[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: 1636711832768.jpg (452 KB, 1200x931)
452 KB
452 KB JPG
What are you working on /g/?
>>
>>101540105
>no previous thread link
>>
>>101540105
dont link it, dont link it right now.
>>
previous thread: >>>/diy/2757813
>>
>>101540105
How do you even put code here, there's a pretty crappy line limit.
>>
File: 1644645722787.png (216 KB, 824x720)
216 KB
216 KB PNG
>>101540135
>>101540227
We had many fights over the previous thread link, i purposely refrained from adding it, the current thread is the current thread no need to talk about the previous thread, i made the previous thread btw
>>
>>101540227
Why?
>>101498438
>>
>>101540105
a dl chrome userscript that filters tranime threads out
>>
>>101540105
osdev niggers want to jedimindtrick you into thinking that acpi is hard
acpi is a JJoke, tranni IIIIIIII IIIIIIIII es just want to write bloated software that nobody will read or use, fuck, Linux doesnt do that... ECI NONsenseIIIIIII . Anyway, I am writing an acpi ""inTerpreter"", the word they use IIIIII IIIIII to trick your FFucking mind fhm IIIIII IIIIII
anyway, not today be IIIIIII IIIII I IIIIIIIIII because i am tired, i speedran writing a UEFI a bootloader in (un)HolyC(++) the past couple of days or so
IIIIIIIIII IIIIIIIII
oh well uhh
>>
>>101540257
first rule of fight club is don't talk about programming
>>
Fuck, Terry was more coherent than you, and he was certifiably schizophrenic.
>>
>>76759441
dont EVAR fucking forget because God... he will fucking blow all you c.i.aniggers on a whim. Yeah.. God could have a whim and wipe away the cia
, bill gates can fuck all fucking fbicattle he wants because i have God
>>
>>101540257
We've literally always had the previous thread link. It's nice to go look at the old thread after waking up or something to see if you were replied to, without having to fuck around with the archive.
It's not something "controversial" like the other crap people try to put in the OP.

Otherwise your OPs are fine.
>>
>>101540387
i am not schizophrenic, i am just having fun
>muh coherence
you live in a world of lies and treachery and this is the nonsense that you dare mention, certifiably goycattlenigger
>>
At least you learned how to write "schizophrenic".
>>
>>101540438
not everybody you talk to is the same person, i just happen to be the same person, you are not smart or anything for having understood that since you didnt even understand* it...
>>
>not everybody you talk to is the same person
>i just happen to be the same person
Cute.
>>
whatever, you are a niggermonkey having fun not giving (You)s to people because it gives you some sort of false conformation that you are better than them, i hope you get gutted by a cartel member
>>
>be the anon from a few threads ago who got sick of trying to learn complex languages like Rust and Zig and just continue to use C because I'm retarded
update, I'm working on a remake of an old project of mine written from the ground-up; a launcher for a video game. I've successfully done some good string handling that allows long strings of user input to be parsed into appropriate arguments.
>>
>conformation
Did you mean: confirmation
>>
>>101540670
deformation
>>
False deformation? I can live with that.
>>
File: t.png (2 KB, 170x33)
2 KB
2 KB PNG
>>101540670
>>
Hey, as long as it's false.
>>
Eh well I already built my hobby site, it was tedious with many details and assests, and today I have begun building websites for my future business endeavours to keep me motivated.

I am working on a game but I am still in the conception and planning stage, next step would be creating assests. I know I can program at least some of the game but I am stuck now because I don't have a computer to support the programs like unreal engine. But I am learning how to use it.

I know games are in C but shit I can't say I know this language
>>
>>101541185
C is simple:
>don't use malloc
>>
File: x4i0zbsw20ed1.png (775 KB, 1282x1206)
775 KB
775 KB PNG
Daily Remind Chatgpt had already replaced most of you and if you think you're in some little niche sooner or later someone will tip off management that you do nothing and you'll be made redundent.

Code monkeys first to go when cost cutting starts
>>
>>101540105
my webphub app is working everything started to work properly, i'll do more work on the ui to make it look prettier
>>
>>101541523
i uses cwebp and other parts of google libwebp, i'll move soon to create a bulk encode and decode functionality
>>
>>101541508
>Daily Remind
Hey, Yevgeny, you've doing a fucking terrible job at demoralization.
>>
>>101541508
>Chatgpt had already replaced most of you
its coding solutions are full of errors and don't work properly though?
>>
>>101541508
Have they managed to solve the halting problem yet?
>>
>>101541879
No, but who needs that in the first place?
>>
>>101541934
So how do you know that AI code does what you asked it to, if you don't even know if it's not going to just get stuck in an infinite loop?
>>
>>101542069
Not my problem.
>>
What do you think of this?
function r(x) if not v then v=0 end v=(v+0.191471894719279)%1 return x*(v*4915.1852717*(v*5612.3232516*(v*9147.1612791%1)%1)%1) end
>>
>>101540105
bundled my first program with an icon
>mfw
>>
I am using Python with Flask.
I have no idea how this alert is even called. How do I approach this? It looks like it must be some kind of javascript crap but I have no idea
>>
>>101542246
forgot the screenshot but already figured it out by myself.
Ignore my dumb ass
>>
File: giga.jpg (94 KB, 1357x960)
94 KB
94 KB JPG
>>101540105
#define maybe if(rand()%2==0)
int main(){
maybe{
while(1)malloc(99999999);
}
...
>>
when using a socket, is there something I can do to query it before I read to ensure that my read call will not block?
if(has_data(sock))
read(sock, ..., ...)

I know there is epoll but that seems like extreme overkill, I just want to know if there is data once
>>
>>101543483
select(), or do you find that overkill too.
>>
>>101543525
but select man pages say not to use it since if you get assigned a fd greater than 1024 it will break?
>>
>>101543616
Do you have 1024 files open then? You don't get a random file descriptor 1024.
>>
>>101543651
oh, I thought they were universal
>>
>>101543682
https://man7.org/linux/man-pages/man2/open.2.html
>The file descriptor returned by a successful call will be the lowest-numbered file descriptor not currently open for the process.
>>
>>101543682
Every new file descriptor takes the lowest number slot possible. And accept() will already fail if there's no more, not select().
>>
>>101543483
You could also just change the socket to non-blocking, then just try reading it to see if there's data or not.
>>
>>101543525
>>101543651
select() is ancient crap. Use poll() in this situation.
>>
>>101543734
No, he's right, accept will keep working but select will fail (but it doesn't matter if you've only got a few sockets). If you've only got a few sockets though I wouldn't even bother with select/poll/whatever, just switch to non-blocking and read every now and then.
>>
The reason select fails btw isn't because there's a limit of 1024 file descriptors, it's because (on linux) select takes a fixed size bitset that's 1024 bits wide, with the positions of the 1 bits signaling which fds you want to check. So if you've got an fd higher than 1024 it won't fit in the set. As another Anon mentioned poll doesn't have this problem.
>>
>>101542229
Congratulations, what language did you write it in? Is it bundled in the traditional installer?
>>
>>101543777
what's the right way to switch a socket to non-blocking? I see people using fctnl() ut every example I've looked at so far uses different arguments...
>>
>>101543777
>every now and then
That is a horrible way to do shit.
The real advantage of poll() and friends is that you get to block in a sensible place while nothing is happening, and you still get to react to events.
"Just make them all non-blocking and check them" as a general solution sounds like a very dumb busy loop. Non blocking definitely has its uses, but that is not one of them.
>>
>>101543766
For most situations (a handful of fds) it works just fine.
>>
>>101543814
my use case is a unit test. I want to write 1 message and I want to read 1 message. if it fails then I don't want to be stuck blocking forever. I wanted to do something like
>write
>sleep 100ms
>try to read
if there is no data consider the test failed
>>
>>101543810
You can just pass SOCK_NONBLOCK when you create the socket using the "socket" function.

