/dpt/ - Daily Programming Thread

File: C.jpg (107KB, 1000x611px) Image search: [Google]

107KB, 1000x611px

First for Pelles C!

File: 6c9.jpg (48KB, 435x571px) Image search: [Google]

48KB, 435x571px

>>61957993

(define/contract (sum lst)
  (-> (listof number?) number?)
  (foldl + 0 lst))

File: 2ab.gif (60KB, 466x367px) Image search: [Google]

60KB, 466x367px

https://support.amd.com/TechDocs/25112.PDF

There's some good shit in Chapter 2.

>>61958010
>be compiler
>tfw void*

>>61958039
It's actually the simplest case, you know.

>>61958031
What esoteric programming language is this?

Someone give me something to do in C++. I'm past the point in Project Euler where I can just brute force everything, so now I'm bored...

File: 1502554078961.png (111KB, 600x800px) Image search: [Google]

111KB, 600x800px

Reminder

>>61958031
>contract
what shitty thing is this?

>>61958088
I just wanted to force a joke.

What does /dpt/ think of Unison?

http://unisonweb.org/2015-05-07/about.html#post-start

The editor looks pretty cool tbqh

>>61958193
Contracts are really nice to be able to quickly create predicates

;; a "reserved thing" is either
;;   a name
;;   a single index
;;   a range:
;;     (list <min-idx> <max-idx>)
;;     (list <min-idx> 'max)  - unbounded
(define (reserved-thing? x)
  (or (string? x)
      (exact-integer? x)
      (range? x)))

(define (range? x)
  (and (list? x)
       (exact-integer? (first x))
       (or (exact-integer? (second x))
           (eq? (second x) 'max))))


;; we can make this much shorter with contract combinators, which end
;; with /c in the name

(define range?
  (list/c exact-integer?
          (or/c exact-integer 'max)))

(define reserved-thing?
  (or/c string?
        exact-integer?
        range?))

You can also put contracts on functions which automatically checks inputs and outputs. It does some comlicated wrapping to guard higher order functions

>>61958241
>yet another irrelevant functional programming language
Don't we have enough of those that go completely unused?

>>61958291
Why bother when you could just use dependent types and make it impossible to send invalid input to functions?

As an aside, Clojure fans seem to miss the point on this too. Cool, you've got Clojure.spec, but it's fundamentally no different from putting a bunch of asserts at the start and end of a Java method. It's still going to blow up at runtime rather than during compilation.

>>61958291
Isn't this just another way of doing type specifiers?

I think Common Lisp has this.

>>61958341
>>61958342
Not that guy, but there's also Typed Racket, however that requires you to specify types at all times, while contracts are an as-you-go sort of deal. Sure, you could specify the types for shit to be Any, but that would just be unnecessary.

>>61958341
Racket contracts and Typed Racket don't aim to be magic catch-all perfect solutions ("dependently typed code has no errors!" is a meme), instead they aim to make it easy to create checks that use idioms you'd usually be using in scheme anyways, like passing some set of symbols as an argument (the contract combinator (symbols ...) or using a symbol as a contract (or/c 'a 'b 'c) ), lists/vectors with a specific number of elements (list/c, vector/c), homogenous lists (listof, vectorof) etc.

>>61958342
Sure, but it produces real predicates, which are objects like any other, making it super easy to create new "types" (they aren't really *types*, they're predicates/contracts). CL uses some weird form of interpreting quoted stuff as a "type" IIRC and has special cases for different types of integers and stuff. Racket's contracts are very intuitive, yet still powerful. Most of them could be naively implemented with just a few lines of code.

How the fuck do I get the Vulkan SDK to work on Debian? Someone already made it? Using a R7 270X btw.

>>61958404
Lmao type specifieres are literally predicates.

I know there's one book in the wiki, but is there any other books or general materal for learning Ada?

>>61958484
Not according to https://www.cs.cmu.edu/Groups/AI/html/cltl/clm/node44.html

You build a weird quoted expression and then pass it to TYPEP

>>61958404
>"dependently typed code has no errors!" is a meme
Of course dependently typed code can have bugs, but dependent types allow you to make a large class of boring bugs simply impossible. Having a compiler check the boring bits of programming for me frees up mental capacity to think about higher-level concerns.

>>61958525
But dependent types also require you to write a bunch of boring confusing proofs

>>61958524
They SPECIFY types, they literally are predicates.

>>61958540
>dependent types also require you to write a bunch of boring confusing proofs
How so? C++ templates are a fine example of dependent types that have the proof built in as the program.

>>61958540
Sadly true at the moment. There's huge scope for improvement.

(Not that anyone forced to target the JVM or CLR in order to satisfy peculiar client fetishes will ever see it.)

>>61958562
We seem to be using the term "predicate" in different ways. I meant it specifically to mean functions that objects and return booleans

>>61958569
Lmao C++ templates aren't depedent types

>>61958586
You mean lisp function objects of type 'function
I mean why can't typep work, it gives a boolean.

>>61958614
With contracts you literally just combining arbitrary predicates into larger predicates

>>61958586
Types are first-class in C++ templates and the template system is Turing complete. That's exactly what's required of a dependent type system.

>>61957652
That would probably have a way better answer on Stackoverflow. Off top of my head:

- force non-negative number always
- doubles your max positive number
- anything that touches hardware like arduino code ought to be unsigned most of the time, the 8th bit is usually very important

For jobs C is used for (low level bit fucking and hardware and systems shit) unsigned is baller. For higher level stuff, well there is a reason Python doesn't give much of a shit about it.

>>61958634
So anything with a turing complete static analyzer is "dependent types" now?
I think you need to do a bit more research. Even if C++ types are dependent, they're seriously lousy dependent types.

>>61958663
There are many Turing-complete languages that are lousy but nevertheless Turing-complete.

>>61958663
>Even if C++ types are dependent
C++ template types are dependent.

What is /g/s opinion of Perl 6?

