We went from Twitter screencap spam, to X screencap spam, to Reddit screencap spam, to now GitHub screencap spam.
And you wonder why the line on 4stats.io is going downwards.

>>108821739
>still some clean up to do
wake me up later

also, it's a win for zig and rust coders since LLMs understand both langs

>>108821739
I'm trans btw

>>108821739
Using automation to translate existing software is pretty much all Rust is good for. It was never intended for human use.

File: file.png (98 KB, 1222x626)

98 KB PNG

>>108821756

>guy who reees about a pull request on a computer-themed gook pictures board
>picrel

also, Line Goes Down because of this retarded captcha

>>108821756
What image do you want instead ? An unrelated anime girl or some slut?

>>108821895
Not about the image, these posts with those screencaps are always 100% worthless garbage. Just go to Reddit with them along with all of the dogshit generals and AI sloppers.

>>108821739
I just started learning both and find this highly amusing

and with "Rewrite" they mean "asked claude to rewrite all of it"
this'll be really, really funny when they try to make this go """""""""""""stable""""""""""""

File: 1u9idbb9g22IgnS6Pz8Wm2Q-1(...).png (541 KB, 1210x642)

541 KB PNG

>>108821739
Like all the rust programmers from last decade learned. You may be saved from all your segfaults, but you aren't saved from your lacking comprehension of your own program. And in fact RUST MAKES THAT EVEN WORSE AS THE TRADE OFF! Equivalent exchange bucko.

>>108821756
fpbp and /thread

>>108821739
ziggers on suicide watch

this shit doesn't even compile

> git clone --depth=1 https://github.com/oven-sh/bun.git
> cargo check
    Updating crates.io index
error: failed to get `lol_html_c_api` as a dependency of package `bun_lolhtml_sys v0.0.0 (C:\Users\nigger\bun\src\lolhtml_sys)`

Caused by:
  failed to load source for dependency `lol_html_c_api`

Caused by:
  unable to update C:\Users\nigger\bun\vendor\lolhtml\c-api

Caused by:
  failed to read `C:\Users\nigger\bun\vendor\lolhtml\c-api\Cargo.toml`

Caused by:
  The system cannot find the path specified. (os error 3)

File: why_wintard_w.jpg.jpg (148 KB, 1344x768)

148 KB JPG

>>108822130
i didn't bother looking for specifics when i checked this out and hit this a couple of days ago,
but some script in the repo probably vendors dependencies from some other repos as a pre-build step. this is not THAT unusual.
the repo is a miss with the zig files still around, which is a part of the reason why i didn't bother sinking time into this.

>>108821739
How's Zig going Kelly? Still going to make another mundane thing hard error again to satisfy your autism?

>>108821739
Jew shills flooding the board with dead internet crap

File: chuddha boondocks.png (19 KB, 678x469)

19 KB PNG

>>108821739
Wait a minute, I'm not an Anthropic employee! Haha!

>>108822022
This

>>108822346
I just checked CONTRIBUTING.md for the build instructions. It is a complete shitshow.
>Configuring a development environment for Bun can take 10-30 minutes depending on your internet connection and computer speed. You will need ~10GB of free disk space for the repository and build artifacts.
>Bun is written in Rust and requires a specific nightly toolchain (pinned in rust-toolchain.toml).
>Before starting, you will need to already have a release build of Bun installed, as we use our bundler to transpile and minify our code, as well as for code generation scripts.

If you want to try it out (presumes zig bun already installed):

```sh
bun upgrade --canary
```

>>108821739
Zig has no reason to exist and it isn't significantly better than C. 99% of zig programmers will constantly produce memory issues but they'll continue on in blissful ignorance. Watching streamers like ThePrimeagen code in zig is honestly hilarious.

why not just port it to assembly?

>>108822533
Because you'll have to write the same code 3 times in differents assemblies flavors

>>108821739
HOLY FUCKING SHIT
HE DID IT
NOW IT'S LE HECKIN MEMORY SAFE.
ZIGGERS BETEEEFOD

File: file.png (75 KB, 1454x538)

75 KB PNG

I reviewed my code and can confirm that it's bug free!

>>108821739
>Conversation (666)
Very fitting.

>>108822396
So that's what he meant by still having to do some optimization & cleanup.