>>101543814
I said if he's only got a very small number (like one or two). If you've got more than a few sockets then yeah obviously poll (or epoll) is better (I still always set my sockets to non-blocking as a general rule though).
>>
>>101540105
I just finished this magic 8 ball program. I kinda wanna add a gui, but I don't know where to start. I've been learning c for a few weeks now
#include <time.h>
#include <stdio.h>

int main(void) {
char * m8b[] = {"It is certain", "It is decidedly so", "Without a doubt", "Yes definitely", "You may rely on it", "As I see it, yes", "Most likely",
"Outlook good", "Yes", "Signs point to yes", "Reply hazy, try again", "Ask again later", "Better not tell you now", "Cannot predict now",
"Concentrate and ask again", "Don't count on it", "My reply is no", "My sources say no", "Outlook not so good", "Very doubtful"};

char ques;

srand(time(NULL));
int r = rand() % 21;


printf("Ask the Magic 8 Ball a question: ");
scanf("%s", &ques);
printf("%s
", m8b[r]);
}
>>
well what do you know
https://godbolt.org/z/MPGE6eTdr
Ask the Magic 8 Ball a question:
Is >>101543894 a faggot?
It is certain
>>
>>101543874
>I still always set my sockets to non-blocking as a general rule though
This is good because even when select and poll return readable, the socket can block in special cases (such as a queued connection on a listening socket that has been removed after the select or poll returned but before the accept).
>>
>>101543810
Non-posix standard, but widely implemented:
Do it at the same time you make the socket
// The question wasn't about CLOEXEC, but it's a idea to set it 99% of the time.
int sock = socket(domain, type | SOCK_NONBLOCK | SOCK_CLOEXEC, protocol);

int client_sock = accept4(sock, NULL, NULL, SOCK_NONBLOCK | SOCK_CLOEXEC);

standard fcntl way:
int flags = fcntl(sock, F_GETFL);
fcntl(sock, F_SETFL, flags | O_NONBLOCK);


>>101543854
You can just use the timeout argument to poll for that.
You could also block inside of recv, and wake yourself up with a SIGALRM.
>>
>>101543894
>scanf("%s", &ques);
Nice buffer overflow there.
>>
>>101543961
it really is magic, huh
>>
File: file.png (3 KB, 376x50)
3 KB
3 KB PNG
>>101543894
whoops, forgot a newline in the final output:
 
#include <time.h>
#include <stdio.h>

int main(void) {
char * m8b[] = {"It is certain", "It is decidedly so", "Without a doubt", "Yes definitely", "You may rely on it", "As I see it, yes", "Most likely",
"Outlook good", "Yes", "Signs point to yes", "Reply hazy, try again", "Ask again later", "Better not tell you now", "Cannot predict now",
"Concentrate and ask again", "Don't count on it", "My reply is no", "My sources say no", "Outlook not so good", "Very doubtful"};

srand(time(NULL));
int r = rand() % 21;


printf("Ask the Magic 8 Ball a question: ");
scanf(" ");
printf("%s\n", m8b[r]);
}

>>101544084
huh yeah, whadda you know
well, i'll just get rid of it. not like i was doing anything with the question anyways
>>
File: file.png (10 KB, 388x95)
10 KB
10 KB PNG
>>101544254
WHY DOES IT KEEP DOING THIS????
>>
File: IMG_0179.gif (45 B, 4x4)
45 B
45 B GIF
>>101543806
Tnx! Golang, just binary in folder, no installer
>>
File: 274569023746.png (507 KB, 1147x1081)
507 KB
507 KB PNG
unsafe
>>
File: 1715017521375986.jpg (3.34 MB, 3500x2500)
3.34 MB
3.34 MB JPG
>>101544566
Tokio soup isn't safe anon, imagine the world being dependent on a third party library because Rust async is borderline unusable
>>
>>101544685
what is boost
>>
>>101543894
Could probably replace the hardcoded 21 with sizeof(m8b)/sizeof(char *), no? Or something like that.
>>
>>101544566
Always hilarious to see people call themselves professional programmers.
>>
>>101545247
Why? Is it derogatory? Is it a lie? Is it a wrong term? Do professional programmers not exist? Do you think they should not call themselves by their occupation? Please clarify.
>>
>>101544685
Why the fuck would you want to write async code in a real programming language. I can understand being forced to write async in javascript or python or some other scripting language, but this is rust?! If you really need asynchronous IO, then write your asynchronous IO without the retarded async/await syntax.
>>
>>101545463
Calling malloc is a firing offense at any real software company.
>>
>>101537897
consteval not constexpr

e.g.
struct dpp8_param{
uint8_t shuffle[8];
consteval dpp8_param(/* params */ ) : /* initializers */ { /* validate */ }
consteval uint8_t operator[](size_t idx) const noexcept { return shuffle[idx]; }
};

i might be misremembering, and might've used it as a non-type template parameter but I think this worked as a const ref, it's the same idea as std::source_location
also there's __builtin_constant_p or whatever that forces constant folding
>>
>>101545238
>sizeof(m8b)/sizeof(char *)
sizeof m8b/sizeof m8b[0]
>>
>>101545238
yeah, i hadn't even heard of sizeof until now, but yeah that's a better option. i also just realized that arrays go from 0-20, and rand starts at 0, so i only needed % 20 in the first place.
>>101545610
what is the difference? does m8b[0] just specifically refer to the m8b char * array?
>>
Why not just set all pointers to 1 instead of 0? No more null pointers!
>>
>>101545543
Can you start using a trip so your dumb shit is easier to filter?
>>
>>101545745
you want your program to stop if it does something unintended
that way you can pinpoint the exact thing thats going wrong more easily.
otherwise you get something thats called a "silent failure".
when your program continues to run, but you dont get the output you wanted.
there you must first begin with finding what went wrong exactly, as opposed to your program stopping at the exact moment something did
and if it stops when you dereference 1
congratulations.
you just recreated dereferencing
>>
>>101545879
>you just recreated dereferencing
you just recreated dereferencing a null pointer
>>
>>101545463
It's because people who start off a post with "as a software engineer/professional programmer" are usually the dumbest retards on the planet and have no clue about what they're talking about.
>>
>>101545934
As a woman, it is very similar to retards and trolls posting with the template "As a <member of group X>, <bad opinion Y>", where X is some elevated/protected group, and Y is a brave statement designed to collect as much upcummies as possible in combination with X. As an ESL, i think it is also grammatically incorrect.
>>
wtf did i just read
no you for you, anons
only the eternal fire of the pit of damnation
>>
File: IMG_1853.png (27 KB, 646x486)
27 KB
27 KB PNG
>>101540105
in honor of dumper-kun (who has thankfully kept the thread cleaner than usual by not talking about the nuances of malloc for 82 replies) i have written a registry dumper for TempleOS, not even sure if i could call it one since all it does is expand DolDoc trees so you can see every entry, but it still counts as "dump" i guess
U8 *MemMCat(U8 *a,I64 i,U8 *b,I64 j) {
U8 *new,*orig=NULL;
if (MSize(a)<i+j) {
new=MAlloc(2*(i+j));
MemCpy(new,a,i);
orig=a;
a=new;
}
MemCpy(a+i,b,j);
Free(orig);
return a;
}