>>61958674
>>61958694
How do you express something like "the result of this function is an integer greater than the argument"

>>61958699
literally who

File: 1498754743800.png (72KB, 948x693px) Image search: [Google]

72KB, 948x693px

>>61958699
Perl's strength is in string manipulation

and the logo is cute

>>61958643
i don't python has any limit on ints now

>>61958699
>I gobble them, while they gobble my filth.
>what do i care!?

>>61958632
You can do that in CL
'(or (and (satisfies prime?) integer) character)

It can be an intersection of types, or a union, etc.

>>61958766
qt

how come we don't have a wiki

>>61958826
sticky you nigger
http://wiki.installgentoo.com/

>upload image of cat in popular blog platform
>click post
>nothing happens
>image is posted

>upload image of nude women
>notification about NSFW, sensitive material, etc
>delete image before posting

how do they do it? AI shit scan every uploaded image posted checking for nude content?

>>61958837
>
>>61958826
There wouldn't be anything interesting in it anyways except for various ways to perform FizzBuzz in languages non one uses.

>>61958771
Yes that's what I'm saying, Python doesn't care if they're signed or not. Hardware and low level bit shit DO care the way an autist cares about the peas being a perfect shade of green and not touching the onion rings, throws a goddamn fit if you screw the tiny shit up.

>>61958694
The fact it's turing complete makes it non-dependent

how come we don't have a gitlab

>>61958908
https://gitgud.io/users/sign_in

>>61958908
https://gitgud.io/users/sign_in

how come we don't have an open-source imageboard template using Elixir?

>>61958953
cause elixir sucks

>>61958188
scripting in shen.... Are you trying to drive people to drink bleach?

>>61958953
>>>/news.ycombinator.com/

>>61958965
It is the future

>>61958953
lainchan

how come we don't have a operating system written entirely in Perl and Python, with a DE using jQuery/Java?

File: Shitty First game engine in java.webm (1MB, 1200x674px) Image search: [Google]

1MB, 1200x674px

Teaching myself java by creating games, so far I made bricks, snake, and now more of a top down zelda like game, I will probably not even actually make much content for it, since I just want to gain some experience from it and create the next one.

Just published a library as GPL, feels pretty good

How come we don't have a deep learning API written in Scheme using a matrix math library written in Io and JSON?

>>61958998
You've got several well documented kernels out there, go to town, build your own os with the only binary being the kernel and your interpreter of choice

Literally nothing is stopping you but your laziness.

Replacing the kernel... that's harder.

>>61959003
Cool make sure you check Game Programming Patterns out too.

>>61959047
gonna check it out, thank you

File: Screenshot from 2017-08-17 22-17-36.png (357KB, 1167x748px) Image search: [Google]

357KB, 1167x748px

I believe I am too stupid to understand SICP

>>61959117
You'll fit in just fine :)

>>61959003
How the fuck do you read the purple text on a black background

Combinatorics is the comfiest math

>>61955539
Reposting.

File: miku cock.png (388KB, 435x437px) Image search: [Google]

388KB, 435x437px

Rate my poker deck simulator, it only uses 52 bytes of memory!

http://ideone.com/XtLT6L

File: ss+(2017-08-18+at+04.30.28).png (702B, 150x17px) Image search: [Google]

702B, 150x17px

>>61959144
it is actually more readable, just in the video its hard to read

>>61959219
>goto

aaand droppped.

>>61959282
do you use linux?

>>61959234
That still seems pretty bad to me anon

>>61959282
Goto is a perfectly reasonable operation in C

>>61959295
yes and no.

the code is perfectly fine. you don't need to use a goto tho

>>61959282
the alternative is to use a clunky flag variable in a while loop
it's the same thing in a roundabout way

File: index6.png (476KB, 735x524px) Image search: [Google]

476KB, 735x524px

I've almost given up on my genitalia detection neural network. There's too many false positives (pic related) likely caused by how I labeled that data or by the lack of data.
I'm most likely going to narrow focus on just cocks. So far I've labeled several thousand cocks in a new dataset and am aiming to label 1000 images a day.

>>61959219
>rand()

>>61959310
sounds like you're using computer programming as an excuse to eat dicks all day

>>61959330
what do you use?

>>61959349
mt::19937 :^)
but seriously, for c: arc4random

>>61959349
complimentary multiply with carry.

>>61959389
rand is faster and in this program there's no need to use a4r

>>61959310
Your false positives are in the ~right area though, you just need to refine it.

>>61959219
rand % 52 won't actually give you a uniform distribution, numbers less than RAND_MAX % 52 will be about 52 / RAND_MAX more likely to occur, or something like that.

>>61959498
but I'm checking for duplicates so it doesn't matter, right?

In java am I going to be made fun of for using

 while (true) 

I read somewhere that it is a joke

>>61959346
I admit I get distracted at times by the images, but I only have an academic interest in this problem. I won't be as distracted now that I added a ton of yaoi to my data.
>>61959420
Nah, I want bounding boxes around actual genitals, not where generals should be

>>61959537
if you want an infinite loop it's fine
while (1) is more common though

>>61959553
Well its not like it really matters for pictures like that since your goal is de-censoring.

>>61959558
>not for(;;)

>>61959537
>>61959558

while(!0)

>>61959582

while (35 > 24)

>>61959523
I think it means you will have more spades in the lower k/2 of your deck. If RAND_MAX is fucking huge you probably won't care, since 52 / RAND_MAX higher probability to be picked would be negligible.

I don't remember the right way to do it off the top of my head. I think something like:

do {
  k = rand() % 64; // divisor needs to evenly divide RAND_MAX or RAND_MAX + 1, forgot which
} while (k > TOTAL);

>>61959592

def fuck():
    while(7 is not 9):
        yield True

while(fuck()):
    print('dicks')

>>61959582

//file.h

// code

#define FALSE "0"

// other code

#include file.h

// Using while to comment out code
while (FALSE) {
  // code
}

>>61959565
If this was a NSFW board I'd post a NSFW image showing the false positives.

