Thread 105783385

Literally the only reason why I decided to learn it over C/C++
If there was an exact copy of C or C++ that had the equivalent of cargo/pip/npm/go mod I would use it in a heartbeat

And believe me, I tried giving C++ a shot, but vcpkg doesnt work with like half of libraries, I get linking errors, undefined functions even after I've properly included everything without error, manually getting the correct version of dependencies for the libraries you want to use is a pain in the ass
Its fucking horrendous, there is no excuse for this shit in 2025

Meanwhile with, say, python you just pip install and use the import keyword and It. Just. Werks. Same goes for cargo, npm, go, etc.

>>105783385 (OP)
The bad thing about modern-day dependency management is that it encourages you to just install modules everytime you have a minor problem.
Take the is-number js library for example.

>>105783617
>encourages you
maybe if youre a literal retard or third worlder
maybe try not being retarded?

I just spent half a day trying to get LLVM working with cmake and failed.
I tried C#'s LLVMSharp and it just worked.

Learn Go instead

>>105783626
Sounds like you passed reading comprehension by giving your English teacher blowjobs instead of studying.

the bad trhing about modern dependency manahbement is it lets you manage dependencies

>>105783699
I already know go and rust/c/c++ (one of them) is something I want to add to my resume/project portfolio

>>105783385 (OP)
Rust trannies hate GPL. This makes it a non-starter.

>>105783734
underrated post

>>105783385 (OP)
I just switched to Linux and then use cmake and everything just works. The only time I have crazy build problems with stuff is windows. But most of people’s complaints about C++ not having a package manager is a skill issue and people who can’t deal with a low level language maybe they should stick with JavaScript or something.

>>105783701
There is nothing wrong with the post youre replying to and youre clearly just seething that it says "third worlder"
Tryharding on capitalization and punctuation is a nice touch, way to out yourself Ranjesh

>>105783822
yeah I'm on windows
I can figure it out but I just know theres a lot of friction I'll run into eventually and I just don't want to deal with it
I've been meaning to install linux on a spare machine I have to turn it into a "work computer" for personal projects, so maybe then I'll get around to learning cpp

> But most of people’s complaints about C++ not having a package manager is a skill issue and people who can’t deal with a low level language
fuck you
everybody knows its shit and rust is also a "low level language" (meme term)
basically what youre saying is
>what, you don't like the taste of shit? cant handle the creamy texture? sounds like a skill issue

>>105783889
Unless you’re building something esoteric with not a lot of documentation it’s really not hard to build anything and link it with a cmake or make file on Linux. On windows it’s a little bit harder but you can still build it on something like visual studios and link the dlls and libs explicitly in cmake or your .prj and .sln files or whatever build system. Not that bad there’s other stuff like meson to check out as well. A lot of cross platforms solutions for building c/c++ projects.

>>105783822
>C++ works but you have to be on linux
and then they dare call C# "locked down"

>>105783822
>low level languages need to have chaotic retard mess build systems
that's now it works midwit-kun

>>105783822
>skill issue
yeah you dont have a job and will never have one

>>105783385 (OP)
>If there was an exact copy of C or C++ that had the equivalent of cargo/pip/npm/go mod I would use it in a heartbeat
Conan exists you retard.

I'm not dependent on most mentally ill men on this planet for basic code I can write myself.

>>105785328
>t. ESL
Capitalizing and punctuating your sentences isnt helping you here
That aside, try writing opengl, glfw, sdl, imgui, qt, or boost from scratch, fucking retard

>>105785313
How is this shit any different from vcpkg

>>105785556
Try writing code that does more than create buttons for retards to click on.

>>105783385 (OP)
>the equivalent of cargo/pip/npm/go
That could be the package manager of your distro if you weren't a faggot.

A lot of people mention package management as one reason they really like Rust and other, newer languages over C and C++. There is vcpkg and Conan for C and C++, but while ecosystem support for those are improving over time, it is still a lot more work than Rust. It is difficult for C and C++ to do well here due to backwards compatibility concerns and existing ecosystem. But, is improving slowly.
Rust is not perfect in regards to package managememt and dependencies, the orphan rule comes to mind, but it is acknowledged widely that it does way better than C and C++ currently.

>>105783385 (OP)
cargo is one of the most, if not THE most vulnerable part of rust
also wtf are you even working on that deps become a problem?

Give me the inside scoop on C++ modules:

1: Are C++ modules trash, and why/why not?

