C# is easy. C++ is comfy. C is alright.

So I've downloaded a bunch of doujins over the past few years and I've finally decided to make some sort of tag search tool for them, luckily they all have some a metadata file in the cbz archive which I can use for tags. I've gotten as far as managing to parse through the metadata, and now I need to actually make some sort of database. I'm guessing SQL is the best choice right? Or is there something more limited but simpler to use? I don't need much, just associating a bunch of tags with a bunch of IDs.
Oh should have said, I'm using python.

>>101076203
I prefer C over C++ purely because C is simpler so for stuff that's a bit theoretical or complex I don't have to contend with language semantics as much.
But if I care about how much time it takes to make something C++ is so much better.

>>101076240
>Or is there something more limited but simpler to use?
A text file. Each line starts with an ID and is followed by the tags.

File: muh_backwards_compatibility.png (334 KB, 1000x800)

334 KB PNG

>another day another msvc ice
god damn I had this compiler so much

>>101076513
they fixed like 3 bugs I reported in their compiler already. they were very quick to react. it's a solid compiler.

>>101076506
I want something more elegant to parse

>>101076203
C forces me to repeat myself again and again and again and again and again and again

>>101076571
Yeah they've fixed a few I've reported as well, I eventually stopped reporting msvc bugs though because there's just too many. I maintain shit that gets built on msvc, gcc, and clang, and I run into at least 20 msvc bugs for every 1 bug I find in either of the other two.

>>101076513
I just stick to C++14 and no problems.
Don't tell me you are trying to use meme modules.
Never. Ever.

>>101076842
But doctor... I am using C++14

>>101076575
What could possibly be more elegant to parse than that? I don't think you know what you want.

>>101076896
I more mean something I can use pre-built functions to parse.

>>101076872
Ah well, I haven't updated VS in 2+ years so who knows what kind of regressions they have incorporated since then.

Can a github workflow that runs a script to generate files save the files to a machine? I know it can save within the repo, but I have a need to save to a local folder.

>>101077007
I honestly think 2015 was the best (using "best" in a very relative way here) msvc. It has all sorts of weird bugs sure but once you've worked around them it's more or less stable at least... I still maintain my msvc 2015 builds and they give me less trouble than the newer versions.

>>101077039
just have it zip up your files and upload them as an artifact.

>>101077105
Aren't the artifacts stored in the repo? I'm generating data for non-technical consumers and I'd prefer they don't have to go to git to access it. No spare machines to just leave on for cron jobs and run the script locally, so I'm hoping github can fill that gap

>>101076965
SQLite is good for small projects that you don't expect to scale beyond a few users worth of requests

File: demo.png (134 KB, 1122x867)

134 KB PNG

Added some new stuff to the cool chat, you can now start and stop the server daemon from the UI, provided you have a binary in the bin folder of course, i do not ship binaries, only source so thats up to you. the idea behind it is to make a chat program that is simple to use and to make it easy for people to run their own servers.

https://github.com/SoftwareWriterDude/coolchat

>>101077157
with github actions, artifacts are only shown on the run instance page along with the logs, I think you have to run some other stuff if you want this artifact to get attached to a new release on your account, like if you're doing automated windows and mac builds or something.

>>101077253
Time to go test shit then. Thanks

>>101077099
That's sort of where I am now. I've figured out most of the bugs in the UI and they annoy me to no end but everything compiles so I don't want to touch anything

rate my C++

float blend = (float)(unsigned char)mask_texture[px] / 255.0f;

>the committee is finally starting to address the retardation that is c locales and the rest of text encoding in C++
it's still annoying seeing the way text conversion was completely neglected in C and C++,honestly they should have just given them the gets treatment with C11 and replaced them with something like iconv

>>101077963
IBM and their retarded insistence to keep using EBCDIC (and the standards committee caring) has done so much damage to the programming community. It could have been so much simpler.
It wouldn't fix locales, but would fix one of the worst parts.

>>101077963
Yeah... if "finally starting to address" it means adding shit like the abomination that is char8_t, I'd rather they just leave things as is. Nobody in their right mind touches C locales anyway.

>>101078373
char8_t would have been fine, and the incompatibility would make it a clear deliminator between programs using implicit character encodings and explicit ones, had the standard library had any level of support for it (and for that matter, char16_t or char32_t) at all and not just here's std::u8string and really shit charconv conversion routines

>>101076153
Working on a new integrated instruction scheduler and register allocator for the 6502 LLVM backend. LLVM's just isn't cut out for such tight register classes, it can't do multi-allocation, and it's questionable good code is really achievable with phase-separated schedule and regalloc.

Some nagging part of me still says "just give up and use integer programming", but that way lies madness.

>>101078461
People can have char8_t if they want. What I'm angry about is that they removed the ability to have char type utf8 literals. It's fucking pants on head retarded, and working around it in a code base that makes heavy use of (pre-char8_t) u8 literals is a huge pain in the dick.

Should I learn C++ through a popular Udemy course, a book or learncpp.com? I've heard C++ is always updating and so is learncpp.com so it's the best option but I'm not sure.

>>101076153
this >>101048131 turned out to be a QEMU bug
it works fine on real hardware
thanks everyone

>What are you working on
I'm reading a book in chinese that puts spaces between each character which makes it a pain to copy paste. So I made a python script to replace all the spaces. Then I rewrote it in C and spent an hour researching UTF-8 because C expects ASCII by default (I think?)
It works now though :)

>>101076490
>I don't have to contend with language semantics as much.
Terry made HolyC because C semantics are beyond niggerlicious levels.

>>101076513
Nowadays they just say "unreproducable"

>>101076688
>forces
Have you heard of macros?

>>101076513
>>101080575
thanks for reminding me of this https://stackoverflow.com/questions/2898228/can-isdigit-legitimately-be-locale-dependent-in-c
>>101080580
show me the macro for adding custom + operator.

>>101080629
>show me the macro for adding custom + operator.
That has nothing to do with repeating yourself.

>>101080687
I accept your concession

>>101080718
That's not me and that has nothing to do with repeating yourself (if you are talking about overloading)

>>101080765
I'm not even anon who said this, I just accept your concession, typing less is always good, and you have no argument, macros don't solve any real problems, they're a cope mechanism, for example when I need to pass a type to a function, I give up writing a function, because only a macro can take a type in any meaningful way and be useful.