File: pickle rick.jpg (74KB, 1280x720px) Image search: [Google]

74KB, 1280x720px

>>61959310
run this thru your genitalia neural network

File: ad75b92ce5d372c1e0ba500ccf64e6a1ff58ed3b36e296c7d71c2f5b94b39c8e.png (294KB, 473x305px) Image search: [Google]

294KB, 473x305px

>>61959310
do you have only positive examples?
less than 10k?
good datasets nowadays have like 1m-10m+ examples

>>61959310
it actually did capture where the genitalia is in that picture, it is just hidden by clothing and animu torso

Is there literally any reason ever to use data structures like records and arrays when they're slower than arrays?
Just write a wrapper function around an array that traverses it through a hash table or something.

Why aren't all data structures just syntactic sugar around arrays? That's how memory works anyway.

inb4 linked list meme

File: 1361918016065.jpg (53KB, 261x257px) Image search: [Google]

53KB, 261x257px

>>61957975
>There will never be a release of Pelles C on GNU+Linux

>>61960067
try Linux maybe?

>>61959854
Will do tomorrow
>>61959897
Only positive examples (i.e. only images that have at least one genital in them). I could add in some images with no genitals.
I'd love to have a bigger dataset, but I have to manually gather and label everything myself. It's a lot of work.
>>61959906
See >>61959553

>>61960040
Space efficiency and reallocation cost. Which is why structures like linked lists exist.

Generally you're otherwise correct, and it's arguable that most common uses of things like maps are just due to laziness and poor planning.

For people programming say upwards of 5 years, do you still get stuck on dumb mistakes? I literally spent 30 mins trying to debug a problem because I wrote if instead of while.

>>61960119
i'd have to do that deliberately
it seems to me you use these two interchangeably and your career consists of manually spewing out boilerplate for a living

>>61960119
I've honestly never gotten stuck on a mistake like that.

If you're really new, then obviously it'll take a bit for things to start structuring and crystallizing in mind. But ideally it doesn't ever crystallize too much.

>>61960091
you could rotate, translate, skew, slightly distort images in your dataset to make it bigger.
Data augmentation I believe it is called.

>>61960129
>>61960135

It happened because I was thinking about doing a recursive implementation of a method but I ended up writing an iterative one.

>>61960119
I spent a good 30 minutes trying to figure out why my changes were not doing anything before I found out I was running an old version from a different directory while making changes to another version

>>61960147
i.e. instead of going

if(x != condition)
    method(x.next)

I was going

if(x != condition)
    x = x.next

>>61958766
preferred the camel

>>61960236
You just need to slow down and better serialize things in mind.

I realized recently, that for various reasons I had developed a "windowed" form of consciousness, which evolved from highly serialized thought processes. Each task is separated into discrete chunks, and there are brief, generally invisible to me, pauses between each chunk where reflection, certain checks and branches are evaluated, and storage takes place. It is designed for rapid iteration, error control, and keep a large amount of working memory in one's conscious frame. It's far slower, but you make wider connections, remember almost everything you see and hear, and integrate new information better. Mistakes like that simply are improbable to the point of impossibility.

And it was then that I realized why every second of every day I felt crippled and as though I was smashing my head against a wall trying to scale to anything of any complexity (the bulk of overall context is dumped between windows). I realized I had to return to the start and remember how to let go of the reins, so to speak, and think asynchronously. Regardless, it taught me a lot. Pieces of that machinery are still used for any linear thought components that are required for whatever task. It's good to know how to do, and have that level of self discipline and internal formalization.

>>61960331
k.

I want to learn how to crack open games and translate them.
I know Java and C++. However only for making enterprise level programs like they teach at uni.

What language should I learn?
Is there a guide?

tfw too dumb for programming

>>61960371
Yep. That's that.

>>61960137
Been doing that. Not doing data augmentation in object detection is very rare.

I remember seeing a massive collection of programming books floating around here. Like guides and exercises do any of you still have a link

>>61959597
You need to reroll while your rand is > a certain multiple of your divisor, then use that as your rand.

>>61960383
Elaborate. You want to write a decompiler?

Writing a disassembler though is fairly trivial, though time consuming to dig through the x86 ISA docs.

>>61960429
I'm still kinda new to programming.
I just know how to make a basic uni level program.

I guess what I want, is to be able to know how to open up games and find the text, translate and put it back in.
Does that require knowing assembly?

>>61959003
I know these kinds of questions are almost always fruitless and don't lead to anything substantial, but I might as well ask.

Would you be interested in having a learning/practice buddy to learn with? I'm not an absolute beginner, but I haven't quite made anything super complex as in more than a couple hundred lines of code.

Are you just using Java's built-in 2D API or something like libGDX?

>>61960401
so few thousand is before this?

>>61959310
https://youtu.be/lgXn_C3vFGs?t=21s

>>61960464
>Does that require knowing assembly?
Yep. Though bear in mind that any text being output by a modern game probably isn't stored in the executable. It might be stored in some .dll, or more likely, it's stored in some files referenced by a script, or map, or otherwise triggered to be loaded by the engine. In which case you have to figure out what those files are, and reverse engineer the format if it's packed in any kind of way. Then repack it into a form the engine is expecting, or able to use.

Look into some of the tools used to work with Square Enix's crystal engine, or whatever it's called, for some example of what that process might entail.

>>61960479
I'm doing data augmentation at runtime. At the start of each training step, the input image is randomly transformed (e.g. saturation changed, flipping, cropping etc.). This process is very fast with numpy.

>>61960558
okay thank you, so where exactly should i go from here?
Should I just start looking at assembly and learning it or should I just look at compilers like you mentioned before.

