g | x | w | all
Bytes Lang Time Link
055JavaScript ES6250310T223847ZSteve Be
016APLNARS250310T090619ZRosario
063Desmos250310T034833ZDesmosEn
180Pascal250306T090015ZKai Burg
009Jelly201127T181550ZUnrelate
044Perl 5 p201127T172832ZXcali
136Javascript140206T191420ZXin
090C#140207T210136Ztdink
142JavaScript140207T120418Zrahulroy
131JavaScript no ES6140207T170042Ziefserge
069ES6140206T225831ZNoyo
120C140206T232215Zwarrenm
045Julia140206T225816Zgggg
044perl6140206T213830ZAyiko
158TSQL140206T193648Zduanxn
089JavaScript140206T200802ZBoann
067PHP140206T185953ZBoann
048Python 2.7140206T085029ZAbhijit
067Ruby 1.9+140206T181056Zdaniero
049Mathematica140206T041631ZDavidC
nanJava140206T165107ZJames We
051Matlab140206T150741ZHannesh
049k140206T084846Zskeevey
056R140206T084206Zplannapu
017J140206T102953ZHoward
029APL140206T101159Zmarinus
065Befunge93140206T085700ZJoel Bos
029J140206T093639ZGareth
065Python 2.7140206T031015Zuser8777

JavaScript (ES6), 55 characters

With a bit of help from 14m2.

s=>(u=d=l=r=0,s.map(l=>this[l]+=2),Math.hypot(u-d,r-l))

Javascript (ES6), 58 characters

Borrowing a few ideas from others. Did not know Math.hypot was a thing!

s=>(u=d=l=r=0,s.map(l=>window[l]++),Math.hypot(u-d,r-l)*2)

JavaScript (ES6), 80 characters

s=>(x=y=0,s.map(l=>l=='u'?y++:l=='d'?y--:x+={l:-1,r:1}[l]),Math.sqrt(x*x+y*y)*2)

Nothing very clever. Takes an array of individual characters as input.

APL(NARS), 16 chars

2×∣+/1i*'ULDR'⍳⎕

Copy the idea of make the powers of the immaginary unity 0J1 or 1i in corrispondence of array 'ULDR'. It is start from x y = 0 0, and each result of partial sum represent the point it arrived. The last result of the partial sum is the point x y of end, of that point I calculate the distance of the start 0 0, and multiplicate that distance by 2.

  2×∣+/1i*'ULDR'⍳⎕
⎕:
  'LULUURRRRURD'
8.485281374

Desmos, 63 bytes

l=(-1,0)
r=(1,0)
u=(0,1)
d=(0,-1)
f(l)=2distance(l.total,(0,0))

f takes an array of direction vectors as an input, sums them, then outputs double the distance of the sum to (0,0).

Try it on Desmos!

Pascal, 180 B

This submission requires a processor compliant with Extended Pascal level 1 as laid out in ISO standard 10206, in particular the complex number support.

function F(S:array[A..Z:integer]of char):real;var P:complex;I:integer;begin
P:=0;for I:=A to Z do
P:=P+cmplx(ord(S[I]='R')-ord(S[I]='L'),ord(S[I]='U')-ord(S[I]='D'));F:=2*abs(P)end

−15 B if you do not insist on having a function but a complete program is acceptable, too. It is otherwise the same algorithm.

program M(input,output);var P:complex;S:char;begin
P:=0;while not EOF do begin
read(S);P:=P+cmplx(ord(S='R')-ord(S='L'),ord(S='U')-ord(S='D'))end;write(2*abs(P))end.