>>101077511
thats not c++
this is c++

auto blend = static_cast<float>(mask_texture[px]) / 255.0f;

>>101080804
macros literally solve the problem of letting c do anything what are you smoking?
>when I need to pass a type to a function
You don't need to do this
>only a macro can take a type in any meaningful way
crack

>>101080834
>You don't need to do this

#define align_up(T, P) ((void*)(((uintptr_t)(P) + sizeof(T) - 1) & ~(alignof(T) - 1)))

Yeah, being a real programmer who gets things done is not an usecase, lol.

>>101077511
>casting to unsigned char
your mask_texture[] type should already be unsigned in the first place if it actually holds values from 0-255. If it actually holds negative values then casting directly to a float is sufficient.

My only guess to why ur doing a cast like that is because it holds regular chars probably because you initialize it with a char array or something at some point. Its better to change the type to unsigned char and then cast any read chars to unsigned before u insert them into the array.

>>101080855
>

#define align_up(T, P) ((void*)(((uintptr_t)(P) + sizeof(T) - 1) & ~(alignof(T) - 1)))

void* align_up(size_t type_size, size_t type_alignment, void* P) {
        uintptr_t p = (uintptr_t)P; 
        uintptr_t aligned_p = (p + type_alignment - 1) & ~(type_alignment - 1);
        return (void*)aligned_p; 
}

Wow that was hard, but again. no usecase.

>>101080912
now I have to repeat myself by typing the type twice when calling the function and this function won't be inlined unless I use [[gnu::always_inline]], good job dunning kruger, you sure showed me how to repeat myself.

>>101080920
>macros don't solve any real problem
>i use macros cause c doesn't have foo bloat for my useless usecase

>>101080950
>allocator that's actually good unlike malloc is not a valid usecase
yeah I know that you're retarded, as you have shown me with your code that is slower than mine purely due to overhead, both cognitive and physical, just look at the generated code.

>>101080960
The types will always allign if you aren't misusing c. nice try though

>>101080950
>>101080960
u guys are both losers stfu

>>101080920
>this doesn't inline
Consider filing a bug report with your compiler.

>>101080976
>can't reproduce as I don't know how to code
Shrimple as

>>101080972
if the type ever changes, I have to repeat myself by changing it twice
>type_size mentioned twice
>type_alignment mentioned twice
>aligned_p mentioned twice
you're a dumb nigger who loves repeating yourself
>>101080976
having more than one compilation unit (aka not fizzbuzz) is not a bug, user not necessarily compiling with LTO is not a bug. Macro is superior, and you're disingenuous nocoder.

>>101080994
>I chose not to inline this code
My bad, I misunderstood.

>>101081019
yeah I can tell that you're retarded, which is why >>101080855 makes you mad, it's like someone being more intelligent than you and having to merely type

align_up(struct x, pointer)

in C which is minimal and actually easy to use makes you mad, it has to be convoluted and verbose so you can feel like an elite fizzbuzzer.

A program that aims to automate 4chan screencapping. It's fucking hard. So I nees to know, would any of you anons use it? What features would you like to have.
Commandline program written in lisp.

>>101081055
>screencapping
you know there is a dedicated key on your keyboard for that already

File: Capture.png (70 KB, 673x233)

70 KB PNG

>>101080902
>mask_texture[] type should already be unsigned in the first place
explain pic related then

>>101081570
stdlib is retarded

>>101081588
applies to every language universally
>>101081570
explain why you are using this instead of read and write syscalls

>>101081611
>applies to every language universally
no in most other languages you literally read a "byte" (0-255 value) or a unsigned char, or a char is simply defined to be unsigned by default.
Its only c++ that has signed chars by default and the standard library function for some reason work with chars instead of unsigned chars.

Its completely retarded because internally it reads unsigned values and just casts them for the type system for the use to then re-cast back.

>>101081637
standard library is not a language, dimwit, in C++ you just call

SYNOPSIS
       #include <unistd.h>

       ssize_t write(int fd, const void buf[.count], size_t count);

Windows users need not apply.

>>101081665
im not critiquing the language retard
im critiquing the standard library specifically

its retarded if i read the codepoint 128 that i get a negative value instead of 128 because they decided to make their IO API work with signed chars

>>101081055
I might care if it was written in anything other than lisp. For such a basic application, which I could probably bodge together with a lua script, ffmpeg and basic input macros, it needs to be hackable enough to adapt it to whatever my specific use case might be and I'd rather learn how to gape my asshole than how to hack a lisp program.

>>101081680
>im not critiquing the language retard
is that because you don't understand it? It's irrelevant that stdlib uses char

>>101081697
no thats because the anon asked why the standard lib does that and i said its retarded. You said "actually every language does that" and i explained it didn't, every other languages standard library isn't that retarded.

And now you've gone into "troll mode" because you're trying to play it off as you were trolling this entire time and not actually serious when you posted that retarded thing here >>101081611

now go on post another baity reply, i'll respond because im bored.

>>101081714
I never used iostream because I never use standard libraries, but if you got filtered by this snippet copied from stack overflow, I'm not sure how to respond.

#include <fstream>
#include <iterator>
#include <vector>

int main()
{
    std::ifstream input( "C:\\Final.gif", std::ios::binary );

    // copies all data into buffer
    std::vector<unsigned char> buffer(std::istreambuf_iterator<char>(input), {});

}

>>101081719
>I never used iostream
thank you for conceding.

>>101081728
I do indeed concede that someone who doesn't use a thing is still more proficient at it than you.

>>101081735
you dont even know what you're arguing lil bro
the irony is the code you posted is exactly what i told the anon to do (use a unsigned char instead of a char)

you've gone full circle and submitted to my answer.

@101081746
>I am responding for someone else because I'm butthurt that my posts weren't read by anybody.
I wonder why nobody cares.

File: b29.jpg (86 KB, 680x1058)

86 KB JPG

haha

>>101081757
Oops, meant for >>101081746

>>101081825
>>101081767

File: Screenshot 2024-06-21 at (...).png (11 KB, 348x83)

11 KB PNG

@101081825
your life must be very sad, first you got ignored by a retard who doesn't know how to use C++, and now you have to beg for my (You)'s, but I ran out.