>>61960563

how did you get started with this stuff? i did some simple work on anime recognition some time ago but i have no idea how to further advance this past what i already have.

writing a bullshit compiler for fun, how would you prefer to declare your functions?

func Hello() is
    puts "Hello World!"
fin

func Hello() {
    puts "Hello World!"
}=>Hello

or

dec Hello() 
    puts "Hello World"
end=>Hello

>>61960717

type name arg1 arg2 return type
    "hello world"

>>61960648
Maybe both, maybe neither. It really depends on what avenue you most readily want to go down.

hex-rays IDA is a good disassembler if you want to start learning assembly that way. It's good to have a real world view of what a modern optimizing compiler will output, while learning assembly. You could also write some short test programs and see what eg GCC will output at various optimization levels.

>>61960750
okay thank you

>>61960753
(GCC's asm output flag is -S, if I recall correct, by the way. Do not recall what the switch to specify AT&T or Intel syntax is)

Been thinking about a new type of kernel. What if one mixed the hypervisor concept with the kernel concept? If every application was contained in a sort of a virtual machine and only was able to see its own resources like RAM, CPU

>>61960808
hasn't that been done?

>>61960664
Took some college courses and did some projects. For you, try http://course.fast.ai/

>>61960817
there's jails but that's for userspace. there's nothing isolating applications as hard as this as far as I know.

>>61960744
how about

!func Main(args:|arr[str]|) {
    -# Prints "Hello World!"
    put "Hello World!"
    return 0
}=>Main:|int|

comparable to

public static int main(String[] args) {
    // Prints "Hello World!"
    System.out.println("Hello World!");
    return 0;
}

?

is it easier to land a job going full javascript (freecodecamp), ruby + rails (odin project) or java back-end? I'm mostly interested in java but I imagine it is the hardest to get a job

>>61960841
Thats pretty awful.
You should really look at the simplicity of functional syntax even if your language wont be.

One thing I do advise regardless is pure-by-default functions.

>>61960850
plenty of remote java jobs. if you're interest in java alot you might become good enough to get a good long term position and not need to look for one for a long time.

>>61960850
no degree
no job

>>61960879
i have no degree and I'm earning a decent income doing remote jobs.

>>61958699
Vaporware. In all seriousness, I met one of the Perl6 designers and said some edgy shit about Perl being "PHP but also used for shitty scripts" without knowing who she was. That was interesting.

Me and my friend have to make a game for university and we're trying to find what a good API would be for something relatively simple.

Why shouldn't we use JavaFX?

>>61960947
>Why shouldn't we use JavaFX?
You should. it's a great library and present in all java installations.

Speaking of remote jobs, how abundant are they as entry level jobs for fresh cs grads? I want to move to a small town but the smaller the town the fewer the tech jobs. Like for example the largest city in a state I'd like to move to only has about 200 software developer listings on Glassdoor, which includes all levels and moving across country in the hope of landing a 1 in 200 is a bit of a stretch. And again, I don't even want to live in that city I'd want something like 30-100k people

>>61960959
stackoverflow jobs and upwork.com have plenty of work for juniors

>>61958109
Clojure. It'a a LISP on JVM but with some cool features like built-in concurrency support. I am currently writing smthing in it, too

>>61960860

I kinda want Ruby/Lua simplicity with Ada/C explicitly. Like a language that can switch between high and low-level functionality without having to fuck with the source. Don't know how the fuck the speed will be, but I don't plan on ever showing this to anyone ever anyway.

!func main {
    -# Prints "Hello World!"
    put "Hello World!"
}=>main
or 
!func main =<
    -# Prints "Hello World!"
    put "Hello World!"
=>main
//Experimentally
!func main 
    -# Prints "Hello World!"
    "Hello World!".puts
=>main

After finishing the java courses at MOOC.fi, what is the next step on getting work in java?

>>61960870
sorry if I'm bothering you too much but is there an estimative of how long it would take studying and practicing java and its frameworks, say 8 hours a day, in order to get a "good long term position"? thx for replying

>>61960879
>>61960907
I plan to get a degree in a near future just not now. thx for replying

>>61960850
Java jobs are everywhere my dude

>>61961020
but why, theres so much needless bullshit in there

main = Print "Hello World"
Or if you want to be explicit, something like
main (With io : Print) = Print "Hello World"

For a web back-end, are the libraries provided by the JVM a good enough reason to use Scala or Clojure over other functional languages?

>>61961020
i sincerely hope you're not designing a language around aesthetic "hello worlds"

>pick up programming book
>starts off ok
>OK NOW INSTALL THIS LIBRARY I WROTE THAT DOES EVERYTHING
>ALL THE EXAMPLES WILL USE THIS LIB EXTENSIVELY
>delete pdf immediately

Why can't they just teach you without a prewritten library?
and NO, they never go into the implementation details of the library either

File: 1502743783049.jpg (315KB, 650x854px) Image search: [Google]

315KB, 650x854px

>>61957754
Hello complete nuub question.

How do I know that someone's module isn't malicious?
I want to do tts for my alarm clock but all the tts modules I found online are not compatible with python3 only 2.

I did find one module called pyttsx3 but I have only one person who has ever reference it. That person is also the author....

I am not advanced enough to audit it myself. ( I am barely past fizz buzz)

How is this usually done?

>>61961112
You must read shitty programming books then. None of the programming books I have read do shit like that. Do you get your books from pajeets in ibook or something?

File: 150283819375.jpg (29KB, 600x600px) Image search: [Google]

29KB, 600x600px

What's your excuse for not using Go?
>super comfy to program in
>strong concurrency
>expressive like a scripting language, fast like a compiled one
Literally no reason to not use Go.

>>61961142
whats your excuse for managing to fuck greentext

>>61961142
Rob Pike is the Randy Pitchford of the programming world.
And Go is awful.

>>61961053
>>61961064

no, the main wouldn't be needed, I'm only including it to show how functions would be laid out. I guess hello world is a shit example, let me use something better.

