/dpt/ - Daily Programming Thread

Welcome home, big sister!

>>62395061
Does anyone know about a 20-byte structure used internally by VC-compiled binaries for heap allocation management?

File: 1483889196969.jpg (46KB, 300x366px) Image search: [Google]

46KB, 300x366px

>>> 0.2 + 0.2 == 0.4
True
>>> 0.1 + 0.2 == 0.3
False

>>62395199
this is why you should never store money as a floating point value

>>62395199
Are you retarded on purpose or it's just a plan to get (You)s?

I always write code in small pieces that I keep checking to make sure it works, to analyze each step and see how it should evolve.

Is this a good coding practice or not?

What is your approach to writing code?

>>62395257
A single compilation unit should never be more than 200 lines of code.

daily reminder to not read HtDP

>>62395257
You have some sort of neurotic paranoia
Seek help

>>62395257
>Is this a good coding practice or not?
They created procedural programming for this purpose.

>>62395061

#include <stdio.h>

int main(){

int c = 5, d;
d = ++c + ++c + ++c + ++c + ++c + ++c + ++c + ++c + ++c + ++c + ++c + ++c;

printf("%d\n", d);

return 0;

}

How much is d?

File: nars2000.png (984B, 232x125px) Image search: [Google]

984B, 232x125px

>>62395199
Huh.

>>62395257
thats more or less how I code. I constantly have to re-prove things about the code to myself. maybe test-driven development would be better, I don't know

>>62395313
UB

>>62395257
If you can't read and comprehend chunks of about 5000 lines of code, you can't call youself a skilled programmer. You're either an intermediate programmer or a total noob. Practice calculation and memorization.

Okay, so here's an idea: What if we made a fully-featured build script language like CMake... but it's actually a DSL of S7 Scheme?

>>62395343
just run a bunch of tests and some expected output.
If a new change broke something, go fix it.

>>62395313
>sequence point baiting
>2017

File: game.jpg (260KB, 1022x1524px) Image search: [Google]

260KB, 1022x1524px

A multiplayer FPS.

>>62395294
I think that a high iteration rate and a constant feedback loop is super important in development.

>>62395280
Why not?

>>62395335
what is govno? Is that a programming language?

>>62395515
>govnocode
That's just Russian for "shit code".

>>62395536
Why Russian? Russian programmers suck. Best programmers are in USA, India, South Korea, China and Japan.

File: 1505029335070.jpg (48KB, 640x677px) Image search: [Google]

48KB, 640x677px

>write a small program which requires bigass runtime library/language interpreter
Can this be called bloatware?

>>62395711
Yes.

>>62395220
53 bits of precision is vastly more precision than you need for any financial purpose. Just round your outputs and don't use == on floats.

>>62395962
or you could just use an integer and count cents

>>62395351
Here's a better idea: modules

Are there any good resources for intermediate-level Haskell? It seems that everything kinda peters out or diverges into a million different (largely theoretical) topics after monads and monad transformers

File: e0dcbf77ea60d1888d690f5450897c42.jpg (21KB, 480x405px) Image search: [Google]

21KB, 480x405px

Followup on >>62374918
I was offered the job even though I fucked up that question. Very tempted to take it.
There is salvation even for plebs like me.

>>62395257
If you're doing this you might as well be writing unit tests, which is pretty much standard practise everywhere these days.

There's absolutely nothing wrong with doing it. Full code coverage means you can guarantee that outputs remain the same after future refactors.

>>62395349
I needs to do euler programming tasks to become a skilled programmer like you?

Microsoft's documentation is shit. I need to learn to do MVCs in Asp.Net Core 2. What's the best resource for this?

Boss is making us use Vue.js + Typescript + ASP.Net core. I'm going to a career fair tomorrow to look for a new job

>>62396504
>Microsoft's documentation is shit.
you are the first person i've ever heard say that, MSDN is fantastic

https://docs.microsoft.com/en-us/aspnet/core/tutorials/first-mvc-app/start-mvc?tabs=aspnetcore2x

>ded american hours
idk if that's a good thing or a bad thing
on the one hand it might mean americans don't give a shit about programming here
on the other it might mean they've grown out of posting absolute beginner shitter questions on dpt like a lot of euro first year CS retards posting fizzbuzz tier bs

i think it might be the second

>>62395962
fractional cent amounts are a real thing

File: fsharp_logo-620x465-100572622-primary.idge.jpg (35KB, 620x413px) Image search: [Google]

35KB, 620x413px

What does dpt think of F#?

What is a good book to practice Data Structures using C?
C is my main language and I know how to program linked lists, but I am yet to program any kind of tree.

>>62397332
depends on .NET

Is there a more efficient way to compress a large dense matrix array than using a general compression algorithm?

>>62396641
It's just a jumbly mess of API calls with no context as to what the fuck these things mean. MS UI is an endless hierarchy of windows and dropdowns. MS cannot into design

Are there any O(1/n) algos?

Does anyone else here feel that programmers are way overpaid? Earning 6 figures for being able to solve shittly leetcode problems seems a bit outrageous. Programmers should be paid no more than what someone could make at dunkin donuts mixing coollattas.

>>62397560
>literally copypasting questions from stackoverflow
https://stackoverflow.com/questions/905551/are-there-any-o1-n-algorithms

>>62397630
I'm as shocked as you are

/dpt/ permitted programming languages:

>Rust
>Scheme
>Idris

Using Python? Go to >>>/sci/

Using JavaScript? Go to >>>/g/wdg

Using C/C++? Move to a safer language.

Using Java/C#? Move to a more expressive language.

>>62397862
so basically you can't have a job

>>62397862
>Using C/C++? Move to a safer language.
lmaoing at your life

>>62397591
Perhaps other professions are underpaid.

File: hqdefault.jpg (17KB, 480x360px) Image search: [Google]

17KB, 480x360px

Why is programming in the modern day so terrible?

Everything is so complicated now. In the 90s you could make a sweet static website using some backend language and build a business out of it. Now everything is about mobile design and frameworks which generate 90% of the real code.

