flax

PyGamer0

5:03 AM

@lyxal can i delte cli-repl

the branch

lyxal

sure

PyGamer0

ok heres is my idea for glyphs:

nilads: subscripts symbols
monads: capital letters / letters from other languages (greek, cryllic etc)
dyads: small letters / symbols
quicks: superscript symbols

i am going to make a change-glyphs branch

@lyxal how is the idea ^^

lyxal

good

PyGamer0

imma copy glyphs from BQN/APL :)

PyGamer0

5:30 AM

    "₀": attrdict(arity=0, call=lambda: 100),
    "₁": attrdict(arity=0, call=lambda: [0, 1]),
    "₂": attrdict(arity=0, call=lambda: 10),
    "₃": attrdict(arity=0, call=lambda: sys.argv[1] if len(sys.argv) > 1 else 16),
    "₄": attrdict(arity=0, call=lambda: sys.argv[2] if len(sys.argv) > 2 else 32),
    "₅": attrdict(arity=0, call=lambda: sys.argv[3] if len(sys.argv) > 3 else 64),
    "₆": attrdict(arity=0, call=lambda: 0),
    "₈": attrdict(arity=0, call=lambda: ord(sys.stdin.read(1))),
    "₉": attrdict(arity=0, call=lambda: [ord(c) for c in input()]),

haha lol wow @lyxal ^

lyxal

Beautiful

PyGamer0

i also have an idea if i need to implement flax in C++

also uh for monads i can only use uppercase letters and letters from greek etc

should i change that cause i see:

    "!": attrdict(arity=1, call=vectorised(factorial)),
    "¬": attrdict(arity=1, call=vectorised(lambda a: 1 if not a else 0)),
    "~": attrdict(arity=1, call=vectorised(lambda a: ~a)),

lyxal

No those symbols make sense

PyGamer0

ok

    "Π": attrdict(arity=1, call=lambda x: foldl1(op.mul, flatten(x))),
    "Σ": attrdict(arity=1, call=lambda x: sum(flatten(x))),

^ change them?

or no?

@lyxal should range vectorise?

@lyxal Jelly uses 1 based indexing

should flax use 0 based indexing?

in The APL Orchard, Dec 30 '21 at 10:52, by Adám

@PyGamer0 APL originally had only 1 due to being mathematical in nature (cf. Mathematica, MatLab, R, Julia, …) but it turned out suboptimal for some domains, and a choice was introduced. Didn't take long before people realised that a choice is worse than a suboptimal standard, and J (which didn't need backwards compatibility) chose 0.

CMP: should flax use 0 based indexing?

CMQ: Should Range vectorise?

lyxal

5:51 AM

0 is good

Range should vectorise

PyGamer0

ok

finally a major diversion from jelly

not including other things :p

@lyxal ok then now how useful is 1...n range?

lyxal

useful enough that it's in vyxal

PyGamer0

examples?

Razetime

@PyGamer0 it's everywhere

literally everywhere

PyGamer0

6:06 AM

BQN doesnt have it :P

Razetime

@PyGamer0 bqn isn't a bleeding golfing language

and when i mean "Everywhere" i mean in questions

PyGamer0

yeah the array languages are all practical

@Razetime should i add you as a collaborator?

Razetime

well what should i be contributing

PyGamer0

\_(○_○)_/

Razetime

if I have something to do on the repo you should add me

PyGamer0

6:12 AM

ok

well technically i am doing everyhting

    "ᵇ": attrdict(arity=1, call=vectorised(lambda a: a % 2)),

ok i guess i can replace ᵇ with Ḅ

    "Θ": attrdict(arity=1, call=lambda x: iterable(x, make_range=True).insert(0, 0)),

uh

10Θ→[0,0,1,2,3,4,5,6,7,8,9]

lol what

lyxal

looks like it does exactly what you tell it to do

although looking at the definition of iterable, tf?