// !func == static func 

func fib(n:|int32|) {
    n <= 1 ? return n : fib(n - 1) + fib(n - 2)
}=>fib:|int32|

put fib( 2 )

>>61961160
>Rob Pike is the Randy Pitchford of the programming world.
What does that mean?

>>61961154
The same u use for u r grammar

>>61961163

fib n : int32 //implicitly requires return to be the same type as the only arg with the option to specify
{
     n <= 1 ? return n : fib(n - 1) + fib(n - 2)
}

put fib( 2 )

Seriously, if you want anyone to use your language in the future, stop with the ocaml-tier syntax.

File: defensive.jpg (34KB, 710x720px) Image search: [Google]

34KB, 710x720px

>>61961188
A-Are you a real girl?

>>61961211
Answer my question and I'll do the same.

>>61961211
yes

>>61961230
>>61961211
My question is here
>>61961123
Incase you cannot tell.

>>61961163
..is the =>name:|type| part of the signature?
it seems rather redundant in all of your examples.

>>61961243
I get it I'll go back to anon forever after this question is done.

>>61961292
jeez you have like no patience at all

File: nervous_cat_girl.png (369KB, 626x885px) Image search: [Google]

369KB, 626x885px

>>61961261
>>61961123
The only way to tell if a program is malicious is to audit it yourself. You could quickly check to see if it is malicious by looking at all the I/O things it does such as check if it connects to the internet or does any file changes etc.

Also, please be my gf ;_;

>>61961261
you read thru the source? look for suspicious imports (ones that don't seem relevant to the problem domain, etc)?

>>61961262
it's an explicit end to the function declared before it. It's supposed to be used to return a particular type ( like void function(args) ), It is redundant when the variable also has defined types, so i'll remove those variable defined types.

func fib(n) is 
    n <= 1 ? n : fib(n - 1) + fib(n - 2)
=>fib:|i32|

//again, or

func fib(n) is 
    n <= 1 ? n : fib(n - 1) + fib(n - 2)
=>fib : i32

>>61961324
decl names are trash, please get rid of them, but you seem pretty set on not changing anything so whatever.

>Self taught since high school
>have mediocre understanding and constantly have to look things up for help
>spend some time building a simple program at my slowly rising level of understanding
>can't post it on github because everyone will call me shit for my working program having shit style

>>61961304
That was someone else posting using my name.

>>61961316
OK checking the imports looks like a good trick. Thanks a lot. I think I am going to try this one.

>>61961311
I cant tell really tell what a program does by just looking at it yet. Especially large ones where everything seems to run together in my eyes.

Ps.
Also you seem to collect images of anime girls looking uncomfortable. That the definition of a red flag.

Right behind cuts the eyes out of pictures with his parents and goes on 4chan.

Besides if I was into greasy cs students i would be getting help from the ones on my campus ratherthan avoiding them to the point of going to ass hole of the Internet for help instead.

Don't mean to be mean I just want to make this clear up front so I can use these threads in piece.

>>61961346
you know, I was mulling over it, but I kinda see what you're saying.

!main is
        -# do something
=>main

is more aesthetically pleasing then

!proc main is 
        -#do something
=>main

or

public static int main(String[] args) {
        //do something
}

>can't concentrate while at home
>can't concentrate while at work
>can only work on personal projects when somewhere outside with a laptop
should I just build a forest shack and do my work there

>>61961385
>That was someone else posting using my name.
lol ok
why use names on 4chan tho

File: pillow.jpg (23KB, 640x559px) Image search: [Google]

23KB, 640x559px

>>61961385
Very good trolling friendo, rate 6/7

>>61961404
Because I want attention. If I didn't say I was a girl you guys would have ignored me. Now that I already got want I want, I'm kicking you beta bitches to the street. Later losers.

>>61961393
I would really leave the bang operator out for now unless you know your language wont have templates or macros or anything else that would be more intuitive with it. Although you could use the ? mark for lazy eval.
And I'd consider trashing

=> thing

as well and have all vital information on the same line as your declaration.

>>61961404
Some one tried to trick me into posting my info a couple threads ago. Used my name ^^^ as a response.

Now my shifty tablet is auto filling it. Didn't know names were faux pas on this site. Will be removing it once I'm not as soon as I stop being lazy.

>>61961385
>Also you seem to collect images of anime girls looking uncomfortable. That the definition of a red flag.
I don't understand, you posted an anime picture and somehow that's a red flag? How do you know I even collect anime pictures? I post two different anime pictures and suddenly I'm a weirdo; I could have posted them using a URL.
>Right behind cuts the eyes out of pictures with his parents and goes on 4chan.
What the fuck does that mean? Use English please.
>Besides if I was into greasy cs students i would be getting help from the ones on my campus ratherthan avoiding them to the point of going to ass hole of the Internet for help instead.
I'm not a cs student, I'm not greasy and you seem like a beta if you don't ask for help from your campus, even from tutors or lecturers, who would provide much more reliable help instead of crying on /dpt/ that no one is answering you.
>Don't mean to be mean I just want to make this clear up front so I can use these threads in piece.
>piece.
Nice English. You're a nice person btw. Totally not superficial.

>>61961440
Don't worry, us girls have got to stick together, otherwise the boys will overtake us.

To. 61961440
Youre really bad at pretending to act like a girl, I hope you know that.

>>61961461
>>61961481
>>61961470
Holy shit Im not lazy anymore see. You'llol never see me again.

Just fuck off.

>>61961434
still prototyping, but more likely than not, macros will probably nullify the static !. As for

=>name:|type|

, I personally feel it does well in specifying a return, and will help the compiler along in the long run. Plus, it looks more complete than an end or a }.

also, sorry for filling the thread with my pseudo-code, but you guys are helping me a lot with how i should go about designing things

(1..100).each do |i|
    x = ""
    i%3==0 ? x = "Fizz"
    i%5==0 ? x = "Buzz"
    put x.empty? ? i : x 
=>each

>>61961493
How old are you btw? How many dicvks have you sucked? How many guys have you fucked?