On top of that, there are these backend frameworks. Nobody wants to just write their own software anymore, they just have rails or django or whatever do all the hard work for them.

Even PhD level Computer Science is boring now. Papers from the 1960s are cool and have useful info that anyone can use. Papers today are all micro-improvements on the latest genetic meta-heuristic for a neural network.

What part of any of this do people find fun anymore? Every problem has either been solved and boiled down to a few lines of some framework or is PhD level work.

All these frameworks are memes too. There's like a gajillion of them out there and next year nobody is going to be using them anymore.

>>62398154
The drive for "simple" programming languages

>>62397591
it's because 90% of the programmers hired turn out to be completely and utterly incompetent
as in, literally can't do the job

>>62398281
Then hire better programmers.

What's that? There's not enough of them? Try offering MORE MONEY. Try not using shit stacks and languages. Try letting them work remotely.

>>62398377
>Then hire better programmers.
the issue is you don't know how good they are until you hire them
obviously nobody would waste their time and money on those 90% of hires if they knew beforehand
>Try offering MORE MONEY.
but the post I'm replying to said they're already making too much

>>62395061
A bash function

>>62395593
>India
>China

>>62398428
Maybe stop the obsession with having easily replaceable developers. You can't have that and at the same time have highly skilled people.

>but what if I want to fire the guy? what if we can't find a replacement?
Why would you want to fire him if you know he's good at his job?

Lads, I'm reading a book on x86 assembly. What's the opinions on this? Am I falling for some meme, or can assembly actually be helpful?

>>62398628
Why?

>>62398628
If you're going to do low level shit like C or embedded or whatever, then knowing how everything works at a memory-cell level is very helpful

>>62398576
>Maybe stop the obsession with having easily replaceable developers.
can't
the 10% who can do the job know that they are special and can get a job anywhere
so they often leave whenever they feel like leaving

>>62395409
Looks breddy gud.
From scratch or are you using some shitty game maker engine?

>>62398671
>so they often leave whenever they feel like leaving
Give them a reason to stay, such as more money.

>>62398645
Honestly for no reason whatsoever. I don't work in anything remotely related to computer science or computer engineering. I just want to see how fast I can make some basic programs. I'm currently good at Python, which is supposedly slow, and from what I heat C is very fast, but you could possibly be faster with asm

>>62398708
>you could possibly be faster with asm
True. Compilers are very clever these days though, anon.

Have you written much C? It's much more painful than writing Python, and writing asm is much more painful than writing C. It gives you even less safety than C.

>>62398377
>Then hire better programmers.
Easier done than said, right?
>Try offering MORE MONEY
If they have the money what's there to motivate them to do more?
>Try not using shit stacks and languages
A programmer worth his salt will be able to develop on any stack or language
>Try letting them work remotely.
So they don't have to be accountable to what is present?

>>62398708
pick up c or c++ first
then you can ease your way into asm because you can see how your c/c++ compiles to asm when you debug, and you can write inline asm in those languages

Does a loop from 1 to n^n count as O(n^n)?

>>62398708
Python is actually really fast if you know what you're doing. Much of the libraries and such for it are also just C/C++ modules.

>>62398842
yes
assuming you're not skipping elements anywhere or anything sneaky

>>62398818
Skilled programming is a seller's market right now. There aren't enough skilled programmers, so you're going to have to make a better offer than other companies. Why would they work somewhere that uses $shitlang when they can take a similar offer from somewhere else that uses $goodlang instead? The same goes for having to commute vs not having to commute.

Why do you think remote workers aren't accountable?

>>62398914
>There aren't enough skilled programmers, so you're going to have to make a better offer than other companies
This doesn't match up with my experiences.

>Why do you think remote workers aren't accountable?
They wouldn't have to show their faces to take the heat should they screw up.

In C# is better to use a multidimensional array or an ADT ? I ask because in C++ it is usually just easier/more flexible to use a vector in the general case unless you specifically need arrays for some reason.

>>62398938
>This doesn't match up with my experiences.
Depends where you are, I guess. In my area companies are desperate for people.

>They wouldn't have to show their faces to take the heat should they screw up.
You know you can communicate with remote workers, right? Email, phone, video conferencing, ...

>>62398979
>Depends where you are, I guess. In my area companies are desperate for people.
What area is this?

I worked at a SV firm before, and despite being able to do N times more than some of the other people the boss was careful to make sure I was paid less. There was an expectation that the high rent prices would intimidate people to not take risks otherwise they might end up homeless. The other SJW companies basically gave me the cold shoulder when I tried to find a better job.

>You know you can communicate with remote workers, right? Email, phone, video conferencing, ...
There is something to be said about face to face communication and presence that isn't present in remote positions.

>>62399120
>There is something to be said about face to face communication and presence that isn't present in remote positions.
It makes it easier to catch diseases from your coworkers?

Trying to learn Haskell (ooh boy)

Where am I going wrong with this? I'm getting an error related to the function composition part at the last part...

--Make EnumToFrom (etf) for different types
module ETF where
eftChar :: Char -> Char -> [Char]
eftChar begin end
  | (begin == end) = end
  | (begin > end) = begin : eftChar (succ begin) end
  | otherwise = (reverse . eftChar)  end begin

Isn't

 (reverse . eftChar) end begin

syntactically equivalent to

 reverse ( eftChar end begin) 

?

>>62399232
No, as eftChar takes two arguments, the types don't match unfortunately

You could do

(reverse . eftChar end) begin

>>62398708
> from what I heat C is very fast, but you could possibly be faster with asm
Theoretically possible, but you have to put in an insane amount of effort before you can generate faster asm than a modern C compiler.

The main reason for learning asm is so that you can read it, not because you'll actually want to write it. Actually writing asm is incredibly rare nowadays; even 8-bit microcontrollers with less than 1K of RAM are typically programmed in C or (increasingly) C++.

