/dpt/ - Daily Programming Thread:

First for Trashkell.

It computes the 100th fibonacci number in more time than D computes the 1000,000th one.

>>60718289
So in other words, trashkell is up to 1000 times slower than D

File: 2017-06-02-223045_658x397_scrot.png (15KB, 658x397px) Image search: [Google]

15KB, 658x397px

Don't know what's going wrong.
Maybe it's the syntax.

>>60718265
I have to use Javascript for something and I'm Greenspunning hard

It's like pottery

Trying to decide whether I should either re-write my webscraper in ruby (from python) or to fiddle with godot

What do ya'll think?

>>60718318
Changed the first :if to :collect .... :and :do (setq ... str nil)

File: please let this suffering end.jpg (41KB, 480x463px) Image search: [Google]

41KB, 480x463px

>>60718265
>spent 20 minutes freaking out on why my code wasn't working as expected
>finally give up and decide to just ask on here
>as i'm writing my post and explaining the problem I realize I used '>' when i meant '<'
>i looked directly at it multiple times and read it as 'less than'
anyone else here retarded? I could excuse myself if it was a huge block of code or something, but it was like 5 lines.

File: dlang_chan.jpg (139KB, 470x545px) Image search: [Google]

139KB, 470x545px

Threadly reminder that dlang-chan is not dead; she's going to have her GC tumor removed (eventually); she's super duper cute; and she's a blast to write in! Say something nice about her, /dpt/!

>Features
https://dlang.org/comparison.html
>Standard library
https://dlang.org/phobos/index.html
>Package manager
https://code.dlang.org
>GC
https://dlang.org/spec/garbage.html
https://dlang.org/blog/2017/03/20/dont-fear-the-reaper/
https://dlang.org/blog/2017/04/28/automem-hands-free-raii-for-d/
>Books
https://wiki.dlang.org/Books

>>60718365
Brainfarts happen, just learn to live with it.

>>60718395
No there's nothing wrong with GC. In theory GC can even run completely in parallel.

The non GC autism should be optional as it is.
And don't post the picture

Made a thread asking the same question, don't think it's going to pick up any steam, so here it goes again.

Hey /g/, new to programming, really seeing myself getting into machine learning and artificial intelligence. To get started, I decided to start with a small project:

A machine learning program made to beat some Touhou game. Simple sounding enough.

But then I came across the idea, is it possible to create a program that can eventually beat any game you give it?

My knowledge of machine learning is small as of right now, but I'm really interested.

can someone help me out? I don't know how to select UTF 8 inside nano

>>60718419
look into neural networks homie

Rust is a great language, imagine Java without the inefficiency

>>60718417
>The non GC autism should be optional as it is.
It's not completely optional which is the problem. But yes, I agree with you otherwise. I posted those GC links because people need to educate themselves and realize that a GC isn't inherently bad in a general purpose systems language or otherwise. dlang-chan's quite fast.

>And don't post the picture
I'm going to update the mascot soon.

>>60718463
>It's not completely optional
You have to use different libraries, that's it. Why would you use standard libraries in the kernel in the first place anyway

File: languages3.png (17KB, 522x384px) Image search: [Google]

17KB, 522x384px

Reminder this is still true no matter how much you wish it weren't.

I'm trying to understand why the with_index method isn't working. When I run it with this I get a LocalJumpError.

class MyClass
  include Enumerable

  def initialize(array)
    @array = array
    each.with_index do |value, index|
      puts "Value is #{value}, index is #{index}"
    end
  end

  def each
    @array.each do |i|
      yield i
    end
  end
end

a = MyClass.new [1, 2, 3, 4]

But when I run it with this, it runs correctly.

class MyClass
  include Enumerable

  def initialize(array)
    @array = array
    each_with_index do |value, index|
      puts "Value is #{value}, index is #{index}"
    end
  end

  def each
    @array.each do |i|
      yield i
    end
  end
end

a = MyClass.new [1, 2, 3, 4]

Why do they run differently? I thought they were basically the same methods?

>>60718491
do people still den ythis?

>>60718524
>>60718491
>Fast development
>Trashkell
1000x slower, thanks to 12Gb/s

>>60718491
Haskell is 1000 times slower than other languages.

>>60718536
>>60718545
Wow, you couldn't even READ the graph

>>60718536
The beauty of languages like Idris and Haskell is you can run your programs written in them orders of magnitude fewer times during development, because you have more than just tests available to check correctness.

>>60718419
Different guy here, I'm interested in this. But, would like to know how to make a bot like this in the first place.

>>60718555
How can you develop anything fast if it takes 1000x more time to crunch numbers?

>>60718486
You're correct of course, but having the standard library have GC-free alternatives for each of its constructs would be nice. As it is now, they only have some alternatives.

>>60718573
i understand that codelets like you might need to test everything because they use such awful languages that don't correspond meaningfully to a simple syntax whatsoever

File: __izayoi_sakuya_and_remilia_scarlet_touhou_drawn_by_odaneru_apart__68511f8cf306fc15718aac14fd6abcdf.jpg (305KB, 1240x1748px) Image search: [Google]

305KB, 1240x1748px

>>60718491
Fact: Fortran has both good performance and fast development.

>>60718589
And not just test it, but test it thousands of times, I doubt the languages they use have good generative testing libraries

>>60718589
I don't think you understand. You need a server PC to run haskel programs to develop because average PCs are too weak to run resource hungry programming languages like haskell

>>60718608
I agree that it has great peformance. I disagree about the speed of developing programs in it.

>>60718589
>Not testing
Is this supposed to be a "plus"?

I think it's just another way to avoid the obvious fact that haskell is like 1200x slower

>>60718491
Assembler is not a language, is this your first week in programming?

>>60718629
Judging by the fact that he "doesn't need to test because the language apparently does it for him", yes.

>>60718626
>>it isn't necessary to test nearly as much
>is this supposed to be a plus

Yes, it is.

