/dpt/ - Daily Python Thread

reimplementing Algol68 in rust

>2017
>C

You have been asked to print the sum of two numbers.

Normal people:

void print_sum(T)(T x, T y)
{
    writeln(x + y);
}

Mentally ill C tards:

void print_sum_ints(int x, int y)
{
    printf("%i", x + y);
}

void print_sum_floats(float x, float y)
{
    printf("%f", x + y);
}

void print_sum_float_int(float x, int y)
{
    printf("%f", x + y);
}

void print_sum_int_float(int x, float y)
{
    printf("%f", x + y);
}

>>60283665
Shitty language, why would you have to write down T?

>>60283665
>silent coercion if you use an int and a float
The C version is safer tbqh

>>60283679
Because any good language will have lambdas which can take types as arguments.

>>60283679
You mean (T) or the T in

T x,
 T y)

?

>>60283721
I applaud your discerning taste. A woman after my own heart.

>>60283665

In Python

def printSum (a, b):
print(a + b)

>>60283734
Both

>>60283721
Seems cancerous.

>>60283721
That's generics not lambdas you cretin

>>60283746
So youve never written a template then.

>>60283763
I did in the past, however as I kept learning about CS I discovered how shitty they are. Seriously, why on earth would anybody use this kind of crap?

>>60283665
>>60283737

print_sum :: (Semigroup a, Show a) => a -> a -> IO ()
print_sum x y = putStrLn $ show $ x <> y

>>60283774
>I discovered how shitty they are
t. brainlet
Theres a reason why templates are called meta-programming.

see>>60283665

>>60283648
wat do

>>60283788
>t. brainlet
I would suggest you to read about type systems. Seriously, templates should not be allowed.

>Theres a reason why templates are called meta-programming.
Because they happen before the compile time?

>see>>60283665
What does this prove?

>>60283665

#define PRINT_SUM(name, t1, t2, f)\
    void print_sum_ ## name(t1 x, t2 y)\
    {\
        printf("%" # f, x + y);\
    }

PRINT_SUM(ints, int, int, i)
PRINT_SUM(floats, float, float, f)
PRINT_SUM(float_int, float, int, f)
PRINT_SUM(int_float, int, float, f)

>>60283746
>Seems cancerous.
Why is that?

>>60283758
>generics
I don't recognize this as a valid word.
>not lambdas
So?

>>60283870
>macros
times_macros_saved_Cs_ass++

>>60283853
>I would suggest you to read about type systems.
I use dependent types though, but we dont live in a world, where i can forego imperative typelet languages yet.

>>60283873
>Why is that?
Because you should be able to do

print_sum(x, y)
{
  writeln (x + y);
}

instead.

>>60283910
What if you need more than one different type?
In this case minimalism will do more harm than good.

>>60283910
In what way does this explain how functions accepting types as arguments is "cancerous"?
>your code
Type inference is for sub 80-IQ people. An addition operator on all number types is even worse.

>>60283910
That doesn't make use of generics. In fact, it's enabling implicit type inference universally, which is not safe for languages that encourages side effects

>>60283932
>What if you need more than one different type?
It will work exactly the same way. This should be a valid function call as long as the + operator can be applied to both x and y with each other.

>In this case minimalism will do more harm than good.
How so?

>>60283945
encourage*

>>60283945
>generics
What is this? Please translate it into non-Java/subhuman talk.

>>60283970
Learn how to use google, dumb mongoloid.

>>60283942
The type is already known as the values that you pass to the function have one.

>>60283945
>That doesn't make use of generics
What do you mean by that?

>which is not safe for languages that encourages side effects
I am interested to know why.
That being said, I prefer languages that avoid them.

>>60283997
>I am interested to know why.
If you are encouraging side effects, you should really keep track of your types. Many functional languages don't have the risk since the type would not change across the project.

>>60283997
So? How does this explain that functions taking types as arguments is "cancerous"?

>>60283994
I tried to, but there was no explanation in non-retard speak. Could you translate it? You seem to know their language.

>>60284030
>If you are encouraging side effects, you should really keep track of your types
Why though?

>since the type would not change across the project.
What do you mean by that?

>>60284040
Because it doesn't make any sense. You already have the type in the parameter, why would you pass it again?

>>60284053
Well in that case you should take English classes, preferably from non-mongoloid "teachers".

How do people usually handle shared model information (i.e. user credentials and cached data, configs, etc.) in MVVM?
My go-to would be something like a Model Locator, like in MVVMLight

>>60283951
Because you have no information about x,y and this really only works for simple two var functions.