U8 *StrMCat(U8 *a,U8 *b) {
return MemMCat(a,StrLen(a),b,StrLen(b)+1);
}

U8 *StrSwp(U8 *s,U8 *what,U8 *with,Bool free=TRUE) {
U8 *cur=s,*res=CAlloc(1),*match;
I64 what_len=StrLen(what),len=StrLen(with),cur_len=0;
while (TRUE) {
if (match=StrMatch(what,cur)) {
//not invoking strlen->100x speedup
//no strlen in hot loops!
res=MemMCat(res,cur_len,cur,match-cur);
cur_len+=match-cur;
res=MemMCat(res,cur_len,with,len);
cur_len+=len;
cur=match+what_len;
} else {
StrMCat(res,cur);
break;
}
}
if (free) Free(s);
return res;
}

U0 RegDump(U8 *reg="~/Registry.HC") {
U8 *s=FileRead(reg);
if (!s) return;
s=StrSwp(s,"\dTR,\"","\dTR-C,\"");
"\dHL,1\d%s\dHL,0\d\n",s;
Free(s);
}

RegDump;
>>
>>101545879
null pointers (and other variables) arent necessarily initialized as zero, if the stack is allocated over previously used memory then they contain whatever the memory used to contain
a lot of people initialize a pointer to NULL, but this takes an extra operation
>>
>>101545745
null isn't zero, it's a symbolic constant representing an invalid address that converts to zero
zero is a valid address on a lot of systems just protected by the kernel
and the guaranteed invalid address can be something else like 0xFFFFFFFF
>>
>>101546211
>if the stack is allocated over previously used memory then they contain whatever the memory used to contain
ik but it garbage values so you dont want that.
unless youre a hakerman
>>
>>101540105
AGI
>>
>>101546171
What's the runtime?
>>
File: IMG_1653.png (665 KB, 811x762)
665 KB
665 KB PNG
any number below the first line of /proc/self/maps on linux may be used as null for professional enterprise purposes
>>101546363
like a hundred microseconds? nothing you can notice, and the HPET latency probably takes up a significant portion of this measurement.
also you could just do
"\dHL,1\d";
Type("~/Registry.HC");
"\dHL,0\d\n";
DocCollapse(FALSE);

but this takes around a few milliseconds because DocCollapse applies for the entire Doc and does a whole lot more.
>>
File: 1697681047561183.png (21 KB, 335x463)
21 KB
21 KB PNG
toylang progress, tables
indexing always gives an optional, you can also define a default value
>>
>>101546534
And the size of the dump?
>>
I made a tool called pepe eyes to encourage me to take a break from my screen every 20 minutes and if I ignore it for 30 minutes it shuts down my computer.

The thing you quickly realise is just how quickly 20 minutes goes by when you're shit posting.
>>
Reading pbrt v4's thread pool implementation so i can use it for my game engine
>>
>>101540105
rolling for total rust troon death
>>
>>101547178
peak retardation
>>
File: love-pepe.gif (555 KB, 498x498)
555 KB
555 KB GIF
>>101547780
>>
File: BOFVN.jpg (53 KB, 429x600)
53 KB
53 KB JPG
>>101540105
>What are you working on /g/?
Attempting to get the cs50 library to work with my IDE, CodeLite. Downloaded from source and installed it and it works from the command line. Attempting the same in CodeLite: go to project settings, passing the same compiler command and I get
"undefined reference to 'get_string"
>>
>>101548197
SOLVED:
In Linker Settings:
Library Search Path: "/usr/local/lib"
Libraries: "libcs50.a"

In Compiler Options:
C Compiler Options:
"-std=c11;-lcs50"
>>
>>101546171
Cute!
>>
>>101540105
>acpi specification withholding information on how data objects are linked to name objects
>"no why would you ever need that goy? just be a good goy and usa acpica"
atrocious, all of these comminties need []
>>
File: maxresdefault (1).jpg (107 KB, 1280x720)
107 KB
107 KB JPG
>>101540105
Just started with Harvard CS50.
Will knowing Binary actually help me be a better programmer?
>>
File: master_baiter.jpg (806 KB, 2048x2048)
806 KB
806 KB JPG
>>101548619
someone post what a real programmer who uses FORTRAN can do.
>>
>>101548619
>Will knowing Binary actually help me be a better programmer?
1
>>
>snatch DSDT data from the Registry,
no wait because the registry dumper guy is actually making money
>>
I want to start contributing to open source, specifically agpl3 code, but I hate c and c++ (subjectively). What should I work on? Language doesn't matter so long as it's not php ruby or the aforementioned.
>>
>>101548619
No, it's bloat that web "devs"(heh) learn to pretend that they're real programmers when they're centering a div. It's about as useful to programming as calculus is, just junk math that you'll never use.
>>
>>101540105
I've just discovered I can use environment variables in my esolang to keep global variables.
What does /g/ think of using such a method?
func setter:
@function that sets a global variable via environment vars
env "gvar", "12345678"

func tester:
@when env gets only one arg, it returns the var value
print env "gvar"

setter
tester
>>
File: DSDT.png (47 KB, 866x988)
47 KB
47 KB PNG
>>101548845
Certainly.
>>
I chiefly write object-oriented code where I work but would like to know how to approach problems with a functional programming mind set. My main language is C#. Where do I start anons?
>>
>>101545543
>Never explains why
Stfu
>>
>>101549247
You know why.
>>
>>101548619
Depends on what you are doing but sometimes yes.
>>
>>101547178
Not a bad idea.
>>
Is Rust community ivory tower?
I don't think an average person could handle Rust.
>>
>>101549504
The average person is too smart to give a shit about static analysis.
>>
>>101549504
no, definitely not
they're terrified of actual system access
i'm like 90% sure the rust community (the loud part at least, i'm sure there's plenty of people who know what they're doing) is a bunch of webdevs who only know surface level shit and larp as systems programmers
the kind of people who jump on the latest new javascript framework, the kind of people who import a library for two line functions
the fact that what i'm pretty sure most people use as the main rust UI framework, tauri, is webview based, does yield credence to that
>>
>>101550054
also everyone knows the ivory tower fags are the pure functional programming fags
>>
File: 1717939644544924.jpg (54 KB, 700x472)
54 KB
54 KB JPG
>>101540105
>What are you working on /g/?
Absolutely fucking nothing. I can't motivate and I don't fucking care no more. I need new purpose in my life.
I WISH IT ALL DEAD.
>>
>>101550119
>I WISH IT ALL DEAD
So why aren't you working on thermonuclear devices?
>>
>>101549063
Functional Programming in C#
How to write better C# code
Enrico Buonanno