Explained:

{ By listing a `program` parameter of the spelling `input`,
  there’s an implicit `reset(input)` which implies a `get(input)`. }
program richBritishPeopleProblem(input, output);
    var
        { `cmplx(real_part, imaginary_part)` returns a `complex` value.
          Here, `polar(radius, argument)` would’ve been another option. }
        position: complex value cmplx(0.0, 0.0);
        step: char;
    begin
        while not EOF(input) do
        begin
            { `Read(input, step)` is short for `step ≔ input↑; get(input)`
              except that `read` cannot read past end‐of‐line sequences. }
            read(input, step);
            position ≔ position + cmplx(
                    { The built‐in function `ord` converts a Boolean value
                      into an `integer` value. The `integer` value is then
                      automatically promoted to `real` because `cmplx` wants
                      `real` arguments. }
                    ord(step = 'R') − ord(step = 'L'),
                    ord(step = 'U') − ord(step = 'D')
                )
        end;
        { `abs` also returns the magnitude of a complex number.
          This is shorter than `sqrt(x*x+y*y)` even accounting
          for the additional declaration and function calls. }
        writeLn(output, 2 × abs(position))
    end.

The constants U, R, D and L are to be interpreted as north, east, south and west respectively, but British rich people aren’t that smart.

Jelly, 10 9 bytes

O:3 ı*SḤA

Try it online!

I fiddled around with other functions to see if I could get rid of the space by finding a different one-byte dyad that would take a two-digit number on the left rather than the right, but it turned out that : doesn't need two digits to produce the desired values.

O            Codepoints of the input
 :3          floor divided by 3.
    ı*       Raise sqrt(-1) to the power of each,
      S      sum,
       Ḥ     double,
        A    and take the absolute value.

Perl 5 -p, 44 bytes

$_=2*sqrt((y/R//-y/L//)**2+(y/U//-y/D//)**2)

Try it online!

Javascript, 136

function z(a){var x=a.split('u').length-a.split('d').length;var y=a.split('r').length-a.split('l').length;return Math.sqrt(x*x+y*y)*2;};

Example:

document.write(z('uuuudrrrwl'));
7.211102550927978

C# - 90 characters

Fresh from LINQPad.

int x=0,y=0;input.Max(i=>i==85?y++:i==82?x++:i==68?y--:x--);(Math.Sqrt(x*x+y*y)*2).Dump();

Where input is a valid string.

>string input = "LULUURRRRURD";

>8.48528137423857

JavaScript - 142 characters - no eval()

function r(a){return Math.sqrt(Math.pow(a.match(/u/g).length-a.match(/d/g).length,2)+Math.pow(a.match(/l/g).length-a.match(/r/g).length,2))*2}

where a is a string like 'uudrrl'

use like this -

a='uudrrl'
r(a)

Test in browser console.

var x = "luluurrrrurd"
r(x)
8.48528137423857

JavaScript (no ES6, no eval) - 131

f=function(h){for(i=0,a=[0,,0,0,0];i<h.length;++i)++a[(h.charCodeAt(i)>>2)-25];x=a[0]-a[4];y=a[2]-a[3];return Math.sqrt(x*x+y*y)*2}

Test:

console.log(f('uuuudrrrl'));     // 7.211102550927978 
console.log(f('luluurrrrurd'));  // 8.48528137423857

ES6, 77 69

Definition:

f=s=>{u=d=l=r=0;for(c of s)eval(c+'++');return 2*Math.hypot(u-d,l-r)}

Usage:

>>> f('uuuudrrrl')
7.211102550927979
>>> f( 'uuuudrrrl'.split('') )
7.211102550927979

(Inspired partially by Boann's answer.)

C, 120

float d(char *p){int v=0,h=0;while(*p){v+=*p=='U'?1:*p=='D'?-1:0,h+=*p=='R'?1:*p=='L'?-1:0,++p;}return 2*sqrt(v*v+h*h);}

d("LULUURRRRURD") -> 8.485281

Julia, 45

f(l)=2*abs(sum([im^(c=='d'?3:c) for c in l]))

Stole the i to powers trick. Also all the characters except d have values that work as acceptable powers for i.

perl6: 44 chars

2*abs [+] i <<**>>%(<U R D L>Z ^4){get.comb}

Yes, I removed all possible spaces.

