Programming thread

[Belisarius Cawl]

What’s wrong, chud? Scared of a few greek letters? You will never reach programming Nirvana if you can only think in the latin alphabet

Greek letters are all played out, basically the new ASCII. I only use the Tibetan Unicode block for my operators.

 

[ADHD Mate]

I like unicode in my code. It looks pretty.

 

[Belisarius Cawl]

I like unicode in my code. It looks pretty.

Broke: anime feet wallpapers
Joke: tiling WMs
Bespoke: https://docs.raku.org/language/setbagmix

 

[Rango Dango]

No thanks, I'll take my future with Mr. God and HolyC.

A helpful table of APL symbols
Ver archivo adjunto 8355198

still better than J - https://code.jsoftware.com/wiki/NuVoc

to be honest i like those types of languages
i personally use uiua from time to time, it has way too much faggotry for my tastes but aside from that the language is interesting

its meant to be something like a stack based APL, but recently the language has been switching from forth-like stack managment like swap or dup or whatever to i guess more idiomatic approach of manipulating the arguments while calling a function like calling a function with order of arguments reversed (n-adic/polyadic commute) or more familiar self modifier which calls a function with all the arguments being the value on top of the stack (in APL also uses the same glyph as commute funnily enough)

---

if you want a more modern APL theres a language called BQN https://mlochbaum.github.io/BQN/
i like it, but i cant use it for the same reason i cant use apl (i cant comprehend the uhh "flow" of data)

 

[General Zoider]

when it comes to unicode and special charaters in source code I only care if it requires a special compiler version or it really fucks up my text editor because its something like the windows xp Japanese locale.

 

[Rango Dango]

when it comes to unicode and special charaters in source code I only care if it requires a special compiler version or it really fucks up my text editor because its something like the windows xp Japanese locale.

shiftjis my beloved

 

[General Zoider]

shiftjis my beloved

at the time it was some c++ project for I think a visual novel engine I was trying to patch so it wouldnt crash on windows 10 but ALL the files were in winxp jap locale.

 

[Rango Dango]

at the time it was some c++ project for I think a visual novel engine I was trying to patch so it wouldnt crash on windows 10 but ALL the files were in winxp jap locale.

ive always used this one for locale emulation https://xupefei.github.io/Locale-Emulator/
always worked with 32 bit japanese locale software

 

[General Zoider]

ive always used this one for locale emulation https://xupefei.github.io/Locale-Emulator/
always worked with 32 bit japanese locale software

true I was just trying to modify the code so it wouldnt need that emulator. Lets just say the japanese are even less adept at writing good comments then amaricans.

 

[Anti Snigger]

Mfw after writing code that places the responsibility on the caller and I am now the caller

 

[ArgonianVoter]

I wish:

def many(f, n):
return (f() for _ in range(n))
#used like:
v, c, d, b = many(list, 4)

or something very similiar
came with default python. I have to do it for nearly every single project.

I don’t want to define helpers just to express something that is conceptually simple and safe.
chained assignements evaluate each expression once, while tuple unpacking does it seperately and neither are very much fun.
there's no standard syntax for evaluationg expressions (O)N to unpack the results.
I think new syntax would be justified.
and no

v, c, d, b = iter(list, None)

is not acceptable because it breaks for ints, bools, strings and floats.
I would also love something for the leftern side of the equation too but asking for that's a stretch that will likely never be met, even lesso than a default 'many'.

 

[WielkiWapiesz2137]

adding syntax for something that can be expressed in one line with map lambda and range is nonsensical

 

[Rango Dango]

I wish:

def many(f, n):
return (f() for _ in range(n))
#used like:
v, c, d, b = many(list, 4)

or something very similiar
came with default python. I have to do it for nearly every single project.

I don’t want to define helpers just to express something that is conceptually simple and safe.
chained assignements evaluate each expression once, while tuple unpacking does it seperately and neither are very much fun.
there's no standard syntax for evaluationg expressions (O)N to unpack the results.
I think new syntax would be justified.
and no

v, c, d, b = iter(list, None)

is not acceptable because it breaks for ints, bools, strings and floats.
I would also love something for the leftern side of the equation too but asking for that's a stretch that will likely never be met, even lesso than a default 'many'.

im pretty sure that goes against the zen of python

 

[ArgonianVoter]

im pretty sure that goes against the zen of python

Yeah well your stupid fucking Zen is wrong, why do you think we don't have constants outside of "social enforcement".

 

[Rango Dango]

Yeah well your stupid fucking Zen is wrong, why do you think we don't have constants outside of "social enforcement".

ion fucking know nigga

python itself doesnt follow the zen all that well
they added "t-strings" (template strings) to the language recently, so now you can format strings in like 5 different ways, which also violates the zen of python

 

[August Levasseur]

I wish:

def many(f, n):
return (f() for _ in range(n))
#used like:
v, c, d, b = many(list, 4)

or something very similiar
came with default python. I have to do it for nearly every single project.

I don’t want to define helpers just to express something that is conceptually simple and safe.
chained assignements evaluate each expression once, while tuple unpacking does it seperately and neither are very much fun.
there's no standard syntax for evaluationg expressions (O)N to unpack the results.
I think new syntax would be justified.
and no

v, c, d, b = iter(list, None)

is not acceptable because it breaks for ints, bools, strings and floats.
I would also love something for the leftern side of the equation too but asking for that's a stretch that will likely never be met, even lesso than a default 'many'.

Imagine not having a personal standard library you include in all your projects for exactly this kind of situation.

 

[ArgonianVoter]

Imagine not having a personal standard library you include in all your projects for exactly this kind of situation.

I avoid doing that kind of thing now to prevent overheard and maintenance burden. I'm not typically working with 200 people, collabs don't usually happen that often for me and when I pass that code I want it to be understood without forcing others to understand my 'personal standard lib', so no that's by design.

 

[Betonhaus]

quick question in a G74 cycle it cannot be used with G02 and G03 movements, only G01 movements with no angular interpolation, right? I've been trying to make it work but just keep getting errors, do I just need to hand write a function to replace G74 but does the proper calculations for positions across an arc?

 

[Belisarius Cawl]

Drunk and watching an old episode of The Computer Chronicles; still worth watching today for the historical knowledge and context

And holy shit:

In pace requiescat

 

[306h4Ge5eJUJ]

adding syntax for something that can be expressed in one line with map lambda and range is nonsensical

Not like that stopped Guido before. But you can even use his beloved list comprehensions or generator expressions for this: a, b, c = ([] for _ in range(3))