August 2017 ISBN 9781617293955 408 pages

printed in black & white

Available translations: Polish, Simplified Chinese

C#
Microsoft & .NET
>>
>>101549063
>but would like to know how to approach problems with a functional programming mind set.
as long as you're not learning it through C++ template metaprogramming (like i did) you'll probably be fine
>My main language is C#.
that makes it easy, F#
same runtime different syntax
i haven't written much of it but i've written a bit and have been meaning to properly pick it up
the tutorials aren't as good unfortunately
>>
how do I learn to make a new coin?
the field is riddled with grifters so I'm hesitant to open youtube
is it as easy as downloading bitcoin source and modify something
>>
>>101540257
Die.
>>
>>101550431
>>>/biz/ was created specifically to contain you kids
>>
>>101545523
Because it's a very nice way to write code.
>>
>>101550431
"coins" are some sort of pattern you can find within a randomized string
You are going to invert the pattern, then simply use a randomization evaluator and boom

It really is this easy

Also fuck, some niggers actually got rich off of this shit
>>
File: squez.jpg (43 KB, 474x459)
43 KB
43 KB JPG
>>101550448
>>101550565
I'm not interested in grifting, never even bothered to learn about cryptoshit, just having some idea rn
>>
>another day wasted thinking of what project to start

How do I get over this? I want to start a long term project but get discouraged when I don't see the fruits of my labor quickly. I suffer.
>>
>>101550798
If you knew how to start you would have done so already. Reconsider what you're actually capable of and go back to grinding your fundies.
>>
Installed emacs on my smart phone.
I'm going to make the next punnet from my smart phone.
>>
can someone give me some advice on this https://pastebin.com/nWdE42mT I want to multithread both producer and consumer but the consumer is reading chunks not just one element at a time. I have a working example but it sometimes shits itself. Also why does it take so long to compile this? Nearly 4s when using g++ -o main main.cpp -lfmt -std=c++20 -pthread -fuse-ld=mold
>>
File: 1491419571379.png (376 KB, 1024x676)
376 KB
376 KB PNG
I'm thinking about writing a driver for an arduino-type board. It's basically going to be a simple sound card. My idea was basically plug it in -> gets recognized as the correct type of device, likely a cobbled-together snd-usb-audio type of device and I can manipulate it with alsa from there.
However, the actual connection is done through a USB-to-TTL cable, meaning that Linux primarily recognizes it as the FTDI FT232R chip. How do I handle this? Do I unload the FTDI driver, do I incorporate it in my own, or is there some override I should use? I still need to use the actual FTDI driver to send my data from USB->TTL, so it seems like I should be able to write my driver so that it boths incorporates and takes precedence over the FTDI driver, without me having to unload anything.

I don't think this should be particularly difficult but I'm curious about the "proper" way to handle this
>>
>>101551242
Where is your synchronization?
>>
>>101551242
not reading it because you waste too much vertical space but stop using 6 headers for a 40 line program and instead use syscalls
>>
>>101551287
I guess like this but I sure there is a better way of doing this https://pastebin.com/x4y9j6EQ I imagine going back to this after a week/month will be a pain in the ass to read.
>>
>>101551567
>>101551287
>>
>>101550468
It's a very nice way to make sure every useful networking library has a badly written async copy that lags 10 years behind on fixing all the same bugs and edge cases, usually with maintainers abandoning the project 3 years in.
>>
>>101551600
its right there
        while( i * chunk_width > data.count) {
std::this_thread::sleep_for(10us);
};

works for my usecase the consumer will wait until producer catches up. The only problem occurs when the producer thread is much faster than the consumer but that doesnt happen for my usecase.
>>
>>101551723
>works for my usecase
Yeah, until it doesn't, because the OS scheduler decided to fuck you over.
Retard.
>>
>>101551278
I guess in a "real" product you'd just rewrite the vendor ID etc on the chip? If that's the case I'll try to figure out some workaround/hack and not bother doing it "properly"
>>
I tried making a CLI program with Swift because I liked it a lot when making an iPhone app.
Turns out it actually sucks. All I was doing was downloading some data from a web API and storing it in an SQLite database. That's when I found out the standard library's JSON functions are somehow slower than Python's.
Apple sisters, it's over.
>>
OS dev anon here, i finally got my setup working with hello world and shit, i need an implementation book now, any reccomendation?
mind you that i already have a theoretical background, the dinosaur book which quite frankly was not that big of a deal, too shallow.
i was thinking about the tanenbaum's MINIX which was the one torvalds used, i also tried https://littleosbook.github.io/ but does't go very deep.
i appreciate any input, thanks.
>>
>>101542226
Is that Julia or Lua?
>>
Say I'm making the exact same GO http request (same body and headers) locally and it works fine but the moment I deploy the code to an AWS lambda function and run it there I get back a 401 error what could possibly be the cause?
It's the exact same code and request body/headers.
Could they have blacklisted us-east-1 or something?
>>
>>101552857
User agent?
>>
>>101552892
Doesn't change any of the outcomes with or without it, but good guess.
Another endpoint required that even though it wasn't mentioned in their docs at all...

Thinking about trying some github library to make the request instead of the default GO "net/http".
Or perhaps setting a proxy?
>>
>>101553020
In that case I'd fiddle with the TLS ciphers. A couple years ago I discovered that Cloudflare distinguished people using the Tor Browser by the cipher suite.

Back then I promised the Tor people that I wouldn't make this public knowledge, but since Cloudflare has changed their mechanism yet again I don't feel like honoring it anymore ...
>>
>>101550256
>>101550260
Thanks!
>>
File: cpp_great_docs.png (73 KB, 814x627)
73 KB
73 KB PNG
>inline constexpr /* unspecified */ split = /* unspecified */;
>(((since C++20)))
here are your docs bro
>>
>>101553592
>clownshow: the C++ standard
>>
File: 1513687910168.jpg (49 KB, 418x500)
49 KB
49 KB JPG
How do I make this also fetch .webms? I thought just adding
 || hyperlink.href.toLowerCase().endsWith('.webm')
to line 25 would be enough but apparently it breaks the script.

