/dpt/ - Daily Programming Thread

>>59446946
First for C(ute)

>off-topic image
No, thanks.

Real thread: >>59446947

File: Anime OP hits redditor's autism.png (4KB, 422x125px) Image search: [Google]

4KB, 422x125px

>>59447005
Welcome to the anime website.

>>59447029
What does each axis represent?

>>59447042
You decide

>>59447064
Burgers and soda

Wtf is up with the amount of shitposters lately?

>>59447165
spring break

>>59447165
They can't get enough people to make Rust popular

>>59447165
>implying /g/ isn't Shitposting:The Board

>>59447181
Dpt at least wasn't this unbearable.
Apart from the anime debates, it was relatively quiet...

>>59447243
Maybe theres a sort of brain drain between /g/ and Reddit, they get the posters who are worth conversing with, we get the edgy l33tists

>>59446946
Bulky computers like these are so sexy

>>59447270
I hope >>59447174 is right and they'll go away after a while, but with the recent Trump and "Meme magic" thing, it may actually be true...

>>59446946
>What are you working on, /g/?
Writing a bot for the Official /dpt/ Discord™

>>59447341
It was inevitable, all the edgy pricks who think they're hardcore cause they can reveal an autofilled password by changing the input box type would naturally flock here due to 4chans perceived reputation. This could then put off some valuable posters who'll try another forum for a change, and so the cycle perpetuates

File: 1489342687666.jpg (297KB, 572x800px) Image search: [Google]

297KB, 572x800px

>>59447181
We need /g/ for guro!

>>59447484
the fuck are you even talking about

>>59447404
I pretty much only talk about programming on Discords these days.

The mods tend to allow funposting, but don't tolerate the shitposts that spam the same phrase there.

>>59446946
Every programming language is just a bug-ridden Lisp in disguise.

sepples question.
I have a string
I need to split it in multiple strings
delimiter char can either be

',' ';' ':'

what's the nicest way to do this?

>>59447545
What does this even mean?

>>59447523
I find myself talking about programming more on various articles, tutorials and the like nowadays as opposed to a forum or board, i tend to get better discussion and valuable information that way, and like you mentioned, mods for these tend to keep a tighter grip of the threads, shitposting gets dealt with quickly

>>59447551
Sane options:
>do it yourself in maybe 10 lines function
>strtok
>istringstream + getline

Insane options:
>use boost tokenizer

>>59447555
Unfortunately, the mods on /g/ are more concerned with deleting something mildly racist and on-topic, rather than an autist spamming the same catchphrase about OOP/C/FP for 40+ posts in one thread.

>>59447554
Lisp is the greatest programming language humankind will ever discover. It can be reprogrammed to become any other language. People do this to turn Lisp into inferior programming languages, like Haskell and Malbolge.

>>59447608
Too bad it can't be reprogrammed to become a good language.

File: anal beads.png (6KB, 659x103px) Image search: [Google]

6KB, 659x103px

>>59447551
Doesn't sepples have an equivalent to this?

var inputArray = input.Split(new char[] { ',', ';', ':' });

>>59447581
Yeah, I don't mind reasonable discussion on any of those topics, I just can't be assed scrolling through 17 different flamewars to find a comment or a thread that looks interesting

>>59447593
Stop false flagging and go to >>>/a/

>>59447523
/dpt/'s discord, or other programming related ones?

>>59447608
Stop posting inane comment Richard.

>>59446946
Shit

>>59447631
Yes.

>>59447593
Akira, Ghost in the Shell

File: 1489269380843.png (166KB, 429x461px) Image search: [Google]

166KB, 429x461px

>>59447593
This is anime website, but not anime board, man.
Anime site means that everyone posts here anime pictures, but most of them are board related, like, on /g/ they post technology anime pictures.
Anyway, Serial Experiments Lain.

>>59447593
Serial Experiments Lain

>>59447593
I like golden boy and cowboy beboob

>>59447593
Neon Genesis Evangelion

File: problem455.png (22KB, 1162x401px) Image search: [Google]

22KB, 1162x401px

>>59447642
>>59447651
>>59447652
>>59447665
>>59447672

thanks my dudes

here's a programming problem in return

>>59447712
Looks more like a math problem than a programming problem.

Wouldn't you just solve this the "math" way and then write it in whatever programming language's syntax?

>>59447712
I am at problem 18 or something :^)
maybe should go on with it

>>59447712
Simplest way:
check all values

Real way:
i don't know

>>59447735
It's a project euler problem

they're normally hard to solve just using the math

>>59447770
>they're normally hard to solve just using the math
They are usually written to be solvable with brute-force, but not in your lifetime. You are supposed to use math to optimize the search. Some of them are actually very easy to solve with pen and paper, I've done it for a few of them.

>>59447770
I really don't understand the purpose of >>59447712

Why would more than two people in the entire world ever need a program that does this efficiently?

What's stopping one person from just showing the other person how to do it once one of the two figure it out?

At this point, it just feels like attempting to complete this is metaphoric autofellatio.

>>59447812
>it just feels like attempting to complete this is metaphoric autofellatio
Wow, are you finally grasping the concept of hobbies and the need to fill the meaningless void of existence with masturbation. Good on you anon.

>>59446946
I have been using Haskell for awhile now, and this is what I have found wrong with the language:
>horrid macro system
>modules are deficient compared to the module system of OCaml or the package system of Common Lisp
>partial functions pervade the language
>broken records (ugly hacks are needed to get around their deficiencies, leading to overengineered libraries like lens)
>most libraries are in alpha and look like a college student's summer project
>library authors tend to pepper their libraries with ugly, meaningless, custom operators, contributing to the overall ugliness of the ecosystem
>String as [Char]
>easily subverted type system with unsafe functions
>laziness makes it a chore even for experienced programmers to reason about algorithmic complexity
>in 30 years of existence, Haskell has yet to become a proven asset in industry

>>59447874
I guess I didn't really consider that some people could enjoy doing this, but I suppose there's a sense of achievement that comes with it. Fuck me; now that I program for a living, I don't want to do it for fun anymore.