T-SQL, 158

IF PATINDEX('%[^UDLR]%', @s)=0 select 2*sqrt(power(LEN(REPLACE(@s,'U',''))-LEN(REPLACE(@s,'D','')),2)+power(LEN(REPLACE(@s,'L',''))-LEN(REPLACE(@s,'R','')),2))

The @s is the input string of varchar(max) type

JavaScript, 89

function f(a){U=D=L=R=0;for(d in a)eval(a[d]+'++');return 2*Math.sqrt((U-=D)*U+(L-=R)*L)}

Example:

<script>
document.write(f(['U', 'U', 'U', 'U', 'D', 'R', 'R', 'R', 'L']));
</script>

>7.211102550927978

PHP, 67

function f($a){foreach($a as$d)@$$d++;return 2*hypot($U-$D,$L-$R);}

Example:

<?php
var_dump(f(array('U', 'U', 'U', 'U', 'D', 'R', 'R', 'R', 'L')));

>float(7.211102550928)

Python 2.7 56 58 56 51 48

With the stolen Number One Dime from Scrooge McDuck, I made my fortune and now have more wealth than Scrooge.

y=lambda s:2*abs(sum(1j**(ord(i)%15)for i in s))

Python 2.7 - 61 53 50 (case insensitive)

y=lambda s:2*abs(sum(1j**(ord(i)%16%9)for i in s))

Implementation

>>> from random import sample
>>> y=lambda s:2*abs(sum((-1j)**(ord(i)%15)for i in s))
>>> path=sample('RLUD'*1000, 100)
>>> y(path)
20.0
>>> path=sample('RLUD'*1000, 100)
>>> y(path)
34.058772731852805

Ruby 1.9+ (67)

f=->s{2*(((g=s.method :count)[?U]-g[?D])**2+(g[?R]-g[?L])**2)**0.5}

Example

f["DRUULULLULL"] => 10.0
f["UUUUDRRRL"] => 7.211102550927978

Mathematica 92 49

Calle deserves full credit for streamlining the code.

f@l_:=2 N@Norm[Tr[l/.{"r"→1,"l"→-1,"u"→I,"d"→-I}]]

Example

f[{"u", "u", "u", "u", "d", "r", "r", "r", "l"}]

7.2111

Java, 185, 203, 204, 217, 226

class A{public static void main(String[] a){int x=0,y=0;for(int i=0;i<a[0].length();i++) switch(a[0].charAt(i)){case'U':y++;break;case'D':y--;break;case'L':x++;break;case'R':x--;}System.out.print(Math.hypot(x,y)*2);}}

I did assume that each "U" was "1 up", so two units up would be "UU"

Edit: swapped out switch for ifs

class A{public static void main(String[]a){int x=0,y=0;for(int i=0;i<a[0].length();i++){int c=a[0].charAt(i);if(c=='U')y++;if(c=='D')y--;if(c=='L')x++;if(c=='R')x--;}System.out.print(Math.hypot(x,y)*2);}}

Moved for iterator

class A{public static void main(String[]a){int x=0,y=0;for(int i=0;i<a[0].length();){int c=a[0].charAt(i++);if(c=='U')y++;if(c=='D')y--;if(c=='L')x++;if(c=='R')x--;}System.out.print(Math.hypot(x,y)*2);}}

No longer takes input as string, rather array of directions

class A{public static void main(String[]a){int x=0,y=0;for(String s:a){char c=s.charAt(0);if(c=='U')y++;if(c=='D')y--;if(c=='L')x++;if(c=='R')x--;}System.out.print(Math.hypot(x,y)*2);}}

Matlab, 51 characters

My Matlab submission, works only with captial letters. This was a fun one! The hardest part was converting the string into an array of complex numbers to be summed.