https://pastebin.com/un9xrYRv
>>
>>101553592
it's an invokable immune to argument dependent lookup referred to as a neibloid
i believe there's multiple ways to do them, and you're not suppose to use the views invokables for anything, aside from actual function calls which is why they're never specified in that way in the standard, but usually they're implemented as a stateless constexpr object with a constexpr operator()
>>
>>101553592
What else is it supposed to say?
>>
File: cpp_const.jpg (274 KB, 1600x1834)
274 KB
274 KB JPG
>>101553674
>niggbloid
I don't even know what is happening with this language anymore. don't even care.
I can't blame zoomers for not wanting to participate in this clown show, chopping their dicks off, and running with rust.
>>
>>101553819
they're named after the person that made them, they're a hack to break argument dependent lookup
it's not something you're expected to know, you don't get to know how these work because if you want their internals you're assumed to be abusing them or working on the STL
it's not relevant information and this is honestly more on cppreference than anything else, they should represent them as normal functions
>>
When are we comfortable admitting that the only reason we continue with C++ is because of the time and effort sunk into learning this prolapse?
>>
>>101553819
bets on what the next constXXX is going to be?
>>
>>101553999
when someone actually makes a decent C++2? like D but without the brainlet pandering and explicitly pandering only to people who use C++
seriously i can't believe it's really only now that something like carbon or cppfront were being discussed
no, not rust
ignoring all the other problems i'm sick of faggots trying to get clever and mixing ML and algol syntax, they're both fine on their own
>>
>>101554113
Jai is a thing. If you don't have access to it yet, then C++ shall suit you.
>>
>>101554300
i've seen his talks on vulkan
johnathan blow is an actually worthless retard who should kill himself
>>
File: zig.png (28 KB, 805x483)
28 KB
28 KB PNG
Ziggers, explain this stupid syntax
>>
File: IMG-20240724-WA0024.jpg (213 KB, 1600x1167)
213 KB
213 KB JPG
Should you let your players do the needful?
>>
>>101554108
constreflect or constasync
>>
>>101554466
or constract
>>101554349
im not a dirty zigger but i think this is like for takes the items and then a lambda, so its like for_each(items by ref or value, [item or &item] ...
>>
>>101554405
>Assigning strings to "\0"
Anon...
>>
>>101540105
>What are you working on
A friend of the family wants to get me a job on his company but I haven't messed with programming since 2015, is learncpp still good?
>>
File: 720.jpg (100 KB, 1280x720)
100 KB
100 KB JPG
>>101553754
nothing, it makes perfect sense. unspecified split is unspecified.
>>
>>101554750
the point is what matters is what it does not how it is implemented
>>
>>101552336
b-bump
>>
>>101554488
How can a lambda be an argument if it's outside the parentheses? And how can it be a lambda in the first place, that's the bitwise OR operator.
>>
>>101555576
its whatever zig wants it to be retard
>>
how do i use c
"just start coding stuff with it" how where would i start
how would i know what to do
how would i know what to type
wtf do i do if i dont have any ideas (ive become retarded)
should i just stick to making garbage in vb.net
what about c++
what about assembly for beloved reverse engineer(ing)
>>
File: 1693565276978052.png (10 KB, 289x216)
10 KB
10 KB PNG
>>101547111
update: made nested assignments work. I had no idea how to approach this initially, especially because I was trying to solve the issue at the evaluation stage, and found little help online. ended up solving it at the parsing step, by backtracking on an indexing expression and saving the indices I find along the way in a vector. by backtracking I mean ???[6] = 7; => ???[5][6] = 7; => x[5][6] = 7;
>>
>>101555972
Codecademy.

Assembly is fantastic but not available on codecademy. Not sure if there are good sites for it outside of YouTuber scams.
>>
File: assembly_is_assembly.png (16 KB, 560x614)
16 KB
16 KB PNG
>>101555972
How fucking hard is it to google "c tutorial"?
>c++
Has bad memory management, even if everyone pretends to ignore it.
>assembly
You can use inline assembly in gcc.
>>
>>101556036
this guy's M68k assembly tutorial is apparently pretty good
https://mrjester.hapisan.com/04_MC68/
>>
>>101555972
K&R
>>
>>101556069
Can't you literally use C code inside C++?
>>
>>101556030
languages like C would evaluate a[b] to a ptr then a[b][c] is just that ptr[c]
similarly in C++ you could have references
>>
>>101556088
You can, but you shouldn't. Not only will C++ devs filet you, but C++ has implicit garbage (constructors, destructors, exceptions, etc) that doesn't exist in C. In C you only pay for what you explicitly use.
>>
>>101556036
will check it out tomorrow, thanks
>>101556069
very hard to google c tutorial saar
also may i ask why you consider it had bad memory management
whats so wrong if it does
>>101556076
i mean i guess ill check that out too since i dont think i did so far
>>
>>101555972
There has never been a less capable or passionate generation than you lazy fucks
>>
>>101556172
>whats so wrong if it does
https://www.rfleury.com/p/untangling-lifetimes-the-arena-allocator
>>
>>101556172
Not her, but the difference between C++ and C is that under the hood, C++ calls malloc and free, which, as you know, are a communist psyop to slow down computers, undermine the Third Roman Empire, and trick people into thinking we landed on the moon
>>
>>101556179
the next one
>>
>>101556119
But the fact you CAN manage memory manually with free/delete, malloc/new, makes it kind of moot I would think.
>>
>>101556172
C/++ attract mad left-leaning political activists, so take their criticism with a grain of salt.
>>
Are python yield generators thread safe? for example if I call this test function from multiple threads/processes they would have all seperate generators and the variables in it aren't shared right?

def test():
total = pre_add = 0

while True:
params = yield

if isinstance(params, dict):
key_1, value_1 = next(iter(params.items()))

if key_1 == "pre_add" and isinstance(value_1, int):
pre_add = value_1
elif key_1 == "add" and value_1 is True:
total += pre_add; pre_add = 0
elif key_1 == "get_total" and value_1 is True:
yield total; pre_add = 0

generator = test()
next(generator)

generator.send({"pre_add": 15})
generator.send({"add": True})
generator.send({"pre_add": 5})
generator.send({"add": True})
print(generator.send({"get_total": True}))
>>
File: 1693168503996945.png (582 KB, 885x711)
582 KB
582 KB PNG
>>101556210
Just curious, but why do you assume I'm female?
>>
haskell is like C++: it has too much damn stuff
>>
>>101556224
But you shouldn't. Like, *really* shouldn't.
>>101556195
>>
>assume
Please
>>
>>101552336
>need an implementation book
nigger just figure shit out as you go, i would stay away from anything linux is doing, those niggers couldnt into arrays and now they are changing everything
>>
>>101556030
celebrated too early, the correct behavior should've been
x[_][3] = 4;
[0: [1: 2], _: [3: 4]]

thankfully it was easy to fix
>>101556099
true enough, if I decided to support pass by reference I could just do this
>>
>not answering the question
I'll enter a plea of "insanity" on your behalf.
>>
>>101553999
C++ is the only language that allows you to not wrap the preprocessor with your own, C projects do that shit all the time
>>
>>101556266
Every modern language is bloated horseshit and keeps adding shit every year. That's why i stuck with C. If only C had memory safety that would be the only feature i would like.
>>
>>101556383
>please suck all the joy out of writing an OS for me because i am a limpdicked faggot
no, if you dont like it give up.
>>
>>101556467
Not everything is targeted at you, insane-kun.
>>
>>101556249
they don't have any thread related behaviour at all, you could reference a shared mutable variable or not, same as a regular function
if you mean the state that says where in the generator you are and what the local variables are that comes from calling the function, if you call it multiple times you get different ones
generator1 = test()
generator2 = test()
but if you did generator2 = generator1 it would be the same
>>
>>101556430
>If only C had memory safety
What would this look like? Austral with curly braces?
>>
>>101556502
4chan is literally a honeypot of federal agents built for me, i play along
>>
>>101540105
Why can't I stop doing Advent of Code puzzles? I want to start and work on other projects, but it is just so FUN!
>>
>>101556604
I looked it up. There's a big list of anti-features. That's the spirit. But i like my curly braces indeed.
>>
>>101553999
Not just the sunk cost of personal time invested, but the availability of best practices that have been developed by people who have spent their whole life figuring out the best way to do things and freely sharing it.
>>
i want a language that optimizes operations like

b = map(a, f)


where a is never used again, to an in-place

for(i = 0,len(a)-1) a[i] = <body of f(x)>
>>
>>101556752
>optimizes
>>
>>101556179
you do not need to hit me with that
>>
>>101553999
There's nothing wrong with C++. Or rather, there are many, many things wrong with it but you can still use it in a way that is mostly correct.