>>101081825
>>101081837
both your lives are very sad quite frankly

@101081845 (You)
Proof?

>>101081853
this very conversation
this isn't the first time i've seen you in dpt "trolling" i can tell its you every time
you legit waste hours upon hours doing this and think you're "winning" because you waste a bit of other peoples time as well not realizing the irony behind that

>>101081955
you must be confusing me with someone else because low quality of these threads usually prevents me from checking it at all

>>101076153
about a million unfinished games

>>101079542
Anyone?

>>101082305
fuck off retard, decide on your own

>>101082305
Don't learn CPP, learn GO or Nim or Zig or Odin or anything with less friction than CPP. learning about classes, inheritance and make files is a waste of your life

>>101081853
nigga seriously @ a post reply bruhhhhhh thats crazyyyy

>>101082321
>learning about useful things is a waste of your life
Cool.

>>101082410
make files have no reason to exist. You have a turing complete language, having a separate language just to compile your program is supremely retarded. and the complication comes from c++ standard practices, making everything into a class, encapsulation and all that disgusting abstraction that makes you a worse, unhappy programmer

>>101082433
>just reimplement a DAG parser yourself
dunning kruger retard kill yourself

>>101076153
Renderer, written in Vulkan.
It's a fairly simple 2D renderer, meant to be akin to a unit in a fantasy console (so, something that's purposefully limited). I'm at that frustrating extremely point with using GPU programming where I'm competent enough to do things, or figure out how to do something, but not competent enough to do them confidently, so it's an absolute slog.

>byte: The byte data type is an 8-bit signed two's complement integer. It has a minimum value of -128 and a maximum value of 127 (inclusive).
you're reply, sirs?

>>101082724
>byte = uint8
>This is an alias for uint8, that is an unsigned integer, 8 bits wide.
Not my problem

>>101082724
Java has bitshift operators that can cope with this, not a problem.

>>101082760
No it doesn't, a negative byte is always negative unless you promote it to int an take the positive modulus of it

>>101082724
>byte: The smallest addressable integral unit for a given system
Then add some qualifier that excludes rare bit-addressable registers

>>101082792
fuck off retarded nocoder

>>101082830
you mad I know java better than your mother benchod

>>101082837
sir...

>>101076153
I m a noob so i am doing the Odin Project

>>101083510
I'm confused each time I read this. This ha nothing to do with the Odin language right?

>>101080920
C++ doesn't have this problem

File: calibretagediting.png (198 KB, 1637x671)

198 KB PNG

>>101076240
download calibre. Tag your books in there.
Its what I do, I've got epubs, pdfs, random ass stupid formats, &c. It supports tagging and file conversion which is rather nice.
You can also bulk edit tags, picrel

>>101081055
Automate it how? Taking screenshots of individual messages?

>>101083724
that's my point

>>101081055
Wouldn't this be very easy if you used selenium? I haven't tried but I assume you could target the coordinates of reply boxes

>>101082321
I'm seriously considering switching from Nim to Swift

>>101076153
how do you prevent clang from turning a switch statement into a jump table?

File: sabotaged learning material.png (380 KB, 818x476)

380 KB PNG

>>101082321
DON'T LEARN, DO NOTHING, LEARN MEMES INSTEAD

>>101082433
Why is a separate language retarded? Scripting languages are fine on the command line, right?

>>101076153
Nothing. I got tired of devoting myself to a hobby where all anyone cares about is making The Next Big Thing. It's too fucking lonely being around you guys or the plebbitors or the wagies. I do not want to drown in apathy anymore. I'd rather drown in unrealized potential.

So seriously what's wrong with negative byte? No troll.

>>101084070
kek, what the fuck is wrong with this diaper wearing fuck?

File: 1712873503201.jpg (157 KB, 700x800)

157 KB JPG

>program uses opengl 1 to display text
>each character is a quad
>each quad takes 8 calls to gl-vertex/uv, along with gl-begin/end
>displaying 2k characters means making 20,000 calls to opengl
comfy

File: pepe-sus.gif (36 KB, 127x120)

36 KB GIF

how to make an app without a web server? (use only on a local pc)
i want to make a cashier (recording transaction) app. what tools do people usually use for this?

>>101084668
it's call a regular GUI software zoomzoom. use any programming language and a gui toolkit library like Gtk, Qt or old Tk

? err   -->  if err != nil { return (zero_values,) err }
? expr  -->  if err := expr; err != nil { return (zero_values,) err }
?! err  -->  if err != nil { panic(err) }
?! expr -->  if err := expr; err != nil { panic(err) }

If it's this easy to fix 99% of complaints about go, why haven't they done it yet?

>>101083793
Anon I have over 8000 doujins. I'm not manually tagging them.

Summer has officially begun

>tfw getting fucking "left operand must be an lvalue" because i fucked with type-casting in gcc so much
sigh, i am going to give it and waste bytes in the binary with 64bit pointers, what will be next?

nevermind, i have to like expand 1 line of casts into 3 lines

(uint32_t) temp = ((uint32_t *)config_address)([0]+32);

makes gcc explode, unaligned accesses are literally impossible

>>101085741
>unaligned accesses are literally impossible
i don't know why you'd want to, but there's at least a couple unaligned load intrinsics like _mm_loadu_si128
and what, gcc doesn't have some equivalent attribute to MSVC's __unaligned qualifier which clang also supports?

>gnu assembler errors
... sigh

File: 1700003919980377.jpg (5 KB, 250x167)

5 KB JPG

how do i into lock-free concurrent data structures? shit seems like magic

>>101084668
>what tools do people usually use for this?
Typically a computer science degree and basic human intelligence. I'm guessing you have neither?

File: 1705480415206954.jpg (56 KB, 656x679)

56 KB JPG

>>101084668
i dont care how advanced gui toolkits get, java swing will always have a special place in my heart

>>101086407
dumb frogposter

>>101086440
no bully pls

>>101086407
double pointers and thread local storage

>>101086454
lock free bullying

>>101086407
Immutable data structures.

