Asal Olmayan Sayıları Bulma


17

Kabul etmeyi seçmeniz durumunda, bir sayı aşağıdaki ölçütleri karşılıyorsa, doğru veya yanlış (veya evet ve hayır gibi benzer anlamlı bir gösterim) döndüren bir işlevi kodla golf etmektir:

  1. Tamsayının kendisi asal bir sayı VEYA
  2. Her iki komşu tamsayısı asal

Örneğin:
girişi 7True döndürür.
Girişi 8de True döndürür. Girdisi False
değerini 15döndürür. (Ne 14, 15, ne de 16 asal)

Giriş, 2 ^ 0 ile 2 ^ 20 dahil sayılar için doğru şekilde dönebilmelidir, bu nedenle işaret sorunları veya tamsayı taşmaları hakkında endişelenmenize gerek yoktur.


32-bit sayı taşmaları, arabellek taşmaları değil sanırım.
kullanıcı bilinmiyor

Hata! "Tamsayı taşması" anlamına geliyordu. Beyin otopilot kullanmaya başladı.
Bay Llama

Yanıtlar:


11

J, 17

*/<:$&q:(<:,],>:)

Süreç dönüş kodları olarak kodlanmış booleans döndürür: true için sıfır, false için sıfır olmayan. Örnek kullanım:

   */<:$&q:(<:,],>:) 7
0
   */<:$&q:(<:,],>:) 8
0
   */<:$&q:(<:,],>:) 15
3

*/0 p:<:,],>:daha kısa ve uygun bir (lambda) işlevi([:*/0 p:<:,],>:)
randomra

9

Haskell, 47 karakter

f n=any(\k->all((>0).mod k)[2..k-1])[n-1..n+1]

6

Python 85 80

def f(n):g=lambda n:all(n%i!=0for i in range(2,n));return g(n)or g(n-1)or g(n+1)

Code Golf ilk kez bu yüzden muhtemelen bazı hileler eksik.


Öğesini kaldırabilirsiniz []. her şey bir jeneratör ifadesiyle çalışmaktan çok mutlu olacak. Eğer kod çirkin olmanın sakıncası yoksa, ayrıca aralarında boşluk kaldırabilir 0ve forve )ve or.
stranac

@stranac Harika. Çok teşekkür ederim.
Kris Harper

3
Birkaç basit değişiklik yaptı, umarım hala çalışır:f=lambda n:any(all(m%i for i in range(2,m))for m in[n,n-1,n+1])
Nabb

@ Nabb Çok hoş. Aferin.
Kris Harper

5

Herhangi bir yolla kod kısalığı açısından gerçek bir yarışmacı değil, ama düzenli ifadeyle önceliği belirlemeye başladığından beri hala gönderme , birçok yönden bükülmüştür!

Python (2.x), 85 karakter

import re
f=lambda n:any(not re.match(r"^1?$|^(11+?)\1+$","1"*x)for x in[n,n-1,n+1])

For döngüsünü kaldırabilir ve "1" * (n + 1) 'i test ederek ancak ^ 1? 1? yerine.
Howard

4

Yakut (lambda olarak 55 veya 50)

def f q;(q-1..q+1).any?{|n|(2..n-1).all?{|d|n%d>0}};end

veya lambda olarak ( g[23]aramak için kullanın )

g=->q{(q-1..q+1).any?{|n|(2..n-1).all?{|d|n%d>0}}}

Kahve (53)

p=(q)->[q-1..q+1].some (n)->[2..n-1].every (d)->n%d>0

<pedantic> "proc" olmalı "lambda" değil </
pedantic

3

Sıkıcı Mathematica, 35 çözüm!