When you got more complicated and need something like

do_something_more(x,y,a,b,c)
{
  //twenty lines of stuff
}

Youre stuck looking through the definition for information about how the types work. When you could do something like

do_something_more(T x,T y, N a,N b, R c)
{
  //twenty lines of stuff
}

>>60284077
If your int turns into a float in some point and you don't know about it you will get incorrect results

Rate my Rust:

mod mylib {

    pub trait Argument<F,O> : Sized
        where F : Fn(Self) -> O
    {
        fn apply(self, foo : F) -> O;
    }

    impl<F,O,T> Argument<F,O> for T
        where F : Fn(Self) -> O
    {
        fn apply(self, foo : F) -> O {
            foo(self)
        }
    }
}


#[cfg(test)]
mod tests {
    use mylib::Argument;

    fn triplicate(x : i64) -> i64 {3*x}

    #[test]
    fn it_works() {
        assert!(6 == 2.apply(triplicate));
    }
}

My pipelines with dot notation will never be interupted again.

>>60284095
noice

>>60284095
not idiomatic enough / 10

>>60284085
>Because you have no information about x,y
You do, you know that + should work with both x and y, thus limiting the sets that x and y may belong to.

>Youre stuck looking through the definition for information about how the types work
Read the documentation maybe? Or use an editor that can tell you the type?
In any case, if someone wants to type down the types explicitly this is fine, however what is not fine is forcing everyone to do it.

>>60283665
>Make parameters float.
>Cast integers to float when giving arguments.
>??????????
>Prophet

>>60284094
I would say that if your ints turn into IEEE floats randomly and the reverse that there is something wrong with this language.
I can't see how writing down the types would help you.

>>60284095
>pub trait Argument<F,O> : Sized where F : Fn(Self) -> O
>impl<F,O,T> Argument<F,O> for T where F : Fn(Self) -> O
eww

>>60284154
>converting integers to float
sign of a bad programmer

>>60284206
How else would you compute interest on that gigantic pile of money that your clients have entrusted you with, anon?

>>60284206
There is a reason, dumbass.
It's only to fix C's handicaps regarding polymorphism.
In good languages like CL, this shit is unheard of.

>>60284259
>It's only to fix C's handicaps regarding polymorphism.
Oh, makes sense then, hahaha

>>60284281
Can I call you Rusty?

Oh god damn it.

Guess what Rust's notation for accessing say the zeroth or the second element of a tuple (0,1,2,3) is.

It's (0,1,2,3).0 or (0,1,2,3).2 , respectively.

>>60284344
Theres a reason rust is named after fungus.

>>60284344
Can you do (0,1,2,3).N where N is a variable?

>>60284344
whats wrong with that?

>>60284344
Rust sounds cool to be honset.

>Does Rust do tail-call optimization?

>Not generally, no. Tail-call optimization may be done in limited circumstances, but is not guaranteed. As the feature has always been desired, Rust has a keyword (become) reserved, though it is not clear yet whether it is technically possible, nor whether it will be implemented. There was a proposed extension that would allow tail-call elimination in certain contexts, but it is currently postponed.

>How do I do O(1) character access in a String?

>You cannot.

>How can I define a struct that contains a reference to one of its own fields?

>It’s possible, but useless to do so. The struct becomes permanently borrowed by itself and therefore can’t be moved.

>What are higher-kinded types, why would I want them, and why doesn't Rust have them?

>The lack of support for higher-kinded types makes it difficult to write certain kinds of generic code. It’s particularly problematic for abstracting over concepts like iterators, since iterators are often parameterized over a lifetime at least. That in turn has prevented the creation of traits abstracting over Rust’s collections. ... [T]here’s no inherent reason for the current lack of support.

Why would anyone use this language, again?

>>60284326
Go ahead

>>60284369
No. They are field names/compiler magic.

>>60284364
>The power of American education

>>60284408
>>How do I do O(1) character access in a String?
>>You cannot.
sjw's picked a shitty language to latch onto

>>60284408
More usable than haskell

>>60284429
What about Krusty?

>>60284453
It's because strings are UTF-8 encoded, and since UTF-8 is a variable width encoding, you cannot get O(1) lookups.

I'm learning Lua for my first language and I like it!

>>60284462
Not really though, Haskell has HKTs.

>>60284472
Haskell doesn't have inline ASM

>>60284448
not an argument.

>>60284453
You mean: the Unicode UTF-8 standard is shit.

>>60284466
That's retarded

>>60284484
then don't apologize for rust making it the basis for their strings

>>60284484
>>60284488
I'm not here to defend Rust, but UTF-8 is by far the most sane text encoding.
Using anything else would be fucking retarded.

>>60284408
>>Does Rust do tail-call optimization?
W-What? Pretty sure that eliminating the tail call is trivial in any case. Or do they talk about changing a non-tail call into a tail-call?

>>You cannot.
Blame unicode.

>>and therefore can’t be moved.
Moved?

>>60284481
Arguing with an illiterate is like arguing with a monkey

>>60284496
It shouldn't be the standard string though. UTF-8 is fine.

>>60284439
Why would anyone use that instead of let's say functions or arrays?

>>60284466
True for any unicode encoding.

File: 1486586820925.jpg (22KB, 500x396px) Image search: [Google]

22KB, 500x396px

>>60284477
B T F O
T
F
O

>>60284493
UTF-8 is the standard string format nowadays. The website you are posting on uses it.