>>101086801
>wordy nonsense
kek, as always
The prolem if you are a noobie is that there is no way out of threads running the same code with the same state, hting is that there is, at the kernel level you have the msr with the logical processor number and like 10 control registers you can do whatever with, at the userprogram level you have thread local storage and descriptors

monitor_device.c:121: Error: operand type mismatch for `ror'
monitor_device.c:123: Error: operand type mismatch for `ror'
monitor_device.c:125: Error: operand type mismatch for `ror'
monitor_device.c:140: Error: operand type mismatch for `ror'
monitor_device.c:151: Error: junk `(%rbx)' after expression
monitor_device.c:151: Error: operand type mismatch for `movdqu'
monitor_device.c:153: Error: no such instruction: `psll $4,%xmm0'
monitor_device.c:154: Error: no such instruction: `psrl $8,%xmm0'
monitor_device.c:214: Error: operand size mismatch for `movsd'
monitor_device.c:226: Error: number of operands mismatch for `cmpsd'

this shitty assembler's operand order is all fucked up isnt it?

>>101076513
MFC is the only thing I hate more than Javascript. I've worked on a C++ MFC codebase which uses like 5 different string types. std::filesystem::path also using wchar_t on Windows only makes things merrier. What happened to the whole zero overhead principle? What if I only wanna use ASCII paths? Whoever decided that GetSafeHwnd should be a thing also deserves to burn in hell.

>gas manual has no instruction syntax portion
can someone lend me a hand?

>wikipedia

leal    8(,%eax,4), %eax         # Arithmetic: multiply eax by 4 and add 8

who is the fucking retard that wrote this? if the bits are 0 eax is used you retard nigger you are doing eax*5 + 8 monkey faggot

>>101084070
>((((kaufmann))))

File: 1569487756654.gif (2.76 MB, 350x254)

2.76 MB GIF

>>101087400
>>101087622
>>101087699

>>101087622
please genius-sama >>101088403

File: video3.webm (536 KB, 1052x500)

536 KB WEBM

>>101088558
Does anyone know how to resize animations in GTK? It seems the only way is to get each frame, resize it separately and then merge them into a new animation. But then you would have to implement the playing yourself! Has anyone else toyed with this? Is there any more straightforward way to do it?

>>101087699
http://microelectronics.esa.int/erc32/doc/as.pdf
page 61

>>101088626
I'm not the SIMD anon btw. I have genuinely never written a piece of inline assembly which compiled and worked as intended. Brainfuck is more readable than GCC inline assembly. It's just easier to compile assembly files separately and then link them together. You just have to be careful about your calling conventions. GCC also doesn't seem to like inline assembly with Intel syntax.

>vector<bool> is le... LE BAD!
>b-b...because you wont to pass iterators somewhere

Why the fuck would I do that if I know this thing does not support it? And why would you need iterators for that in the first place?

>>101088953
2024 and still no std::bitset<std::dynamic_extent>

>>101076513
MSVC is a horrible garbage.
>can't optimize basic shit
>it's STL is unoptimized(80+ critical section initialization/leaving calls for a fucking locale)
>since MSVC can't optimize shit it also have to manually optimize its STL making it more unreadable garbage
>libstdc++/libc++ still performs better without manual optimizations
>proprietary

Why people still use this garbage when you have Clang or even MinGW? They are available on windows, they are free and open source. What the fuck, why.

>>101089008
Not to say about the amount of features the GCC/Clang provide

Also, one important thing that totally kills the MSVC for me.

/arch:SSE2 invokes a fucking AVX2 implementation of std::rotate.

Jesus christ why is MSVC so retarded.
GCC/Clang STRICTLY conform the -march=whatever it's just you that never link libraries with different instruction set. MSVC just breaks everything.

File: 1718834819687548.webm (3.23 MB, 870x540)

3.23 MB WEBM

A few weeks ago, I made a post asking for a design opinion on the app I'm building, and then some anon told me I should share the app. Well, here it is: it's a Windows app for live producing electronic music. It lets you record and loop MIDI messages to control drum/synth plugins or external hardware synthesizers. If you record some drums and notes for other instruments, and loop them at different periods or whatever, you can build up something resembling a song. It's not great, but it surprisingly works. I need to add more features to make it easier to control.
If anyone wants to try it, there's a download link at this pastebin: https://pastebin.com/R46kVMzN (It also has a guide to setup some free software if you don't have a DAW with plugins already)
Here's the same webm with audio: >>>/wsg/5592808 (it almost sounds like a real song)
https://is2.4chan.org/wsg/1718834405676750.webm

I'm automating applying to jobs, currently my code supports most manner of greenhouse and lever applications though every so often I have a captcha issue, in the future I intend to charge to use my service once its robust enough though for now are there any decent captcha solvers that aren't too sketchy? I'm currently using nopecha for testing for now but its a security risk I wish to avoid

I hate python and I hate the assholes who wrote this pile of garbage. I hate how modules work, I hate the import functionality, I hate the package management and I hate unit tests.

How am I wasting my life today? Trying to figure out the perfect magic incantation to mock.patch(...) to monkey patch some garbage in a garbage application because I'm supposed to magically know exactly how and where something was imported 10 modules away that I have no control over. Thank you based python gods for your wisdom

Also what a scumbag community, basically every site/article/blog post saying anything about python is adware riddled garbage.

>>101089936
how do you have your project set up?

Which AI do you use to troubleshoot stuff?
I'm learning programming and so far Phind or ChatGPT 3.5 seem the best. I use it to explain stuff to me instead of searching on the internet, and to tell me why my code isn't working.
Should I get GPT4?

>>101087212
You can always do Clojure or Haskell.

>>101090197
There's this interesting new AI that will help you troubleshoot C and shell programs. It's quite popular, it might even already be installed on your Linux distro. If you type "man" into your terminal and then the name of the function or command you're interested in, you will get answers very quickly.

>>101076203
This is exactly how I feel. I can do so much effortlessly with C#. C++ is where I challenge myself. C I’ve only used in school hardware engineering. I was thinking of learning Go since it’s in this same sort of lineage.