>>62399232
>>62399330
There is the alternative of
((reverse .) . eftChar) end begin
(I think)
But I would rather do
reverse (eftChar end begin)

>>62399344
>>62399330
Okay I think I might be a retard, because I'm still getting errors:

--Make EnumToFrom (etf) for different types
module ETF where
eftChar :: Char -> Char -> [Char]
eftChar begin end
  | (begin == end) = end
  | (begin > end) = begin : eftChar (succ begin) end
  | otherwise = reverse (eftChar end begin)

Returns the following error when I load it in ghci:

Prelude> :l EnumToFromTest.hs 
[1 of 1] Compiling ETF              ( EnumToFromTest.hs, interpreted )

EnumToFromTest.hs:7:22: error:
    • Couldn't match expected type ‘[Char]’ with actual type ‘Char’
    • In the expression: end
      In an equation for ‘eftChar’:
          eftChar begin end
            | (begin == end) = end
            | (begin > end) = begin : eftChar (succ begin) end
            | otherwise = reverse (eftChar end begin)
Failed, modules loaded: none.

So I'm guessing it's getting confused with the type of 'end' or something?

>>62399395
If you look at the first case, you're returning "end" which is a Char.
Did you mean [end] ?

>>62395220
Floats are unnecessary anyway, fixed points are better

>>62399436
Yup. I'm a moron.
Thought I could pull a Lisp and just return the value and have the list-syntax automatically take care of it.

Thank you very much, though!

>>62399232
> eftChar :: Char -> Char -> [Char]
Note that this is equivalent to:
> eftChar :: Char -> (Char -> [Char])

"reverse" has the type
> reverse :: [a] -> [a]
and "." is:
> (.) :: (b -> c) -> (a -> b) -> a -> c

So the LHS function's argument type must match the RHS function's result type. But reverse's argument type is [a] while eftChar's result type is (Char -> [Char]), and those can't be unified.

If you uncurried eftChar:
> eftChar :: (Char, Char) -> [Char]
then you could use:
> | otherwise = (reverse . eftChar) (end, begin)

But there really isn't much point in using composition here. If you want to get rid of the parentheses, use $:
> | otherwise = reverse $ eftChar end begin

File: prop vs trad members.png (10KB, 451x275px) Image search: [Google]

10KB, 451x275px

What is the advantage of using properties as opposed to traditional variables, accessors and mutators? It seems to give me more boilerplate by forcing methods I don't need. Is private get ever not a waste of space? It doesn't feel better to me.

>>62398154
Why are automobiles in the modern day so terrible?

Everything is so complicated now. In the 20s you could make a sweet auto-gyro using some some helicopter parts and build a business out of it. Now everything is about interior design and industrial-scale machine shops that produce 90% of what goes into a car.

On top of that there are these automobile oligarchies. Nobody wants to just build their own car anymore, they just have Ford or Subaru or whatever do all the assembly for them.

Even driving a car is boring now. Cars for the 1960s are cool and have handling characteristics that anyone can enjoy. Cars today are all micro-improvements on the FWD commuter car.

What part of any of this do people find fun anymore? Every car has either been neutered into a Prius-shaped faggotmobile and boiled down to an eco-friendly automatic econobox or is GT-R level expensive.

All these aftermarket cupholders are memes too. There's like a gaillion of them out there and next year nobody is going to be using them anymore.
[spoiler]sorry[/spoiler]

>>62399549
Properties are garbage, don't fall for the bullshit.

>>Python 3.6.1
How to if statement?

if opt == 1:
    print("You picked option" +opt)
else:
    print("Siike... Das the wrong number.")

It just bypasses the entire command after typing in any numner, What gives!?

>>62399700
>Siike
What did you mean by this?

File: python.png (10KB, 802x128px) Image search: [Google]

10KB, 802x128px

>>62399700
https://repl.it/LBAI

>>62399764
sike

>>62399451
It depends whether you want roughly-constant absolute error or roughly-constant relative error. Each has its uses.

Fixed-point still doesn't help with the original problem, which is essentially the use of the wrong base (using binary where you want decimal).

And C does (optionally) have decimal floats:

#include <stdio.h>
int main(void)
{
    printf("%d\n", 0.2dd + 0.2dd == 0.4dd);
    printf("%d\n", 0.1dd + 0.2dd == 0.3dd);
    return 0;
}

>>62399791
Son of a bitch how could I have missed that. Thanks anon.

>>62397041
Why are people retarded? Like this fellow. Does he not know how big 2^53 is?

Start using C++

>>62399857
No problem anonkun

So how do you go back to the top?
>>Python 3.6.1

# Startup
print(
    "Dank Software\n"
    "By Anon |2017\n"
    )

opt = int(input("Type an Option: "))
if opt == 1:
    print("You Picked The Right Option")
else:
    print("Sike... Das the wrong number")

>>62400242
https://repl.it/LBAI/1

File: pythonn.jpg (16KB, 248x199px) Image search: [Google]

16KB, 248x199px

>>62400261
But it doesn't work in my shell.

>>62400330
Nvm I see it thx!

File: Thinking_Face_Emoji_grande.png (143KB, 600x600px) Image search: [Google]

143KB, 600x600px

Can I use my old GPU's VGA output as a bit banged high speed COM port?
It just pushes ones and zeros as pixel values, so I could use it to send lots of data one-way very fast, right?

>>62400483
VGA is an analog format. It doesn't push discrete digital data at all, but a continuous stream with "interrupts" for new line and new frame.

Can someone help me understand this? I'm new to C#.

String input = File.ReadAllText( @"c:\myfile.txt" );

int i = 0, j = 0;
int[,] result = new int[10, 10];
foreach (var row in input.Split('\n'))
{
    j = 0;
    foreach (var col in row.Trim().Split(' '))
    {
        result[i, j] = int.Parse(col.Trim());
        j++;
    }
    i++;
}

I know it's supposed to load a text file into a 2D array, but the parts I don't understand are:

foreach (var row in input.Split('\n'))

foreach (var col in row.Trim().Split(' '))

I know what foreach does and what .Split does, but where the fuck did row and col come from and how are they interacting with the foreach loop?