>>60284477
Inline ASM is not portable, so it's useless.

>>60284477
>>60284514
wall.org/~lewis/2013/10/15/asm-monad.html

>>60284496
you just gave a bunch of reasons utf-8 is shit compared to ascii, and then say utf-8 is better than ascii. you have to give more reasons for it being better than you gave for it being worse

>>60284522
>Inline ASM is useless
This what a brainlet would say.

>>60284484
All the Unicode standard is retarded.

>>60284496
>Using anything else would be fucking retarded.
Why? It's not even ASCII compatible.
The best would be a UTF21 without retarded shit like emojies or other characters that are made up of multiple codepoints.

>>60284509
>It shouldn't be the standard string though.
What on earth are you proposing?
UTF-16, which is infinitely more retarded than UTF-8 and still doesn't have O(1) lookups?
UTF-32, which takes 4 times as much space to encode english text?
ASCII, which makes you have to deal with garbage like code pages or locales if you want to go beyond it?

>>60284531
>Inline ASM is not useless
This is what a codelet would say.

>>60284525
That's not inline assembly, that's using Haskell to generate assembly.

>>60284540
The first 128 chars of ASCII, nothing else. The extra bit in each byte can be used for parity checking.

>>60284535
>Why? It's not even ASCII compatible.
UTF-8 is ASCII compatible, fuckface.
>UTF21
>Non-power of 2 encoding
Wow, that would be so much better.

>>60284540
>What on Earth are you proposing
A string being an array of characters, like most languages. You can also have a utf-8 package if you want to use it.

>>60284555
>The first 128 chars of ASCII
That's literally all there is to ASCII. ASCII is a 7-bit encoding.
>The extra bit in each byte can be used for parity checking.
It's been decades since anybody has used it for that. It's not like we're using it over a lossy medium.

>>60284578
>A string being an array of characters
Are you fucking retarded? That doesn't answer the question at all.
How the fuck would those characters be encoded?

>>60284561
>UTF-8 is ASCII compatible, fuckface.
No, it isn't. ASCII is 7 bit. UTF-8 is 8 bit. You can't read ASCII files with a UTF-8 decoder nor you can read UTF-8 files with an ASCII decoder. Please learn more about a topic before you hurry to insult someone.

>Non-power of 2 encoding
I see no reason on why it should be one.

>>60284593
It doesn't matter, as long as they have constant size for constant access time.

File: asm.png (9KB, 266x216px) Image search: [Google]

9KB, 266x216px

>>60284553
jit :: [Word8] -> IO ()

>>60284541
Entry# 0905.60283641
>I claim Haskell dosn't have inline ASM
>Hasklet responds inline ASM is "useless"
>I infer it has not made any substantial projects yet. Probably spends all day in /g/ gate keeping /dpt/
>I call him a brainlet for needing portability
>It responds and calls me "codelet"
>Brainlet is not even making any sense anymore
>I realise it lost its arguments and I call it a victory
>I will post this entry and see it reacts
>It will be fun
end

http://www.strawpoll.me/12925856

>>60284600
>No, it isn't. ASCII is 7 bit. UTF-8 is 8 bit. You can't read ASCII files with a UTF-8 decoder nor you can read UTF-8 files with an ASCII decoder. Please learn more about a topic before you hurry to insult someone.
This is so retarded, you don't even deserve a proper response. (You).

>>60284613
Basically what you're proposing is UTF-32, which is incredibly wasteful. The downsides of UTF-32 outweigh the downsides of UTF-32.

>>60284553
https://hackage.haskell.org/package/harpy

>>60284593
He is actually pretty smart actually as this seems like the most sane solution. This or storing the positions of the start of each character separately in the string struct inside an array.

It does not matter how the characters would be encoded, any encoding would work fine, as long as one full character was stored. Sadly this is not possible if you consider a character to be one unicode codepoint.

>>60284626
Then ASCII it is

>>60284620
I'm saying that it is a poor programmer who believes inline ASM is useful.

I realize that, being a poor programmer, you may lack the cognitive capacity to understand this.

>>60284626
>This is so retarded, you don't even deserve a proper response. (You).
How so? It seems correct to me.

>Basically what you're proposing is UTF-32
No, he is not.
1: UTF-32 still does not guarantee constant-access and
2: You only need 21 bits.

>>60284618
Oh, I see. It just looked very similar.

>>60284626
>The downsides of UTF-32 outweigh the downsides of UTF-32
The downsides of UTF-32 outweigh the downsides of UTF-8*

someone make a "pajeet my son" meme with pajeet being rust, and having to choose between beautiful constant time character access in strings, or emojis

>>60284642
+
>it doesn't know that portability is for people that are too lazy/afraid to do programming
>I'll let the brainlet live inside the bubbled world it lives in

>>60284657
post the comic and i will

>>60284657
I've seen pajeets using C more than Rust

>>60284644
>1: UTF-32 still does not guarantee constant-access and
Yes it does, you fucking retard. Do you not know anything?
>2: You only need 21 bits.
The major hassle of using an encoding that does not fit in 8, 16, or 32 bits far outweighs any advantages. Nobody would use that garbage.