PyGamer0

and if i run it in the repl it errors lmao

AttributeError: 'NoneType' object has no attribute 'xreplace'

Razetime

then search for xreplace in your folder

PyGamer0

no its from sympy

ill see what is the problem

Razetime

do you have a debug mode for your interpreter

PyGamer0

6:26 AM

like?

wdym

Razetime

like when you run the interpreter there is a flag that you can use

PyGamer0

no

Razetime

which prints the stack every time a command is called

your golflang is in the early stage so make debugging easier now

PyGamer0

stack?

Razetime

oh yeah nvm

just prints the ast and what is going into each atom

PyGamer0

6:28 AM

ok

uh did vectorisation break

because of sympy?

      10⍳+1
>
>
Traceback (most recent call last):
  File "C:\Users\a\AppData\Local\Programs\Python\Python310\lib\ru
npy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "C:\Users\b\AppData\Local\Programs\Python\Python310\lib\ru
npy.py", line 86, in _run_code
    exec(code, run_globals)
  File "c:\Users\c\Desktop\stuff\flax\flax\__main__.py", line 19,
 in <module>
    cli_repl()
  File "c:\Users\d\Desktop\stuff\flax\flax\main.py", line 31, in
cli_repl
    flax_print(variadic_chain(chain, *args))

this is dumb

what if i remove sympy

Razetime

go ahead

wait what do you use it for

rational number type?

PyGamer0

yes

Razetime

you'll have to manually implement a rational type then

PyGamer0

hmmmmmmmmMMMMMM

@Razetime i think i implemented a debug flag

Razetime

6:44 AM

cool

good job

PyGamer0

and for some reason now it works lol

what it fixed itself

I am confused

Razetime

bruh

PyGamer0

      10I
>
>
range(0 10)

sympy

why

Razetime

i mean

not very surprising

sympy works on symbolic math

PyGamer0

why does it do that

Razetime

6:46 AM

you have to always change things back to ordinary stuff

i hope you have read at least one sympy tutorial?

PyGamer0

yes

i have read atleast one

actually i read liitledocs

ok what tf

WTF

@Razetime this is random, one time 10I+1 works the other time it errors

and similarly 10I works other times it range(0 10)

@lyxal ^

ill go take a break

Razetime

@PyGamer0 how can it be nondeterministic if the operation is deterministic

have you pushed the changes that cause this? I wanna try it out

PyGamer0

@Razetime pushed

its in the glyphs branch

Razetime

so i need to clone the glyphs branch?

PyGamer0

or checkout to it

Razetime

7:00 AM

ah yeah

what are the running instructions

PyGamer0

i run it like: python -m flax

and for debugging: python -m flax d

i think it doesnt error when debugging

Razetime

from which directory

flax/?

PyGamer0

yes

but i guess i can do that from anywhere

Razetime

$ python -m flax
Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/home/razetime/Documents/Code/fun/flax/flax/__main__.py", line 3, in <module>
    from prompt_toolkit import print_formatted_text as pft, HTML
ModuleNotFoundError: No module named 'prompt_toolkit'

PyGamer0

oh poetry install?

then python -m flax

Razetime

7:08 AM

you really need some build instructions from scratch

:60277505 $ poetry install
The currently activated Python version 3.8.10 is not supported by the project (>3.9).
Trying to find and use a compatible version.
Using python3.10 (3.10.2)
Creating virtualenv flax-YjccEFaI-py3.10 in /home/razetime/.cache/pypoetry/virtualenvs
Installing dependencies from lock file
Warning: The lock file is not up to date with the latest changes in pyproject.toml. You may be getting outdated dependencies. Run update to update them.

  SolverProblemError

  Because flax depends on sympy (*) which doesn't match any versions, version solving failed.

PyGamer0

hm poetry update?