>>60718629
No, I've been programming for 9 days, shitlord

>>60718641
>it isn't necessary to test nearly as much
But one test of haskell takes 1000 times more than a typical test in other languages

>>60718641
By the time haskel finishes 1 test, my language can finish 1000 of them

File: 1496276268739.jpg (68KB, 700x700px) Image search: [Google]

68KB, 700x700px

stop bullying hasklel

>>60718640
You really don't need to test a function with type

forall a b c. (b -> c) -> (a -> b) -> a -> c

if you're using a consistent language. Really. Honestly. There's only one behavior that an implementation of that type could have in such a language.

File: 關我屌事.jpg (690KB, 2031x2652px) Image search: [Google]

690KB, 2031x2652px

>>60718623
>I disagree

>>60718655
>>60718663
You can write knowably reliable Haskell code without using any tests whatsoever.

>>60718667
>if you're using a consistent language
I am though. It's consistently faster than your scripting language

>>60718675
Coincidentally there is no haskell programs that one relies on

>>60718666
no

>>60718671
If nothing else Haskell/Scheme/etc. code will generally be shorter than corresponding FORTRAN code, making it quicker to write.

>>60718675
>relying on haskel
>NASA puts haskel in their space shuttle
>space shuttle comes back to land in the US
>emergency situation due to weather condition
>oh let's run that haskell program to compute the safest distance from the tornado
>haskell program freezes for 1000 minutes
>space shuttle lands in white house and kills trump
win win

>>60718683
>>60718718

Someone sure shoved a Haskell stick far up your ass for you to be this pissed off

>>60718718
I don't like America so this would be a plus for me.

File: POO.jpg (68KB, 709x895px) Image search: [Google]

68KB, 709x895px

Java is the most POWERFUL language

File: 12506712512.jpg (41KB, 960x960px) Image search: [Google]

41KB, 960x960px

>>60718718

Studying dynamic programming by watching ucdavis' algorthm design playlist.

>>60718697
>shorter code = faster development
If by "development" you mean "writing buggy code" then sure.

>>60718718
>ALEX WHY IS THE PROGRAM TAKING SO LONG?
>S-sorry you can write reliable hasklel code without using any test
>AND WHY IS THE PROGRAM FROZEN
>I-I didn't test it to see how many hours does it take to compute fib(n)
kek

IS HASKELL PART OF A BRITISH FUNCTIONAL PROGRAMMING CONSPIRACY?

Haskell was developed by Simon Peyton-Jones, a KNOWN BRIT.

ML was developed by Robin Milner, also a KNOWN BRIT.

Idris was developed by Edwin Brady, yet another KNOWN BRIT.

>>60718773
>britcuck lang
>1000 times slow
yeah

File: 1485173668037.jpg (106KB, 593x578px) Image search: [Google]

106KB, 593x578px

>>60718767
>>60718718
My sides

>>60718765
Shorter code means fewer locations for bugs to reside.

File: 1eddd581ff9612c621455b6f6c6eb380.jpg (244KB, 800x573px) Image search: [Google]

244KB, 800x573px

>>60718718
>>60718767
kekd

>>60718773
Britain is just superior and you're going to have to accept that at some point. Haskell is an Anglican language.

>>60718718
Legend says the space shuttle finished counting 1000 minutes after it crashed on the white house

File: smug_ran.jpg (632KB, 709x1063px) Image search: [Google]

632KB, 709x1063px

>>60718794
By buggy code I also mean not getting the results you want, not just compile errors/seg faults/etc.
I presume you've never done any scientific computing?

>>60718823
An advanced type system helps in that regard. Scheme might not have one, but Idris certainly does.

File: boa-constrictor-closeup.ngsversion.1412351496655.jpg (509KB, 1600x900px) Image search: [Google]

509KB, 1600x900px

Which programming language should I teach my boa constrictor? Python? ASP? He's keen to learn.

File: 1486245487926.png (145KB, 500x500px) Image search: [Google]

145KB, 500x500px

Hey guys! I just wanted to chime in--

If haskell is a thousand times slower than D, how many times is it slower than C?

File: just_like_life.png (113KB, 369x245px) Image search: [Google]

113KB, 369x245px

>>60718844
*engages damage control mode*
The time saved developing in Idris cannot compare to the time saved by running heavy-lifting programs in Fortran.

>>60718868
D is dead, so that must mean Haskell is dead and buried.

>>60718876
You mean *rekt, dead and burried.

>>60718871
It might be faster overall (development + runtime) to first develop a program in Idris and verify its correctness, then translate it to FORTRAN.

File: 16387944.jpg (32KB, 552x446px) Image search: [Google]

32KB, 552x446px

What's more funny is the hasklel faggot himself started this in the previous thread.
How ironic.

>>60718889
If only it was that simple anon.

>>60718876
>Getting tucc'd by a dead language
That's a new low, anon!

File: 1445762626403.jpg (36KB, 300x360px) Image search: [Google]

36KB, 300x360px

>>60718901
STOP BULLYING HASKELL

>>60718876
>D is dead
https://dlang.org/blog/2017/05/31/project-highlight-excel-d/
Here's the latest example on why that's false.

>>60718926
Stop false-flagging, faggot.

File: Picture 3.png (821KB, 588x795px) Image search: [Google]

821KB, 588x795px

Which programming language should I learn?

Fuck haskell, I want to see how does D compare to C in performance.

File: ran_rape.png (156KB, 387x309px) Image search: [Google]

156KB, 387x309px

>>60719046
Fortran.

>>60718265
Programming language waifus unironically help me stay motivated.

>>60719053
D > Haskell > C > FORTRAN > SBCL > Julia > Rust

File: 1472704534317.png (19KB, 595x577px) Image search: [Google]

19KB, 595x577px

is there a way to extend a object[] in c#?

I need to add some functions that operation on the array but be able to load values into the object as if it was a object[] from functions.

>>60719046
python

File: yuuka_unamused.png (275KB, 568x310px) Image search: [Google]

275KB, 568x310px

>>60719084
Push_back??

File: dead esoteric languages.png (56KB, 705x556px) Image search: [Google]

56KB, 705x556px

>>60719025
>https://dlang.org/blog/1997/05/31/project-highlight-excel-d/
Wow, D's been dead this long?