2: Is the specification and design of C++ modules the issue, or is it just the ecosystem and tooling that have to catch up, and all we have to do is just Be Patient™ and help out?

3: Was the specification of C++ modules changed officially/unofficially to fix it, like mandating that there may be no "import" declarations in any text included through the preprocessor? What's the story of C++ modules after it got accepted into C++20?

4: Why is there so much disagreement on the status of C++ modules, and what is right or wrong with them?

>>105786198
something that requires a GUI library (immediate mode is fine), plotting (somewhat optional, using rust I'll prob have to roll something myself), async, http, json unmarshaling, and websockets
>THE most vulnerable
dont care
this is a tool for personal use
also how is it more vulnerable than literally any other dependency management system (including c/c++)? using external libraries always carries risks, just dont be a retard/pussy

>>105786198
True, there have been complaints that projects in the Rust ecosystem have a tendency to pull in way too many (transitive) dependencies, making vetting of dependencies way more difficult, and there have also been cases of malicious packages that tried to mimic names of popular packages or expected package names.
>This week, GitHub user Askar Safin identified a malicious Rust crate 'rustdecimal' found on the crates.io registry. This package is a typosquat of the real 'rust_decimal' package and contains a rather elusive kind of obfuscation to download malware, as confirmed by the Sonatype security research team. The malware targets both Linux and macOS users.
It can be difficult to encourage more modest usage of powerful features in an ecosystem, and that can end up having practical consequences. One can of course avoid depending on a lot of projects, but that may limit what options a project has for its dependencies, including depending on popular projects that may themselves have a lot of transitive dependencies.

>>105786242
>a whole lot of webshit
yeah, thats definitely not the usecase for C
id expect more of sepples, doe. theyre the ones who want to be a jack of all trades master of none
>also how is it more vulnerable than literally any other dependency management system
recursive deps
younger ecosystem
worthless community workethics

so not only theres more to audit
not only theres less tools to do that
but also the community doesnt give a fuck bc youre a chud if you want to do your job right
amazon for instance doesnt use cargo. they have their own solutions. and this is the proper way of using rust
---
otherwise why dont you just go with python or something of that tier instead of learning a new language, and a turbo-bloated on at that?

>>105786294
theres a dead simple fix:
since youre dling the source instead of a binary, just recursively extract the live code form the deps- start at the top, and follow function calls
im beyond surprised and a little bit annoyed that this isnt a feature yet
annoyed because of the shilling, when it turns out the whole of rust community cant come up with such a basic idea

Real devs just use plain JS for everything.

>>105786327
if i used plain js in my project id be waiting one year for my computations to finish instead of several hours
you webshitters are completely devoid of imagination

>>105786333
Works on my machine.

>>105786341
would work on mine too
only too slowly...

>>105786317
How would that approach handle versioning? And updates? Would the downloaded code be vendored? What if dependency A depends on dependency C 2.0, but dependency B depends on dependency C 3.0?
Not that I know much about this topic, nor have I given it much thought, to be honest.

>>105786345
V8 is a speedy boy. Stop being poor and buy more ram if you're having problems.

>>105786305
because my app needs to be very fast and reliable
python doesnt suit my usecase at all
Im also pretty familiar with go and it would actually be better for this but I want to add rust to my toolbelt (its actually relevant to some jobs I want to apply to)

>>105786346
>What if dependency A depends on dependency C 2.0, but dependency B depends on dependency C 3.0?
how would that be a problem?
you would have one
somefunction_C_2_0(...)

and a
somefunction_C_3_0(...)


the code could then be re-packaged, but youre killing versioning then...
oooor
you save your "extracted code", and whenever a new version happens, you rerun the process
diff the two versions
and now you have the changes "marked in red" so to speak
making auditing subsequent versions much easier. for the maintainer, but also for the rest of the community
and you can then repackage again

makes the wort of auditors quite easier
and cuts down on bandwidth, and even compile times, for the end user i suspect

>>105786367
>reliable
rust doesnt guarantee code correctness
it only enforces memory security
and its guarantees are quite iffy to be honest
but you do you
esp since you seem to care abt employement, so you will talk with hr roastis so you need to fill that resume up

>>105786389
>rust doesnt guarantee code correctness
No shit
Im not one of these rusttroons who thinks itll magically make me write good code
But increased memory safety does increase reliability
and the fact that your edge case to produce a seg fault is exactly that, an edge case, kind of reinforces that point

>>105786483
Garbage collection increases reliability. Use that.