>Implement a queue by a singly linked list L. The operations E NQUEUE and D E -QUEUE should still take O.1/ time.
do you think i'm allowed to keep a reference to the end of the list or is that cheating and i need to do something fancy?

>>60284664
This is how I know you're not a professional programmer.

>I use assembly language because I'm a real programmer and such a l33t h4x0r

>>60284667
on my work laptop so i dont want this shit saved to my hd
http://imgur.com/9dK42gz

Rust lets you convert back and forth between a unicode string, and an array of u8 ints which does have random access. It also gives you char literals for u8 integers.

If you know that all your unicode characters happen to fit in ASCII, you can treat it as an u8 array just fine.

>>60284670
because pajeet doesn't have enough spare money to evangelize for a DOA language

>>60284673
>Yes it does, you fucking retard. Do you not know anything?
Sure I do, more than you it seems.
Consider U+0065 and U+0301, two different unicode code points that results in é, one character.

>The major hassle of using an encoding that does not fit in 8, 16, or 32 bits far outweighs any advantages
Doubtful.

>>60284642
>I'm saying that it is a poor programmer who believes inline ASM is useful.
Okay then, write a Hello World program in C that does not have any external library dependencies.

>>60284795
gcc -static

>>60284664
>>60284691
Inline assembly is not the same as using assembly as your language of choice. It's for situations when you require features that simply aren't present in the language.

how the FUCK do i do this?
>Give a ‚.n/-time nonrecursive procedure that reverses a singly linked list of n elements. The procedure should use no more than constant storage beyond that needed for the list itself.
creating a new list to reverse through appending to the front would take linear storage, and i can't think of any other way to do this in linear time

not homework, classes are already over

>>60284812
Doesn't count. You still need the library file at compile time.

I wish we had ternary computers with seven-trit bytes. Then the 21 bit unicode encoding would have fit just fine in two bytes.

Challenge: Reverse a singly linked list in O(1)

>>60284841
with constant storage, fuckface

>>60284841
Can I assume that I have a quantum computer and the ability to copy a qbit without measuring it?

>>60284841
autism

>>60284841
>>60284849

void reverse(list* thelist) {
  thelist->reversed = !thelist->reversed;
}

>>60284841
>Modify n elements in constant time
Hmmmm

>>60284866
Stallman-tier challenge mode: accessing a pointer is O(n)

>>60284841
Using a GPU with infinite parallel operations, and assuming that the linked list is the sole occupant of a memory pool:

For each pool element, check that it is occupied by a link. If it is, dereference its .next pointer. In the node pointed to, change that to point back to the node you had. The node with a pointer that was null is your new head.

Use your old head pointer to set the pointer of the old head to null. You are now done.

>>60284921
i surrender
take me to your leader

>>60284948
>with infinite parallel operations
Just n should be enough.

>>60284981
>>60284921
>>60284866
in the first place, that solution only works with a doubly linked list.

>>60284915
This. Physically impossible

>>60285019
no

>>60285097
Unless your ``singly linked list'' is an array, there's no fucking way the nodes know what came before it.

>>60284822
bubble sort anon ;) But it's n^2.
I do not understand your notation. What is the time complexity limit wanted?

>>60283664
reimplementing rust in Algol68

>>60283873
wtf is this cancer posting?

>>60285174
What did you mean by that?

>>60285165
it got fucked copy pasting from the pdf. i looked it up online and the "solution" is to not use a true singly linked list, but to use one with a sentinel end node

>>60285157
and?

>>60285361
>implying an ArrayList is the same thing as a List

>>60285602
what have arrays got to do with anything i wrote?

File: 1371956854343.png (190KB, 550x550px) Image search: [Google]

190KB, 550x550px

>>60285668
>solutions are totally independent of implementation, haven't you ever heard of academia :へ)

>>60285716
what? what even is an arraylist?

>>60285602
>ArrayList
wut

>>60285726
I believe it's what Java calls vectors.

Slowly learning how to Python so that I may extract potential black hole gravitational lenses from this database:
http://cdn.gea.esac.esa.int/Gaia/gaia_source/csv/
I've yet to learn file handling, I'd also like to do the downloading automatically
I'm a complete noob who is just starting to code this semester tho

>>60285737
Why would anyone call vectors like that?
And nobody talked about java either, I don't get it.

>>60283665

C does not have function overloading, as this would require name mangling, and could lead to ABI issues between compiler versions as is the case with C++. C++ can do this trivially though...

void print_sum(auto x, auto y)
{
    auto sum = x + y;
    std::cout << sum;
}

>>60285772
Disgusting.

>>60285772
Ruby, my man! Hope the beard's going well. Hey, I got a question for you, since you're actively in academia whereas I've been outta the loop for more than half a decade. Is Rust gaining any traction in those ivory towers in which you dwell? Or is it seen as just another uninteresting language created by and for industry, full of mistakes and design flaws?