>>61961503
>the neckbeard asks at 2am with is hand cocked 'n loaded

>>61961510
A man has need y'know. It's hard to get fresh OC.

>>61961500
Nah i love discussing lang-dev/design, and its not like were cutting into any hot programming discussion anyway.
And I dont know man, after messing around in functional land for a bit, the way you do things there just feels right. Although i hate the arrows, theres really no reason why you cant just parse the right-most thing as the return type. As for that one.

|i| is a pain to type, any specific reason for the jail bars?

I suck dicks. My favorite systems language is javascript.

>>61961528
Ruby compatibility, using |i| is purely subjective.

//Using Macros & showing variable decl
mac add(a, b) 
        a = a
        b = b

        a + b
=>add

put add(1+2)

//Using classes
//notice the flip-flop use of "is", "is" is optional
cl Person is
 
    init(name) is
        @name = name
        @age = 21    
    =>init
    
     //Instead of having to declare functions for variables to be used outside of classes, use >> or .ppush
    @name>> //or name.ppush
    @age>>

=>Person

tom = Person.new "Tom"
put tom.name // "Tom"
put tom.age //  21

>>61961654
>Ruby compatibility
So why the baggage, and why not just use Ruby?

>>61961654
looks like yet another 0/10 shitlang

File: PongScreenshot.png (5KB, 644x513px) Image search: [Google]

5KB, 644x513px

I made a minimal pong clone in opengl, what do i make next?

>>61961700
what are you using to learn it?

>>61961673
again, pet project, but hopefully this will be a very low-level Ruby, with no preprocessor or VM.
>>61961679
How do i make it good senpai?
>>61961700
nice

>>61961643
i hope you suck as hard as your language.

>>61961706
bits and pieces from books and sites i guess

>>61961643
Fuck off. I do actually know some javascript though lol.
>>61961503
No way I'm telling you, 1, 2.

>>61961679
What is a shitlang?

File: Screenshot at 2017-08-12 06:25:31.png (8KB, 644x513px) Image search: [Google]

8KB, 644x513px

>>61961735
JavaScript is a shitlang

When I have an XML that contains xml-encoded HTML entities, then a proper marshaling framework should work just fine with it, correct?
Example: <Name>&amp;sup2;</Name>

I am asking because a framework I am working with throw as an exception and to fix it I am thinking about whether I want to fix the framework or my file.
But I believe the file is legit the way it is and thus the framework has to be at fault.
It seems like it unescapes the the input and then tries to interpret the entities.
But maybe that is standard behavior for JAXB?

>>61961772
>&amp;
Isn't this supposed to mean & only when inside attribute values?

>>61961779
Hm, no idea to be honest.
The parser does seem to make a difference between attributes and elements - it escapes the attributes but not the elements.
So you might be onto something here - but that would just mean the framework is fucked?

But still, I think underlying is just some JAXB implementation, so maybe that is still standard behavior.

Intellij's XML editor also automatically shows a & instead of the &camp; - so maybe the above xml example is indeed faulty?
This has me co fused.

>>61961743
Why do you guys use so much slang?

It's hard enough to learn all these new words for programing without a whole additional informal language too.

File: I just use this one for every image post I make.jpg (78KB, 640x640px) Image search: [Google]

78KB, 640x640px

>>61957754
>Haven't seriously coded in years after the HL2/Source Engine modding scene died out
>"Well maybe I'll just run some old practice problems back when I was in college just to brush up now that I've got free time"
>Spend 4 hours trying to think up solutions for something as simple as "read a C-string without passing null"
>Give up, go to sleep
>Wake up because I remembered <cstring> library exists
What the fuck is wrong with me, I was doing shit like

while( s[a] || s[b] || i < MAX ) 
{
a = s[ i ];
b = s[ i + 1 ];

I feel so stupid now and want to go back to not having to program anymore.

>>61961836
No, you're right, I checked few online XML validation tools. Entities do go into xml text just fine. The document you posted is valid XML, if the framework can't work with it, the framework is broken.

>>61961859
Your formatting triggers me

>>61961879
Can't help it, it's the only way I'm able to clearly see where brackets are. I even do it when I'm working the spreadsheets and word processor at the office.

>>61961859
Tiger so cute!!!

What is file name can't read on mobile.

>>61961875
Alright, that's what I thought too - thanks for the input anon.

>>61961893
He suffers. What's cute there?

>>61961893
"phoneposters btfo"

File: 20170818_002951.jpg (2MB, 2592x1944px) Image search: [Google]

2MB, 2592x1944px

It's all over for me. I am 1000% hacked. Was nice knowing ya. Gonna take me hours to figure out what this shit means.


>>61961931
>>61961931
He is fat and cuddly.

>>61957103
Emacs with Brin Theme and Fantasque Sans Mono font

has anyone worked with the pillow library for python? i dont quite understand it all and i would like to transform images with a sort of implode, wave, stretch, or some other kind of morph effect. is this possible?

>>61960879
not in webdev

File: 1503008757738.jpg (35KB, 603x645px) Image search: [Google]

35KB, 603x645px

>>61957754
What's an elegant way to check if a number has multiple occurences of the same digit?

Example: 1377286 -> 2x7

>>61961397
yes

>>61962082
Consecutive or any?

1737286 -> 2x7 ?

>>61962082
do you just want to check if a duplicate exists or do you want to count all duplicates

>>61962082
You could use an array to iterate through, using the index for the digit. For example roughly in C.

int arr[10], i;
for (i = 0; i < 10; i++)
    arr[i] = 0;

for (i = 0; i < numbers; i++) 
    arr[numbers[i]]++;

Then any integer in the array greater than 1 has duplicates.

>>61962120
not necessarily consecutive

>>61962144
both

>>61962082
>elegant
You just want to do the job first.
An array indexed on the digits of your base to count their occurences.
A loop to iteratively apply euclidian division by the base, the remainder being the current digit.
That's all you need.