>>62395061
I was trying to figure out how to draw a rudimentary shape, probably a parabola and scale it off a float to make a side by side comparison of scale. But then I got bored reading about turtle and realized I don't even remember how to make a parabolic arch.

I was also planning on integrating a table to compare against, running on a timer to flash through comparisons. But this is peripheral to the above.

>>62400537
>I know what foreach does
Do you?

foreach (var foo in bar)

means that an iteration of the foreach loop will run for each foo which is part of the collection bar.

>>62400568
but where did col come from? input is just a string

>>62400616
col is declared in the inner loop condition.

Input is a string.
Input.split('\n') is a collection of strings.
The outer loop will iterate over that collection of strings, with each member being called "row".
The same thing will happen in the inner loop, with each member of row.Trim().Split(' ') being called col.

>>62395199
>not using Perl 6

File: Spectacle.J13523.png (35KB, 1137x301px) Image search: [Google]

35KB, 1137x301px

Hi, it's the operator overloading fag again. So How can I make the operator- to work in a way that n-x and n-x are understood?

#include <iostream>
using namespace std;

class Digit {
 public:
  int value{0};
  Digit(int x) : value{x % 10} {}
  friend ostream& operator<<(ostream& os, const Digit& d) {
    os << d.value;
    return os;
  }

  friend istream& operator>>(istream& in, Digit& d) {
    in >> d.value;
    return in;
  }

  friend Digit operator-(int x, Digit& d) { return Digit{d.value - x}; }
};

int main() {
  Digit x{17};
  int n = 4;
  cout << x - n << "\n";  // won't work, n-x will
}

>>62400819
nvm got it

>>62395271
Fuck no, a practical limit would be closer to 10000 lines due to editors slowing down. Editing is generally simpler with less files and the less files, the faster the compile times.

File: solution.jpg (78KB, 460x566px) Image search: [Google]

78KB, 460x566px

>>62400829

>>62400870
kek

>>62395280
Also am wondering why not?

Writing a very small compiler, from Scheme to MIPS assembly.

(define (compile->ssa e)
  (define orig e)

  (match e
    [(e:dat s d)
     (let ([r (fresh-reg!)])
       (ssa/k r (λ (next) (s:dat orig r next d))))]

    [(e:var s x bnd)
     (match (hash-ref (current-bindings) bnd)
       [(b:reg r)
        (ssa/k r identity)]

       [(b:globl g)
        (let ([r (fresh-reg!)])
          (ssa/k r (λ (next) (s:globl orig r next g))))]

       [_ (fail/unbound s x)])]

    [(e:if s c e1 e2)
     (match-let* ([r-phi (fresh-reg!)]
                  [(ssa/k r-chk k0) (compile->ssa c)]
                  [(ssa/k r-e1  k1) (compile->ssa e1)]
                  [(ssa/k r-e2  k2) (compile->ssa e2)])
       (ssa/k r-phi
              (λ (next)
                (define s1 (s:mov orig r-phi next r-e1))
                (define s2 (s:mov orig r-phi next r-e2))
                (k0 (s:br orig r-chk
                          (k1 (s:jmp orig s1))
                          (k2 (s:jmp orig s2)))))))]))

trying to architect a large program is hard

File: 1502988638907.jpg (41KB, 315x527px) Image search: [Google]

41KB, 315x527px

How to improve this glibc patch?

$ src/ls --godbless -l
total 1.1 milebyte
-rw-r--r-- 1 root root 526 byte Feb 15 00:46 application.ini
-rw-r--r-- 1 root root 6.0 yardbyte Feb 6 16:48 blocklist.xml
-rw-r--r-- 1 root root 4.0 inchbyte Feb 6 16:50 crashreporter.ini
-rw-r--r-- 1 root root 1.9 footbyte Feb 15 00:46 libprldap60.so
-rw-r--r-- 1 root root 1.1 milebyte Feb 15 00:46 libxul.so

*coreutils

>>62395711
It depends how common the runtime is on your target system.
Most Windows users have .NET installed
and most Linux users have Python installed.

File: 5eb.jpg (193KB, 1112x978px) Image search: [Google]

193KB, 1112x978px

Friendly reminder that you should return reference to an object iff:
1. The object itself is passed to the scope as a non-const reference
2. If the object is created within the function scope, it is static
3. In no other situation

Reminder that you don't also need multi-return functions, create the expected values first and pass them as non-cost references into the function. These are called out params.

>>62401590
dumb akariposter

>>62401590
>you don't need multiple return because out params
dude aliasing lmao

Complete noob here

I'm in an intro programming course. We're using Dev C++, but I've got linux at home so I've been trying to get a functional IDE alternative.

So far I've settled on codelite, and everything seems to be working, except that when I run my program it just instantly closes, whereas in Dev C++ the same program will pause and ask me to press a key.

I tried using

system ("PAUSE");

But this didn't work. Neither did

cin.get;

.

Is there something simple I can add to the program, which will work in both Dev C++ and codelite and causes the program to pause before closing?

>>62395061
ever spent hours trying to reduce ifs in your shitty algorithms?
this is a shitty feel my dude.

>>62401652
Use codeblocks if you really need a C/++ IDE.

>>62401624
But that's no akari

>>62401652

char ch;
cin >> ch;

struct Foo {};
struct FooBar : Foo {};

struct Baz { virtual Foo* factory() { return new Foo(); }};
struct Qux : Baz { FooBar* factory() override { return new FooBar(); }};

How come this works?

File: ba5.jpg (294KB, 768x625px) Image search: [Google]

294KB, 768x625px

Trying to optimize an algorithm, which would yield more results, cache optimization of data structures, or a faster floating point math library?

File: example.png (107KB, 1920x1035px) Image search: [Google]

107KB, 1920x1035px

>>62401652
Okay, I realized I was actually just running a debugger instead of the running the program

Now I'm trying build and run, but instead of running my program it's running some other program that prints hello world. pic related

>>62401695
Is it easier to use than codelite?