Shithub won't even load the comments

File: Capture.png (98 KB, 848x574)

98 KB PNG

>>108822916
FFS

 rg 'unsafe \{' | wc -l
10428
 rg 'unsafe \{' / -l | wc -l
736

>10428 unsafe blocks in 736 files
it's all good though, see >>108822781

>>108822935
vibe infra for a vibe future

Sucks that I had high hopes for bun. Looks like it shat down the drain after anthropic acquisition with practically bunch of unlimited AI slop code.

File: 1575707025351.jpg (10 KB, 270x190)

10 KB JPG

>make major new program
>acquired by AI company
>slopify 1M lines of code into another language for the hell of it, minimal review aside from automated tests
>discussion is limited to bragging about it on xitter
>don't even bother releasing benchmarks or the promised blog post
>merge into main
how fucking retarded can you be

>>108823032
again. the code was 100% unsafe as rust defines it. from the multiple valid reasons why this is retarded, "hurr durr unsafe{}" is not one.

jeethub is shitting the bed. I want to read those 800 comments.

>>108823142
it's not "hurr durr unsafe{}", it's "claude slopped unsafe{} all over the code"

>>108822916
>>108822935
>>108823035
https://desuarchive.org/g/thread/106230818

>>108823157
it wouldn't have been much different, in that aspect, if it wasn't slopped. because the retarded dev SPECIFICALLY asked for mirroring logic 1-to-1, file-by-file.

>>108823216
that's the best claude can do though

>>108822130
>C:\Users\nigger
there, found your problem. rust is for white women only

File: 1763588663004900.jpg (32 KB, 326x321)

32 KB JPG

Rustfags wished more projects got rewritten in rust and the monkey paw heard them.
The future of rust is slop, enjoy.

>>108821739
This was an AI re-code and nobody is going to accept it on its face

>>108822130
You need the package, Rust is a faggot npm-like language that needs 10,000 cargos to work

>>108823142
If most of your code is unsafe keywords than you've invalidated the one good thing about rust as a language.
AI cannot tell if this code is safe or not, the mechanism that makes it safe doesn't work and was disabled.

>>108823349
dumb retard

>>108823355
illiterate shitstain. deep throat a shotgun if your brown country lets you have one.

>>108823366
deep throat me yourself you fucking coward

>>108823366
dumb retard, cargo automatically fetches dependencies.

>>108823378
SAAR SAAR BENCHOD SAAR

>>108823381
Not when it relies on compiling the relevant crate
https://github.com/cloudflare/lol-html/tree/main/c-api
This is the crate. It's probably in the build instructions to include it.

>>108821739
>Shittiest, laziest Rust rewrite possible
>No optimization yet
>Still faster, still the binary is smaller

Ziggas... why dhey doin' dat to us poor ziggas we jus' tryna survive

>>108823390
dumb retard, the build instructions are "build bun with bun". see >>108822396

>>108823354
neither is most of the slop code unsafe, nor is "safety" the one good thing about rust as a language.

{
  "functions": {
    "safe": 6480,
    "unsafe_": 256
  },
  "exprs": {
    "safe": 619556,
    "unsafe_": 47345
  },
  "item_impls": {
    "safe": 6636,
    "unsafe_": 346
  },
  "item_traits": {
    "safe": 522,
    "unsafe_": 20
  },
  "methods": {
    "safe": 22674,
    "unsafe_": 651
  }
}

>>108823397
Who cares, you're a big boy, you can figure out that it needs that crate.
Rust as a language is perfect.

>>108823415
> "safe": 6480,
> "unsafe_": 256
That's fucking absurd, can you not see why that's excessive? Most codebases maybe use unsafe a few times for known-safe use cases.

>>108821756
X and sites like github are where real software news comes from now. Wired magazine meanwhile has a front page article by pajeet paresh about butt cushions.

Whining about twitter screenshots should be reserved for bait comments from literal whos, not legitimately relevant news posts.

>>108823419
with 2 billion more parameters claude could have figured out that it can just do

cargo add --git=https://github.com/cloudflare/lol-html lol_html_c_api

Bun is pretty much dead anyway. It was never popular but made headlines as a fast JS runtime. But it's pretty much useless unless it's 100% compatible. It started off not being able to run most projects, slowly got better, but also slower, but never reached 99% combability. As it turns out Node.js already did most optimizations and bun has to slow down to become compatible.

Then anthropic bought it and they're only going to do this kind of slop, for marketing purposes of their masters.
>>108823354
>>108823341
Are correct.