Don't get a job programming, anons. It's a trap.

>>59447892
What do you propose better?

>>59447812
http://wiki.c2.com/?MentalMasturbation

>>59447896
>Don't get a job programming, anons. It's a trap.
I write automation software at an engineering firm. I have no problem coming home to do something like that, or work on my processor. So maybe it's more specific than that.

Of course I'm usually too tired after work to do much of anything anymore. Living for other people is a terrible bore, but it's important. Fuck.

>>59447896
That's what you get for turning your hobby into your job.

hey traps what's the best source for learning python for someone that's already into programming? (if the source is good it's ok if it's for total newbies)

>>59448040
docs.python.org

>>59447892
I have to agree with everything. Still, it's a good language worth learning just to broaden your horizon.

>>59448055
I think people should learn to reason about abstract machines more than languages.

>>59447712
I tried but it takes forever to run so I either did something wrong or the challenge wants you to waste time, either way here's a broken solution :^).

LIMIT = pow(10, 9)

def f(n):
    curr = 0
    while curr < LIMIT:
        chunk = pow(n, curr, LIMIT)
        if chunk == curr:
            return curr

        curr += 1

    return 0

def sum_fun(fun, a, b):
    return sum([fun(n) for n in range(a, b)])

print('f(4) =', f(4))
print('f(10) =', f(10))
print('f(157) =', f(157))
print('Σf(n), 2 ≤ n ≤ 10^3 =', sum_fun(f, 2, pow(10, 3)))
print('Σf(n), 2 ≤ n ≤ 10^6 =', sum_fun(f, 2, pow(10, 6)))

>>59446946
Traps are degeneracy, jump into an oven you subhuman.

>>59448164
there is no traps so far

>>59448128
Yes, but learning without practicing is meaningless and you practice this kind of reasoning by writing programs, so the language you use influence the way you're reason about things profoundly.

>>59448171
OPs image is fag shinji dressed like a girl.

File: 6OJc4SBQvNo.jpg (73KB, 582x728px) Image search: [Google]

73KB, 582x728px

>>59448197
That's where you are wrong.
THIS is Shinji

>>59448197
It's not.
She's Maya Ibuki.

>>59448212
Forth reich can't come soon enough. Traps and animefags need the rope.

File: 1489156283362.gif (132KB, 640x480px) Image search: [Google]

132KB, 640x480px

>>59448239
Japan was the ally of the Third Reich.
Anime is the art of ubermenschen.

File: 1470743205649.jpg (136KB, 960x1280px) Image search: [Google]

136KB, 960x1280px

>tfw no average k/g/f

File: 1395793945881.jpg (305KB, 1261x1000px) Image search: [Google]

305KB, 1261x1000px

look what i've found

>>59448278
roll

File: emacs_user.jpg (39KB, 500x375px) Image search: [Google]

39KB, 500x375px

>>59448267
>Emacs
Don't wanna raise your hopes up anon...

>>59448288
Enjoy your sine wave generator.

>>59448257
Until the degeneracy ends, the purging continues. Maybe japan needs another bomb dropped on them.

>>59448319
>reichfag
>american
/0

>>59448319
>japan gets dropped a bomb
>goes from good fun quality shit to the modern garbage
>retard thinks dropping another bomb will fix that
its a downward spiral not a fucking on off switch

>>59448278
rolling

>>59448278
rolling for terry davis

>>59448349
Then Japan has lost all hope. No longer are they proud warriors but degenerates who follow yid marxist brainwashing and become social diseases. More and more japs are letting their race decline.

>>59448278
give me the hello world

>>59448382
well that's Americas fault :))

>>59448278
rolling

What guides what should be put in the .h/.c file?

Are header file for the public interface?

>Visual Studio 2017 removed UML diagrams

They're finally learning.

>>59448407
Japs don't generally care about foreign opinions, they killed themselves from the inside.

>>59448436
>they killed themselves from the inside.
>all those american bases
>foreign ban of Japanese army

>>59448466
The jews ordered the bomb and the jews manipulated Japan. The jews are entirely responsible for Japan decline and now Japanese people are brainwashed enough to do it themselves as programmed.

>>59446946
I've fixed my fizzbuzz code in haskell. I forgot about

 fizz <-  divByThree num 

will return string or int. So in order to use

 (Right a, Right b) -> Right $ a ++ b 

I had to either use let instead of <- or put them in monad again

 case (return fizz, return buzz) of 

module Fizz where

type FizzBuzz = Either Integer String

divBy :: FizzBuzz -> Integer -> String -> FizzBuzz
divBy num by fizzString = case num of
                          Right str -> Right str
                          Left a -> if a `rem` by == 0
                                     then Right fizzString
                                     else Left a
divByThree :: FizzBuzz -> FizzBuzz
divByThree num = divBy num 3 "Fizz"

divByFive :: FizzBuzz -> FizzBuzz
divByFive num = divBy num 5 "Buzz"

divByFiveAndThree :: FizzBuzz -> FizzBuzz
divByFiveAndThree num = do
  let fizz = divByThree num
  let buzz = divByFive num
  case (fizz, buzz) of
   (Right a, Right b) -> Right $ a ++ b
   _ -> num

printFizz :: FizzBuzz -> IO ()
printFizz fizz = case fizz of
             Right str -> putStrLn str
             Left num -> putStrLn $ show num

main :: IO ()
main = do
   mapM_ (\x -> printFizz $ (divByThree . divByFive . divByFiveAndThree) (Left x))  [1..100]

how do i source good papers for a given topic?
where should i go looking if i dont want to pay out the arse for papers?

>>59448278
might as well

File: 1474114334189.jpg (145KB, 800x642px) Image search: [Google]

145KB, 800x642px

>>59447712
I feel like Euler's Theorem might prove useful to solve this problem. It's an intuition, but it's the only thing my brainlet can think of a possible solution.

