

[versed eagle]

ah

thats probably why i didnt notice

[long fulcrum]

packing is the stupidest way to reduce chars and I hate that it works

like this does everything my code does in one line

exec(bytes('浩潰瑲猠獹刊爽湡敧儊刽⴨ⰱ⤲䬊攽畮敭慲整䜊䔬✽⸣ਧ㵃敬੮㵁䕛椫䔫映牯椠椠祳⹳牡癧ⵛ崱献汰瑩✨湜⤧੝㵪⩅⡃孁崰਩䄪㴬ⱪ䄪樬昊牯䈠椠⡒ⰱ⡃⥁ㄭ㨩牰湩⡴✧樮楯⡮䕛䜬孝ⴲ䄨䉛孝嵆㴽⥇猼浵䜨㴽孁⭂嵎䙛伫晝牯传椠⁑潦⁲⁎湩儠椠⁦籎⥏㐼晝牯䘠椠⡒ⰱ⡃孁崰⴩⤱⤩','u16')[2:])```

[gleaming linden]

and it's also longer (if you count bytes instead of chars)

[meager zinc]

I love it when there's a \✨ in the compiled (encoded) code

just magical

[long fulcrum]

yeah thats its only saving grace

[meager zinc]

You have to add an extra two bytes at minimum to encode it in UTF16

0xFF 0xFE DATA