The alternatives are all literally worse. Including C.
>>
>>101556747
>who have spent their whole life figuring out the best way to do things and freely sharing it
Can you point to some of these best practices? And what do I get if my practices beat theirs swimmingly?
>>
>>101556808
>The alternatives are all literally worse. Including C.
Riiiiiiiiiiiiiiiiiiight.
>>
>>101556752
And how do you know this isn't already happening in cases where it is practical?
>>
>>101553999
i haven't used sepples since, i don't know, 2010?
>>
>>101556808
Java mogs your language when it comes to literally anything except embedded or gamedev, and even then most people still don't gamedev in C++
>>
>>101556829
that is the fault of cretins using the standard library, not the language
>>
>>101556752
kinda pointless instead of just having an explicit map assign function
>>
File: IMG_20240725_000027_903.jpg (124 KB, 1280x720)
124 KB
124 KB JPG
I finished the the ui i just need, few tweaks and i'll be done
Now give me a head pat for my hard work
>>
>>101556890
What alternative would you propose?
>>
>>101556814
Link to your conference presentation?
>>
>>101556909
yo good job
>>
Is Rust the future?
No bias, what do you think? Is it to become the C++ of 2040? Is it inherently better as a language and just merely limited by lack of development of features and compiler optimization?
>>
>>101556814
Not that trannon, but the best practices of C++ is literally:

>No STL
>No templates at all
>Write your own containers if you need to, just no fucking templates
>Stay the fuck away from TMP
>Don't use fucking smart pointers everywhere, they are completely unnecessary most of the time
>In fact, understand RAII but don't blindly follow it
>Focus everything on data, your specific data, and how to transform it. Programming is just transforming data. Nothing else.
>OOP is short for OOPs I ruined your codebase.
>Seriously, focus on data.
>And make as few memory allocations as possible, that way you won't have to worry about lifetimes, ownerships, fragmentations and won't suffer from cache misses which slow down your code heavily.
>No fucking polymorphism. The 'P' in Robert C. Martin stands for "Professional Programmer".
>Seriously, stay the fuck away from TMP.
>>
>>101556927
Post best practices.
>>
>>101556829
Looks like someone is stupid. And it isn't the language.
>>101556877
Java is a Pajeet language for tards who goes "OOPs" all over your codebase.
>>
>>101556951
As long as they're running with the "object equals allocation" paradigm I'm not interested.

>>101556954
Why not use C at that point?
>>
>>101556914
handle memory manually, i like dynamic allocation on per-type arenas with a side array of bytes to check for free spots faster
>>
>>101556979
Sure, but I don't understand why you're talking about yourself in the third person.
>>
>>101556909
good shit bro
>dicks you down

>>101556954
>no polimorphism
you are retarded and cant into unions
>>
>>101556987
Because C++ is more flexible and allows for more tools than C does. There is nothing wrong with Objects just because OOP sucks. There is nothing wrong with using parts of the STL as long as you don't blindly do it and know what the fuck you are doing, OR for a very good reason don't give a shit about compilation time and performance. Etc.

C++ is a good language filled with retarded stuff and always makes the wrong thing by default.
>>
>>101557029
>Because C++ is more flexible and allows for more tools than C does
Like?

>There is nothing wrong with Objects just because OOP sucks
I never said anything about objects, my concern has always been memory allocation.

>as long as you don't blindly do it
Orrrrrr I can simply stick to see and keep my life simple.
>>
File: Capture.png (580 KB, 1224x629)
580 KB
580 KB PNG
>>101556962
https://www.youtube.com/watch?v=nZNd5FjSquk
>>
>>101557055
>>101556418
>>
>>101557071
What about these are specific to C++?
>>
>>101557071
>that image
fucking cancer
>>
>>101557011
Ok, let me rephrase that. No virtual. But if you know about tagged unions you don't need the best practices.
>>
>>101557087
I don't even know what that means, and considering that I don't I apparently don't care, neither.
>>
>>101556837
i expect rust's compiler does it

>>101556907
i don't even want to think about it, i just want to express some code in the idiomatic way. if i can see that 'a' is obviously never used again in a scope, then it should be just as obvious to SSA-based passes. and surely every map can be replaced by a loop and a lowered function, right?
>>
>>101557098
>>101557108
lmao now he's pissed, look you have NOTHING to share, nobody cares
>>
>>101557055
I used objects as an example.
What's your concern about memory allocation?
>>
>>101556979
>[X] is SHITTY and is SHIT for SHITTY SHIT EATERS!
Wow yes very sophisticated opinions there
>>
>>101557131
>everyone is the same person
Are you being mentally ill again?
>>
>>101557140
I don't need to explain why OOP is shit, do I?
>>
>>101556951
It's like Haskell, aka a research language. Not for real use
>>
>>101557138
Have you ever overloaded malloc/calloc/realloc/free in any program of your choice, logged every single instance of them being called into a file, and then seen how completely fucking retarded the allocation patterns are? Because I have.

>also multithreading
>>
>>101557155
>organizing your code is uhhh bad because i said so, okay?? modularizing is bad because I SAID SO!!!! STOP PUTTUNG YOUR CODE IN OBJECTS I DONT LIKE IT!!!!!!!!!
Sorry buddy but real software is object-oriented for a reason
>>
>>101557148
you're projecting and deflecting. just admit you have nothing to share and nobody cares about your opinion compared to somebody who has worked decades in the industry at a high ranking position and has generously shared their knowledge with the world free of charge even if you disagree, nobody cares because you have no proof just baseless narcissistic claims of superiority with nothing to back them up
>>
>>101557192
You sound easily impressionable.
>>
>>101557192
>mindbroken
which alphabet agency do you work for? Dont worry, this is 4chan noone will come and find you
>>
>>101556951
rust is the future in the same way that the next year is always the year of the linux desktop.
>>
>>101557183
Becuase of overreliance on STL, OOP-mindset and RAII? No, I haven't done it myself but I can quite imagine. Point is, you don't have to do it like that.

>>101557185
In ten years or so you'll realize how horribly wrong you are and how much time you've wasted being a bad programmer.
>>
>>101557306
>Point is, you don't have to do it like that.
Then why would I use C++ in the first place? I still don't know what this means: >>101556418
>>
File: software1677853442497504.png (718 KB, 1500x1084)
718 KB
718 KB PNG
>>101557185
i'm tired of real software
>>
>Sepplesfags arguing about "my subset of C++"
That's one of the main reasons it's such a shit language.
>>
>>101557345
my personal and private assembly implementation of virtualalloc replaces all of that.
>>
>>101557332
Because I have more tools available. Just because the STL is shit doesn't mean that I can't use it in moderation if I'm smart about it. Just because OOP is shit doesn't mean that combining data and methods are necessarily a bad idea.

I don't know what that means either, didn't write it.

And honestly, main reason for using any language is just that you learn it for a job, project or whatever and get used to it. I don't have any reason to move over to C or anything else that doesn't provide any great benefits over what I'm already used to.
>>
>>101557391
>he doesn't know you can't access kernel structures even with assembly
Cool story, bra.
>>
>>101557400
>Because I have more tools available.
>>101557055
>Like?
>some preprocessor garbage I don't care about and apparently no one can explain either, and objects, which are traditionally retardly allocated on C
There are no tools, are there.

>main reason for using any language is just that you learn it for a job
For you, maybe. For me it's also a pride thing. Do your fucking job properly, or don't do it at all.
>>
>>101557345
350M lines of code for Office??? How???
>>
>>101557454
premature encapsulation and enormous teams
>>
>>101557345
>50TB
God fucking damn.
>>
>>101557410
i did virtual alloc with sepple
>>
File: doeshe.jpg (10 KB, 844x475)
10 KB
10 KB JPG
>>101556987
>the "object equals allocation" paradigm
>>
>>101557444
>Such as
Any of the many features C++ has over C. They are all tools that may be used. I don't see any reason why this needs to be elaborated on. Just because you don't see any value in it doesn't mean that others don't.