>>59448481
>The jews ordered the bomb
Are you aware that the same Jews run your country?

>>59448543
The jewish problem will be over soon. They know their days are up and they're scared.

>>59448598
>soon
>7000 years and no solution
good luck

File: anal beads.png (149KB, 843x985px) Image search: [Google]

149KB, 843x985px

>Vim

>>59448521
libgen or sci-hub

Doesn't making a C++ destructor private prevent it from being called when the class goes out of scope? Somehow mine is called twice (even though I only call it once).

>>59448485

fizzBuzz = mapM_ (putStrLn . toFizz) [1..100] 
  where 
    toFizz a = case (a `mod` 3 == 0, a `mod` 5 == 0) of  
      (True, True) -> "FizzBuzz"
      (True, _)    -> "Fizz" 
      (_, True)    -> "Buzz" 
      _            -> show a

>>59448650
Don't call destructors yourself. They're supposed to be called automatically.
The only exception is when you're writing your own deleters.

Which language can i get a job with? like i don't want to waste my time, i already know some c++, which one will get me a job for sure?

>>59448705
COBOL

>>59448630
I got used to it and then I realized separating system clipboard and copy buffer in editor is a nice thing to have.

>>59448687
>The only exception is when you're writing your own deleters.
Which is exactly what I'm doing.

>>59448705
Honestly, Java for ENTREPRISE development or Javascript for web.

>>59448636
man, you are a fucking lifesaver, thank you

>>59448677
Just do (a `mod` 3, a `mod` 5) and you can pattern match on 0s

>>59448726
You jest but companies still have software they can't shut off written in this. Want to make 200k+/yr? Learn COBOL

>>59448851
I wasn't joking actually, because I know man. I KNOW!!! Make bank while staring at COBOL, slowly losing your sanity and wondering what the hell happened to your life.

>>59448809
I use Java in Android as a portal that can unleash the unlimited superpower of C and C++ through JNI.

>>59448419
Yeah, put typedefs, structs, function declarations, inline or macro functions, and configuration constants and defines in the header. The C files should contain the definitions of the extern functions declared in the header, and any private functions, typedefs, or variables used by those extern functions.

Header is interface, C file is implenetation.

Without batteries library so I have to implement range.

let rec list_range a b = 
    if a > b then
        []
    else
        a :: (list_range (a+1) b)
in let fizzbuzz n =
    match (n mod 3, n mod 5) with
      (0, 0) -> "FizzBuzz"
    | (0, _) -> "Fizz"
    | (_, 0) -> "Buzz"
    | (_, _) -> string_of_int n
in List.map fizzbuzz (list_range 1 100)

>>59449057
That's cool I've wanted to do something like that for a few weeks now.

>>59448804
then don't call delete/detroy in your deleter. use p->~T() then deallocate(p)

>>59449171

#.(loop for i from 1 to 100 do
        (if (not (or (zerop (mod i 3)) (zerop (mod i 5))))
            (princ i)
            (progn (when (zerop (mod i 3)) (princ "Fizz"))
                   (when (zerop (mod i 5)) (princ "Buzz"))))
        (terpri))

new retard at C here, is there some sort of way (in unix systems) that i can parse options separately from other arguments, i.e. instead of doing

if (!strcmp(argv[1], "-c")) {
  /*stuff*/
} else if (!strcmp(argv[1], "-f")) {
  /*stuff*/
} else {
  /*stuff*/
}

i could do something like

switch(option) {
  case 'c':
    /*stuff*/
    break;
  case 'f':
    /*stuff*/
    break;
}

>>59449236
strtok the argument on '-', then grab the character succeeding it.

/dpt/ threads are awful.

>>59449276
>daily programming threads threads are awful

>>59449276
Report /pol/ and ignore

>>59449284
>/daily programming thread/ threads are awful

>>59449236
>>59449258
Is there any difference in machine code between switch and IFs?

>>59449236
getopt for getting option
and optarg for getting its argumant

 while ((c = getopt (argc, argv, "abc:")) != -1)
    switch (c)

https://www.gnu.org/software/libc/manual/html_node/Example-of-Getopt.html#Example-of-

>>59449276
they'd be fine without all the creepy japanese pedo cartoon stuff…

>>59449297
http://stackoverflow.com/a/98024

>>59449297
If switch has few cases then it's usually going to be conditional branching (like if). If many then it uses a switch table. Switch tables can blow dicks performance-wise if the blocks are huge/very different in size.

File: 1468833221815.gif (80KB, 500x287px) Image search: [Google]

80KB, 500x287px

>>59449316
pedo website

>>59449320
>Switch tables can blow dicks performance-wise
no hablo ingles
Does it mean switch tables are better or worse?

>>59449423
Overall it's always going to be at least as good as a ton of if-else. However if you do something like one big switch table, your code will be very inefficient.

>>59449446
C doesn't optimize that to a search tree / hash table?

>>59446946
some stupid fucking CS homework. I don't know why I didn't choose a real career.

>>59449459
If you jsut started, get out while you can. It's a worthless degree. Go get something useful and learn how to program on the side.

>>59449297
Depends on if the compiler can infer certain constructs from the switch statement. If the values in the table are non-contiguous it will generally create a branch chain (just like if - else if - else), but if the values are contiguous it might generate a jump table, similar to putting a bunch of function pointer in an array and indexing it to call a particular function. However, if the number of cases are small then it will probably generate a branch chain because it would tend to be faster than a memory access and indirect jump, which could incur up to two cache misses. It's probably one of the reasons I hate switches, the code they generate is not deterministic. C should have kept computed gotos in C89.

>>59449451
It might, the problem is it gets harder to branch predict and put the correct instructions in icache the bigger the entries' content get.

>>59449451
Why the fuck would it do that? Those things are more expensive than a small set of branches or a table lookup. If you want more performant code, write it yourself.