Honestly, I don't want to have to set up a workgroup, project, whatever. I just want to open my .cpp file and then click "build and run" and it builds and runs.

vim, emacs, and apparently codelite, are too complicated; Dev C++ is really nice and simple and I'm wishing linux had an equivalent.

>>62401719
>cache optimization of data structures
almost always that one

another trick you may be able to apply is to actually do all the math in integers and just display the result as a float (may not be possible with your algo)

also try to figure out if your algo can benefit from memoization

File: 1485652958941.jpg (131KB, 591x800px) Image search: [Google]

131KB, 591x800px

Have you loved Lain today, /dpt/?

>>62401652
>>62401773
>We're using Dev C++,
jesus why

>>62400483
dumb emojiposter

>>62401773
>I just want to open my .cpp file and then click "build and run"
youll like it then, you just open the project file and then click build and run. Though do yourself the favor and just learn a TE now. Because thats where youll end up eventually.

Fucking around with mongoDB is pretty fun.

Trying to make a simple GUI program in Scala and I'm learning that sbt is a giant pile of diarrhea disaster. Who the fuck thought it was a good idea to remove swing from the standard scala library?

What book would u recommend for learning c++?

>>62401913
that scala is a giant pile of diarrhea disaster.*

>>62401773
Follow these steps, anon:

>xbps-install vim
>vim
>:make

>>62395349
What if I want to call myself a computer master?

>>62401797
Yes,I figured that would be the case.
Working on somebody else's code, and the chucklefuck used a record type to store variables of the same data type (floats), instead of an array with enums as indexes. Just for syntax sugar reasons, I guess, he wanted to access the variables through coordinate names.
My particular environment does absolutely no cache line optimization for record types. In fact, I think I think records might be implemented as pointers to variables.

Then the chucklefuck put the records in small array, increasing memory fragmentation even further.

>>62401652
Go on a mass shooting in your school. Your stupid teachers deserve it.

>>62401889
I agree, but for now, I need something simple
I'll worry about more advanced editors once I'm more comfortable with what I'm already doing.

P.S. codeblocks is working like a charm

>>62401997
>codeblocks is working like a charm
Yeah, it's kind of a pain when adding 3rd party libs though. But if you only use one or a set for everything just set it in the global imports instead of having to set each project.

>>62395335
cyka blyat, my dude

>>62395335
This is both unsettling and arousing.

File: 53oGhwBMmiA.jpg (36KB, 604x346px) Image search: [Google]

36KB, 604x346px

>>62396095
good job, senpai!

>>62401112
source code release when?

have you guys figured out the real life equivalent of lain yet? pics please

>>62401811
yep yep

File: 1500550027667.png (12KB, 192x226px) Image search: [Google]

12KB, 192x226px

>>62402219
>3DPD
>equivalent to 2D

>>62402219
I'm sure you could find one in your local asylum.

File: 1293437661490.jpg (22KB, 550x825px) Image search: [Google]

22KB, 550x825px

>>62402306
nm found it

>>62402363
shitty edit

File: [angry loli noises].png (193KB, 274x303px) Image search: [Google]

193KB, 274x303px

>>62402363
Repent, heathen!

>>62402393
gimme the good stuff

Multiple inheritance seems like a good idea until you try to implement it.

>>62402569
Are you trying to say that only the first child should get the inheritance?

>>62402210
When it's done, hopefully thursday or so, depend how fast I can figure out how to do register allocation

>>62402219
Yeah I met her once

>>62402651
can you introduce me to her?

>>62402664
I wish I could

What happened to the parenthesis girl that used to be standard for the OP?

>>62403082
>parenthesis girl
Sweet merciful jesus

>>62403082
it's mostly either Lain or Yuki

procedure FSplit
  ( Path : PathStr; VAR Dir : DirStr; VAR Name : NameStr; VAR Ext : ExtStr );
    VAR
      i,j : integer;
      disk : char;
    Begin
      i := pos(':',Path);
      if i = 0 then
        disk := ' '
      else
        begin
          disk := UpCase(Path[1]);
          Path := copy(Path,3,255);  { strip off C: }
        end;

      i := RPos('\',Path);  { find the rightmost \ }
      If i=0 Then
        Dir := ''
      Else
        Begin
          Dir := disk+':'+Copy(Path,1,i-1);  { strip off trailing \ }
          Delete(Path,1,i);
        End;

      j := Pos('.',Path);
      If j=0 Then
        Begin
          Name := Path;
          Ext := ''
        End
      Else
        Begin
          Name := copy(Path,1,j-1);
          Ext := copy(Path,j,length(Path)-j+1)
        End;
    End;    (* FSplit *)
[/code}

Messing around

File: 1481398378150.png (389KB, 934x1000px) Image search: [Google]

389KB, 934x1000px

>>62403133
>>62403151
You know what I mean

Does FP make unit testing easy? Any pure function should be trivially testable, since you don't need to check any files but just that the return value is correct.

>>62403190
yes, especially in Haskell with stuff like QuickCheck

>>62403177
No, you're a fucking disgrace.

Shell scritpz

# siapir.sh
#
# see if a program is running
# sh siapir.sh progname
#
if test $1
then
  fn=$1
  x=`ps acux | egrep $fn$`
  if test "$x" 
  then
    echo $1 is running
  else
    echo $1 is not running
  fi
else
  echo "Usage:  sh siapir.sh programname"
fi

>>62403190
I guess, but unit testing is a meme.
Real programmers formally prove their programs to be correct.

>>62403177
You don't know what you mean

File: 1480974118479.png (518KB, 974x974px) Image search: [Google]

518KB, 974x974px

>>62403082

>>62403232
Off yourself, fag.

File: trap-programming.png (1MB, 1127x1600px) Image search: [Google]

1MB, 1127x1600px

>>62403251
>he's not a little girl crossdressing as a guy crossdressing a girl
I bet you're a shit programmer.