Function:

f=@(s)abs(sum(fix((s-76.5)/8.5)+((s-79)*i/3).^-99))

Usage:

>> f=@(s)abs(sum(fix((s-76.5)/8.5)+((s-79)*i/3).^-99))
>> f('UURDL')
ans =

     1
>>

k (50 49)

{2*sqrt x$x:0 0f+/("udlr"!(1 0;-1 0;0 -1;0 1))@x}

Example

{2*sqrt x$x:0 0f+/("udlr"!(1 0;-1 0;0 -1;0 1))@x}"uuuudrrrl"
7.211103

R, 86 74 56 characters

Ok it's actually way shorter with imaginary numbers indeed:

2*Mod(sum(sapply(scan(,""),switch,u=1i,d=-1i,l=-1,r=1)))

Usage:

> 2*Mod(sum(sapply(scan(,""),switch,u=1i,d=-1i,l=-1,r=1)))
1: u u u u d r r r l
10: 
Read 9 items
[1] 7.211103

Old solution at 74 characters with xy coords:

2*sqrt(sum(rowSums(sapply(scan(,""),switch,u=0:1,d=0:-1,l=-1:0,r=1:0))^2))

Usage:

> 2*sqrt(sum(rowSums(sapply(scan(,""),switch,u=0:1,d=0:-1,l=-1:0,r=1:0))^2))
1: u u u u d r r r l
10: 
Read 9 items
[1] 7.211103

Takes input as stdin, need to be lower-case and space-separated. Use x-y coordinates starting from (0,0).

J, 17 characters

2*|+/0j1^'urdl'i.

Uses the fact, that the powers of j represent the proper directions.

Example:

> 2*|+/0j1^'urdl'i.'uuuudrrrl'
7.2111

APL (29)

{|+/2 0j2×-⌿2 2⍴+/'URDL'∘.=⍵}

e.g.

     {|+/2 0j2×-⌿2 2⍴+/'URDL'∘.=⍵} 'UUUUDRRRL'
7.211102551

Explanation:

Befunge-93 (65)

It has 65 non-whitespace characters (217 with whitespace, though that can be reduced by a more compact layout (for 69/176 chars)). It takes some liberality with the output format, but is undeniably accurate. Doesn't seem worth the effort to implement/steal a square root implementation.

v                  >$:*\:*+88*4*5-2.,.@
               >3-:|
           >6-:|
       >8-:|
>~"D"-:|
       $   $   $   $
           \   \
       1   1   1   1
       -   -   +   +
           \   \
^      <   <   <   <

echo 'UUDLLUU' | ./befungee.py ../man outputs 2√13 (actually implementation seems to have issue with the extended ASCII though).

J, 29 characters

+:+&.*:/-/_2[\#/.~/:~'ruld'i.

Only works with lower case directions and any characters other than r, u, l, and d will cause it to give a wrong answer.

Usage:

   +:+&.*:/-/_2[\#/.~/:~'ruld'i.'uuuudrrrl'
7.2111

Explanation:

'ruld'i.'uuuudrrrl' The dyadic form of i. finds the index of items from the right argument in the left argument. In this case:

   'ruld'i.'uuuudrrrl'
1 1 1 1 3 0 0 0 2

/:~ sorts this list into ascending order:

   /:~'ruld'i.'uuuudrrrl'
0 0 0 1 1 1 1 2 3

#/.~ counts the number of occurrences of each number:

   #/.~/:~'ruld'i.'uuuudrrrl'
3 4 1 1

_2[\ chops it into 2 rows:

   _2[\#/.~/:~'ruld'i.'uuuudrrrl'
3 4
1 1

-/ subtracts the bottom from the top

   -/_2[\#/.~/:~'ruld'i.'uuuudrrrl'
2 3

+&.*: borrows a trick from another J answer I saw this morning, and squares the items, then sums them, then performs a square root. See under &. documentation:

   +&.*:/-/_2[\#/.~/:~'ruld'i.'uuuudrrrl'
3.60555

+: doubles the result:

   +:+&.*:/-/_2[\#/.~/:~'ruld'i.'uuuudrrrl'
7.2111

Python 2.7 - 65

Nice and short one, this uses complex numbers to step through the plane:

x=lambda s:2*abs(sum([[1,-1,1j,-1j]['RLUD'.index(i)]for i in s]))

Props to DSM and Abhijit in other questions that showed me the use of 1j to calculate this.