File: IMG_3045.jpg (215KB, 902x789px) Image search: [Google]

215KB, 902x789px

How the everloving fuck do I learn about Neural Networks and deep learning processes without spending four years going to a university and winding up 100 grand in debt?
Are there any (good) resources lying around I can look into, or am I fucked?

>>59449524
>Why the fuck would it do that?
It should do that when it increases the performance of the switch block, like other sane languages that perform this optimization.

>>59449482
too late, i'm already 2 years in

nothing but suicide is waiting for me

>>59449561
Yeah it's called community college.

>>59449579
I'm in my local one already but the heaviest shit I've learned ot seen offered was how to make android studio not shit itself. The college here focuses more on hardware side of things.

>>59449572
>hash table
>more performant than a few branches
Nigga u high. Not to mention you might have to try several hashing functions to find one that perfectly maps everything. Or are you about to tell me a chained hash table is a good idea...

A search tree would be more performant, but only for HUGE switch statements. Otherwise branch chains or jump tables would be much more performant.

>>59449561
http://quickdocs.org/clml/

File: IMG_2688.jpg (119KB, 600x480px) Image search: [Google]

119KB, 600x480px

>>59449626
Cheers, I'll give it a read

File: sample_8a1635d189f5f13524fb22c1c7e68071.jpg (137KB, 850x1092px) Image search: [Google]

137KB, 850x1092px

Starting with HtDP.

File: 1485397730268.jpg (2MB, 2347x3295px) Image search: [Google]

2MB, 2347x3295px

Traits are objectively superior to OOP style inheritance.

i have several buttons that do different things, and i let the user use them depending on a variable i call 'state', so when i change state i call a function called setState(nextState) that enables or disables them depending on the state, in a switch clause.

to add more functionality i want to let the user change 'context', with each context having different ui.

but my problem is each context may have different states, so the setState function now contains nested switch clauses, which is fucking ugly and complicated. what should i do instead? is there a pattern or something that solves this? thanks if you read this far and for any answers

File: Untitled.png (54KB, 1844x798px) Image search: [Google]

54KB, 1844x798px

>>59449626
The weirdest part was when I knew I was a bit off (or so I assumed) and then the time ran out and it said perfect anyway.

File: OOP strikes back.jpg (67KB, 1006x692px) Image search: [Google]

67KB, 1006x692px

>>59449863
I agree
I wish I was that anime girl

File: 1482772013270.jpg (708KB, 2560x1600px) Image search: [Google]

708KB, 2560x1600px

Why are Rust executables shared libraries instead of normal ELF executables?

File: 59092287_p0.jpg (464KB, 826x1169px) Image search: [Google]

464KB, 826x1169px

>>59449863
>prototype-based programming with type inference

Why does my book on c++ keep referring to client code? What is the definition of client code? The book is just about simple console applications.

>>59446946
Rate my FizzBuzz:

#.(loop for i from 1 to 100 do
        (format t "~:[~a~;~2@*~:[~;Fizz~]~:[~;Buzz~]~]~%~@*" (or (zerop (mod i 3)) (zerop (mod i 5))) i (zerop (mod i 3)) (zerop (mod i 5))))

Nothing I haven't typed a line of code in months and don't know why I'm here

>>59450290
A little bit simpler:

#.(loop for i from 1 to 100 do
        (format t "~:[~:[~a~;Buzz~]~;Fizz~:[~;Buzz~]~]~%~@*"  (zerop (mod i 3)) (zerop (mod i 5)) i))

File: 1474107798032.gif (2MB, 281x203px) Image search: [Google]

2MB, 281x203px

What's a better field to work with computers and not directly working as a programmer?

>>59450470
Garbageman

>>59447373
What's the discord?

>>59450470
sys admin

>>59450866
some rich profession?

File: 0ca.jpg (261KB, 900x900px) Image search: [Google]

261KB, 900x900px

DASKETE

>>59450941
Pharmaceutical industry.

>>59450984
Prerequisites?

File: 82-84.png (487KB, 1566x1528px) Image search: [Google]

487KB, 1566x1528px

Doing Project Euler and am in a Path-Sum dilemma...

I solved the top-most problem (81, two-path sum) by taking the upper and left-most sums and seeing which one was lower (and storing that value in the "current" index)

I'm assuming I can't do that for 82 and 83, as I'd need to take into account more ways to get to a particular index. I tried using Dijkstra, but it ended up being a bit slow (though I will fully admit it may have been my fault)... Any other possible path-finding algorithms I could use to solve 82/83?

Here's the code for 81:

long getMinSum(vector<vector<long>>& matrix)
{
    long maxRow = matrix.size(), maxCol = matrix[0].size();
  
    for(int row = 0; row < maxRow ; row++)
    {
        for(int col = 0; col < maxCol ;col++)
        {
            long min = 0,   current = matrix[row][col];
            
            if(at(row-1, col-1, matrix) != -1) //If both up and left are in bounds
            {
                long left = matrix[row][col-1];
                long up = matrix[row-1][col];
                
                (left < up)? min = left : min = up;
            }
            else    //One of our neighbors is OOB
            {
                if(row == 0 && col  == 0)
                    min = 0;
                else if(at(row-1, col, matrix) == -1) //Can't go up;
                    min = matrix[row][col-1];
                else //Can't go left -> go up
                    min = matrix[row-1][col];
            }
            current += min;
            matrix[row][col] = current;
        }
    }
    return matrix[79][79];
}

>>59450998
Having a Jewish surname.

>>59451016
What if I am going to immigrate in Israel?

>>59451072
One does not become a Jew.

>>59451087
...and accept Judaism?

Is there C library that provides interface to native mobile stuff like gps and stuff?

>>59451133
There was NDK but Google doesn't recommend to program with NDK.

I bought windows 10 can I get visual studio for free or something without pirating?