>>62403251
I'm straight

>>62403208

#!/bin/sh
pgrep -l "^$1$"

I wrote a client-server application in C#, it uses Sockets for communication and data exchange. Currently both server and client share some common classes (a managed .DLL is referenced in both projects). Those classes are used for data exchange, before being sent to client(or server) each object gets serialized to byte[] array and de-serialized on receiving side to it's original state.

The question is - how can i make effective communication between C client and C# server? I plan to write client in C instead of C# for both Linux and Windows (currently it only works on Windows). First and simplest thing which comes to my mind is communication using simple string commands but this is not what i want, it's too trivial and not suitable for large data.

I am looking for good examples and any advice will be appreciated, thanks.

>>62403355
Have a standardized protocol, use structs ro something
also
>C#
trash

I'm currently learning programming fundamentals in college with visual basic.
Currently reading about polymorphism

>>62403365

Why is C# trash? its good for what i use it for. Why would i code server (which is just GUI) in C using GTK or something when i can do it easier using winforms

>>62403381
>server
>GUI
What are you doing?

>>62403393

Remote administration tool lel

>>62403355
>it's too trivial
>I don't want to do it because it's too easy
???

>>62403399
>>>/hackforums/
Fucking neck yourself, using C# is the worst possible choice due to dependencies.
Anyway, you're writing it oddly.
>stub
write in C, should only handle connecting back to control server and acquiring one (use bittorrent tracker, dns generation, etc)
>control server
listens on port 443 or whatever, installed on a server, takes commands from GUI client and queues them, takes result from stub and queues them for connection from GUI client
>client
installed on your computer, connects to control server, interacts with builder
>builder
compiles new stubs from source code, switches out libraries for signature detection evasion, interacts with crypter

>>62403430

I'm not writing some skid RAT for hacking or anything, just for educational purposes on network programming. I wouldn't use C# if i wanted to write l33t hax.

Anyway thanks for suggestions

File: 1470446490627.png (368KB, 483x531px) Image search: [Google]

368KB, 483x531px

The compiler is not really helping me with this one. Which lines caused this error?

#include <iostream>
using namespace std;

class Avg {
 private:
  static int total;
  static int count;

  float getMean() const { return count == 0 ? 0 : this->total / this->count; }

 public:
  Avg() {
    total = 0;
    count = 0;
  }

  Avg(int t, int c) {
    total = t;
    count = c;
  }

  Avg operator=(const Avg& given_avg) {
    return Avg{given_avg.total, given_avg.count};
  }

  friend Avg operator+=(const Avg& average, const int n) {
    return Avg(average.total + n, average.count + 1);
  }

  friend ostream& operator<<(ostream& out, const Avg& average) {
    out << average.getMean();
    return out;
  }
};

int main() {
  Avg avg;

  avg += 4;
  std::cout << avg << '\n';  // 4 / 1 = 4

  avg += 8;
  std::cout << avg << '\n';  // (4 + 8) / 2 = 6

  avg += 24;
  std::cout << avg << '\n';  // (4 + 8 + 24) / 3 = 12

  avg += -10;
  std::cout << avg << '\n';  // (4 + 8 + 24 - 10) / 4 = 6.5

  (avg += 6) += 10;          // 2 calls chained together
  std::cout << avg << '\n';  // (4 + 8 + 24 - 10 + 6 + 10) / 6 = 7

  Avg copy = avg;
  std::cout << copy << '\n';

  return 0;
}

>>62397862
>/dpt/ permitted programming languages:
Says who.

>>62403466
Why are total and count static? You're using them in nonstatic context later.
Also, you haven't overloaded operator<< for printing.

>>62403509
But I did, see the last method

>>62403466
use printf, don't use oo, it's not needed here

File: ....jpg (18KB, 200x256px) Image search: [Google]

18KB, 200x256px

>>62397862
Remove yourself, peasant.

>>62403574
It needs to be a free function.

>>62403454
Make it free and modular, a FOSS RAT would cause tremendous butthurt among (((AV))) authors. Remember DarkComet, now imagine if it was actually good.

TLS is good enough for networking, use it over p443 with LE cert+cloudflare. Send b64 encoded structs is good enough, C# and C can share header files right?

>>62403177
this "parenthesis girl" has a name, you know. That's Yuki. The one that's in OP's post right now is Lain

>>62403232
is this scheme

>>62403579
that faggot caused so much damage to the software industry

he's a war criminal responsible for global warming

>>62403691
>he's a war criminal responsible for global warming
and financing terrorism

>>62403677
But it used to be >90% yuki and threads without were considered invalid.

Why do we need heap memory in 2017?

>>62403748
size_t x[50000000];

>>62403737
yeah but Lain actually writes in common lisp, which is THE /g/ language. Also there is literally a cult around her persona

>>62395257
I divide everything into a function of its own. For example if I have to make a program mimicking a toilet, I'd write different programs for flush, flush knob, drainage, seat, tank etc , test them each separately then convert them into functions and integrate them into one big program

File: wow dude.jpg (178KB, 934x1000px) Image search: [Google]

178KB, 934x1000px

>>62403677
What about Hime?

>>62403795
In what episode? I've only seen her write assembly and C.

>>62403795
>which is THE /g/ language
Stallman isn't /g/, and never will be /g/.
Even the stallmanites here don't maintain the purist craziness their leader does.
Lisp is trash

Is there some C interpreter like for python? Just for testing/quick experiments, #import a .c file and then call random functions etc

>>62403763
Memory is memory, especially if it's known at compile time.

>>62396001
>No financial calculation involves a value smaller than a cent
Do you know the term 'interest'?

>>62403823
Stallman has literally no relation to common lisp

>>62403861
It's his favorite language.
>The most powerful programming language is Lisp. If you don't know Lisp (or its variant, Scheme), you don't know what it means for a programming language to be powerful and elegant. Once you learn Lisp, you will see what is lacking in most other languages.

Comes from his site.

https://stallman.org/stallman-computing.html

>>62399805
>2017
>not using rational bcd

>>62403883
Common lisp != emacs lisp
Common lisp != scheme
Yea, all of them are lisp family, but they are not CL, and i think we are talking about CL. Common lisp existed lonh before emacs and stallman saying "that's my favourite lanhuage"

>>62403915
What even is the difference between Lisp and CL?