File: file.jpg (35 KB, 620x480)

35 KB JPG

Been watching Reboot and wondering if there's any free resources for the study of viruses

>>101077007
I was reading this on my phone earlier and saw "negressions" instead
which I am calling them from now on

>>101090771
this Reboot?
https://en.wikipedia.org/wiki/Reboot_(2022_TV_series)

>>101090840
No, this Reboot
https://en.wikipedia.org/wiki/ReBoot

>>101076153
just got told my ui sucks and old and got compared to a website,I DONT GIVE A FUCK IT WORKS

>>101090958
i don't give a fuck. it looks and like a website. how are some programmers able to learn all this and then refuse to learn some visual shit?

File: microsound.jpg (4 KB, 176x200)

4 KB JPG

>>101089813
Could you autotune and/or LFO the rhythm?

>>101090988
people are always reluctant to learn, they just dont when it seems like there is no reward (or even because of the feeling of shame that comes from doing something so "plebeian")

>>101091266
Well, you could, but that would have to be configured in the synth plugin you use. My app just sends MIDI messages to play notes. There are also MIDI CC messages which can be used to change parameters in the plugin (such as LFO), but the plugin would have to be configured to listen to that CC message and tie it to the right parameter. My app will forward CC messages from an external MIDI controller, but it doesn't have a way to generate them in-app yet. It also doesn't record them, so if you change a CC value it won't get looped.

>>101076240
lanraragi is what I use for my doujins. I wrote a script that pulls new doujins from nhentai and takes the tags in the metadata json and writes them to a separate file that is then parsed by a plugin I wrote for lanraragi that will retrieve tags from said text file when the doujin is scanned in. I think lanraragi already has native plugins for dealing with metadata inside cbz archives.

>>101091287
>people are always reluctant to learn
It's nothing new, but you'd expect somebody in IT to be actually willing to learn. I'm learning programming, there are obstacles everywhere, and even I can learn new stuff that I don't want to. Literally giving up at the first obstacle is insane. That guy must hate his job.

>>101088558
>>101088651
>>101088777
IT COMPILED AND ASSEMBLED! :DDDDD

IT DOES WHAT I WANT (love 'volatile')

it segfaults because linux is gay

>>101091453
>give up at the first obstacle
he literally made the application and the UI, viper.

>>101091415
>I think lanraragi already has native plugins for dealing with metadata inside cbz archives
I doubt this counts, since it's a metadata file pulled by the installer and then packaged up. There's like two different formats of it too, a json one for the older version of the program and xml for the new one

>>101091453
like >>101091660 said, i already made the thing, they just want me to also make it look like a modern website and i won't since its not my fucking job to pretty up shit that works, what >>101091287 said is total bullshit only a total noob could spew, learning has nothing to do with this specially with such a boring thing also again its not my fucking job

>>101091584
Isn't the whole point of inline assembly that you don't have to do weird volatile hacks and the inline assembly syntax is supposed to tell the assembler what it can and can't do with registers and how they map to variables?

>>101091772
>learning has nothing to do with this specially with such a boring thing
you are in denial kek

>>101091819
no, the point is to go faster

>>101091984
Then what do you need C for. Stack frames are bloat. We live in a spoiled RISC world with 32 registers and SIMD

>>101092022
because assembly doesnt have structures and enums, although to be fair i should just write a parser to make all
of that stuff macros

>>101092048
They made Load Effective Address just so you could use structs in ASM. And now people misuse it for arithmetic

>>101092071
no? in what situation is that useful? it was made to not waste 4 cycles for the cpu to figure out the most efficient way to do mul

File: Veblen.jpg (9 KB, 196x257)

9 KB JPG

>>101090988
Learning how/making it look good is deliberately wasteful. UIs go out of fashion just like clothes. In a few years today's modern UI will say "I'm poor, utilitarian or frugal" just as well as today's ugly UI.

>>101092112
It's in the name

They laughed. They said it couldn't be done. I'll show them. I'll show them all.

std::vector<char> zLib_decompress(const char* start, unsigned long src_size, unsigned long dst_size)
{
    std::vector<char> dst(dst_size);
    int error = uncompress(
        reinterpret_cast<unsigned char*>(dst.data()), 
        &dst_size,
        reinterpret_cast<const unsigned char*>(start),
        src_size
    );
    assert(dst_size == dst.size());
    assert(error == Z_OK);
    return dst;
}

>>101089008
> can't optimize basic shit
clang/gcc use unsafe optimizations that break your code.

>>101092852
... if you rely on UB....

>>101092934
Its pretty much impossible to avoid UB in anything but the most trivial of software.

>>101093015
what UB can you not avoid? Genuinely curious.
All I've ever seen about UB and the sins of optimising compilers is "this piece of code has the right to fuck your girlfriend and because you used undefined behaviour it's completely legal"

When I do 100 GET requests to a server, is there a mechanism in the network stack that shortcuts something to speed up the responses if it can infer all these requests are from me and coming at roughly the same time? Does it cache something?

>>101093295
keep alive to skip the tcp handshake, in the real world the socket is simply kept open and polled for a bit

File: 1704364893718778.jpg (72 KB, 400x400)

72 KB JPG

>>101076153
I'm working on my python script. i got it running faster than my C script in some cases now so i have switched back to it

>xor hash for sets
that is magic
it's astonishing the things people invent to make fp langs a fraction of the speed of imperative languages

Ok, anons, please don't bully.
I never programmed before, and I wanted to start. Specifically wanted to learn the Dart language/Flutter, because it seems it is most popular locally to me right now.
Now, that InstallGentoo Wiki has some very helpful articles, links and resources for various programming languages. But not a single mention of Dart or Flutter on the entire wiki. Why? Is it unspeakable for some reason, or not popular enough, or I'm a retard that isn't searching right way?

>>101094042
>not popular enough
this
learning meme shit when you learn to program is a bad idea. there are no resources to help you learn, you'll discover that you need to learn popular stuff anyways (like javascript if you want to do web shit) and it usually turns out they're a meme for a reason.

>>101093044
>what UB can you not avoid?
you're underestimating the incompetence around here

Currently in the early stages of making a Youtube desktop application. The newest layout just sealed the deal for me.