>>61962153

function check(n) {
    var map = 0;

    while (n != 0) {
        var digit = n % 10;
        n = Math.floor(n / 10);

        var bit=1<<digit;
        if(map&bit)
            return digit;

        map=map|bit;
    }

    return null;
}

check(1737286) returns 7
check(1377286) returns 7
check(137286) returns null

>>61962152
Use calloc instead of manually setting values to zero.

>>61962082
For an actual numeric value:

while(n>10){
    digit = n%10;
    n/=10;
}

Then just store the counts of 0 -> 9 and see if any are above 1

>>61962166
>>61962152
A batter way would be to use

int arr[10]={0}, i;

>>61962165
that's a nice solution

>>61962157
>>61962152
I had the same idea, but it's not elegant enough, because I have to check billions of numbers

>>61962165
this one looks good. I will try it out

>>61962165
why is the bit = 1 << digit necessary instead of just bit = digit?

>>61962343
Because then check(51) would return 5.

(1<<5) & (1<<1) = 0
but
(5) & (1) = 1

>>61962359
isn't (1<<5) 1010, and (1<<1) 0010, resulting in &'ing them being 0010?

>>61962432
No, 1 << 5 is not 1010, that's 5 << 1.

>>61962474
oh, I thought it was the other way around. thanks anon, neat solution.

>>61961859
hours later that file name is not very satisfying.

File: 1421165704604.png (137KB, 1631x1571px) Image search: [Google]

137KB, 1631x1571px

I license my source code to anyone except Normies.
Is this legal or will I get sued?

Can we finally put an end to the "javascript is programming" meme?

13 + !0; // 14
"13" + !0; // '13true'

>>61961211

Needs attention of a tripcode, ofc it's a girl.

>>61962735
Python gets dynamic typing right:

13 + (not 0) // 14
"13" + (not 0) // TypeError: Can't convert 'bool' object to str implicitly

>>61962692
Define what normies are, and I'll tell you if it can be considered discrimination or not.

To avoid accusations of discrimination, you'd rather close your source code and only license it on demand. This way you can control how exactly your code is distributed and you don't have to justify why you don't license it to persons you consider to be normies.

>>61962758
This.

>>61962763
If a user wants a license to the source code, they contact me on IRC or by email and I ask them questions to determine if they're Normies or not.

So I guess it's on demand

>>61962735
Javascript is programming.

>>61962788
It's okay then, you don't have to justify the denial of license, as long as you don't make it too obvious why you're refusing to grant it.

>>61960183
Holy fuck this. 10 years in the business too.

>>61962792

[] + [] // empty string
[] + {} // = [object Object]
{} + [] // = 0
{} + {} // = NaN
"1" + 1 // = "11"
"1" - 1 // = 0
"1" - "1" // = 0
"1" + + "1" // "11"
"foo" + + "bar" // = "fooNaN"
"1" + - "1" // = "1-1"

For the last time. Javascript is not programming.

>>61962938
None of what you listed is anywhere near enough to make language not a programming language.

Wew visual studio 2017 finally added some of the nicer c++17 features.

File: the great fedora in the sky.gif (801KB, 360x203px) Image search: [Google]

801KB, 360x203px

>>61962982
Like Christianity, JavaScript is a nihilistic religion.

Thanks for the insight and I tip my fedora to you.

>>61962982
Wait what? Is "\t" == 0, true?
That makes no fucking sense.

So how the fuck do I make Vulkan samples run? Shit keeps telling me thinks like demo_init_vk: Assertion `!err && gpu_count > 0' failed despite having the latest mesa drivers built and everything.

>>61962982
Hey, that ain't right!

0==""

but

0!="0"

>>61963036
are you sure, though?

0 != "0"
false

>>61963029
Whitespace is ignored with type-coercing equality.

>>61963051
Yeah, you're actually right. I don't know why the js console gave me 0=="0" to be false at first, but whatever.

File: images.jpg (8KB, 163x160px) Image search: [Google]

8KB, 163x160px

>>61963062

>>61963071
Actually, it didn't like the fact that I backspaced. Shit prompt to be honest.

File: javascript good parts.jpg (54KB, 600x450px) Image search: [Google]

54KB, 600x450px

>>61963076
at least

"\t" == ""

still gives false. But honestly Javascript is a joke.

>>61963088
Has its quirks but it's a proper programming language. You seem to be focusing on its worse parts. Remember that it was Javascript that popularized closures, and now every language is adding them, even fucking C++.

File: sleepy_smug.jpg (133KB, 1275x715px) Image search: [Google]

133KB, 1275x715px

>he thinks programming will get rid of the loneliness

>>61963102
>it was Javascript that popularized closures
No it was scheme.

>>61963125
Scheme wasn't, isn't, and (likely) will never be popular, and because of that it is not possible for Scheme to popularize anything.

>>61963102
>even C++
I don't know why this is a surprise, C++ has everything whether it belongs or not.

>>61963134
We understand that you know only one language but you should go retarded somewhere else.

>>61963138
C++ didn't have closures until C++11. My opinion of C++ formed about 10 years before C++11 was released.

>>61963153
it's OK, sepplesfags hate everything before C++11 as well.

>>61963150
I know C, C++, Perl, Javascript and Java very well. I am also somewhat proficient in python in C#.

>>61963134
even if Scheme is/was not popular, SICP is/was popular

>>61963035
Well, apparently it's another case of AMD being shit and not giving working drivers. No longer surprised.

>>61963168
Hardly. It was taught in universities. And we know it had no impact because scheme didn't gain popularity. If it was popular anywhere, it's on /prog/, as a meme.