>>62403818
Somewhere in the first six, can't really remember

>>62403931
Lisp is not common

>>62403814
who that

>>62403931
Lisp refers to familiy of languges that include Common Lisp, Emacs Lisp and the Lisps before them like maclisp. You should remember though that clojure and scheme are not lisp.
Sometimes when people talk about common lisp they just say Lisp.

>>62403931
Lisp is language family: clojure is lisp, scheme is lisp, CL is lisp. Common lisp is a standard, that tried to unify other old lisps into one. SBCL and ClosureCL are CL implementations with some of non-standard things on top

>>62403961
>clojure and scheme are not lisp.
>Sometimes when people talk about common lisp they just say Lisp.
>>62403979
>clojure is lisp, scheme is lisp

>>62401773
just fucking install codeblocks holy shit cunt faggot

>>62403961
> Clojure is a dialect of the Lisp programming language.
> Scheme is a minimalist, multi-paradigm dialect of Lisp

>tfw c++
Do I use struct or class

>>62403823
Only a smelly PAJEET would think this way.

Java is a poo poo language designed for poos who can't find the loo.

Is there some autocomplete plugin for vim? Preferably that lists arguments too so you don't need to alt-tab and `man 3 <function>`.

>>62403961
>Sometimes when people talk about common lisp they just say Lisp.
so what stallman was doing then?
And from my understanding Scheme is just an experimental lisp, right?

Why does a Lisp have to strictly follow the standard to be considered a Lisp? Scheme and Clojure get called Lisps for a reason, as they are most like it by a wide-margin.

Lisp purists are weird.

>>62404007
yeah dude, and like JS is Lisp with C syntax and C++ is lisp because it now has lambdas, bro!

>>62404008
If you have to ask, it doesn't matter which.

>>62404017
YouCompleteMe, but aldo just google it or something, not that hard to find

File: K&R hime2.png (1MB, 1280x720px) Image search: [Google]

1MB, 1280x720px

>>62403947
Hime Arikawa, the crossdressing C programmer of shimoshina academy!

>>62404008
use struct if you're making a lightweight datatype with no constructors or destructors, no access protection and maybe a few member functions and static members
use class otherwise

>>62404030
fuck Mozilla false flagger.

>>62404019
no they're not and that's obvious

>>62404045
Yes, scheme and clojure are not lisp, that's obvious.

File: you wouldn't cast malloc.png (119KB, 563x378px) Image search: [Google]

119KB, 563x378px

>>62404030

>>62404008
Use C instead.

>>62404017
Also would be nice to have something that catches the basic syntax errors so you don't have to waste the few seconds compiling.
>>62404024
I'm asking for recommendations. WIll check it out, thanks.

>>62404061
They are *dialects*

File: 1486160754675.png (22KB, 370x320px) Image search: [Google]

22KB, 370x320px

Let's say we have an input stream for c++. A line in the input data looks something like this.

Month Day, Year    Movie title Here

I want to put the year the movie was released on one string, and then the movie title in another. But if you do

data >> string

it will only go up to the next whitespace. Getline gets the whole line.
What I'm wondering if there's an "easy" way to do this or do I just need to a lot of work around?

>>62404236
Take the entire thing as one string. And then just parse the chunks.

>>62404236
Put '\b' in-between.

File: mp.png (17KB, 510x636px) Image search: [Google]

17KB, 510x636px

>>62398680
From scratch.

>>62404236
I said year, but I meant the whole date. As such that

string foo = Month Day, Year;

and

string bar = Movie title Here;

.

>>62404249
>>62404252
I will look into both these suggestions, thanks.

>>62404265
You better open source that shit, anon.

>>62403961
>clojure and scheme are not lisp
Nice opinion you've got there, buddy.

>>62404285
Keep poking fun. We both know it's a lot better than you can do. :)

>>62404300
Did reply to the wrong person, bud?

>>62404278
I meant '\n', it was a typo

>>62404314
Thought you were being ironic. I don't see why anyone would care about the source code for this thing.

Anyone here working on a cool program in Rust or Go?

>>62404326
Do you mean in the input data itself?

>>62404339
I was being serious. If you're building a game and its engine, you should open source that good stuff for both dick lengthening purposes and for other people (like me) to be able to get a copy and fuck around with it.
Not related, but don't use emoticons on /g/ regardless if you're being ironic or not.

>>62404379
>you should open source that
I probably will. I've been planning to start a public repo for ages, but I never get anything sufficiently done to show it off.

>don't use emoticons on /g/ regardless
Been posting here on and off for almost a year and you're the first one who complains about that.

>>62404370
No, just break your input in two and make '\n' output to switch to the next line.

>>62404438
>I probably will. I've been planning to start a public repo for ages, but I never get anything sufficiently done to show it off.
I feel you. Well if you do push this to public repo, I (and other anons I'm sure) would be happy if you notified us. Always nice to an anon with a good project going.

>Been posting here on and off for almost a year and you're the first one who complains about that.
That's because /g/ died nearly a decade ago and what we use today is nothing more than a corpse being perpetually raped by retards. Hence why /g/ is mostly composed of awful threads and posts that violate both global and board specific rules. Along with enforcement by hot pockets and jannies, the community must also self-regulate. If you don't, you get /pol/, /v/, /tv/, here, etc. So what I'm trying to say is that there is a tiny faction of users left on /g/ that give a shit. The rest want to meme it up with emojis, emoticons, brainless tech support, frogposting, feelsposting, and (of course) stupid flame wars.

>>62404505
Fair enough.

>>62404460
I'm not sure what you mean but I'll google it. Thanks for the suggestion.

>>62404549
I don't understand you.
You want your input in two strings.
Why wouldn't you make it in two strings?

>>62404584
Classes. The input would go into a class. The separate them up into class variables of their own, something like string release, and so would the movie, something like string name.

>>62404236
scanf

>>62404008
Always use struct. class has broken default behavior.