>For you, maybe. For me it's also a pride thing. Do your fucking job properly, or don't do it at all.
Language is not a job, it is a requirement for a job. As a programmer you don't decide what languages are used in the codebase.
>>
File: 112.jpg (44 KB, 624x480)
44 KB
44 KB JPG
>>101557497
Sure thing.
>>
>>101557306
>In ten years or so you'll realize how horribly wrong you are and how much time you've wasted being a bad programmer.
In ten years java will have added value types, made the GC 90% faster, added more FP principles and will still have Clojure/Scala/et al. when you feel like being more fanciful than practical.
Meanwhile, you will still be ignoring the standard library of C++ because of how horrible it is and somehow think you're still winning.
>>
>>101557547
>Any of the many features C++ has over C.
Since you won't list them I'll assume that you know how shittily they're all implemented, which tells me that the only reason you'd ever use C++ is if you're literally too stupid to do it properly in C.

>As a programmer you don't decide what languages are used in the codebase.
No, but as a project lead you do. Guess what I am.
>>
>>101557562
>made the GC 90% faster,
OK lol.
>>
>>101557570
I don't need to list them. You know what they are.
>No, but as a project lead you do. Guess what I am.
No you don't, unless you work at a small company or on a greenfield project and most of those times the decision is made far above the team lead. I've been a developer, team leader, project leader, product owner and now I'm a manager.
>>
>>101557548
>malding about his low iq
top fucking lmaos
>>
File: file.png (62 KB, 234x216)
62 KB
62 KB PNG
>>101557585
In 2034 Java will actually be faster than C++ because Oracle pumps hundreds of millions into making Java faster, stronger, and better, meanwhile Bjourne will be jerking himself off over what completely unnecessarily complicated new thing he will come up with, throw into the standard and force compiler devs to implement
Your language is a joke, at best merely a happy accident. You will die doing embedded or as an overworked gamedev because C++ has no reason to exist in any other space
>>
>>101557653
>No you don't
So now you're at the stage of reality denial.
I rest my case.
>>
>>101557667
Let's say that you wanted to implement your next project in Ruby. And next one after that you'll give Zig a go, or maybe Fortran. That wouldn't raise any eyebrows, would it?

You are free to make those decisions yourself, right?
>>
File: double_regger.png (304 KB, 571x525)
304 KB
304 KB PNG
>>101557665
>In 2034 Java will actually be faster than C++
That is possible, but not due to the virtues of Java (as the original statement suggested), but due to the fuckups of C++.
>Your language is a joke
I'm regger anon. I despise both Java and C++.
>>
>>101557712
>Let's say that you wanted to implement your next project in Ruby.
No, because I know what I'm doing.
>And next one after that you'll give Zig
No, because I know what I'm doing.
>Fortran
No, because I know what I'm doing.
>>
>>101557740
You mean no as in no eyebrows raised or that you wouldn't ever do that?

Because one of them is a lie and the other is a concession.
>>
i stumbled upon this
https://uncyclopedia.com/wiki/C%2B%2B
>>
could you two go and take your meds now, please?
>>
>>101557763
>You mean no as in no eyebrows raised or that you wouldn't ever do that?
Both. Since I'm the lead I get to decide that. But I wouldn't switch to any of these three languages.
>>
>>101557796
>two
bring back the ip coutners, it had been me all alogn
>>
>>101557800
so what do you program in, larper-tan?
>>
>>101557812
Nah, my orthography is marginally better.
>>
>>101557822
>larper-tan
Why don't you guess, considering you're baking your own version of reality again?
>>
>>101557800
No, you wouldn't get to decide that. You are in the lead position because you wouldn't decide that. Because you act like a good boy who thinks he calls the shots when he does what the masters want.

That is literally one major part of effective leadership. Gather everyone under you around the goal and mindset so that even when they think that they have good ideas on their own, in reality they are just already made for them.

But go ahead. Try it as an experiment, see how long it takes before you're fired.
>>
>>101557829
>orthography and posting snippets of a game he didnt program is all he has
damn
>>
>>101557853
i know you use c++ dumb fuck
>>
>>101557863
Curious. I've been doing that for the last five years, and somehow - *somehow* - I'm still not fired.

The reality denial is strong in these threads, holy Christ.
>>
>>101557880
>i know you use c++
Wrong. Care to take another guess?
>>
>>101557909
You've been changing programming languages in your projects willy nilly? No, you haven't. What you have done is looked at existing tech stack, talked with managers and other stakeholders, and "together" come to the conclusion to either use the same languge as before or "boldly" try a new language as an experiment for a shorter project that is either similar enough or basically the same as what you already have.

But go ahead. Suggest Ruby next time and see how that flies on your meeting.
>>
>>101557974
>You've been changing programming languages in your projects willy nilly?
Certainly have. Right now I'm up to four.

>"together" come to the conclusion to either use the same languge
Nope. I made those calls. Competency breeds trust.
>>
>>101557974
pajeet here is chest-beating about the food he ate to generate the stool samples he produces for stakeholders
>>
>>101558017
>>101557909
>>
>>101557345
Bloat begets bloat.
>>
>>101558014
That means that either you work at a smaller company that doesn't maintain or support your projects (for long) after release and certainly don't have future development of either, or you work with non-technical people. Because no way in hell would any other company eat that cost.
>>
>>101558086
The fact that this seems so unbelievable to you tells me you've never worked in a company that trusts you. I wonder why.
>>
if you don't use the language *i* learned as a freshman and the text editor and other tools that *i* feel comfortable with, then you've never worked on *real* software with big boys like me
>>
>>101558143
It's real funny you say that, because I started out with C++ and C#. And then I dropped them because they were garbage.
>>
>>101558113
Companies that trust their employees are increasingly rare. Maybe she wasn't around for the good old days before the Nixon Shock anally unbirthed the entire economy.
>>
something something I don't know. I embraced the lifer life.
>>
>>101558166
Lol major cope
>>
What did the passive-aggressive schizo work on today?
>>
as i thought.
>>
>>101558323
Yeah, that's what your post is.
>>
>>101544685
I thought Tokio was bad, but then I started looking into C++ coroutines... Async programming is generally a hard to problem to solve in any native/compiled language because you need to have some kind of global runtime/thread pool/etc. It definitely violates zero overhead, compared to threads, because the cost of thread scheduling is minimal. I am still a fan of async though, even in C++, because it really encourages good practices like structured concurrency and cancellation.

At least the Rust standard is flexible enough you can even use async in bare-metal applications: https://embassy.dev/
>>
>>101558502
You've probably flown in hundreds of planes where your life is literally in the hands of C++. These are some of the best programming languages in the world.
>>
>>101558575
>You've probably flown in hundreds of planes
do americans really?
>>
>>101558575
I probably haven't because I don't fly.
>>
If it's not segmentation faulting, I'm not going
>>
I don't get why c++ is supposed to be bad when combined with clang-static tidy and format you get all the perks of new managed languages while losing none of flexibility. It's not like crazy shit ever gets merged into develop (formerly master) nowadays without them and I've seen enough initiatives of changing language vs updating and patching the codebase to see latter wins.
>>
>>101558706
>(formerly master)
>formerly
not on my watch
>>
>>101558706
>I don't get why c++ is supposed to be bad when combined with clang-static tidy and format you get all the perks of new managed languages while losing none of flexibility
Mate, what the fuck do you want from us? The arguments are all there: >>101556195
>>
>>101558757
I don't understand how that's relevant. That's about C and you can have an arena if you need it in C++ just as well. I was more speaking against C++ is bad because rust linter does things that are supposedly unthinkable.
>>
>>101558113
No, it is because I've worked with maintainance and support. I know the price of introducing new technology in an existing codebase in larger applications. It doesn't come cheap.