File: 1361125257555.jpg (11KB, 327x388px) Image search: [Google]

11KB, 327x388px

>>59451556
>He bought Windows

>>59451562
So should you Microsoft is a great company

>>59451584
Fuck off Bill

>>59451556
https://www.visualstudio.com/vs/community/

>>59451486
Nobody gives a shit what google recommends and what would be the alternative, making web app?
There's million projects trying to provide framework to make cross platform mobile applications. Kind of sad that they all are big projects instead of little libraries providing some functionality that you could easily include into your program.

File: 1362765098504.gif (1MB, 150x113px) Image search: [Google]

1MB, 150x113px

>>59451584

>>59451556
The standard version of Visual Studio (community edition) is free as in free beer, regardless of whether you bought Windows (though using it without Windows would be a bit difficult).

>>59451584
>>59451587
Microsoft was good back in 1998 or so. Now it's fully pajeetified.

File: halp.png (146KB, 1440x900px) Image search: [Google]

146KB, 1440x900px

why the non empty bracket never get used?

File: 1478106018832.jpg (211KB, 640x640px) Image search: [Google]

211KB, 640x640px

>>59451600
>and what would be the alternative, making web app
As far as I am aware, main programming language on Android is Java.

>>59451744
Now making android specific java application isn't exacly cross platform nigger.

File: 1487526024182.jpg (40KB, 450x600px) Image search: [Google]

40KB, 450x600px

>teach coding to kids
>put a trillion mini-games in scratch together
please kill me

>>59451711
Your first one should be

[(empty? a) 0]

.

>>59451849
thanks, but now It complains about null errors.

if I try to put a second check, it still gives me null errors.

trying to make a recursive function to find the biggest number in a list.

weird question maybe.
in c++ I know you can override global

::operator new
::operator delete

which under the hood call malloc/free.
is there any way to override malloc/free instead?

>>59451876
and by override, I mean override for one specific type of object...

>>59451871
Might be because if you are on the last number of the list, it tries to do (> last-number ()). Just a guess.

>>59451776
The original question was:
>Is there C library that provides interface to native mobile stuff like gps and stuff?
Did it imply cross platform application?
What did it ever imply?

>>59451871
>>59451917
So what you should probably do is on the first one,

[(empty? (rest a)) 0]

>>59451926
Quote the original post then dump nigger.

>>59451600
>Nobody gives a shit what google recommends and what would be the alternative, making web app?
>There's million projects trying to provide framework to make cross platform mobile applications. Kind of sad that they all are big projects instead of little libraries providing some functionality that you could easily include into your program.
> to make cross platform mobile applications.

File: halp.png (161KB, 1440x900px) Image search: [Google]

161KB, 1440x900px

>>59451917
>>59451938
yeah, I'll try thanks.

>>59451964
All right.
It's not like you can have cross-platform applications in C.

>>59451996
Don't forget there is a max function in Racket.

File: Untitled.png (94KB, 1014x647px) Image search: [Google]

94KB, 1014x647px

So I want to search for x in the given array of columns synchronously to for maximum efficiency. When x is found on column 2 I want all other search iterations to stop.

How can I achieve this? Sorry for the stupid question

>>59452151

(mapcar (lambda (n) (find x n)) (apply 'mapcar 'list array))

>>59452151
I'm confused. Do you want to spawn a new process for each column? Why?

File: halp.png (154KB, 1440x900px) Image search: [Google]

154KB, 1440x900px

>>59452023
is for a homework.

right now It works, but it doesn't show 9, as I want.

any idea?

>>59452151
you first create 4 threads
each thread checks one number
after that it checks a variable that can be accessed by them all if x is found
if it is found end the thread if not continue checking the next set of numbers

>>59452317
>>59452151
there is literally no benefit doing this in 4 threads versus 1

http://ec-lang.org/
What is this esoteric language?

>>59452445
>OO
dropped

>>59452424
that depends on the location of the x
and the size of the array
assuming you´d only have one thread
you at the very least would have to check 5 times if you wanted to reach the x

>>59452511
no it does not

>>59452315
You don't want to put your recursive call outside of the cond. It should be the last one like you had before.

File: halp.png (131KB, 1440x900px) Image search: [Google]

131KB, 1440x900px

>>59452679
thanks lads.

functional programming is fun.

>>59452770
What do square brackets do in Racket?

>>59452806
is a cond.

is like a switch, sort of.

>>59452806
shit up the syntax.

>>59452630
of course it does
if you use 1 thread you have to check everything up to the x in this case 5 comparisons
if you have 4 threads you are checking the first 4 numbers at the same time
and the second row once the first is done which means in time wise you made 2 comparisons

>>59452818
I do know what's cond.

>>59452823
Oh, so, nothing special?

>>59452770
Also some more advice: the coding style in lisps is different from languages like C.
Try something like this.

(define (mayor a)
  (cond [(empty? (rest a)) 0]
        [(< i_mayor (first a)) (begin
                                 (set! i_mayor (first a))
                                 (mayor (rest a)))]
        [else (mayor (rest a))]))

>>59452894
> ))]))
makes me sick

>>59452917
I know, I was simultaneously vomiting and shitting myself while writing it. >>59452770 was just begging to be fixed though.

I remember seeing a link to a course on making a c/c++ language. anyone remembers something like it (doesn't have to be the same)? I figure if I do something like that I'll be the ultimate h3ck8er
also, I see a lot of languages but rarely they go too in-depth on how I'd code a similar lang

>>59452946
It's called compiler theory and implementation.

File: 1486296847325.jpg (44KB, 450x370px) Image search: [Google]

44KB, 450x370px

Is Scala a good FP language?
What's the most practical FP language after Erlang?

>>59452992
yeah I know
but it's hard and I'm dumb

>>59453001
Don't know after Erlang. But before you can have a look at OCaml.

>>59453001
Elixir
Scala

best way to learn functional programming for an already professional programmer?

would prefer a book.