>>60719053

Any C fag here? Can you fix this? I just want to find out the remainder of 10 of the computed fibonacci number

#include <stdlib.h>
#include <gmp.h>

int main(int argc, char **argv)
{
    mpz_t buf[2];

    mpz_init(buf[0]);
    mpz_init(buf[1]);
    mpz_set_ui(buf[0], 1);

    if(argc != 2) return 1;

    for(int i = 0; i < atoi(argv[1]); i++)
    {
        mpz_add(buf[0], buf[0], buf[1]);
        mpz_swap(buf[0], buf[1]);
    }

    printf("%Zd\n", buf[1]%10);    //<- here

    return 0;
}

File: dlang-chan is cute as fug.png (12KB, 693x106px) Image search: [Google]

12KB, 693x106px

>>60719107
>implying D isn't /g/ as fuck

>>60719100
the function I am dealing with always returns an object[] { (int)id, "name", Object("object") }

I have multiple classes that have similar object array values and I am not sure the best way to avoid reusing code. I'd rather define my own class and have it be able to be case as a object array.

>>60719165
>have a conference to have a meeting to convene to decide to go ahead with the plan to make a plan to plan the new planning of the D logo

>>60719185
>>60719162
Let me do that properly:

printf("%lu\n", mpz_mod_ui(buf[1], 10));

File: smug_ran6.jpg (114KB, 802x1240px) Image search: [Google]

114KB, 802x1240px

>>60719174
In Mathematica this is just

object = Append[object,x];

File: Screenshot from 2017-06-03 13-52-39.png (137KB, 1324x695px) Image search: [Google]

137KB, 1324x695px

>>60719195

>>60719214
and adding ; still doesn't work

>>60719214
Oh I didn't read the documentation properly. There was another argument.
Just read it yourself: https://gmplib.org/manual/Integer-Division.html

Someone post the books picture

File: js.jpg (792KB, 3264x1836px) Image search: [Google]

792KB, 3264x1836px

>>60719231

File: Screenshot from 2017-06-03 14-03-49.png (77KB, 1420x665px) Image search: [Google]

77KB, 1420x665px

>>60719214
At any rate. Here is D taking 1/5th of a second in my machine

>>60719327
Try this

#include <stdint.h>
#include <stdlib.h>

int64_t fib(uint8_t n)
{
    if (n <= 1) return n;
    if (n > 92) return -1;
    n -= 2;

    uint64_t a = 1, b = 1, c = 0, d = 1, e = 1, f = 0;

    while (n) {
        uint64_t g, h;
        if (n & 1) {
            g = a * d + b * e;
            h = b * d + c * e;
            f = b * e + c * f;
            d = g;
            e = h;
        }
        g = a * b + b * c;
        a = a * a + b * b;
        c = b * b + c * c;
        b = g;
        n >>= 1;
    }

    return d;
}

>>60719344
Wtf is this?

>>60719344
that's not the same algorithm but let me try

>>60719363
A function to calculate the nth Fibonacci number (as long as it fits into a 64-bit signed integer).

What language is the most fun for just doing random projects that pop in your head? Probably just text based simulations and games, but 2D vector graphics would be fun. I have experience with Java and C++ already.

>>60719245
Not that one you cuck.

>>60719344
overflows, fag

#include <stdint.h>

#include <stdlib.h>
int64_t fib(uint8_t n)
{
    if (n <= 1) return n;
    if (n > 92) return -1;
    n -= 2;

    uint64_t a = 1, b = 1, c = 0, d = 1, e = 1, f = 0;

    while (n) {
        uint64_t g, h;
        if (n & 1) {
            g = a * d + b * e;
            h = b * d + c * e;
            f = b * e + c * f;
            d = g;
            e = h;
        }
        g = a * b + b * c;
        a = a * a + b * b;
        c = b * b + c * c;
        b = g;
        n >>= 1;
    }

    return d;
}

int main(int argc, char **argv)
{
    uint8_t  count;
    scanf("%d", &count);
    printf("%lu\n", fib(count));
    return 0;
}

Fix this

>>60719421
uint8_t can't take 100,000

>>60719409
The one with the edited book covers, including "Oh look a horrid little money"? I don't have it, but if I did, I probably wouldn't post it because you sound like a stupid /pol/ poster

File: 1495663187091.jpg (84KB, 600x787px) Image search: [Google]

84KB, 600x787px

>>60719428

>>60719428
No I mean the one listing all the good programming books.

>>60719428
>stupid /pol/ poster
That's redundant, anon. Just like "dumb frogposter" is.

>>60719374
n := 0..2^64 ?

>>60719421
>overflows
What do you expect? It's C

>>60719469
the dumb is for emphasis and has a spirtual dimension too it

>>60719506
Take a look at the code and tell me if you think that's a correct interpretation, anon.

>>60718265
>Anime
Go to hell.

>>60719564
Mine can do 2^16 in in about 0.2 seconds.

>>60719585
Say that to my face faggot.

File: textonglass_8.png (251KB, 401x368px) Image search: [Google]

251KB, 401x368px

Is it possible to create aero blurred transparent using GTK or Qt (without relying on windows API)?

>>60719547
Well shit, I haven't thought about it that way.

>>60719793
by calling the frogposter dumb, even though everyone already knows he is dumb, you literally become better as a person

>>60718318
Honestly you should probably put most of that loop crap in a loop body

>>60719806
So what you're saying is that I not only have a moral obligation to call frogposters dumb but also an obligation to myself?

Is lisp dead

>>60719879
And not just an obligation morally, and not just an obligation to yourself, but an obligation to THEM, so that they can learn the error of their ways

>>60719895
Depends on whether you consider Clojure a lisp

>>60719900
>so that they can learn the error of their ways
But is that even possible? These are the people that would (and have!) wear shirts with the frog on them in public. They have no shame, and being shunned by other anons doesn't seem like it would do much. Hell, most anons seem to tolerate with dumb frogposters.