>>105786483
>But increased memory safety does increase reliability
yeah but it comes at the cost of increased complexity
if you dont have problems with mem safety, rust becomes counterproductive
yes, it took me something like 200klocs in C before i stopped making mem mistakes
but people like me, and ones that are even more experienced do exist
and in our case rust would be a downgrade

you cant talk in absolutes because usecases vary and so the metrics of the problem of correctness

>>105786490
See >>105786367
Kill yourself, nigger.

>>105786483
The less 'unsafe' your code and your transitive dependencies in Rust uses, the better. Unfortunately, the Rust standard library uses quite a lot of unsafe, and has also had UB a fair number of times.
https://materialize.com/blog/rust-concurrency-bug-unbounded-channels/
>This experience reaffirms that even in a language like Rust, where memory safety is a cornerstone, the presence of unsafe code and relaxed atomics introduces the potential for subtle and severe errors. Our findings underscore the critical importance of exhaustive CI, robust diagnostic tooling (e.g., ASAN, Valgrind, Miri), and adversarial stress testing.
Pattern matching is neat in Rust overall, despite some relatively minor quirks and issues. Funnily enough, pattern matching in itself might be doing more for code correctness than the borrow checker in Rust.

>>105786525
Stop coping and start rewriting your project in C#.

>>105786516
^this is the namefag without a name field who spams all Rust threads with his C edits all day and derails them with inane comments. After writing 200klocs of C he claims that:
- malloc is a syscall
- templates incur a runtime cost
- borrow checking incurs a runtime cost
- you should just use half of your ram upfront for an arena memory pool instead
- he knows how to always write memory correct code (after a debugging session)

>>105786580
>a character assasination psyops just for me
you made me laugh to tears anon
also youre signaling that my posts REALLY fuck your shit up
how is that supposed to make me stop???

youre such a bunch of fuckups lamao

>>105786580
>>105786603
yeah
self btfo again
such is the curse of the crab

this screenshot is gonna resurface at the most inopportune moment, i can assure you

>>105783385 (OP)
>the equivalent of cargo/pip/npm/go mod I would use it in a heartbeat
People like you and the people who invented that cancerous shit need to be flogged and executed.

>>105786617
(You)

Flamebaiting samefag. I guess that image really will be posted.

>>105786636
>damage control
why do you engage then, retards?
as far as youre concerned, getting btfod by the
"no-name namefag" is a constant in the universe

youre actually self btfoing, and im only helping you in that

>>105786603
>>105786617
Buckbroken. The best part is that these bare metal performance conscious cniles most often don't even know assembly. They're literally btfo by the Asahi rust troons.

>>105786676
>

>>105786658
You're confusing me with someone else. Unless you're just the flamebaiter.
Also be more kind to yourself.

>>105786718
>cant even follow a discussion
and so you need rust

>>105786709
Show me your apple silicon Vulkan driver written in C in about a month. By the time those rust troons finished you'd still be debugging triangle drawing in GDB and Valgrind.
>>105786727
If you spent all that time you put in flamebaiting, larping and making retarded rust edits in learning the language instead you'd be a pretty t-girl by now, programming socks and all

>>105786772
>flamebaiting
>>105786580
everything you say is projection
rust is a lang for the developmentally retarded ep 123456789

>>105786789
Those Asahi troons achieved more in life than you ever will. Go paint some walls

>>105786794
but by painting walls i already achieved more than them
->i got paid for that
L

>>105786816
>tactical retreat
no
you started
got btfod like the retard you are
and you are now seething, trying to save face

you wont
the board, and even normies know you by now
theres nothing to save

>>105786531
>The less 'unsafe' your code and your transitive dependencies in Rust uses, the better
Wrong

>>105786901
>shartypedo = opinion automatically retarded

Rust is not a perfect language, but damn, it is better than all of the rest.
crates I use often:
- thiserror
- slab
- tokio_postgres
- axum
- petgraph
- itertools
- chrono

>>105786984
>filtered by threads
>doesnt understand the concept of usecase
this is your brain on rust
just say no!

>>105787005
I've built my own language with green threads
you are the one that is filtered

>>105787034
>green threads
>emulation
so you made pretend threads in your pretend lang, right?

>>105783617
Why is it a problem?

>>105787047
I'm done talking with you
there is zero point to engaging with someone that thinks they know everything

>>105787066
i dont know everything but yeah, fuck off, illiterate d&k

>>105787059
yeah, i wonder why

>>105787088
?