>>101094042
I recommend you start out either with Visual Basic, C or assembly.
The last one you should only undertake if you don't expect to write for modern systems, although technically you absolutely could use assembly on modern systems.
I think you should still learn assembly either way, because it's nice to have a bit of a grasp of what the fuck the computer is doing underneath everything, but I understand that raw logic can be a bit overwhelming at first. If you want to do so, go pick up 86box, Turbo Assembler and MS-DOS off of WinWorldPC, and set yourself up a virtual 386 machine.

>>101076153
>gif
that guy's literally me

I asked chatgpt what other languages are there that are better than python for cpu bound tasks and it said java, is this true? both aren't compiled languages, I thought performance langs were compiled ones

>>101096143
Java is compiled though

>>101076571
>it's a solid compiler.
lol

>>101096152
It's compiled into bytecode, the same way Lua and C# is. I guess that means it's not assembled, but still.

>>101096143
java is JIT compiled
python is also very, very poorly made and very slow as a result, it has a lot more than the virtual machine penalty slowing it down

>>101096143
>I asked chatgpt what other languages are there that are better than python for cpu bound tasks
literally all of them

>>101096143
>I asked chatgpt
You are an insect. Pathetic waste of natural resources.

>>101085048
You didn't care about tagging when you had 7999 doujins, nor 7998 doujins, not 7997 doujins... Fuck off dumb coomerstain.

>>101085741
>unaligned accesses
>1 line of casts into 3 lines
or you could just use memcpy which is 2 lines and require 0 casting, memcpy is a special function in gcc and will be inlined for things like this.

>>101092934
How can you "rely" on UB you retard?
Literal NPC board holy shit, kill yourself.

File: 1592686766017.gif (247 KB, 269x350)

247 KB GIF

>>101092805
>mixed pascal and snake case
>c++ wrapper except you take the exact same c like arguments so whats even the point
>char type as byte leading to a bunch of reinterpret_casts
>asserting runtime error
>not shrinking the buffer afterwards

this is how i would write it:

using Byte = std::uint8_t;

std::vector<Byte> ZLibDecompress(const std::vector<Byte>& compressed, size_t decompressedSize = DEFAULT_BUFFER_SIZE)
{
  std::vector<Byte> decompressed(decompressedSize);
  int error = uncompress(decompressed.data(), &decompressedSize, compressed.data(), compressed.size());
  
  if(error != Z_OK)
    throw std::runtime_error{/*...*/};

  if(decompressedSize < decompressed.size())
    decompressed.resize(decompressedSize);

  return decompressed;
}

>>101096623
span instead of vector arg so you canjust call it like this if you only have a c-style buffer

auto decompressed = ZLibDecompress( {buf, size} );

but yeah

>>101096324
I mean people wouldn't have to resort to chatgpt if fucking discord wouldn't be the nowadays go-to tool for help, not to say people on discord is more focused on joking and chitchatting than actually helping. And I don't even have to talk about stackoverflow, do I?

Good morning dpt

>>101092022
In truth, C is too readable to give up on it, in C i can give a program a new feature in 1 hour, in assembly this would take me a day. The problem is the compiler and the solution is writing my own that says "fuck your gay standard" bit my plate is too full right now

>>101076153
writing a kernel module to periodically check for keyboard input

>>101096623
>std::vector
babyduck's first function, std::vector has so much overhead that convenience is never worth it, its only usecase is "unknown size" allocation, except mmap/VirtualAlloc of your total RAM achieves exact same thing in one allocation, page faults were going to happen one way or another

File: output.png (72 KB, 1024x768)

72 KB PNG

it's so over

>>101076153
is it really impossible to run a static binary on macos? do I always have to link dylib?

>>101097362
every time your program does anything useful, it requires syscalls, nowadays this is done via vsyscall through pages that are dynamically linked into your executable by the kernel, the only way to have a static binary that contains everything necessary is to write your program the way you'd write a bootloader.

>>101097336
erm, MAP_POPULATE

>>101097382
Not useful for unknown size allocations.

>>101097381
>nowadays this is done via vsyscall through pages that are dynamically linked into your executable by the kernel
by that drivel did you mean libc?
>the only way
nigger those are wrapping around the syscall instruction, bootloaders dont have syscalls, you can just
mov rax, 7
syscall //mmap
which is, by the way, what i thought the syscall function did in gcc, instead it is actually a function from glibc that doesnt get inlined, crazy.

>>101097341
saving that as a wallpaper

>>101097387
what do uknown size allocations have to do with prefaulting pages? Also if you are going to be a retard and call mremap and such nonsense you should have just allocated more memory to begin with

>>101097400
no, I meant the kernel, tranny

>>101097411
nobody ever mentioned mremap ITT except you, nocoder, learn to read, and learn to use your worthless brain

>>101097412
i think... you dont know what you are talking about and instead just repeated what some confused stackoverflow kid said

>>101097419
>giganigger is using mmap with raw addresses
>thinking that will fucking work
kek you are a nigger, you should be doing mremap always
>uhhh but no actually there is a variable for which i am doing a mapping
well, still there is no reason you wouldnt be able to do MAP_POPULATE
Are you a bot?

>>101097421
that confused kid's name is Linus Torvalds https://github.com/torvalds/linux/blob/16f73eb02d7e1765ccab3d2018e0bd98eb93d973/Documentation/x86/x86_64/mm.txt but sure, some /g/ tranny knows better

>>101097430
I never needed mremap in my entire life, but I am not a nocoder so maybe that's my problem.
>well, still there is no reason you wouldnt be able to do MAP_POPULATE
I wonder what would happen if firefox which mmapped 200G of virtual memory on my 64GB system, did MAP_POPULATE, not that a nocoder would need to be worried about such things since he will never be allowed to write any code for things that matter.

File: copilot234.jpg (10 KB, 578x174)

10 KB JPG

>>101097244
chatgpt/copilot is based don't listen to the zoomer

>>101097432
>vmalloc space is lazily synchronized into the different PML4/PML5 pages of
the processes using the page fault handler, with init_top_pgt as
reference.
what he said here is that there is a queue for page allocation instead of giving your process pages instantly you retard nigger, how the fuck did you come to the conclusion this had anything to do with system calls?

None of you will ever be a real woman

>>101097465
could you repeat your niggerbabble in english?