>>60719846
Expand on this, I don't understand what you mean.

File: 1347345865532.jpg (259KB, 1275x1275px) Image search: [Google]

259KB, 1275x1275px

>compiling and debugging code with Visual Studio Code
Holy fucking nutballs, is this what fossturds go through every time they want to run a program they wrote? I would literally just kill myself instead of putting myself through this for every single program.

>>60719972
because one day they will look back at themselves and see the frog and it will inspire in them a hatred for everything they were

>>60720019

I do all of my compiling and debugging within a terminal. It's really pretty simple.

Is it a 32bit datatype if it can be stored in 32bits but needs to be loaded into a larger register to be properly used?

>>60720026
Are you a former dumb frogposter, anon? You seem pretty insightful about this.

File: frog legs.jpg (18KB, 250x250px) Image search: [Google]

18KB, 250x250px

>>60720077
No, thankfully, though I did save some images that happen to have frogs in them, most of which I have deleted except for the ones which are either anti-frog or unrelated to the reddit frog

File: 1486855146158.jpg (193KB, 796x1280px) Image search: [Google]

193KB, 796x1280px

Anyone know of a good tutorial to templates for C++? I get the idea of using them to compute average of interest and floats with one function, but beyond that the syntax and explanations get more and more obtuse.

>>60720035
>It's really pretty simple
I can't tell if you're just lying to push some sort of agenda or only pretending to be retarded for your own comedic effect, but how is jumping through 5 hoops simpler than pressing ctrl+f5 in VS?

>>60720122
You shouldn't be using templates, let alone C++ at all.

>>60720133
Why

>>60720103
I feel you, anon. I think we all downloaded some frogshit at one time. It was once a quaint fad overtaken by normalfags and /pol/. I hope /r9k/ goes to hell for popularizing this shit.

>>60720125
Not him but "jumping through 5 hoops"? How about:

make all

Or with Visual Studio's command line utilities:

nmake all

>>60720133
I'm a C programmer but one of the useful things I can see C++ has are templates. Debugging large macros is a pain, I would expect templates would make it an easier process (and not the least because of the syntax.)

>>60720155
You don't even need to type "all" with make.

File: Ötaku.Chiqué1998.jpg (128KB, 600x600px) Image search: [Google]

128KB, 600x600px

There is females pretending to be guys ITT right now.

(By not explicitly stating you're a girl you're pretending to be a guy)

File: Screenshot from 2017-06-03 15-33-59.png (36KB, 766x281px) Image search: [Google]

36KB, 766x281px

C++
Will it ever get modules?

>>60720169
If I recall correctly, if you don't give a target to nmake, it will pick everything, but an "all" target doesn't necessarily mean "every target described in the file" because that could mean targets such as "clean", too. But yes, if "all" is your only target.
GNUmake on the other hands picks the first target if no target is given.

>>60720195
>That image
AHAHAHHAHAHHAHAHAHHAHAHHA
No.

File: 1453832145764.png (686KB, 836x720px) Image search: [Google]

686KB, 836x720px

>>60720185
Is this a surprise to anybody? I thought we all tacitly agreed that we were all pretending to males.

File: Screenshot from 2017-06-03 15-38-07.png (52KB, 769x377px) Image search: [Google]

52KB, 769x377px

>>60720203

>>60720125

>5 hoops
I just type make and hit enter.
Then I type the name of the program and hit enter.
If I'm doing these very often, I'll just hit up to go to the last command and hit enter.

>>60720185
shh you'll ruin my cover

File: yukari_face.jpg (179KB, 637x900px) Image search: [Google]

179KB, 637x900px

>>60720195
It will be in C++69.

>>60720228
Hi Ruby

I was trying out C++ because I want to use Qt creator. Why does C++ feel so ancient anf shitty? It doesn't even have modules! eww
And the ::<T>?
EWW

How do I get used to it?

>>60720185
I'm a girl (male)

>>60720256
That's not nearly enough nesting; I don't believe you.

>>60718510
array.each_with_index do |v, i| { do stuff in the block }
not
each.with_index

>>60720254
C++ is garbage. Use something else.

>>60720195
Define module

>>60720265
girl (male (boy (attack helicopter (fritata))))

>>60720265
I'm a girl (male (female (transexual (demiqueer (male (female (male (male (bisexual (mentally ill))))))))))

>>60720271
http://ddili.org/ders/d.en/modules.html

>>60720223
> You can write so much better C++11 than you could write in the styles of the 1990s.
Just create a better C++ based of the so called C++unicornfrom ground up then, fucking dipshit

>>60720276
>>60720277
have trannies made their own lisp dialect?

>>60720301
No, but c+= was a thing for a minute.

File: wrong.png (251KB, 1603x912px) Image search: [Google]

251KB, 1603x912px

https://www.youtube.com/watch?v=hoKJAAFmfZc
better than mikmod

>>60720332
Why didn't they just host that shit on another git platform?

currently implementing a coring system into my simple math game.

used this to do it:

(I == result ? ri++: ri--);

where I = user input of an integer, result is the actual value of the number they are guessing, ri = number of right answers.

It gives me this error:

Multiple markers at this line
    - The left-hand side of an assignment must be a variable
    - Syntax error, insert "AssignmentOperator Expression" to complete 
     Expression

when i run it gives me this error:

Exception in thread "main" java.lang.Error: Unresolved compilation problems: 
    The left-hand side of an assignment must be a variable
    Syntax error, insert "AssignmentOperator Expression" to complete Expression

    at Math.Math_Master.main(Math_Master.java:24)

File: language triangleFIXED.png (17KB, 522x384px) Image search: [Google]

17KB, 522x384px

>>60718491

>>60718265
I have GUI programs written in C++ with QtCreator. They need a shitload of DLLs to work on other computers though. Some of the DLLs are 100MB or greater.

I just want to be able to include everything in one .exe, without it being massive.

How do I build the necessary DLLs into the executable?

Is there a better GUI framework I can use for this?