You need:
Developers who are proficient in the language both for development and long term support.
Compilers, tools and frameworks for build, test, static quality etc.
Set design and design rules and works well with the language chosen
Which means that the language needs to make up for all this overhead cost compared to what is already known, used and working.

This is not something done willy nilly unless you work at a small company, or in a greenfield project. Knowledge isn't cheap.
>>
>>101558816
>he doesn't know how C++ works under the hood
News flash: new/delete go through malloc/free. If you don't believe me: >>101557183
>>
>>101558826
>anon doesn't know what interfaces are
This is getting embarrassing for you ...
>>
>>101558847
Alright, you're just going deeper, that's fine. I was talking about one thing and you're promoting the other. I'll even agree, but job requires the subject and will for long. If anything I do like being pulled into a project with void* params and abstracting them away :).
>>
>>101558882
>job requires the subject and will for long
... I guess I understand now why other anons in this thread don't have their companies trust them. :) Setting up a Windows server with a C++ backend is the sort of decision that wouldn't instill trust in me, either.
>>
>>101558926
And I also agree there, but it is what it is. Why would you set up a Win Server for anything new nowadays?
>>
>>101558874
No part of my text had anything to do with interfaces. I do not want to work in the company you work in.
>>
>>101558961
Coincidentally I perused a job website today, and you wouldn't believe how many companies out there have done or are planning to and need help with exactly that. Azure, MS SQL, C#, Kafka, the works.
>>
>>101558973
Yeah, that makes sense. We only hire good people, after all.
>>
>>101559040
You, of course, being the one exception.
>>
the jobb( hav)er thread is this way
>>>/g/twg
>>
>>101559024
I've been writing about the current project, but I've worked on a monolith where the idea was to pinch away at things it doesn't have to know always and else be deployed as dockers wherever. First steps were hard but we got a good framework and a bunch are now communicating over web. All linux there though and all C++, also the microservices. Don't know about their overhead lol. But that's web, I'm on desktop now.
>>
File: 432476457567.jpg (39 KB, 800x600)
39 KB
39 KB JPG
>>101559070
Naturally. Hence my track record of five years and the complete trust that I enjoy.

But I've already said, /dpt/ grip on reality isn't so much loose as it's been flung around like a piece of ape poop in the jungle.
>>
>>101556951
there is always a best tool for the job, rust fills in the gap where people want high security from memory issues while still having high performance.
unfortunately it seems like performance is not as important as people think, and languages like python or javascript are fast enough for most jobs, rust is not actually as easy as JS or python, and the security given by rust for memory actually gets in the way of writing high performance code since a lot of performance that could be extracted in high performance code is actually locked behind custom memory allocators and structures and rust and C++ stl containers / RAII is not as easy to use / customize as C in this aspect (but C lets you shoot your foot, but when something happens in rust or C++, you blow your whole foot off because you need to understand what your code is actually doing, but this also applies to C and C isn't an easy language either).
>>
>>101558513
Just because C++ is worse doesn't mean Rust gets a pass for being bad
>>
>>101559117
You never contradicted me on that you are either working on a small company or with non-tech people. You never contradicted me on the overhead cost of introducing a new language. You never contradicted me on, well, anything of importance.

You're a joke.
>>
shut the fuck up you retards
>>
>>101559167
>>>101559167
mad
>>
>>101559162
>You never contradicted me on that you are either working on a small company or with non-tech people
Why would I? It's just another piece of fantasy your feeble mind has made up to protect yourself from the ugly truth that you're middle class at best. It doesn't require acknowledgement.

>You never contradicted me on the overhead cost of introducing a new language.
No need to. Because you're middle class you don't know what interfaces exist that allow components to be rewritten. The fact that you didn't even make the connection told me all I needed to know: >>101558973
, and it didn't require acknowledgement.

>You're a joke.
OK, middle-class-at-best.
>>
As a kid, like many others, i played videogames with my friends, in that time i quickly came to realize that he who cannot compete with skills competes with money.
>>
>>101559213
What's wrong with being middle-class, your ideal is hard to come by at job.
>>
>>101559248
How am I supposed to know? I'm not his mind which is trying to protect himself from that realization. Not everyone can be a super hero.

But at least listen to people that are better than you, rather than ignoring reality time and time again. I didn't do it either: >>101558166
>>
File: 1718252730333439.jpg (200 KB, 1024x1024)
200 KB
200 KB JPG
Just make a new thread anons
>>
>>101559362
fuck off
>>
>>101559213
Like I said, my post had nothing to do with interfaces and don't get solved by interfaces. If I have a code checker for a specific language, that sure as hell ain't helping me if I decide to write in another language. If I have a tried set of design rules for a language, that sure doesn't help me for another language that may or may not work different in key areas. Etc.

Like I said, you're a joke.
>>
>>101559417
>fuck off
Why are you upset?
>>
>>101559419
>Like I said, my post had nothing to do with interfaces and don't get solved by interfaces.
They do, though, and the fact that you ignore this simple fact repeatedly really shows not just me, but everyone in this thread just how middle-class-at-best you truly are.
>>
>>101559449
>but everyone in this thread just how middle-class-at-best you truly are.
you are coping for something, care to share?
>>
>>101559449
Pray tell, how is a set of design rules designed for one language helping me when I write in another language? What does the "interface" do?
>>
>>101559468
>you are coping for something
>I don't know what though
Real judge of character, aren't you. No wonder you don't instill trust.
>>
>>101559485
>What does the "interface" do?
Providing a set of rules that act as glue between different languages or modules, like XS for Perl, or a calling convention for C.
>>
>>101559485
>>101559449
And adding to this, lets say that we have Sonarcube as a static code analyzer, but for whatever reason you come along and decide that the next project will be done in Ada of all languages. What does the "interface" do?
>>
>>101559575
>for whatever reason you come along and decide that the next project will be done in Ada of all languages
I love your strawmanning, but I've already said that I know what I'm doing: >>101557740
>>
>>101559522
Which is not what I asked for. Like I said, my post had absolutely nothing to do with interfaces. This thing you have written here, this is just basic stuff that everyone knows.
>>
>>101559604
Just answer the question. What does the "interface" do?
>>
>>101559608
>Which is not what I asked for
It's exactly what you asked for:
>>101559485
>What does the "interface" do?

I'm done with you. Unlike you I have to work tomorrow.
>>
>>101559657
Nope. It is not. Try again.
>I'm done with you. Unlike you I have to work tomorrow.
Yeah, that company ain't gonna bankrupt itself.
>>
>>101559657
>claims he will finally leave
>copes
>doesnt leave
>>
>>101559702
I think he got mad and left. But yeah, his idiotic responsed just prove what a joke he is. The "interface" he keeps bringing up is not solving anything regarding the overhead, it is just extra overhead required to make different languages mix well, it doesn't take away any tooling or rules needs.

If anything it just proves that you are limited in what languages you introduce. Well, unless you are incompetent.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.