How do I become a true code artisan?

>>59453094
Do you read French?

>>59453094
Dr. racket

>>59453094
>for an already professional programmer?

>>59453095
H A S K E L L

>>59453073
>Elixir
Wasn't it just Erlang with fancy syntax?

Also, are Scala advantages grow from the fact that it's built upon JVM?

>>59453094
Common Lisp.
-On Lisp
-Practical Common Lisp
-Common Lisp Hyperspec
-Common Lisp the Language 2

Haskell.
-Programming in Haskell
-Haskell Programming from First Principles

Idris. Agda. Shen.

>>59453162
Pretty much but erlang is like the disgusting homeless looking guy on the street that can juggle and hop on one leg at the same time while elixer is the successful, good looking, younger sibling.

>>59453094
easy way:
-programming in lua
-on lisp

>>59453094
>59453235
forgot to mention:
[spoiler]sicp[/spoiler]

>>59453235
>>59453181
>>59453137
thanks for the recommendations.
leaning towards Haskell. I'll look into common lisp. I know a guy who programs in Scala, is that another viable one?

>>59453094
meanwhile...

>>59453249

>>59453271
Great, another pointless general thread with a whole bunch of shit in the OP.
If it failed before, I don't see why it isn't going to fail now.

File: cdd.jpg (20KB, 480x420px) Image search: [Google]

20KB, 480x420px

Can someone give me a good neovim config?

Or should I be using vim 8.0?

>>59453128
Why so ?

Is there any good scripting language for android? I just want something that can read my sms's. Tried qpython but i dont think it has a library for working with sms.

>>59453305
just use atom

Ok, so I'm building a thing to measure resistance, and have been posting this on diy/ohm/. By measuring the voltage across a known resistor with an ADC, I can calculate the resistance of an unknown resistor that the known resistor is in series with. The device will have the highest resolution when the unknown resistor is near in value to the known resistor. But that's just background information.

I wrote a piece of python code to simulate this, which takes a known resistance, converts it to a voltage rounds it to the resolution of the 24 bit ADC (by rounding to the nearest 1/(2^24-1)), converting it back to resistance again, then comparing it with the original value, both rounded to the 4 digits I want to display. I made the code iterate through a shit-ton of values looking for the false comparisons closest to the reference resistor value to know how many ranges/different known resistors I'd need to switch between. With a small step size (x += x/10) I found a nice wide gap of accuracy about 6000* (80mΩ-600kΩ on either side of 500Ω) to either side of the reference resistor value. But when I made the step size so fucking small (x += x/100000) then I get a tiny gap of 490 to 530, which is stupid.

After getting that out of the way, is this a problem with python's shitty rounding? Because all of the problematic R values close to the reference value (500Ω) ended in 0.050000 or 0.049999, at varying orders of magnitude. Is there some way of fixing this? I can post the code if I have to, but you're not going to like it.

>>59453264

As pointed out by >>59453271 in >>59453249
https://0x0.st/pbp.pdf is an excellent resource for haskell, maybe the best one out there.

File: smegnefiying.png (457KB, 860x389px) Image search: [Google]

457KB, 860x389px

>Prof has this extremely strange (to me) way of handling objects
>For instance, if you're making a class to read a DNA sequencing file, the reader itself will be a single class and you pass the BufferedReader in the constructor, while read() takes no arguments
>read() itself typically will not return anything and instead just store the info in an instance variable
This is really fuckin weird to me. I usually use instance variables infrequently, only for when they truly need to be used (I.E. caching values)
He seems kind of odd as he was an ecological simulation tech who worked in /Java/ and this implementation feels weird as shit but it also kind of makes sense
This isn't standard, is it? In almost every instance, all the info is passed to the class in the constructor and then the actual methods almost never take args.

>>59453458
Yes, it is retarded.
This is what happens to people who fall for the OOP meme.

>Repeating code is bad, it results in failure to update sub classes and unintentional code branches
>You can only inherit from one single superclass; if you want to "inherit" from multiple different super classes, make them interfaces
>Interfaces cannot have default behavior in them, forcing you to hand-repeat code
...what? Why?

>>59453305
Vim 8.0 has been shitting itself lately so I switched to Neovim
90% of my stuff still works fine but some of the defaults are different

>>59453413
>>59453517
I like atom, but /g/ keeps meme'ing it and I wanted to see what the hype about (neo)vim was.

So far it's a lot more time consuming to customize it...

>>59453454
>I can post the code if I have to, but you're not going to like it.
Post please, I brely understand the problem
And here no one likes anyone's code, that's normal.

>>59453458
>>For instance, if you're making a class to read a DNA sequencing file, the reader itself will be a single class and you pass the BufferedReader in the constructor, while read() takes no arguments
Like this?

DNAReader reader = new DNAReader(File f); 
DNA result = reader.read(); 

that makes sense.
>>read() itself typically will not return anything and instead just store the info in an instance variable
Like this?

DNAReader reader = new DNAReader(File f); 
reader.read(); 
DNA dna = reader.getResult()

That makes no sense.

whats the best approach to sort a struck that has a single char and a number in c++, i need to sort it by char and number

const int SIZE = 14;

struct card {
    char colour;
    int number;
};

card Array[SIZE];
//from
Array.colour = {'M', 'G', 'M', 'M', 'M', 'G', 'M', 'R', 'M', 'G', 'M', 'M', 'J', 'M'};
Array.number = { 9,   3,   1,   1,  10,   2,   3,   6,   2,   1,  11,   4,   7,   4};
//to
Array.colour = {'G', 'G', 'G', 'J', 'M', 'M', 'M', 'M', 'M', 'M', 'M', 'M', 'M', 'R'};
Array.number = { 1,   2,   3,   7,   1,   1,   2,   3,   4,   4,   9,  10,  11,   6};

>>59447518
>newfags
Why do they come here? We've made it perfectly clear you don't belong.