>>60720276
>>60720277
I'm going to write a java script to see the maximum version of this I can get while staying inside character limit. It's 2000 right?

>>60720455
>How do I build the necessary DLLs into the executable?
You have to build qt from source and statically link everything.

But the better option is to stop being a retard and stop using qt altogether.
You actually using minuscule percent of Qt anyway.

>Is there a better GUI framework I can use for this?
Why you are using C++ in the first place, if you need so many layers of abstraction?
I think you've got the wrong door, the java club's two blocks down.

>>60720586
Not him but what's better than Qt?

>>60720653
Anything that's not GTK.

>>60720455
What kind of a program are you building? If it's a single window program, I would recommend trying something like dear imgui - imgui for example doesn't have external dependencies other than for stdlib and you can link it statically. Though, the UI system doesn't use your desktop environment's / system's provided framework.

>>60720691
What if I don't suck pajeetella's cock dry every wednesday?

>>60718318

 
(loop for x across y do
   loop-body-code) 

>>60720784
Meant this to respond to >>60720001

>>60720743
/wdg/

>>60720795
o, thank u

>>60720653
Qt is not necessary bad.
Everything depends on the task.

Any framework is just a layer of abstraction.
If you are not a cretin, you should learn how to use windows.h/xlib directly first.
It would be a good learning experience.
After that, with knowledge you gain, you will be able to make reasonable decision when you should use framework or work with API directly for a given task.

If you are a cretin and don't want to learn/think, Java club is two blocks down. C++ is not for you.

>>60720267
Yeah, I figured it out. .with_index expects you to return an enumerator object, so I just went

def each(&block)
  @array.each &block
end

for it to work with or without .with_index. As a note, is there a huge performance difference between the two? I would think each_with_index would be faster because it doesn't have to call two methods.

>>60718265
>Anime
Hi.
Go to hell.

File: fragezeichen yui.png (94KB, 396x395px) Image search: [Google]

94KB, 396x395px

What's the deal with numale langs like Rust and ES6 Javascript cherry-picking a whole bunch of functional features to get people excited, but then not guaranteeing TCO?

>>60721184
dumb frogposter

>>60720653
There's wxWidgets.

My own observations from some GUI libraries (your own experience might vary):

Qt:
+ works faily well on all platforms and often even looks mostly native
- huge size, sometimes slow due to excessive OOP, can hog memory
- compilation issues (moc and all that crap that shouldn't be needed)

GTK:
+ significantly smaller than Qt
+ easier to compile, uses less memory, somewhat faster
- looks and behaves like crap on anything that isn't Linux (GTK programs in Windows, for example, suffer very badly from "death by a thousand cuts" type problems)
- being a C library, very clunky to use (there are C++ wrappers, but they don't help much)

wxWidgets
+/- larger than GTK, smaller than Qt
+ in Windows, just a wrapper for the native controls
+ in Linux, just a wrapper for GTK
- somewhat clunky in design, but less than GTK

Both Qt and wxWidgets suffer from the "not just a GUI library, but a platform that takes over all of your code" problem. Partly because they were origianlly designed when C++ didn't have proper container libraries (std::vector, list, map, even std::string), so they had to have their own. But they can't let go of them now, so they take over everything you have.

>>60718889
>in the total subset Idris
Fixed that for you

>>60720879
>C++ is not for you.
What is C++ good for again?

>>60721455
are you suggesting it would be faster to develop solely in FORTRAN than in non-total idris -> FORTRAN?

>>60721347
What are "Functional features"?
Also, does Yui have autism?

>>60721521
I'm suggesting that non-total languages are for plebeians.
It would probably be faster to develop in a non-total language if you don't care about code correctness, but I hope such people aren't in this thread.

What's the best python book?

>>60721398
Did you actually build/used wxWidgets or you just copypasted that shit from stackexchange or something?
It's a fucking mess. Whole build pipeline is macros into macros into macros.
How can you actually recommend that bloated garbage?

>>60721509
C++ is one of the languages that actually have 0 use cases

>>60721347

Guaranteeing an optimization may negatively impact portability to obscure platforms that behave differently. Knowing how LLVM works, you should nonetheless feel free to use and abuse tail recursion in Rust, as it's going to be optimized regardless, unless the platform is super weird and tail call optimization is impossible.

>>60721565

AAA video games and game engines.

>>60721597
Just use Rust senpai.

>>60721630

Name a good Rust game engine.

>>60718867
C++, so he becomes a boa::constructor.

>>60721597
>Circular logic
You're pathetic

>>60721699
>replying to a tr*pfag who thinks he's a female horse
You're pathetic

>>60721730
He's a girl pretending to be a dude

Guys, would you suggest doing the MIT course of SICP or use this one here?
composingprograms.com

The C++ Programming Language (4th Edition) is shilling windows. What's the alternative to that book that doesn't shill windows and visual studio?

>>60721641
http://arewegameyet.com/

Is there any way I can connect Java code with Outlook template?

>>60721746
>he

>>60721597
Made better with C

>>60721528
immutable variables, maps/folds, first class functions, anonymous functions, pattern matching, type signatures

not all of them are inherently "functional," but almost all functional languages make heavy use of them, and imperative languages of the past generally did not have them implemented in the standard library

>>60721595
really? that's encouraging
It's especially frustrating for javashit though, because TCO is specified for ES6 to the best of my knowledge, but no browser's engine will honor it yet (and it's looking like they never will)

>>60721564
It's slightly less garbage than GTK and the topic was "better than Qt". Whether it is better than Qt, I'll leave that to everyone's own judgment.

Stack exchange/overflow didn't exist yet when I first learned wxWindows (as it was called back then).

>>60721912
https://www.edx.org/course/how-code-simple-data-ubcx-htc1x
and
https://www.edx.org/course/how-code-complex-data-ubcx-htc2x

>>60721995
>immutable variables, maps/folds, first class functions, anonymous functions, pattern matching, type signatures
So basically almost every language is "Functional"? I see. So it's a word with no inherent meaning.

>>60721912
Just do the SICP book.