PrimeQ[n-1]||PrimeQ[n]||PrimeQ[n+1]

15
En azından içine golf olabilir Or@@PrimeQ/@{n-1,n,n+1}.
Howard

Bu bir işlev değil.
Martin Ender

@MartinBüttner: I don’t know Mathematica, sorry.
Ry-

2
Using Howard's version, Or@@PrimeQ@{#-1,#,#+1}& (the slash in his code isn't needed)
Martin Ender

3

C, 112 82 72 characters

Following Ilmari Karonen's comment, saved 30 chars by removing main, now P returns true/false. Also replaced loop with recursion, and some more tweaks.

p(n,q){return++q==n||n%q&&p(n,q);}P(n){return p(-~n,1)|p(n,1)|p(~-n,1);}

Original version:

p(n,q,r){for(r=0,q=2;q<n;)r|=!(n%q++);return!r;}
main(int n,int**m){putchar(48|p(n=atoi(*++m))|p(n-1)|p(n+1));}

You could save 2 chars with main(n,m)int**m;.
Ilmari Karonen

...and besides, the challenge says "code-golf a function".
Ilmari Karonen

3

Mathematica, 24 bytes

Don't know why this old post showed up in my list today, but I realized Mathematica is competitive here.

Or@@PrimeQ/@{#-1,#,#+1}&

Unnamed function taking an integer argument and returning True or False. Direct implementation.


PrimeQ threads over lists, so Or@@PrimeQ@{#-1,#,#+1}& (or Or@@PrimeQ[#+{-1,0,1}]&) also works, for -1 byte. (Although, I guess I don't know if PrimeQ threaded over lists in 2012.)
Misha Lavrov

3

Stax, 6 bytes

Ç▀<╝ºΩ

Run and debug it

Explanation (unpacked):

;v:Pv<! Full program, implicit input  Example: 7
;v      Copy input and decrement               7 6
  :P    Next prime                             7 7
    v   Decrement                              7 6
     <! Check if input is >= result            1

2

JavaScript (71 73 80)

n=prompt(r=0);for(j=n-2;p=j++<=n;r|=p)for(i=1;++i<j;)p=j%i?p:0;alert(r)

Demo: http://jsfiddle.net/ydsxJ/3/

Edit 1: Change for(i=2;i<j;i++) to for(i=1;++i<j;) (thanks @minitech). Convert if statement to ternary. Moved r|=p and p=1 into outer for to eliminate inner braces. Saved 7 characters.

Edit 2: Combine p=1 and j++<=n to p=j++<=n, save 2 chars (thanks @ugoren).


You can use for(i=1;++i<j;) instead of for(i=2;i<j;i++) to save 1 more character.
Ry-

1
@minitech: !j%i won't work because of precedence. A working alternative is j%i<1.
Nabb

@Nabb: Wow, you're right. That's silly.
Ry-

How about p=j++<=n? If Javascript is like C here, it should work.
ugoren

@ugoren: Looks like it worked, thanks!
mellamokb

2

Regex (ECMAScript), 20 bytes

^x?x?(?!(x+)(x\1)+$)

Try it online!

The above version does not correctly handle zero, but that only takes 1 extra byte:

^x?x?(?!(x+)(x\1)+$)x

As an added bonus, here's a version that gives a return match of 1 for one less than a prime, 2 for prime, and 3 for one more than a prime:

^x?x??(?!(x+)(x\1)+$)x

Try it online!


The range the question speaks about it is "between 2^0 and 2^20" so 1..2^20 so 0 there isn't...
RosLuP

@RosLuP Which is exactly why my primary answer is 20 bytes and doesn't handle 0 correctly. I find value in going beyond the precise specifications of the question and giving more robust answers, along with the answer that minimally matches the question's spec.
Deadcode

Sometimes I do the same (write 'unnecessary' test) but this seems goes against codegolf way of thinking, and people that write them are not considered "serious"...
RosLuP

1
@RosLuP But what's the harm, as long as I give the minimal answer as my primary answer? And can you give any examples of people who actually think that way? I could understand it if I gave my only answer as the robust one, but I'm not doing that.
Deadcode

1

C#, 96

It returns -1,0,1 for true, anything else is false.

Any suggestions to make it shorter would be wonderful!

int p(int q){var r=q-1;for(var i=2;i<r&r<q+2;i++){if(i==r-1)break;if(r%i==0)r+=i=1;}return r-q;}

Expanded form:

int p(int q){
    var r=q-1;
    for(var i=2;i<r&r<q+2;i++){
        if(i==r-1)break;
        if(r%i==0)r+=i=1;
    }
    return r-q;     
}

I'm not entirely sure, but I think you could remove the if(i==r-1)break; and change the middle of the for loop from i<r to i<r-1. It would bring you down to 82.
Ciaran_McCarthy

1

GolfScript: 26

)0\{.:i,{i\%!},,2=@|\(}3*;

Explanation: The innermost block {.:i,{i\%!},,2=@|\(} determines if the top of the stack is prime by checking if there are exactly 2 factors less than the top of the stack. It then disjuncts this with the second item on the stack, which holds the state of whether a prime has been seen yet. Finally, it decrements the number on the top of the stack.