>>59453662
this is unusual

>>59447712
So you should firstly know about the exponentiation by squaring algorithm.
The arithmetic should be done in the ring of integers modulo 10^9 as we only care about the last 9 digits

xi is the i-th bit in x (starting at the least significant bit)
n^x = Σ xi * n^(2^i), 0 <= i <= 31
x = Σ xi * 2^i, 0 <= i <= 31

n^x = x
n^x - x = 0
Σ xi * (n^(2^i) - 2^i) = 0

We precompute all the n^(2^i) - 2^i then we have to pick the highest number x such that that sum is 0. Remember that all computations are modulo 10^9. The best way to compute x is with memoization. We precompute a table of possible sums for the last k bits of x and then do backtracking on the rest of the bits. k should depend on the memory constrains.

Also because in the most expensive step (the backtracking) you only use addition and modulo you could code everything in BCD and replace the modulo with a bitwise and. Here are efficient ways to do operations on BCD coded numbers: http://homepage.divms.uiowa.edu/~jones/bcd/bcd.html
I'm not 100% sure if it's actually faster than modulo

>>59453621
Use std::array and std::sort, also the way you define Array and then assign elements makes no sense.

>>59447042
the learning curve

>>59453513
OOP is a mess.
People have shifted from the idea of multiple inheritance because it's a bad idea (like all of OOP).
It's only a matter of time before the POOs catch on and start programming in an imperative style again.

The most constructive thing I think a programmer can do now days is to ignore people who complain about lack of code reuse. Recognise that even if there's parts of code you could reuse, their similar behavior is just coincidental and it's very possible (even likely) that future code iterations will make them more dissimilar (because the interactions will be more specific and hence more different). Inheritance is spreading that coupling of these two dissimilar parts throughout the code base which makes it difficult to modify.
People consider this protecting against bad programmers. Which is true, if your code is more difficult to modify effectively you will make bad edits less likely. But you also make good edits hard.

>>59453732
i know, its just a way to show how i want it sorted

>>59453761
FP makes for excellent code re-use, but sometimes it's slow

>>59453761
>protecting against bad programmers
"You should just not hire bad programmers." Jonathan Blow

>>59448630
Don't see the problem here

>>59446946
Why is Shinji dressed like a girl?

>>59453621
>I need to sort it by char and number
So one has presidece over the other?
What you're asking for is a stable sort. There's plenty of options.

So just sort your array based on the least significant sort key first and then the most significant.

>>59453777
Capitalism aims to deskill programmers, in order to be able to flood the market with cheap, unskilled labor, so it can drive down wages

>>59453761
>People have shifted from the idea of multiple inheritance because it's a bad idea (like all of OOP)

I want to make a level editor for a game.

I have things that make other things do work (triggers, and triggerables). An example of a trigger would be a button, a trigger zone (when the player walks into it), a physics pressure sensor, and a damage plate. Examples of triggerables include doors, sounds, kill zones, enemy spawners, etc.

I want to make it so that if the player presses two buttons (triggers) that they open a door. But wait! If I want to not make a custom, single-use class, I would have to have the buttons trigger not the door, but a trigger logic gate; each time it gets triggered it increments a counter, triggering its target(s) when it goes over the specified amount.

This logic gate is a triggerable and a trigger.

What is your proposed method to fix the above issue without duplicating code?

>>59453305

if !has("nvim")
    runtime ftplugin/man.vim "for some reason this breaks neovim
endif

noremap <F2>:terminal<CR>

"a bit like K in normal mode but better (note the space at the end)
nnoremap <leader>k :terminal man 
nnoremap K yiw:terminal man ^R0<CR><CR>

>>59453847
ADTs and typeclasses

>>59453773
I'd say FP makes programming easy in general.
>its slow
Yes. That's its downfall. And most FP languages are quite high level. Most low level and fast languages are relying on their imperative nature.
I do think FP is the future but not yet.
>>59453777
I don't think Jonathan Blow has experiences that can be applied broadly. But its probably true for him. He's the designer and lead programmer. He doesn't have to interface with bad people because he determines the scale and he will only need to do what he manages himself. The parts he doesn't manage himself are done by specialists.

Basically JB is the worst programmer on his team and he's good enough, so he doesn't have the problems others face.
>>59453813
I disagree. It's most likely just that lots of people want to do big software projects. So there's not enough skilled programmers.

Should I use emacs or vim?

>>59453847
just use arrows transformers senpai

>>59453873
Neither, they're a waste of time in the long run even for those that are proficient in it.

>>59453561
Fuck, it's too long

>>59453871
>So there's not enough skilled programmers.
Wrong. There are plenty of skilled programmers, just not at the wages companies want to pay.

>>59453873
there is no right answer to this question, try using both for a week and decide yourself

>>59453904
I've been using vim a little so far, I like the very quick motions and stuff, but the way of configuring it seems horrendous.

>>59453873
vim will be more useful - vim ships with linux, emacs does not

>>59453915
>vim ships with linux
vim, or vi?

>>59453847
>without duplicating code
Well you realise that if you remove this constraint your problem is trivial right? You could easily construct a trigger class that takes two inputs.

The best programmers are women who dress like men.

>>59453897
Don't we agree that most skilled (productive) programmers are employed? At least the ones that desire employment in general.

>>59453932
most of the popular distros have both

>>59453915
>vim ships with linux
Linux is a kernel. Nothing "ships with" it.
It's up to the distribution whether they ship Vim by default.
For example, Archlinux doesn't ship Vim nor Emacx by default.

vi is the program that gets shipped by default, because it's a standard POSIX program. vi != Vim.