>>105787104
>1.1 gig mem usage for a fukken webpage
its not really a problem in itself
bc browsing the web is a frivolous activity
but its symptomatic of a larger ailment in the industry
fukken windows menu written using react is another one

>>105787126
you think the problem is dependencies. you are wrong.

>>105787158
no
dont try to devine what im thinking. its a midwits move
ask it instead-

and what i think is not that deps are the problem
OVERRELIANCE on them, is

>>105787126
And what has it to do with programming language dependencies?

>>105787169
that, again, is not the problem.
I suggest you look at the DOM of Youtube and attempt to surmise the problem.

>>105787169
You're a no coder and have no idea how dependencies work.

>>105787175
>>105787169

>>105787191
>>105787186

>>105787186
fine argument but you still drink cum
and that it comes from a plastic jug doesnt make it any less gay.
>>105787183
im not a webshitter
tell me whats the problem then
how possibly can you go from a state where 1.1gigs used to be enough for the whole system- to be merely sufficient for one sole webpage?

>>105787195
>>105787191
>>105787183
>3 webshitters
>all amgery
>all btfod by simply asking them whats their opinion
this is bad, lads
sad, even
get laid or something

>>105787212
It's a perfectly appropriate argument for someone who accuses a feature that has nothing to do with the problem he points to.

>>105787274
it a non argument retard
also that i dont do webdiarhea doesnt mean i cant understand how a computer works
in fact
i start to believe webshitteryt is a mental illness which prevents you from achieving that understanding

so now,
if it isnt dependencygasm
why did youtube mem footprint multiply like salmonella?

>more silence
ok
so i guess i know more about webshittery than you, despite avoiding the subject like an electric fire

ODIN fixes this

>>105785313
With rust, you can git clone anything, run the same cargo command, and it just werks. Can't do that with Conan because most C++ projects don't use conan.

>>105786901
Why?

>>105788691
its a rustnigger
trying to reason with him is a waste of time
you'd get better luck trying to reason with a rock
at least it wont try to gaslight you

>>105786984
>thiserror
I heard complaints about boilerplate code when doing error handling in Rust, and that a common alternative, 'anyhow', had drawbacks like losing out on more narrow types.
What are your experiences with error handling in Rust?

One alternative that has been proposed was https://github.com/rust-lang/rfcs/issues/294 , which is reminiscent of C++'s std::variant but with pattern matching and some extra functionality for merging different tagged unions (also something about untagged unions). Another alternative might have been https://github.com/rust-lang/rfcs/issues/2414 .

>>105789132
going back to a more serious tone:
what are the rust devs thinking?
people are creating fucking libs to be able to cope with the language?
what more do they need to understand the lang is in dire need of streamlining?

>>105789132
>>105789162
(cont)
theres an easy fix
and a lazy one-
just say certain select libs are part of the standard now
everyone will focus on them so the issues will be ironed out pretty quick

theres a deficiency in leadership in the rust team id say
this is not the first idea i share about how to improve on the language
and it is a minimum-effort one, like the rest ( of ideas i share)

>>105783385 (OP)
>cargo/pip/npm/go
Please stop it. Compiler devs don't have to make a package manager and every time they do it, it's shit. Package manager is absolutely not related to programming language, they shouldn't be doing it just like they shouldn't make their own IDE or their own console or browser, despite that you need to use those for programming with their language. If you don't like your package manager just use a better one. Microsoft doesn't provide you with a good one and you don't want to use linux? Great, write a feature request on their forums or whatever. Please stop, you are making things worse.

>>105789162
>>105789184
flamebaiting

>>105789132
yes, error boilerplate can be annoying. I have heard of anyhow, but have not tried it, nor do I have an interest in trying it.
the crux of the issue with error management is assigning identifiers to each failable code path within your codebase.
I give each function that can return an error its own error enum (identifies the function that failed). I give each failed code path its own member in the enum (identifies the path that failed). `#[from]` is used liberally to link to downstream errors.
tl;dr: error handling in general is a bitch (if you care about being able to determine what exactly went wrong). I think spending the time writing the boilerplate initially is worth it, because it prevents me from needing to go back to old code and update its error handling shenanigans when I find that I actually am interested in figuring out what exactly went wrong.

>>105783385 (OP)
i learned bazel for the same reason
its sad that the only sane build system for c++ is a java tool made from whatever internal code scraps google bothered to publicly release

>>105789264
that was ironical.