>>61962938
Type coercion craziness aside, I would unironically love to hear someone defend the "logic" behind this:

Number.MAX_VALUE + 9e100 // = Number.MAX_VALUE
Number.MAX_VALUE + 9e200 // = Number.MAX_VALUE
Number.MAX_VALUE + 9e300 // = Infinity

>>61963186
dude floats lmao

File: bday_cat.jpg (114KB, 394x500px) Image search: [Google]

114KB, 394x500px

>>61957754

I'm developing a Bjarne Stroustrup AI in C. I want to see if it comes to the conclusion that C is superior, and OOP is terrible. Also, I want to see if it realizes it fucked up the private and protected keywords again if it does go with OOP.

For some reason all it's doing is saying, "Use Gentoo." I knew I shouldn't have let Stallman in to see my AI...

>>61963226
>Also, I want to see if it realizes it fucked up the private and protected keywords again if it does go with OOP.
How so?

>>61963226
>writing AI in C
I see you're really fond of wasting your time, anon.

File: 1491054918558.jpg (70KB, 800x343px) Image search: [Google]

70KB, 800x343px

>>61963117
Lifting didn't so I'm hoping programming will.
But now that I lift and program I feel much more productive compared to before when I played vidya all day

>>61963213
And that explains MAX_VALUE either not overflowing or becoming Infinity on addition how, exactly?

Speaking of retarded numbers implementation, all numbers in JS are supposed to be 64-bit, except they're silently truncated to 32 bits on all bitwise operations:

Math.pow(2,31) // = 2147483648
1 << 31 // = -2147483648

>JS fags will defend this

File: forehead_feels.gif (6KB, 200x240px) Image search: [Google]

6KB, 200x240px

>>61963380
>tfw too poor to go to the gym
I guess I'll just work out my brain instead.

File: 1387390220.png (14KB, 456x442px) Image search: [Google]

14KB, 456x442px

>>61963380
>tfw full-time programmer, started lifting again recently, reading relevant books, started learning piano and thinking of getting driving lessons

NO TIME FOR ANYTHING ELSE

have random picture from 2013

>>61963402
lol, why even have bitwise operators at that point

>>61963402
All ints are 32 bit, all floats are 64 bit? What seems to be the problem?

>>61963284
What would you suggest instead?

>>61963475
JS doesn't have ints, just a single "Number" type which is a double precision float.

>>61963284
Should've used lisp

>>61963510
Any of modern NN libraries with easy support for GPU. Theano, Tensorflow, Torch.

I have a question about data-oriented programming versus OOP.

Lets say I have a game where there are multiple instances of an NPC, like a town guard. The guards might all share a sprite set, but would obviously have independent health and other statistics like coordinates in the world.

Would there be a really noticeable difference in performance if I decided to represent these guards as, instead of say 30 individual objects with references to static strategies and sprites, a set of 30 unique "guard" objects containing their unique information that are passed around collections ala existence-based programming depending on their state?

Correct me if I'm wrong, but when it came to actually drawing these to the window, assuming I was already using buffers, there wouldn't be a difference in how expensive the drawing process, would there? (of course that wouldn't mean that actual game logic wasn't a consideration)

t. noob who should probably just write programs to test this but can't be arsed

can one live as a self employed programmer ? I have a decent source for contracts, but is anyone living with this kind of job ?

File: 1502146404171.jpg (135KB, 1200x677px) Image search: [Google]

135KB, 1200x677px

Vala

I'm making a cli utility that both prompts for password and reads from stdin. But I can't continue reading password from stdin once I finish with what's piped in.
OpenBSD has some library for that which reads directly from /dev/tty or something like that. Is this portable on Linux? Any resources how I should solve this?

>>61963700
I assume you're thinking about data locality when you talk about data oriented programming and not the paradigm where you load your program behavior as external data. In general, OOP and what you describe aren't exclusive at all. Depending on how complex the interaction between your actors are, it might (fairly sure) not even matter in the end. If you happen to notice iterating over your actors causes trouble, you can switch to something more cache friendly with an object format where you have the often accessed unique information plus a reference to the not-so-often-used data.
Food for thought, commercial engines such as UE4 or CryEngine don't have much of this autism and the actor update loop isn't the most expensive thing.

>>61963734
Gnome discourages it's use. The language is dead.

>>61963700
I'm gonna go with no. References are the same (unless you copy the whole entity to pass it). You may probably save up on some memory though.

>>61963700
>>61963766
Follow up to the answer, if you read Game Programming Patterns there are two chapters that explain very well how to handle that kind of structure (data locality and object pool patterns).

>>61963766
Yes, that's what I mean. Thanks!

>>61963784
So this might be more relevant with, say, 300 guards on screen at once than 30

I found out about data-oriented programing a while ago, and really want to get into the habit of writing something other than pure OOP where everything is an independent object (although as the first replier said I realize they aren't necessarily exclusive and I should probably find a better way of articulating this)

File: steppedonsnek.jpg (36KB, 640x427px) Image search: [Google]

36KB, 640x427px

>Note that Python 3.5+ cannot be used on Windows XP or earlier.
How shit a language must be to not work on Windows XP?

>>61963821
>Windows
kek

>>61963821
>windows XP
is it 2002 again?

>>61963821
XP a shit.

>>61963821
A lot of languages that target a newer version of the mscrt won't run on xp. (Vista is the minimum version). so XP, 2000, 98, etc).

New thread:

>>61963889
>>61963889
>>61963889

>>61963732
I'm technically self-employed but for me it means contracting for someone else while doing own taxes &c. So basically you sign contracts, either 6 months or indefinite or whatever. It's not like those shitty freelance sites you see where it's "need to fix website, paying 200 bux" and you're out of work after a day.

Updating a program written by a guy who died. The comments are a shadowy flight into the world of an autism who does not exist. It's mostly just musings and him LOLing at how he needs to "// optimize dis shiiiit before someone else works on dat mane". I hate this, I'd rather overfuck a capacitor in my urethra man. How the fuck do I pass array variables between methods in a c# program? Just define a global array variable at tge beginning of the environment and pass it around between the public and private voids like the whore that array is? Or what

>>61963955
use a List<T> or use ref.