>>108823147
yeah wanted to read them too, no chance

>>108823439
i didn't argue whether it's excessive or not. just correcting false claims.
the zig code was 100% unsafe, as rust defines it.
the slop code is not "mostly" unsafe.
mirroring logic from an unsafe language 1-to-1 results in seemingly excessive unsafe usage, *irrespective* of the conversion method.
there are many nice things about rust, beyond safety features. the complaints about "leaks" in particular from the retarded dev relate to RAII, which is not a safety-exclusive feature.

>>108823415
>>108823439
Zed has 1000 unsafe in total, in the whole project

>>108823493
zed started as rust code, so comparing it with bun is flawed from the get go.
it's also partially-slopped crap anyway.

both projects will have a non-zero amount of unsafe anyway, since they either bind to native graphics apis for zed or to a c++ runtime (JSCore) in the case of bun

>>108823492
>the zig code was 100% unsafe, as rust defines it.
Could have had 0 bugs, though. That's a useless metric.
>mirroring logic from an unsafe language 1-to-1 results in seemingly excessive unsafe usage
Perhaps, but this is a major drawback of rust, that the borrow checker is not advanced enough to check what's actually correct, and the rust devs aren't interested in improving it and would rather force you to follow their standards.
>there are many nice things about rust, beyond safety features.
No. It's a painful language, that asks you to submit and suffer all of it's idiosyncrasies, slow down your software, and in return get the holy "memory safety".

so, is bun kill now or what?

>>108823572
>Could have had 0 bugs, though.
no one is arguing that this slop move was good.
>the borrow checker is not advanced enough
this has nothing to do with borrow-checker limitations. unsafe languages' code lacks the necessary type/lifetime info/annotations/constraints necessary to prove correctness, which converted-to-rust-code could mirror.
>It's a painful language
i'm sorry you got filtered.

>>108823664
>this has nothing to do with borrow-checker limitations
> lacks the necessary type/lifetime info/annotations/constraints necessary to prove correctness
No, this is a big limitation of it. Rust fangirls(male) always shift definitions to imply rust is good and everything else is inferior, like "memory safe"-> "correct and vurn free".
It's possible for a human to figure out, it's possible for a machine to test, it's logic and not subjective opinion so rust should be able to figure out if it's correct or not. It should not be my job to prove it. Im not saying it's easy, just like rust team admits it's not easy to write approved-safe performant rust. But they should do that job, not the user of rust.

>>108823880
>nocoder screeching
thank you for your input

>>108823633
the anthropic purchase was the kill and this slop conversion is the nail in the coffin

>>108821756
I only somewhat mourn lit and ck which are both in (less rapid but still sad) decline
Let this shithole crash and burn, desolate and pointless like /fit/

>>108824173
the comments on github are wild so I guess it is really the nail in the coffin lmao

>>108821756
>We went from Twitter screencap spam, to X screencap spam, to Reddit screencap spam, to now GitHub screencap spam.
So it's improving. This is /g/, shit from github is very topical. Unlike random OF whores opinion on phones which is posted on X.
>And you wonder why the line on 4stats.io is going downwards.
It's been great. I've had a much better time here. And talked to real people. And been nice to each other.

>ziggers
for a moment i thought this was about russia ukraine war kek

File: 1000006370.jpg (102 KB, 846x900)

102 KB JPG

>>108821739
>drama surrounding zig
>its actually horribly embarrassing for the other party and not andrew
got my fucking hopes up
you win this time, andrew...

>>108821739
stupid slop coded github can't even handle opening slop code PR kek

>>108822372
Nice boondocks ref. Funny how a show can be so popular with both niggers and racists

>when dogfooding goes too far

https://github.com/oven-sh/bun/issues/30719
Get slopped. This is unsafe.

>>108825335
>also hire a real rust dev
Jared did hire a real rust dev, his name is Claude.

>>108823142
Retarded poster

>>108825335
*unsound*. as in unsafe+broken invariants. unsafe alone is not a problem. otherwise, the zig codebase was 100% problematic. but since this was 100% slopped, it's all shit anyway.

>>108823492
It takes 1 (one) incorrect use of 'unsafe' to invalidate the rest ot the Rust's safety mechanisms.
And considering the code was generated by a LLM based on code that was known to be incorrect (due to the bugs encountered) it's extremely likely 'unsafe' is used unsoundly here and multiple of times at that rendering any benefit from it being in Rust moot.