Start by incrementing the input, initializing the prime-seen state, and repeat the block 3 times. Since this will decrement twice, but we started by incrementing, this will cover n+1 and n-1.


1

C#, 87 97 chars

bool p(int q){return new[]{q-1,q,q+1}.Any(x=>Enumerable.Range(2,Math.Abs(x-2)).All(y=>x%y!=0));}

I don't think this works with 1 or 2 as input
Ben Reich

@BenReich It didn't. I had to add ten chars to fix it :(
Steve Clanton

1

CJam, 12 bytes

CJam is much younger than this challenge, so this answer is not eligible for the green checkmark (which should be updated to randomra's answer anyway). However, golfing this was actually quite fun - I started at 17 bytes and then changed my approach completely three times, saving one or two bytes each time.

{(3,f+:mp:|}

This is a block, the closest equivalent to a function in CJam, which expects the input on the stack, and leaves a 1 (truthy) or 0 (falsy) on the stack.

Test it here.

Here is how it works:

(3,f+:mp:|
(          "Decrement the input N.";
 3,        "Push an array [0 1 2].";
   f+      "Add each of those to N-1, to get [N-1 N N+1].";
     :mp   "Test each each element for primality, yielding 0 or 1.";
        :| "Fold bitwise OR onto the list, which gives 1 if any of them was 1.";

1

F#, 68 bytes (non-competing)

let p n=Seq.forall(fun x->n%x>0){2..n-1}
let m n=p(n-1)||p n||p(n+1)

Try it online!

This is why I love code golf. I'm still very green with F# but I learn so much about how the language works and what it can do from these kinds of challenges.


Why is it non-competing?
Nit

1
Because I'm not sure if I'm using anything in F# today that wasn't around when the question was asked in 2012. I admit it's pedantic - even paranoid. But I write pharmaceutical software for a living. Paranoia is healthy. ;)
Ciaran_McCarthy

1
Look at F#'s version table in Wikipedia. Depending on the version you need, it may be older than the question.



1

Java 8, 83 bytes

n->n==1|p(n-1)+p(n)+p(n+1)>0int p(int n){for(int i=2;i<n;n=n%i++<1?0:n);return--n;}

Returns true / false as truthy/falsey values.

Try it online.

Explanation:"

n->                    // Method with integer parameter and boolean return-type
  n==1                 //  Return whether the input is 1 (edge-case)
  |p(n-1)+p(n)+p(n+1)>0//  Or if the sum of `n-1`, `n`, and `n+1` in method `p(n)` is not 0

int p(int n){          // Separated method with integer as both parameter and return-type
  for(int i=2;i<n;     //  Loop `i` in the range [2, `n`)
    n=n%i++<1?         //   If `n` is divisible by `i`
       0               //    Change `n` to 0
      :                //   Else:
       n);             //    Leave `n` as is
                       //  (After the loop `n` is either 0, 1, or unchanged,
                       //   if it's unchanged it's a prime, otherwise not)
  return--n;}          //  Return `n` minus 1

So int p(int n) will result in -1 for n=0 and non-primes, and will result in n-1 for n=1 or primes. Since p(0)+p(1)+p(2) will become -1+0+1 = 0 and would return false (even though 2 is a prime), the n=1 is an edge case using this approach.


A single loop without separated method would be 85 bytes:

n->{int f=0,j=2,i,t;for(;j-->-1;f=t>1?1:f)for(t=n+j,i=2;i<t;t=t%i++<1?0:t);return f;}

Returns 1 / 0 as truthy/falsey values.

Try it online.