>>101097448
maybe you should learn to fucking program instead of relying on the kernel to unfuck your nonsense allocations

>>101097473
ok, post a real useful program you wrote that uses MAP_POPULATE
hard mode: I must find it useful to myself

>>101097471
go write a kernel you retard nigger, going through a list of pages and checking which ones are occupied by a process and all of the state and complications that come with that is slow (although not slower than restoring 4 tss's from the other threads in the data structures... anyway linux is a retarded kernel) so these niggers are waiting for the processor to throw an exception and just signing that "we owe this process a page", when the fault happens they are going to check if they can even give you the page and if they cant well though luck

>>101097484
>useful
i guess you win, even with no actual arguments.

>>101097495
could you repeat this while staying on topic and not rambling about irrelevant things and in english instead of your aboriginal dialect?

>>101097506
Ok so you literally cannot show me an usecase of MAP_POPULATE but I'm the nocoder.

>>101097381
dude I already have it on linux now I want the same on macos
just tell me how to execute a static binary ffs

>>101097507
no, you can go cut your dick off and fucking kill yourself for all i care.
Go no and ban me retard tranny.

>>101097516
maybe not wasting time on thousands of context switches for no reason?

File: carving freetype.jpg (1.1 MB, 3024x3024)

1.1 MB JPG

>>101097410
t-thanks guy

>>101097517
>I already have it on linux
you literally do not, it's impossible to run a program in a modern OS without it being dynamically patched.
>>101097527
>there's only one program running on my system
ok dunning kruger

>>101097537
>>there's only one program running on my system
you are actually a fucking retarded commie and this conversation is over, the handler for pagefaults does not take control away from your program and if it does all linux developers should be stoned.

>>101097537
>you literally do not, it's impossible to run a program in a modern OS without it being dynamically patched.
how do check if it's "dynamically patched"?

$ file a.out
a.out: ELF executable, 64-bit LSB arm64, static, stripped

>>101097559
>muh control
your userspace program isn't in control of anything, and even if it was, it wouldn't do anything useful, nocoding larper
>>101097564
man vdso

>>101097576
I know what vdso is, my program is not linked to anything

$ ldd $PWD/a.out
error: "/home/anon/a.out" has unexpected e_type: 2

>>101097576
my program is in control of your processor and i can do
while (1);
at any time to not give it back.

>>101097587
>I know what vdso is
The "vDSO" (virtual dynamic shared object) is a small shared library that the kernel automatically maps into the address space of all user-
space applications. Applications usually do not need to concern themselves with these details as the vDSO is most commonly called by the C
library. This way you can code in the normal way using standard functions and the C library will take care of using any functionality that
is available via the vDSO.

This is first paragraph in vdso manual, proving that you know nothing.
>>101097596
ok show me a program that will take control of all processors and will prevent context switches from ever happening, you have one try.

>>101097604
you are so ignorant yet so confident i am amazed

>>101097604
I don't use C library at all and invoke syscalls via the svc instruction
you sure you know what you're talking about rajesh?

>>101097622
one try as in the code must be in next post, I accept your concession
>>101097627
post code, we can talk then

File: output.png (22 KB, 838x134)

22 KB PNG

>>101097341
I wrote some text rendering code once, using Freetype2, harfbuzz, and fontconfig, and I wanted to properly render non-English text, but honestly I barely remember any of the details.

>>101097632
do it yourself.

>captcha: gatsb

>>101097641
You parents may not have accepted you, but I will accept your concession every time.

>>101097648
post the back of your hand

>>101097658
you're in control over what I do less than your userspace fizzbuzz is in control of your system

>>101097632
>post code

.global _start
.global msg
.text
_start:
    mov x0, 1
    adrp x1, msg
    add x1, x1, :lo12:msg
    mov x2, 5
    mov x8, 64
    svc 0
    mov x0, 0
    mov x8, 93
    svc 0
.data
msg:
    .string "poop\n"

$ as -o tst.o tst.s
$ ld -static tst.o
$ ./a.out
poop

File: camera_wont_work....png (251 KB, 697x666)

251 KB PNG

>>101097641
>>101097648
>>101097658
>>101097675
Take your meds samefag

>>101097604
NTA, and I work exclusively in pre-386

; Initialization
call getAddr
mov cx, cs
xor bx, bx
xor ds, ds

purge:
mov [bx], 0
inc bx
jc purge
; Increment segment position.
clc
inc ds
jc lock
cmp ds, cx
jnz purge
; Should be safe enough to clear our program.
add dx, endOfCode
mov bx, dx
jmp purge

; Lock time :)
lock:
jmp lock

; Get our exact address.
getAddr:
pop dx
push dx
ret
endOfCode:

Probably not the best way of going about this, and I'm not testing it, but... Y'know. It was a nice exercise.

File: Screenshot 2024-06-22 at (...).png (49 KB, 352x168)

49 KB PNG

>>101097701
this still has vdso
>>101097716
>take your meds
don't speak to the mirror in this thread, use your mouth IRL

>>101097701
>arm
you will NEVER be a woman.

>>101091772
well your first post sounded like you were just lazy and did minimum effort. it didn't mention that the visual side wasn't part of your job.

>>101097724
>outed for phonesposting
get run over by a car

>>101097723
>code that literally won't even work on my CPU will take control over it

>>101097724
>this still has vdso
proof?

>>101097736
this stupid nigger was also european, i hate europeans so goddamn much

How can I make neovim ignore a command in a specific directory? For example I am in some directory, editing a file I can NOT :w or :wq in. How do I just get it to ignore those commands?

>>101097739
>using firefox screenshots is phoneposting
when I phonepost, I use Brave btw

>>101097741
The only reason it doesn't work on your computer is because your computer enabled the A20 line.
Disable it anon, it's fun :)

>>101097754
not interested in your tranny larp, write C program that compiles for x86_64 in 2024 and takes control of processor in userspace

>>101097754
>thinking whatever you tell bios in regards to gpio will ever matter

>>101097762
The only way to do so is to exploit a flaw with the OS, which means that any code I write will be OS dependent. There is no (known) CPU flaw that allows you to do this for all computers.
Now excuse me, I'm going back to my assembler. I need to stop being a dumbass and figure out how the parser is going to work.