(btw since i dont get hate, i will post my o-ekakis more often? feel free to borrow the style, the idea is to express as much as you can, with as little lines as feasible. also use a mouse, if you use a pen thats too easy)

>>105789494
(but the square eyes are mine)
(thats my signature, find yours)

>>105789355
Imagine if the first result instead looked something like:

#[derive(Debug, Error)]
#[error("password too small")]
pub struct PasswordTooSmall {...}

#[derive(Debug, Error)]
#[error("password too big")]
pub struct PasswordTooBig {...}

#[derive(Debug, Error)]
#[error("password has invalid character")]
pub struct PasswordHasInvalidCharacter {...}

#[derive(Debug, Error)]
#[error("password not equal")]
pub struct PasswordsNotEqual {...}

fn check_passwords() -> rust_variant[
PasswordTooSmall,
PasswordTooBig,
PasswordHasInvalidCharacter,
PasswordsNotEqual
] {...}


More verbose for a single function, though the more each error struct is used across more functions, the better this approach would probably be relative to the current state in Rust. In the example you give, some of the error options appear in multiple enums.
But, as the Github issues show, something like rust_variant may not be that easy to figure out well.

>>105783675
pacman -S llvm

>>105789853
>passwords not equal

>>105786198
It's more secure than relying on random packages provided by myriad of random package managers in random distros.

>>105783385 (OP)
Everything you said is retarded and I hope you fucking die

>>105789132
>What are your experiences with error handling in Rust?
Use thiserror for libraries and anyhow for programs.
I don't find this bothersome, it just werks. Pattern matching thiserror is easy and anyhow makes mixing error types trivial.

>>105783385 (OP)
>vcpkg
Conan and meson my guy. turns out hijacking pip for c++ is great

>>105790023
thats wbshit considerations
and you can do better, just earmark a couple libs as standad and then everyone concentrates on them to make em bulletproof

BUT NO
JUST FUCKING TRUST THE TOOLCHAIN? YOU FUCKING GOY
except
im smarter tthan you
and thats why your women outbreed so much because jewish blood is poison
it kills your children's future.

>>105790136
Learn English

>>105790158
no, you will learn my creole instead, half human

>>105789698
its obvious that aint the original
but im willing to admit that squares SEPARATE from brows is another signature
but you have to make that style distinctive then, no?

>>105789698
the idea of expressing an emotion is also important

>>105790084
Not everyone likes anyhow, even for programs.
https://felix-knorr.net/posts/2025-06-29-rust-error-handling.html
https://reddit.com/r/rust/comments/1lnbr0g/on_error_handling_in_rust/

>>105789698
its part of the whole idea
emulate anime art
caricaturize emotions with the least strokes possible
bc you have a studio of 20 fukken people and have to animate 10k hours of anime
in a way its a hommage, too

>>105790292
>Not everyone likes
There are only two kinds of languages: the ones people complain about and the ones nobody uses.
It works for me.

>>105790136
>>105790158
samefag

105790513

>>105789355
>>105789853
https://gist.github.com/osa1/38fd51abe5247462eddb7d014f320cd2

no (you) btw
in case you missed it btw

>>105790577
devilish

105790158
105790136
105790529
105790577
vpn

>>105790486
Zig. I've never met a person that uses Zig complain about Zig.

>>105790672
I have seen quite a lot of complains about zig in their issue tracker. It seems that all this criticism is quickly silenced too.

>>105783385 (OP)
The dependency management is literally one of the worst parts of these languages.

You got filtered by .mk, truly not my problem.

>>105790702
Only thing I can think of off the top of my head was when people were complaining that they wanted their own compiler instead of LLVM, how many packages it would break. I've heard people say the package manager is a bit rough to use too, but it's still in production.

>>105790960
RAII, better closures, no tabs support, etc

>>105790672
No one uses zig

Have you ever seen any zig thread? They are 100% artificial shilling like hare threads. Literally no one ever posts any personal projects. They just circlejerk around shitting on C++ and Rust.

>>105791297
Bun is written in Zig. Though Bun is a scam kind of since they shilled themselves on having faster than rust HTTP and websockets, and it turned out that's just because they were using uws.js which is just using a C++ library lmao. Then the uws.js creators themselves called out Bun on their own Github package saying it works 80% faster on Node.js and the bun creator started posting crying emojis on twitter lmao

>>105783385 (OP)
"Modern" dependency management is just Wintoddlers discovering package managers, refusing to install Gentoo, and then bolting a new package manager to every language they want to use. Every other system has this figured out, even Macs have brew and it just works, but for some reason cargo is this magical revelation to Windows users because they refuse to use winget or chocolatey.