>>60285819
Literally nobody talks about Rust, except for the zealots on the internet.

>>60285884
The internet is more important than real life these days.

>>60285903
That's what the zealots wants you to believe.

>>60285903
Deep

>>60285819
This is true

>>60285772

printSum x y = print (x + y)

>>60285884
This
t. phd student

>>60285819

None of my professors really talk about Rust. The one professor who has talked about it is mostly a C and NetBSD fan who doesn't like it. I can't remember his reasons, since it's been a while since I brought it up.

I remember last quarter, all of the master's degree students had to review the new incoming professor candidates as part of our seminar class (which normally has us watching some video on a topic in computer science and then discussing it for an hour once a week). One of the candidates was talking about his research using Prolog to prove correctness in programs. Among the things he was able to prove was that there was a fundamental flaw in the Rust type system. I can't remember if he went over what that flaw is, but I do know that they haven't fixed it and can't really fix it.

>>60285987
>there was a fundamental flaw in the Rust type system
>they haven't fixed it and can't really fix it
Sad, but not really surprising. Thanks for the info bruh.

Sorry for calling you a woman before.

http://ix.io/t9s/prolog
killing myself trying to get this damn shit to work. maybe i should just declare that my type system is undecidable and give up?

>>60285987
What uni?

>CLion regularly starts using 200% CPU
I'm not even using it for C++, just plain C, what the fuck is it doing??

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

139KB, 470x545px

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

File: project.png (108KB, 1532x607px) Image search: [Google]

108KB, 1532x607px

I've been writing an implementation of Ext2 in Haskell which is finally starting to work. I'm so happy, I've put so much time into this.

>>60286056
>she's going to have her GC tumor removed (eventually)
Does she live in a country with socialized healthcare? She's been waiting 16 years for it so far

>>60286027
It was rebuilt using electron.

>>60286078
Well shit.

File: d died.png (662KB, 940x590px) Image search: [Google]

662KB, 940x590px

>>60286056

>>60286071
She was born and raised to work with a GC, and her parents have been dragging their feet on forking over the money for surgery. Supposedly her current self, Phobos, is going to be the one where they remove it.

I want to prototype a flowchart-esque program I had an idea for, except my GUI experience is extremely lacking, what language/tools would be best/easiest?

>>60286021

I am currently finishing up my Master's Degree in Computer Science at Western Washington University. I will be starting my PhD at Washington State University in the fall.

>>60286056
She kinda looks like a boy.

>he doesn't like garbage collection
It's because it collected you, didn't it.

>>60285772
Why would you have to write std::cout if you already have imported std?

Is it to prevent the possibility that another abc::cout may exist?

>>60285987
>Among the things he was able to prove was that there was a fundamental flaw in the Rust type system
Elaborate?

>>60286145
GUI? C#

>>60286163
Are you a memorylet who can't manage your own memory?

Thinking of doing ncurses client to this shithole.
My current problem is displaying posts.
Somehow, I'm supposed to scroll the posts, who vary in dimensions, in a smooth way.
Maybe I should make the posts's dimensions fixed, and have a (read more) keybinding.
Thoughts?

>>60286216
Inane.
I'd rather think about the program than idiosyncracies of obsolete languages.

>>60286216
This is a bad argument

write a more optimized ring buffer

BUF_SIZE        EQU 16

        ORG $2000-(BUF_SIZE+1)

buf:
        DCB.B   BUF_SIZE,$00

push_to_buf:
        ;; takes param in d0
        ;; modifies no registers
        move.b d0, buf.W
        addq.b #1, push_to_buf+3
        beq.b reset_push
        rts
reset_push:
        move.w #buf, push_to_buf+2
        rts

pop_from_buf:
        ;; takes no params
        ;; returns value from buffer in d0
        move.b buf.W, d0
        addq.b #1, pop_from_buf+3
        beq.b reset_pop
        rts
reset_pop:
        move.w #buf, push_to_buf+2
        rts

>>60286017
Why are you using a classical system?

>>60285987
So he was able to "prove" his theory to his gullible students and not to the actual project maintainers? I'm sorry that statement of yours holds nothing of value, as that supposed proof

>>60286250
what would you suggest I use in its place?
i see no particular issue with Prolog that has been better solved by another language

>>60286153
She's a genki tomlang!

>>60286163
Garbage Collection is not a bad tool. It only becomes detrimental when you have realtime and intensive shit to process and/or you're lacking in system resources.

>>60286174

>if you have already imported std
Technically, you don't import a namespace. And I never wrote anything along the lines of:

using namespace std;

It's not something I tend to do, given the number of other things in namespace std that are possibly names for other things (string, vector, count, sort, etc...).

>>60286184

>I can't remember if he went over what that flaw is
Important shit here. It was a 1 hour lecture that was several months ago. I'm sure if and when the guy's work gets published, all of the Rust fanboys will hear about it.