>>60721912
Run as far away from Python tutorials as you can
It's fine if you already know computer science and want a hassle-free kitchen sink language, but it is terrible to learn on. You will develop so many bad habits.

Scheme is literally meant for learning computer science. Stick with SICP.

"I built C++ on C because I did not want to build from scratch"
Cause we stop pretending C++ is not related to C?

>>60722018
>we
Speak for yourself. I don't want to be associated with either shitlang.

>>60722008
In the past decade lots of major languages have adopted features pioneered by functional languages yes. I thought I inferred that in the previous post, but it looks like you're autistic/itching for some pointless argument.

>>60722018
>related
Yes, C++ is C's retarded cousin

>>60722035
But what are these "Functional languages"? You didn't even give a definition.
Is C "Functional"? Is sepples"Functional"? Is SML "Functional"? Is Lisp "Functional"?
What do you need to know to give me an answer for an arbitrary language?

>>60722061
A functional language is one designed to foster a functional style of programming where mutable state is avoided and referential transparency is guaranteed. The features I mentioned earlier make doing so painless for the programmer.

C is imperative.
Sepples is object-oriented with features cribbed from functional languages.
SML is functional.
Lisp is multi-paradigm but easy to write functional code in.

This isn't rocket surgery.

File: WIN_20170603_02_16_37_Pro.jpg (129KB, 1280x720px) Image search: [Google]

129KB, 1280x720px

so did you guys know its possible to use the bash on the windows console to use most of the regular bash programs?

I'm not trying to start a little thing but this means I have no need to open those boxes. This laptop so far can play the one game I play, I have consoles, and can also do my coding in a style I"m accustomed to.

That baby crying sun thing never meant something before.

Also, this laptop has touchscreen...


oh and the fallout shelter idle music effect is great.

And I have 300mbps internet now and its awesome. Because it's fast. I don't even need to code. It all happens in the wire between destinations. For every available shift in the current, we get to put an entire list thing. And that's all after.

And don't you forget it!

>>60722061
>But what are these "Functional languages"?
Use google

>>60722061
Learn Haskell

>>60722179
No. It is 1000 times slower than other languages

>>60722117
>referential transparency is guaranteed
So neither SML nor Lisp are "Functional"?
>Sepples is object-oriented
So being a poolang is mutually exclusive with being "Functional"?
>The features I mentioned earlier make doing so painless for the programmer.
Why did you even mention them if they aren't what makes a language "Functional"?
>This isn't rocket surgery.
Then why did you contradict your own definition?

>>60722169
I don't use government-run websites.

>>60722179
What makes you think I don't know "Haskell"?

>>60722184
except its not.

>>60722184
calm down sonic

>>60722188
Having a bad day?

javascript

>>60722216
State your purpose.

>>60722188
Drink some H2SO4 please.

>>60722252
Why would I be trying to actively harm myself?

>>60722188
>So neither SML nor Lisp are "Functional"?
Referential transparency is guaranteed if you write referentially transparent code. SML and Lisp are easier to do that than in C, so they're known as functional languages. A functional style doesn't have to be enforced; it just has to be immediately accessible.

>So being a poolang is mutually exclusive with being "Functional"?
Not necessarily, but the conventions of OOP generally don't promote referential transparency.

>Why did you even mention them if they aren't what makes a language "Functional"?
They are. If the language was designed to make it easy to write functional code, it's functional.

>Then why did you contradict your own definition?
I didn't. You have middle school tier reading comprehension.

File: 5.jpg (10KB, 300x100px) Image search: [Google]

10KB, 300x100px

Working on my imageboard, 4kev.org

Writing Linux Device Drivers by Jerry Cooperstein is such a bad book.

>>60722231
I was just wondering why you were so aggressive in that post.

>he doesn't have a skill graph on his resume

how do you evaluate your experience into a graph like this guy? https://jeaye.com/
it's like saying: "i am a 8/10 in c++ but a 6/10 in opengl" how do you rate yourself on 10?

>>60719162
>>60719195
wrong, read the manual on formatting strings with gmp

I started learing C++ yesterday. I know I know too little to come to conclusions but it feels awful lot like Java. But why does /g/ hate Java so much, anyway? Care to elaborate?

>>60722262
Because your continued existence is a waste of oxygen.

>>60722264
>Referential transparency is guaranteed if you write referentially transparent code
I see, so C is "Functional".
>SML and Lisp are easier to do that than in C
"easier" isn't a valid word in this context without additional information unless you can somehow know what's "easier" for every single programmer.
>A functional style doesn't have to be enforced
So it's a meaningless word?
>it just has to be immediately accessible
I see, so every non-trivial language is "Functional"?

>Not necessarily
So no? Why did you mention it then?
>If the language was designed to make it easy to write functional code, it's functional.
How can you possibly know the intent of the designers? And this contradicts your previous statement about referential transparency.

>I didn't.
You claimed - "A functional language is one [in which] referential transparency is guaranteed"
Which blatantly and obviously makes "SML" into a "Non-Functional" language.

>>60722298
Oh well in that case...

>>60722264
>functional code
Oh, and also. What is this "Functional code"?

>>60722321
"[a language] designed to foster a functional style of programming where mutable state is avoided and referential transparency is guaranteed" and "a language where referential transparency is guaranteed" are not the same thing.

If you're going to be autistic as fuck about something, at least learn English first.

>>60722349
Referentially transparent code.

>>60722321
not that anon but

c is not a functional programming language, the language doesn't have any notion of what a function is. functional programming is when your program with functions and a functional programming language is a language made to specifically help you in doing that. the only pl that is legitimately a fpl is haskell.

that's doesn't mean you can't do functional programming with a PL that is not function oriented but it would be tedious and limited.

>>60722369
How can you "foster a style where referential transparency is guaranteed"?
How can you "guarantee" something without enforcing it? That doesn't make any sense whatsoever.

>>60722376
I see.

>>60722383
>functional programming is when your program with functions
So any referentially transparent language?
>the only pl that is legitimately a fpl is haskell
This seems like a retarded claim to make when we're talking about a potentially infinite list of languages.