>>62404646
Are you talking about default private inheritance or something else?

>>62404657
Yep. I get why C++ does it that way (for consistency) but really they should just force always declaring what kind of inheritance you want instead.

File: a career in game development.jpg (100KB, 610x537px) Image search: [Google]

100KB, 610x537px

Quit your job and enter the lucrative field of mobile game development

>>62404712
lol mobile dev
>need java program to launch your android application
>no GUI library with C API, have to use something shit like QT
>the whole development ecosystem is so fucked it's just easier to make a fucking web app to get cross platform application
>limited resources so your shitty scripting language using slow application barely even runs on mobile devices

I just finished making a program that solves 8-puzzles/15-puzzles (any arbitrary size sliding puzzle) and I'm trying to figure out ways to optimize it and improve the speed.

It works well for most puzzles, I have yet to encounter one that caused it to run out of memory and it can solve some difficult 4x4's with 40-50 step solutions. But for really massive ones it hangs for too long, and I think I could speed it up on medium size problems too.
Is it even possible to make an efficient sliding puzzle solver using A*? What are the alternatives?

>>62404703
>(for consistency)
eh, access protection on members isn't really the same as inheritance type. Private and protected inheritance have such limited use cases that it should have defaulted to public in all cases.

But this is C++, it's full of "should have"s.

vdhl and usb1.0, could anyone recommend me (or share) some good bibliography about those topics? thanks.

>>62404008
Struct is on the stack, class is on the heap

>>62404827
Not in C++.

>>62404827
your object will be on heap if you use new. Has nothing to do with struct vs class.

What did C# turn to??

static (string first, string last) Foo(string n)
{
    string[] arr = n.split;
    return arr[0], arr[1];
}

Top kek, python.

>>62404827
That's only in C#.

>>62404827
Wrong.

>>62404878
What is split here, a property?

>>62404904
*split(' ');

File: 1504703003348.jpg (297KB, 1030x1500px) Image search: [Google]

297KB, 1030x1500px

New Game!! makes me want to be a gamedev but I don't know anything about games or programming

ヽ(〃･ω･)ﾉ

>>62404878
>they don't require parens to construct tuples
Fuck you C#

also >python invented tuples

>>62404907
So is this default argument + paren dropping?

File: 1443818149546.png (2MB, 1200x1726px) Image search: [Google]

2MB, 1200x1726px

>>62404909
Programming isn't as cute as NG portrays. But if you're heart's into it, then get some thigh-highs, cute panties, a skirt, and a copy of SICP (https://mitpress.mit.edu/sicp/ or https://github.com/sarabander/sicp-pdf/raw/master/sicp.pdf). After completing SICP, you can branch off into whatever (like gamedev).
Remember: if you're stuck, wear more frilly clothing.

>>62395199
>

>>> 0.1 + 0.2 == 0.3

>False
>0.1+e1 + 0.2+e2 + e4 - 0.3 - e3 > u
>e1, e2, e3, e4 < u
>e1 + e2 - e3 + e4 < 3*u
>

>>> 0.1 + 0.2  < 0.3 + 3 * numpy.finfo(float).eps

>True

File: 1446332888264.png (711KB, 1600x2162px) Image search: [Google]

711KB, 1600x2162px

>>62404966
>>62404982
Stop spamming your mental illness here, you fucking faggot.

>>62395199
Never use == for floats.

File: 1485182031460.png (706KB, 594x924px) Image search: [Google]

706KB, 594x924px

function ( blankspaceoneachside )

WHY C++?!

>>62405010
because there is no consistent style

>>62404008
Struct has fields public by default, class - private. Otherwise they're identical.

>>62405010
thats not a bad style though.

>>62404992

0.1 + 0.2 >= 0.3 and 0.1 + 0.2 + 3 * numpy.finfo(float).eps <= 3

*

>>62395199
Decimal type exists you know.

File: 1449759967675.jpg (164KB, 626x640px) Image search: [Google]

164KB, 626x640px

>>62405022
>>62405032
in all other languages you don't do that kind of jank

>>62402219
Activate your tulpa

>>62405065
>dualbooting your brain
Hell no.

File: why do you people talk to eachother.jpg (27KB, 500x324px) Image search: [Google]

27KB, 500x324px

>>62402219
Lain is real tho.

>>62405058
I do it in C# and scheme

>>62405058
>jank
how

File: Facepalm (3).jpg (35KB, 500x500px) Image search: [Google]

35KB, 500x500px

>>62405086

Why are languages that aren't lisps allowed to exist?

>>62405077
>not overclocking your brain for double the performance
baka

>>62405107
You're halving your performance by this preemptive multitasking.

>>62405091
what's more readable

function( parametersandother, bol shit )

or

function(parametersandother, bol shit)

the second one is a single uninterrupted line

>>62405133
He clearly has a spaces after the function name and the args. It's consistent and readable.

who wants to do my java hw? girl btw =)

>>62405225
What is it

>>62405225
My services start at $100USD per hour.

File: 1497335093747.png (890KB, 1550x1123px) Image search: [Google]

890KB, 1550x1123px

>>62404999
It's not my mental illness, anon, it's /dpt/'s mental illness.

>>62405261
programming a thing that's like a calendar app?

>>62405273
lol

>>62405225
Post it and I might help.

>>62405370
I already said it's a calendar app duuuuh!!
when can you get started?

>>62405225
>>>/out/

>>62405386
import java.util.Calendar
done :^)

File: 1492369274754.jpg (310KB, 750x850px) Image search: [Google]

310KB, 750x850px

>>62405225
>>62405361
>>62405386
Reminder.

>>62405404
do you have one that says "replying to shitposters is also shitposting"

File: 1495273859977.jpg (197KB, 640x910px) Image search: [Google]

197KB, 640x910px

>>62405445

NEW:

>>62405687

>>62405687

>>62405687

>>62398154

>Why is programming in the modern day so terrible?

Because modern developers suck. All these abstractions are an expression of their incapability to solve any problem.

The amount of bloat is directly proportional of technical amateurship