>>60286239

I've had Java programs crash with out of memory errors that were fixed by forcibly calling GC.collect(); These sorts of problems would not exist in a language with RAII. As soon as the memory is no longer being used, it should be available for use again.

>>60286275

>So he was able to "prove" his theory to his gullible students and not to the actual project maintainers?
As I recall, he did prove it to the project maintainers. The problem was that it's a fundamental flaw in the language's design.

>I'm sorry that statement of yours holds nothing of value, as that supposed proof
Again, my memory of the lecture is very hazy.

>>60286303
>>60286333
This is why you have a language that allows you to change the implementation of gc.

>>60286275
>Being this ass-hurt over a tool

What is the fastest functional programming language?

I am talking about one that's actually faster than python

>>60286410
>python
>functional

>>60286421
Top quality reading comprehension skills there

>>60286427
It was implied.
But try Haskell or Common Lisp.
Both are faster than Java.

>>60286279
>what would you suggest I use in its place?
Coq. And I've heard Twelf is good for proving meta-theorems about programming languages.
>i see no particular issue with Prolog that has been better solved by another language
Affirming LEM is a pretty big issue. Also I suspect proving things about a constructive system would be more natural in a constructive system.

>>60286432
>faster
you mean slower, right

>>60286457
i'm pretty sure he meant "faster", as in "not slower"

>>60286462
Well, he is absolutely wrong

>>60286410
x86 ASM is pretty fast.

redpill me on the CR3 register, stat!

File: 1453303517244.png (48KB, 336x280px) Image search: [Google]

48KB, 336x280px

What's the best general purpose stable sorting algorithm?
I've been looking at timsort, but I want to know if anybody knows of anything better.

>>60286475
i don't know, i was just correcting your misunderstanding of his words

>>60286442
irrelevant to the use-case.
Coq and Twelf are more work than it's worth to get something working since i'm gonna rewrite this in another language anyways

>>60286514
DOGGO

How's your OCaml-killer going?

>>60286544
it's not. it's an Erlang killer

>>60286563
so in other words it's shit?

>>60286475
Explain why I'm wrong.

>>60286482
Intel dev manuals

>>60286576
at the moment, yes

>>60286514
>irrelevant to the use-case.
It's relevant to every use case.

>>60286626
anything which is related to Erlang will be shit independent of time.

File: 91153.jpg (28KB, 225x350px) Image search: [Google]

28KB, 225x350px

I have a program that takes an input as a string, and chooses one of these strings as random to output. Anyone have any ideas to make it more interesting/features to add.

>>60286706
why is that?
either way, i could easily sell my language as a commercial product at the very least (assuming that I complete it and it's not totally retarded when im done)
to a software company trying to create extremely stable software, the combination of the BEAM VM, OTP, and complete static typing of multiple simultaneously-communicating actors seems like it could bring in quite a few dollars

>>60286735
>random
Lol

>>60286776
???????

File: consider_suicide.png (628KB, 1280x620px) Image search: [Google]

628KB, 1280x620px

>getting weird ass segfault errors
>litter my code with print statements and still can't find the problem
>turned out I put <= inside a for loop instead of <

>>60286860
><=
This shouldn't really be a separate operator.

>>60286868
is it
>=
or
>>=

>>60286919
>=>

File: 2017-05-05-234521_581x685_scrot.png (17KB, 581x685px) Image search: [Google]

17KB, 581x685px

apparently I'm retarded, cannot seem to replicate this image in python

>>60286776
>good pseudo-random algorithms aren't effectively random for the end user
Can we end this meme? Why are you assuming he isn't using radioactive decay as the source of random data anyway?

>>60283641
is there a way to compile libtorrent without dealing with the clusterfuck of boost bullshit?

>>60286919

Which is more unsightly,

This:

a <= b

Or this:

(a < b) || (a == b)

>>60286492
Heapsort a best

>>60283641

(define (sum-of-squares a b c)
  (display (+ (* a a)
              (* b b)
              (* c c))))
(sum-of-squares (3 4 5))

>Exception: attempt to apply non-procedure 3
b-but the syntax is correct...
Chez Scheme Version 9.4 btw.

>>60287153
It is not, you want to do

(sum-of-squares 3 4 5)

instead.
Right now you are trying to call a function named 3 with the arguments 4 and 5.

>>60287091
>==
This should be banned and outlawed.
The more correct version would be:

(a < b) ∨ (a = b)

Also, even '+' would be better than "||".

>>60287150
It's not stable.

>>60287173
And?
Also, pretty sure that there are stable versions of it. If not go with a binary tree sort or something.

>>60286992

import numpy as np
import matplotlib.pyplot as plt

stepCnt = 1000

xs = np.empty((stepCnt + 1,))
ys = np.empty((stepCnt + 1,))

xs[0], ys[0] = (1, 1)

for i in range(stepCnt):
    xs[i+1] = 1 - abs(ys[i] - xs[i])
    ys[i+1] = 1 - abs((1-xs[i])-ys[i])

plt.plot(ys, xs)

plt.show()

wat am I doing wrong?