>>60722321
>"easier" isn't a valid word in this context without additional information unless you can somehow know what's "easier" for every single programmer.
Easier in the sense that control procedures and interpreter/compiler behavior that ensure performant functional code are immediately available in ML/Lisp, and not in C.

It seems your argument is that since anything can be computed, everything is actually everything. In which case congratulations on the earth-shattering revelation.

>How can you possibly know the intent of the designers?
Really easily. It's clear that ML/Scheme/Haskell were designed with a functional style of programming in mind because all the formal specifications say as much.

>>60722405
The programmer guarantees his code is referentially transparent by writing referentially transparent code. The language doesn't guarantee it.

The language fosters the programmer's guaranteeing by providing him immediate access to control structures/compiler behavior that allow him to guarantee his code is referentially transparent.

>>60721641
What programming languages do you know and what are you favorite programming languages and why?

>>60722191
It is. See the previous thread. Counting the 1000th fibonacci number takes more time in Haskell than D needs to count the 1000,000th number.

So Haskell is 1000 times slower than D

>>60722446
>It seems your argument is that since anything can be computed, everything is actually everything.
No. I just need a convincing argument of why "x language is functional" is decidable and a decision algorithm for it. Or alternatively a list of all "Functional" languages.

>Really easily.
Do you have mind reading capabilities?
>all the formal specifications say as much
Oh, so there is a formal definition of what it means to be "Functional"? Link?

>>60722470
>The language doesn't guarantee it.
So the word is basically meaningless?
>that allow him to "guarantee" his code is referentially transparent
In what way?

File: 2017-06-03-055743_510x294_scrot.png (10KB, 510x294px) Image search: [Google]

10KB, 510x294px

Finally got my shit to work.
Is there a shorter and more efficient solution?

C++11 extends the syntax of the for statement to allow for easy iteration over a range of elements:

int my_array[5] = {1, 2, 3, 4, 5};
// double the value of each element in my_array:
for (auto& x : my_array) {
    x *= 2;
}

My question is, how do I get the index?

>>60722492
I don't think you're going to get the answer you seek.

"Functional" is like the "blue." There's no hard and fast rule for what qualifies as blue. In fact Japanese people say that traffic lights and apples are blue. But even then, most people know blue when they see it.

How do I tell whether a char is a tab or a space in my vim?

>>60722543
What programming language is this?

>>60722499
God Lisp is so much uglier than Scheme.

>>60722535
>I don't think you're going to get the answer you seek.
I'm working on a definition myself and I think I'm almost there.
>In fact Japanese people say that traffic lights and apples are blue
Because "blue" in Japanese translates to two different words in English.

>>60722520
You can't, C++ is a garbo.

auto my_array = [1, 2, 3, 4, 5];
// double the value of each element in my_array:
foreach (index, ref x;  my_array) {
    x *= 2;
}

This is what a well though-out language created by actual compiler engineers looks like

>>60722558
What's your definition?

>>60722557
>(define (dick penis))
>crippled macros
>caring about surface details
>>>/plebbit/

>>60722499
wtf so you took a whole day for writing a program that extracts a quoted substring?

>>60722585
I was watching Breaking Bad lol.

>>60722590
>2018
>Still hasn't finished walter bright
Get with the times my man

>>60722573
Many Schemes have classic lisp macros on top of (syntax-rules)

>>60722607
*defines his own standard library because schemes have like 3 functions that come by default

>>60722520
C++ doesn't have real foreach loops

>>60722624
>le 1,000+ page spec face

>>60722520
There's no "enumerate" method if you're leading to it, you have to iterate classical style with an index variable.
>>60722559
>Ad-hoc syntax and conventions just to iterate with index
Yeah, this was definitely created by a compiler engineer and not by a language designer. Actual good design needs only one additional method and builds on existing concepts like tuples and iterators:

let mut a = [10, 20, 30];
for (i, v) in a.iter_mut().enumerate() {
    *v *= 2;
    println!("{} {}", i, v);
}

>>60722571
I don't have one yet.

>>60722640
>what is a keybinding in your vim to open the hypersec

>>60722661
>what is a 40MB binary

File: >.jpg (55KB, 258x360px) Image search: [Google]

55KB, 258x360px

>>60722640
>70 page spec

>>60722668
What is ecl.

>>60718265
I have a struct Ptype

struct Ptype{
  int l;
};

What is the difference betweeen this

Ptype p1 = 0;

and this

Ptype p1 = {0}

>>60722649
>additional method overhead just to get index
of course it isn't written by compiler engineers

File: 1496425515646.jpg (200KB, 1127x1079px) Image search: [Google]

200KB, 1127x1079px

>>60722669
Why are you making frog noises?

>>60722266
he's a good bloke

>>60722266
Good chan for dead dropping cp.
Thanks senpai.

>>60722676
>Ptype p1 = {0}
shouldn't compile, what shitty language allows that?

>>60722682
I'm just memeing senpai.
I don't actually hate Scheme.
I need to use it for SICP anyways.

>>60722676
The first is initialising it with just an integer, which is wrong.
The second is initialising the fields of the struct in order.

Which language should i learn first?

I just want to learn thing

>>60722745
D

>>60722745
Scheme.

>>60722679
>he thinks the method is actually called
So you're not close to being even a compiler engineer.

>>60722729
what if I had other members after int l and I ONLY included 0;

struct Ptype {
int l;
char b;
char* a;
char** argv;
char*** somecmd;
};

Ptype p1 = {0};

>>60722768
>Just because I wrote method() doesn't mean I called it

>>60722745
dont listen to >>60722767
D is obsolete and disgusting.
Also made by Google, so it has adware.

>>60722703
It coompiles

>>60722775
>D
>Also made by Google
>language has an adware.
Are you clinically retarded?

>>60718265
I want to cum inside shell, Scala, Perl. and ActionScript [spoiler] over and over [/spoiler]

>>60722773
All unmentioned elements are initialised to zero.
So if you want to zero-initialise a struct, it's a pretty common thing to just use {0}.