Explanation:

n->{              // Method with integer as both parameter and return-type
  int f=0,        //  Result-integer, starting at 0 (false)
      j=2,i,      //  Index integers
      t;          //  Temp integer
  for(;j-->-1;    //  Loop `j` downwards in range (2, -1]
      f=          //    After every iteration: Change `f` to:
        t>1?      //     If `t` is larger than 1 (`t` is a prime):
         1        //      Change `f` to 1 (true)
        :         //     Else:
         f)       //      Leave `f` the same
    for(t=n+j,    //   Set `t` to `n+j`
        i=2;i<t;  //   Inner loop `i` in the range [2, t)
      t=t%i++<1?  //    If `t` is divisible by `i`:
         0        //     Change `t` to 0
        :         //    Else:
         t);      //     Leave `t` the same
                  //   (If `t` is still the same after this inner loop, it's a prime;
                  //   if it's 0 or 1 instead, it's not a prime)
  return f;}      //  Return the result-integer (either 1/0 for true/false respectively)

1

Japt, 7 bytes

´Uô2 dj
´Uô2    // Given the range [U-1..U+1],
     dj // sing "Daddy DJ", uh, I mean, check if any are a prime.

Try it online!


0

R, 68 chars

f=function(n){library(gmp);i=isprime;ifelse(i(n-1)|i(n)|i(n+1),1,0)}

Usage (1 for TRUE, 0 for FALSE):

f(7)
[1] 1
f(8)
[1] 1
f(15)
[1] 0

1
I don’t really know how R works, but could you just do i(n-1)|i(n)|i(n+1) instead of ifelse(i(n-1)|i(n)|i(n+1),1,0)?
Ry-

You are right: g=function(n){library(gmp);i=isprime;i(n-1)|i(n)|i(n+1)} - down to 56 characters! ;-)
Paolo

0

C++

k=3;cin>>i;i--;
while(k)
{l[k]=0;
  for(j=2;j<i;j++)
   if(!(i%j))
     l[k]++;
  k--;
  i++;
}
if(!l[1]|!l[2]|!l[3])
     cout<<"1";
else cout<<"0";

Welcome to CodeGold.SE. If you look at the other answers you'll notice a common format used for answers to [code-golf] questions. You might want to apply it to your answers as well.
dmckee --- ex-moderator kitten

0

Q, 43 chars 36

{any min each a mod 2_'til each a:x+-1 0 1}
{any(min')a mod 2_'(til')a:x+-1 0 1}

0

J, 16 chars

   (_2&<@-4 p:]-2:)

   (_2&<@-4 p:]-2:) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 1 1

0

Python, 69 67 chars

any(all(i%j for j in range(2,i))for i in range(input()-1,8**7)[:3])

8**7 > 2**20 while being slightly shorter to write


0

Ruby, 47 chars but very readable

require'Prime'
f=->x{[x-1,x,x+1].any? &:prime?}

0

C++ 97

ugoren seems to have beat me to the clever solution. So he's a shortish version on the loop three times approach:

P(int k){int j=1;for(int i=2;i<k;){j=k%i++&&j;}return j;}
a(int b){return P(b)|P(b+1)|P(b-1);}

0

Forth (gforth), 104 bytes

: p dup 1 > if 1 over 2 ?do over i mod 0> * loop else 0 then nip ;
: f dup 1- p over 1+ p rot p + + 0< ;

Try it online!

Explanation

Prime check (p)

dup 1 > if          \ if number to check if greater than 1
   1 over 2 ?do     \ place a 1 on the stack to act as a boolean and loop from 2 to n
      over i  mod   \ take the modulo of n and i
      0> *          \ check if greater than 0 (not a divisor) and multiply result by boolean
   loop             \ end the loop, result will be -1 if no divisor was found (prime)
else                \ if n is less than 2
   0                \ put 0 on the stack (not prime)
then                \ end the loop
nip                 \ drop n from the stack

Main Function (f)

dup 1- p             \ get n-1 and check if prime
over 1+ p            \ get n+1 and check if prime
rot p                \ rotate stack to put n on top and check if prime
+ + 0<               \ add the three results and check if less than 0 (at least 1 was prime)


0

Jelly, 5 bytes

‘r’ẒẸ

Try it online!

How it works

‘r’ẒẸ    Monadic main link. Input: integer n
‘r’      Generate range n-1..n+1
   ẒẸ    Is any of them a prime?
Sitemizi kullandığınızda şunları okuyup anladığınızı kabul etmiş olursunuz: Çerez Politikası ve Gizlilik Politikası.
Licensed under cc by-sa 3.0 with attribution required.