>>60286992

Hrm...

# Assumption 1: X is as true as Y (X = Y)
# Assumption 2: Y is as true as X is false (Y = !X)

# +-------+-------+-----------+-----------+---------------+
# |     X |     Y | Assump. 1 | Assump. 2 |     Result    |
# +-------+-------+-----------+-----------+---------------+
# | True  | True  | True      | False     | Contradiction |
# | True  | False | False     | True      | Contradiction |
# | False | True  | False     | True      | Contradiction |
# | False | False | True      | False     | Contradiction |
# +-------+-------+-----------+-----------+---------------+

>>60287153

(apply #'sum-of-squares '(3 4 5))

>>60287208
the point is that you iterate it silly

>>60287171

>∨
Not ASCII, not available on any keyboard, complete trash.

= is the assignment operator, and should not be used for equality comparison.

@60287208
>Truth table
Immediately discarded. Try better next time.

File: 2017-05-05-230617_573x694_scrot.png (47KB, 573x694px) Image search: [Google]

47KB, 573x694px

>>60287244
>>60287208
>>60287200
heres another example

http://ix.io/t9T/prolog
I'm getting closer to having a working thing but still, FUCKKKKKKKKK
>>60287258
why does your language have assignment?

>>60287258
>not available on any keyboard
Your keyboard is complete trash.
>= is the assignment operator
This is blatantly false. := is the assignment operator. = is and has always been propositional or sometimes definitional equality.

>>60287170
dumb me, thanks senpai.
>>60287238

(apply sum-of-squares '(3 4 5))

not sure what the # is supposed to do yet.

>>60283665
No, actually the C way would be to take two void pointers which are interpreted as the appropriate types based on a string or flags variable.

>>60287385
Implement it using the true C way. Preferably with more than 1 star.

File: 1350594293147.jpg (109KB, 500x500px) Image search: [Google]

109KB, 500x500px

Is it possible to add two integers in C?

>>60287419
No ;_;

>>60287419
only if they're both 0

>>60287348
how do you get the function value of a symbol if not with #'

>>60287419
I don't understand this joke.

>>60287419
Maybe.

>>60287462
It's not a joke...

>>60287419
Define add.
Define integers.

>>60287489
The below average user, myself, would say that you can do:

int a = 9;
int b = 4;
int c = a + b;

This would appear to be integer addition to me, but you imply it not to be so. What is the problem?

File: 1453657295316.jpg (112KB, 1920x1080px) Image search: [Google]

112KB, 1920x1080px

>>60287594
>What is the problem?

>>60287385
No, actually the C way would be to take two null pointers and produce another security vulnerability.

The true C way.

>>60287419
No, not really

>>60287494
>Define add.
An operation which given x and y of type integer applies the successor function to x y times.
>Define integers.
The set of integers from set theory or anything equivalent will do.

>>60287594
Yup, that seems right assuming the answer is 13 of type int. Thanks!

>>60287621
But didn't this poster just do it? >>60287594

File: __flandre_scarlet_kirisame_marisa_and_remilia_scarlet_touhou_drawn_by_yamato_damashi__e1f72569fe22a25a237bf6b108cd2716.png (342KB, 1100x768px) Image search: [Google]

342KB, 1100x768px

>can't even add two integers in C
>while you can multiply an int and a complex number in C++ without any trouble
Why do people think C > C++ again?

>>60287633
That program can't really add anything but a certain range of numbers

>>60287300
Because my language doesn't suck.

>>60287308
>Your keyboard is complete trash.
My keyboard is standard US QWERTY. Every other keyboard is trash.

>:= is the assignment operator
Fortran was the first real programming language, and it uses = as the assignment operator. C uses = as assignment as well, and numerous top 10 languages base their syntax on C's. The standard assignment operator is =, and the standard equality operator is ==. Everything else is hipster shit.

Does SHITKELL have pointers?

File: chen_rape_face.png (54KB, 241x235px) Image search: [Google]

54KB, 241x235px

>>60287633
>the set of integers
>set
Undergrad detected.

>>60287664
Yes, but it's an unsafe operation.

>>60287664
No

Is there a algorithm like shortest path algorithms that can be used for finding the most target nodes you can go through without going over a certain amount of distance?

Something like Dijkstra or Bellman-Ford but that is not based on visiting all nodes. Instead I need to start at the starting node then end at the ending while traveling through the most target nodes possible.

>>60287644

For both C and C++, it's all really a library issue for adding arbitrary precision integers. For fixed width, it's baked into the language.

>>60287664
Yes, and it even has NullPointerExceptions

https://hackage.haskell.org/package/structs-0/docs/Data-Struct-Internal.html

>>60287655
Is there a type which will allow me to add anything I want as long as I have enough memory?
>>60287682
Yes, I know sets are laughable garbage, which is why I said anything equivalent.