>>60722775
>language
>adware
So this is what /g/ looks like when it's flooded with /v/ cross boarders

>>60722775
Fuck off idiot
>>>/b/

>>60722774
Where have you been in the last 20 years, m8? There is this fancy new thing called "inlining", you might like it, see https://is . gd/RJVmSP .

>>60722789
oh thanks

>>60719046
none, quit while you're ahead.

I feel retarded, gcc compile errors are just impossible to read and make anything useful out of them.

>>60722856
Later versions of GCC really stepped up their games in terms of error messages.

>>60722856
Try clang.

>>60722908
Writing Linux drivers for school assignment so can't.

>>60718265
Isn't every /dpt/ anime edition?

File: rust.png (868KB, 3840x2160px) Image search: [Google]

868KB, 3840x2160px

>>60719327
Here's Rust taking ~0,01 seconds on my laptop

File: Screenshot from 2017-06-03 12-58-18.png (865KB, 3840x2160px) Image search: [Google]

865KB, 3840x2160px

>>60722980
Fuck I can't read here's the same program taking 0,1 seconds with the correct argument

File: letherman.jpg (34KB, 617x462px) Image search: [Google]

34KB, 617x462px

>>60720586
>I think you've got the wrong door, the java club's two blocks down.

Fuck you

>>60722145
I used to do that too. The issues (can't even use ifconfig for example) started to bother me too much though and I switched to a Windows Linux dual boot. Ubuntu works really well with a touchscreen btw (source: am using Ubuntu on Dell XPS 15 with GTX960M and 4k touchscreen atm)

How many different ways are there to initialize a stack object in sepples?

Just uploaded my vr apk to the Google Play Store

https://play.google.com/store/apps/details?id=com.Nexpedia.VrZombieBunker

>>60723148
not enough
t. Bjarne

File: fragzeichenmadchen.jpg (111KB, 500x500px) Image search: [Google]

111KB, 500x500px

For hash tables what is more important: keeping the size of the table small or avoiding collisions?

If my table has to be twice as large as the maximum amount of keys to ensure no collisions, does that mean the hashing algorithm is bad?

Of course algorithms that eliminate collisions can be more expensive.

What's the best compromise?

>>60718289
Lel you're probably doing something very wrong with laziness, optimized haskell is usually within one order of magnitude of C speed, especially with something so trivial as fibonacci.

>>60723381
There is no "best" general compromise. Also those aren't your only concerns - access patterns and the CPU cache are another one.
For example, I had a project that had a lot of small, static hashtables, so the only collision handling I needed was assertions - crash the program if a collision happens. If a collision happened during debugging, we just changed the source of the hash.
So yeah, it all depends on your use-case.

>>60723381
Hashmaps are pleb-tier. Tries are the patrician data structure.

Why are bad programmers bad? Why are there so many of them? Is it laziness? Stupidity? Lack of community? Bad planning skills?

>>60723547
the ideal programmer is a maths major sitting in front of a computer being paid 400K starting to do what programmers progdont

the truth is most """programmers""" are better classified as computer niggers. a computer nigger throws problems at hardware like a simian throws shit against the wall

if I had a company I would weed out computer niggers by forcing junior devs to work on hardware from 20 years ago. if they can't get shit to run fast enough: it's proof of their melanin and they're fired

New thread:
>>60723587
>>60723587
>>60723587

>>60723578
>if I had a company I would weed out computer niggers by forcing junior devs to work from 20 years ago.
Just forcing them write everything in assembly will be enough.
It's pretty hard to be a monkey with assembly.

>>60723619
Assembly is literally tied to computer architecture though. If your code is more performant because it's less abstract, that doesn't necessarily mean you have good algorithms. Bit diddlers can also be hardcore computer niggers.

We need people who can reason about algorithms divorced from computers. People who aren't niggers.

>>60723578
I can see where you're coming from but if that was the standard nothing would get done.

The problem is when computer niggers can't even do things like write decent tests or documentation or split up their code into tiny bits or make things reusable. None of these things are hard and they keep the software world moving by allowing morons to write code, yet even that's too much to ask now.

>>60723645
People are not actually that stupid.
The biggest problem is that they are using layers of abstraction without understanding them.
You cannot reason about system, if don't understand it.
This is the reason why we have webpages that use 3gb of ram nowadays.

>>60723749
>reason
stopped reading right there

Writing everything in assembly would be too much actually.
Just teaching new programmers assembly would dramatically reduce the problem.

>>60723760
Why?

Let me show you a real fibonacci calculator:

require "./matrix.cr"
require "big_int"
require "benchmark"


def fib(n : BigInt)
  return 0 if n == 0
  return 1 if n <= 2

  number = Matrix.rows([[BigInt.new(1), BigInt.new(1)], [BigInt.new(1), BigInt.new(0)]])
  final  = Matrix.rows([[BigInt.new(1), BigInt.new(1)], [BigInt.new(1), BigInt.new(0)]])

  while n > 0
    final = final * number if n % 2 == 1
    number = number * number
    n /= BigInt.new(2)
  end
  return final[1, 1]
end


firstArg = BigInt.new(ARGV[0])
puts Benchmark.measure {
  fib(firstArg)
}
number = fib(firstArg)
puts "\n#{firstArg}th fibonacci number is:\n#{number}"

matrix.cr form https://github.com/Exilor/matrix

Output for 100,000th number:

>>> ./fib 100000
  benchmark: 0.010000   0.000000   0.010000 (  0.003770)

100000th fibonacci number is:
2597406934722172416615503402127591541488048538651769658472477... and ends in 62349895374653428746875

File: download.jpg (42KB, 410x410px) Image search: [Google]

42KB, 410x410px

>>60718265
That image is not very inclusive, tbqh. Not including Rust in 2017 is frankly unacceptable. The author should redraw this to include Rust.

>>60724041
>should
Haskell, Elixir, Crystal, Fortran and Lisp dialects aren't there either.

>>60724079
But those are obscure or niche languages.

>>60724121
Neither or them are "obscure" or "niche".