>>59453932
>vim or vi
What's the difference? Is one CLI and the other GUI?
.t person who doesn't use vi*

>>59453454

from math import log10, floor

adc_bit = 24
reference_resistance = 500
sig_figs = 4
test_range = 10000
step_quotient = int(input("Step Quotient: "))
test_resistance_i = reference_resistance / test_range
test_resistance_f = reference_resistance * test_range
test_r = test_resistance_i
innacurate_first = 0
innacurate_rounded = 0

def res_to_v(resistance, reference_resistance):
    voltage_fraction = 1 - resistance / (resistance + reference_resistance)
    return voltage_fraction

def v_to_res(voltage_fraction, reference_resistance):
    resistance = reference_resistance*(1 / voltage_fraction - 1)
    return resistance

def v_to_rounded(voltage_fraction, adc_bit):
   voltage_binary = voltage_fraction * (2 ** adc_bit - 1)
   voltage_binary_rounded = round(voltage_binary)
   voltage_fraction_rounded = voltage_binary_rounded / (2 ** adc_bit - 1)
   return voltage_fraction_rounded
    
def round_to_sig(number, sig_figs):
    rounded_number = round(number, (sig_figs - 1)-int(floor(log10(abs(number)))))
    return rounded_number

#code starts here

while test_r <= test_resistance_f:
    test_fraction = res_to_v(test_r, reference_resistance)
    rounded_test_fraction = v_to_rounded(test_fraction, adc_bit)
    rounded_test_r = v_to_res(rounded_test_fraction, reference_resistance)
    significant_test_r = round_to_sig(test_r, sig_figs)
    significant_rounded_test_r = round_to_sig(rounded_test_r, sig_figs)
                              
    if significant_test_r != significant_rounded_test_r and test_r < 500:
        innacurate_first = test_r
        innacurate_rounded = rounded_test_r
        
    if significant_test_r != significant_rounded_test_r and test_r > 500:
        print(innacurate_first, "-", test_r)
        print(innacurate_rounded, "≠", innacurate_first)
        print(test_r, "≠", rounded_test_r)
        break
    test_r += test_r / step_quotient

>Please close Visual Studio now to reduce the chance that a computer restart will be required later.


So it's like a lottery or what?

>>59453983
vim is an improved vi, both are CLI. gvim has a GUI

how do I initialize a std::stringstream with a stateful allocator?
is that even possible?

File: Untitled.png (12KB, 1103x639px) Image search: [Google]

12KB, 1103x639px

Can someone explain to me how I managed to fuck up win-builds this badly?

>>59447651
>technology anime pictures
how is technology related to mental illness?

>>59453954
Most of them are employed at depressed wages.

>>59454130
>he doesn't use msys2

>>59454046
Thanks

>>59454130
PROTIP: don't use cmd as a terminal

>>59454040
Yes. Microsoft developers don't know anything about their OS.
They have guesses all over the place.

Is learning multiple programming languages at the same time a bad idea?

I feel like learning one programming is in itself a huge undertaking (depending on what you mean with "learn") and I don't want to wait with picking up the other languages I'm interested in learning.

File: Untitled.png (126KB, 1190x902px) Image search: [Google]

126KB, 1190x902px

>>59454130
Here's a verbose dump if it helps at all

>>59454231
It was run from Vim, since I fell for the meme.

>>59454280
Only if you're a brainlet.

>>59454280
I think it's a good idea to keep them seperate, because if you're learning them so close together that the syntax gets mixed up then you're fucked. But learning one language one day and another language another shouldn't be a problem once you've got the syntax basics down.

>>59454526
>>59454625
Alright, thanks.

I want to draw something over time, take for example a fractal tree which branches slowly grow on screen. what is a proper way of doing it?
can I keep drawing to a texture?
what are buffers and would they apply here?
how do I into computer graphics?

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

94KB, 396x395px

Will we solve chess in our lifetime?

>>59454802
I don't think so, nobody has ever completed a game of chess before so I don't think anybody ever will

File: intredasting.jpg (112KB, 716x716px) Image search: [Google]

112KB, 716x716px

>>59448278
I shall give this a rollerino

>>59453561
>>59454035

>>59454802
Chess with best play is a draw.

>>59448278
roll

>>59454802
I already solved chess in my free time, but I don't care enough to publish my results.

Not gonna lie, former C programmer here. This is fucking hilarious watching C segfault and crash. But in all seriousness we can't let these buffers get more overflows.

>>59454231
Why?

>>59455282
That's not hilarious at all, you sick fuck!

File: 1489394414673.jpg (38KB, 600x332px) Image search: [Google]

38KB, 600x332px

>>59455282
What on earth are you on about?

>>59455433
Its an edit of the pol pasta "former trump supporter[...] cant let him get the nuclear codes"

>>59447968
>Living for other people is a terrible bore, but it's important. Fuck.

Start working for yourself you fuck.

File: mfw.png (44KB, 256x233px) Image search: [Google]

44KB, 256x233px

>>59448212
>shinji

will this thread hit the bump limit?

>>59455807
Maybe.

>>59455807
Impossible

>>59447896
Maybe get another hobby?

Move over to
>>59446947
>>59446947
>>59446947

We really need less of this stupid duplicate thread shit going on.

>>59455850
>we really need less of this stupid duplicate thread shit, so move over to this duplicate thread that was created by someone who's so new to 4chan that they're triggered by anime

>>59455867
Well, create yet another duplicate thread isn't going to fix it.
I wish janitors would delete that stupid shit, like they used to.

>>59455878
>create
creating*

>>59451006
>[79][79
Is your implementation even Dijkstra?
I think your original matrix problem should first be converted to a graph. Then apply the Dijkstra algoritm.

I would use a map graph implementation, but you can use a matrix graph. If you don't wish to implement this, just use a another function to return the neibour nodes of a node.

Just get a Dijskra working with good variable names and good comments, and then optimize it. I recall once going from five minutes to 3 seconds just by optimizing my Dujskra for speed.

>>59452151
Build 4 son processes. Each one searches for X, while father process waits for pipe delivery of result. Process that founds X, push to pipe X location and dies. Father reads X location and then kills all children.

The the guys saying that having more process doesn't make this faster in amulti core processator is a retard.