File: chen_face.png (68KB, 289x398px) Image search: [Google]

68KB, 289x398px

>>60287739
Integers are a ring you fucking mong LMAO smfh desu if you don't know how to prove the fundamental theorem of arithmetic using categories and Zorn's lemma you have no business pretending to be in STEM lfmao

>>60287746
sets are a semiring

Does C have auto?
auto x = a + b

>>60287777
are sets semigroups though?

>>60287419
Depends on what you mean by integer.

>>60287419

int add(int x, int y)
{
    return 0;
}

valid for 4,294,967,293 inputs (implementation-dependent). if that seriously isn't enough for you, i mean come on. at that point you're just being nitpicky

>>60287777
Wrong, freshman.

>>60287825
Wrong, preschooler

>>60287399
https://pastebin.com/V1CeAndX

>>60287851
Wrong, embryo

>>60287269
>>60286992
>>60287200
pls someone help

>>60287781
No. C does have an auto keyword, but it basically means allocate variables on the stack, and it's only legal within a function (where variables are allocated on the stack anyways). So it has no actual effect, it's left over from the B programming language I believe.

>>60287867
Wrong,「 」

>>60287711
yeah. use the same techniques used in those other algorithms

>>60287602
BEST GIRL

>>60287949
>desu eyes
back to /a/

File: 1470050121990.jpg (1MB, 2048x1536px) Image search: [Google]

1MB, 2048x1536px

>>60287857
>public domain

>>60287664
Yes but use IORef instead

>>60287746
>Integers are a ring
There is definitely a s*t of integers. I don't need to mention additional irrelevant information when I'm trying to convey an idea to someone who might not be familiar with other terminology.
>you have no business pretending to be in STEM
I don't study anywhere, I'm not subhuman.

>>60287975
but 4chan is /a/

C
The c(ulprit) of all security bugs

>>60288069
Not anymore, gramps. Times have changed.
It's time you accept the reality.
/pol/ and the alt-right domm now.

File: 1494198077023.jpg (32KB, 480x360px) Image search: [Google]

32KB, 480x360px

>>60288151
>t. rustfag

>>60288177
>not rustacean
Gent your pronouns right, creepe!

>>60288177
>Rust is the only language I know of focusing on safety
Really telling, Anon.

Why in the world is it so hard to find file formats for audio files like mp4, m4a, mp3, etc? Google only gives me shitty programs to convert shit but that's not what I want at all.

>>60288597
yeah, i would like to bump you... if you catch my drift

File: 1473378705661.png (24KB, 500x500px) Image search: [Google]

24KB, 500x500px

>>60288597
>>60288642

File: 1467754418255.jpg (35KB, 560x420px) Image search: [Google]

35KB, 560x420px

>>60288654
you got me

>>60289060
>>60288642

File: dasit.jpg (181KB, 1200x1800px) Image search: [Google]

181KB, 1200x1800px

Administering CPR to ded thread

>>60289156
i want that body desu

does this work?

File: 1460303555163.png (425KB, 589x564px) Image search: [Google]

425KB, 589x564px

>>60289188
Yes!

>>60289177
Big floppy titties seem like such a hassle.
They get in the way, they hurt when you run, they hurt your back, they don't even look good past 25

>>60289212
don't be jealous

>>60289212
Non-issues when you're an anime girl

File: 1447809875611.png (229KB, 615x870px) Image search: [Google]

229KB, 615x870px

>>60289156

File: lucoa_sweater.jpg (199KB, 850x1202px) Image search: [Google]

199KB, 850x1202px

>>60289262
Someone sounds jealous

File: 1475132037502.png (183KB, 308x539px) Image search: [Google]

183KB, 308x539px

>>60289272

File: __quetzalcoatl_kobayashi_san_chi_no_maidragon_drawn_by_sky_freedom__91848c399dde66a7b889c8896a7afc6a.jpg (107KB, 753x1200px) Image search: [Google]

107KB, 753x1200px

>>60289285

New thread:
>>60289330
>>60289330
>>60289330

>>60289339
It's not even at the bump-limit yet, faggot.

>>60287739

>Is there a type which will allow me to add anything I want as long as I have enough memory?
In the language? No. But libgmp is one of the more commonly used libraries for this, in which case, mpz_t is the type you are looking for.

File: vomit.png (88KB, 389x213px) Image search: [Google]

88KB, 389x213px

>>60283870

>Macros
You're better off using C++ templates and nothing else.

don't know if this belongs in stupid questions but I'll try my luck here

@pokeBot.command()
async def upload(**kwargs):
    botchat = 99556354253783043.0 #Destination is NOT supposed to be a string, integer or floating point. SO WHAT THE FUCK IS IT THEN!?
    destination = botchat
    print(kwargs)
    return await pokeBot.send_file(destination , "C:/Users/Xenon/Desktop/1491784016530.jpg", filename="1491784016530", 
    content="It begins", tts=False)

I'm shit out of luck when it comes to this I just get the error "Destination is supposed to be a Channel, User or Server str given or floating point given or int given." WHAT AM I SUPPOSED TO PUT THEN YOU FUCK

>>60283665
Fuck C anyway