>>105793164
>brew
>just works

I'm pretty sure it's the most complained about package manager I've ever heard of.

>>105793177
Works on my computer. It's retarded that it's written in Ruby, but it's certainly not the worst I've ever used. apt is universal shit, pacman has a bunch of retarded corner cases and Arch jannies are fools that package some of the worst vomit I've ever seen in a package manager, and portage shits the bed whenever python updates in its vicinity. Brew's pretty smooth sailing by comparison, but maybe I should chalk that up to only controlling a handful of packages that mostly live on the command line.

>>105783385 (OP)
Since when was rust a chud lang?

>>105791297
If they can untangle their shit from llvm, Zig is the only nulang that has any potential. If I was going to use llvmslop I'd probably use C3 instead. Hare's only a non-starter because nobody trusts Drew not to be a malevolent dictator for life.

>>105793250
The ziggers were all upset and whining that untangling from llvm would break all their shit. Particularly the tranny writing Mach game engine, which is way behind schedule and seems like a hopeless project.

>>105793250
They have made the first steps towards that. x86 debug mode is self hosted.

>>105793224
>Works on my computer
Well try to install package for other user

>portage shits the bed whenever python updates in its vicinity
Literally never happens

>>105786188
>it is still a lot more work than Rust
work harder or you cant be successful

>>105783385 (OP)
Bruh u just download random zip files into your directory and extract them in esoteric and complicated directory hierarchies and then do some cmake magic like it's not hard.

>>105793224
brew is unwashed ass and anyone who says otherwise has either barely used it or just assumes the most macos-adjacent option must be the best because apple

>>105783385 (OP)
>vcpkg
use CPM.cmake
https://github.com/cpm-cmake/CPM.cmake

>>105786229
>or is it just the ecosystem and tooling that have to catch up, and all we have to do is just Be Patient™ and help out?
think this the issue.

I'm just waiting to see if it's even worth it. a lot of shit is not on paper and not in the compilers or this case the libraries/modules

>>105793164
install packages, use fucking pkg-config

>>105787354
is that rust derivate thing that has automatic c/c++ binding? no.. that's carbo-something+

>>105788444
can't you do that with meson, ninja and htose guys?

shit like cargo and maven are cringe

just script the dependency paths string concatenation and feed it into a one liner compiler

if you need makefiles and yaml and xml and shit to compile what the fukc is the point

literal make work horseshit. MUH WORKFLOOOOOW

IM GONNA FLOOOOW

fuck off baseddevfaggotniggers

>>105794706
>if you need makefiles and yaml and xml and shit to compile what the fukc is the point
You don't. Cargo does most of the work for you.
The actual issue is cultural because webshits are obsessed with putting everything into a tiny micro-package and never collaborate. Plus they make literally everything async.

>>105794734
>Cargo
still needs a toml file full of makework semantics retardation

>>105794749
Yeah, no. A few lines of text in a glorified ini file is almost no effort at all compared to make. If you're too buckbroken by 70s software design to see that, that's your problem.

>>105794762
compared to dependencies += path it is absolutely makework bloated horseshit semantics id rather not deal with

>>105783385 (OP)
Just use header only libraries with thick non-leaky wrappers, fuck dependency management.

>>105794762
>at all compared to make.
imagine getting filtered by this
and then the fucking crabs larp as le elite brogrammers

i highlighted actual code btw, the printout clause (its actually called a recipe) is a printout of some stats about the project

>>105794796
also cringe and gay

>>105794817
i use em just bcoz youre supposed to
i dont even have object files in how i compile so i might as well do that in a shell
but point still stands
make is easy
even if it was written by an autist

>>105794831
reminder this was wwirtron inr ust

>>105794037
Nope, those don't take care of dependency management which is the most painful part.

>>105784876
if something is exclusive to linux, that means it's open.
if something is exclusive to mac or windows, that means it's locked down.
happy to have educated you today

>>105783385 (OP)
If I were to redesign C++ from scratch, it would look more like Rust than C++23.

>the eunuchs are fighting
>AGAIN
Pascal never had that problem.

>>105783385 (OP)
Rust's abysmally slow build times eat into productivity more than managing a handful of dependencies.

>>105786242
Just use https://www.ultimatepp.org/

It has GUI, plotting, async, http, json and websocket support.