>>101097776
not my problem, you said you can do it

>>101097762

---nigger.c---
int main(){ while(1); return 0;}

gcc -o nigger nigger.c -w

>>101097778
I didn't.
>NTA, and I work exclusively in pre-386
I don't know why you thought I was the same guy.

File: Screenshot_2024-06-22_12-18-56.png (4 KB, 348x54)

4 KB PNG

Huh?
Is this real, chat?

>>101097780
I just started this and my system didn't freeze up, seems like context switches are happening...

>>101097787
schizos think the world is against them so to him all people are one and the same, just like his dick and a vagina

>>101097746
no proof? was it a lie?

>>101097795
because you have 30 processors nigger, try with one

>>101097796
Yeah I guess. I don't know how else you'd look at a program which so obviously overwrites literally everything in memory except itself starting from the interrupts and go "hurr, this must be the same guy and he can't do a thing!1!!!1"

>>101097808
ok I started nproc amount of these programs and I can still post on 4chan, seems like context switches are still happening...

>>101097824
because linux is an even more gay kernel and when you give interrupts through the keyboard it will treat it as a system call instead of giving control to my fucking program, stone them all.

>>101097824
That is because the CPU can do things about that, such as limiting execution time. However, in a vacuum, this code?

main:
        push    rbp
        mov     rbp, rsp
.L2:
        jmp     .L2

This right here?
Cannot be escaped.

>>101097840
>>101097846
yes yes, system just working fine is le bad... I accept your concession.

you guys have fun, i am out

>>101097851
That's not even a fucking "this is bad" statement, this is me explaining why even if you managed to purge every NMIs from RAM, the stated goal is impossible unless the OS does something really, REALLY stupid.

>>101097866
not my problem, I am in a nocoder thread asking for a program they claimed to be capable of writing.

>>101097873
You did get one that worked, just not for the processor you were hoping for. It'd sabotage DOS, and that's the only branch of operating systems that matter.

>>101097884
x86_64 is the only relevant architecture, Linux is the only relevant kernel, as evident by you being unable to break it.

>>101097890
>x86_64 is the only relevant architecture
It's also nearly impossible to program for, which is why we have high level languages that obscure it as much as possible.
>Linux is the only relevant kernel
Unix is, Linux is just a cheap imitation.

@101097905
weak bait

come on niggers tell me how to execute static binaries on macos

>>101097915
I don't need to make bait to make my point. Linux has to copy from a mainframe operating system in order to be relevant, and your tools are exactly what I promise you: A pretty veil to cover up the ugly cracks and mangled mess that is x86_64.

@101097930
weak bait once again

>>101097748
what

>>101097942
Deny it all you wish, anon. You keep replying because you know all too well that the truth looms over you, just as it looms over me.

>>101097922
Static in which way? I know it doesn't support static libraries; if you wish to treat something as statically linked, it has to be baked into the kernel already.

How do I print "Hello, world!" in Python?

@101097958
you keep coping because you filtered by ASM syntax that doesn't cause eye cancer

>>101097986
>that doesn't cause eye cancer
What, you think I specialize in ARM? 65816? Z80? Itanium?
No. AMD x86_64 is a blight upon the beautiful simplicity combined with elegant flexibility that is the 80286 instruction set. Results have yet to be posted for the 80386 but I suspect that it'll be too far gone for me to like it.

@101098009
nobody ever had a problem writing x86_64 assembly except you

>>101097958
>Static in which way?
compiled with -static and not linked to anything

learning to use C++20 modules, only using GNU Make because every other build system and generator of is proving that they're inflexible garbage.

>>101097973
print("Hello, world!")

>>101097335
>tutorial makefile wont work because arch is a meme
>forced to learn gnu make
i just wanted to program, what the fuck is wrong with linux? just let me fucking privilege escalate to kernel mode from root for fuck's sake.

>>101098529
filtered

>>101098619
temporarily because there really is no other way out

Hate how every GHC release you have to wait a million years for cabal to fuck around before releasing a new version with a usually very minor change

>>101098808
Why don't you help then release it faster?
Oh wait, don't tell me, nocoder.

>>101098835
>nocoder
Get out of /dpt/

>>101098808
What does the newest GHC have that's worth the trouble?

>>101098974
https://gitlab.haskell.org/ghc/ghc/-/blob/ghc-9.10.1-release/docs/users_guide/9.10.1-notes.rst

What language server do people use for python? goto-reference on fullcontrol.Point fails to find fullcontrol.distance while going the other way works just fine (goto-definition on the use of Point in the type hint). So far I'm not impressed with the architecture.

any decent courses/books on networking / programming network applications?
im currently reading Beej's Guide to Network Programming but I would some more

>>101099058
look for the one that is not written in js/python

>>101096623
Snake case and mixed cases are common in C++

>>101099003
RequiredTypeArguments is barely different than TypeApplications. You can probably get better behavior with the latter if you constrain it with a closed type family whose last fallback case is an GHC.TypeError.ErrorMessage. If that's actually true, all the extension does is let you write vshow Int instead of vshow @Int and the type signatures are also slightly prettier.
>Linear let/where
>GHC now collects backtraces for synchronous exceptions
>C in webassembly/js
These are good news. It's possible to finish the imgui/raylib app first though, unless it's supposed to be the linear types version.

>>101099307
its better to avoid ambiguous type complaints in cases where it can't be inferred
plus there's type lambda

>>101098107
Solved in CMake :P

>>101099505
doesn't work in shitmake at all

>>101099505
no.
>>101099617
this. modules are useless until they solve them in cmake.

File: output.png (53 KB, 1024x768)

53 KB PNG

>>101097341
we are so back

>>101099914
other way around, tranny, it just works in GNU Make

File: pylizer.png (237 KB, 800x529)

237 KB PNG

>>101099206
Pylizer does the same thing.

File: LwZJF.png (68 KB, 1543x717)

68 KB PNG

>>101100013
Cute! I've ran into this confusion to a much lesser degree in Xlib. Fonts have too many damn metrics
https://tronche.com/gui/x/xlib/graphics/font-metrics/

>>101099914
They work in CMake.

b

>>101101055