Razetime

    File "/home/razetime/.cache/pypoetry/virtualenvs/flax-YjccEFaI-py3.10/lib/python3.10/site-packages/pip/_internal/utils/logging.py", line 13, in <module>
      from pip._internal.utils.misc import ensure_dir
    File "/home/razetime/.cache/pypoetry/virtualenvs/flax-YjccEFaI-py3.10/lib/python3.10/site-packages/pip/_internal/utils/misc.py", line 40, in <module>
      from pip._internal.locations import get_major_minor_version, site_packages, user_site
    File "/home/razetime/.cache/pypoetry/virtualenvs/flax-YjccEFaI-py3.10/lib/python3.10/site-packages/pip/_internal/locations/__init__.py",…

PyGamer0

○_○

Razetime

i dunno what's causing this

PyGamer0

even me

Razetime

7:22 AM

what is your python version

PyGamer0

can you remove the poetry.lock and try poetry install again?

@Razetime 3.10.0

Razetime

ok same

@PyGamer0 same modulenotfound error

PyGamer0

manually install the dependencies?

pip install sympy prompt_toolkit pyhof more_itertools

Razetime

ugh

ok it works

so

$ python -m flax
flax REPL version 0.1.0
      10I
>
>
[0 1 2 3 4 5 6 7 8 9]
      10I
>
>
range(0 10)
      10I
>
>
range(0 10)
      10I
>
>
range(0 10)
      10I
>
>
range(0 10)

see a pattern here?

after the first run it seems to always return range(1 10)

And it always does this

so you're doing something wrong where I is messing up after first time it's used

PyGamer0

hm ok

i pushed a potential fix

it works on my machine

Razetime

7:29 AM

what was the problem

PyGamer0

ah the joys of using custom libraries

@Razetime pyhof

Razetime

aha

PyGamer0

apparently compose(list, range) refused to work

ill keep sympy

and remove pyhof

Razetime

why?

pyhof is made for your convenience

in fact you should patch pyhof to make sure that this works

I would recommend lambda x: list(range(x)) rather than [*range(x)]

PyGamer0

why?

Razetime

7:32 AM

because it's more readable

PyGamer0

ok

Razetime

[*range(x)] is best kept within golfing

PyGamer0

ok

ah the range(0, 10) problem is with pyhof

ill fix it later ○/

lyxal

7:49 AM

@PyGamer0 the problem is because of janky iterator behaviour inside wrapped functions

I'm working on a fix and a way to better describe it

But deepsource might give a better description

PyGamer0

where are you fixing it? pyhof?

lyxal

Yes

I'm running an analysis now

PyGamer0

ok

lyxal

Huh, deepsource doesn't have a term for what I suspect is happening

But basically, the first time you call a composed function, it has all the functions in func

The second time you call it though, there's no function in funcs

Because funcs has been exhausted because it's a generator

I'll or a

I'll pr a fix later

Razetime

compose looks like it has an easy recursive definition

PyGamer0

7:58 AM

@lyxal oh a generatoelr

Razetime

maybe that is my haskell self showing up

PyGamer0

oh i totally forgot to install haskell lol

lyxal

8:14 AM

@Razetime you could also just reduce by 2 function composition

Razetime

@lyxal ah yeah pretty much

lyxal

Use the higher order function stuff to construct the higher order function stuff

PyGamer0

8:32 AM

ato.pxeger.com/…

@Razetime @lyxal ^ does that work?

Razetime

have you run it

lyxal

I think it looks fine

PyGamer0

@Razetime ato.pxeger.com/…

looks like it works

Razetime

ok you have run it and it seems to be fine

basically never modify the original array

make a copy/deepcopy function or sumn and use that

PyGamer0

ok

lists are mutable, generators can get exhausted

i dont have the access to my computer rn so ill do it later.......

also we have passed 1000 messages

Feeds

8:47 AM

PyGamer0 has added Razetime to the list of this room's owners.

PyGamer0

9:03 AM

ill just open issues in flax and pyhof reminding me to fix em

Transcript for