Given a non-negative number n
, sort the digits of n
by their first occurrence in pi.
Input can be taken via function cli argument, or STDIN and as a string, char[] or integer. You may output via return value, exit status or STDOUT.
-
\$\begingroup\$ Related \$\endgroup\$Stewie Griffin– Stewie Griffin2017年04月29日 16:38:42 +00:00Commented Apr 29, 2017 at 16:38
-
\$\begingroup\$ Can we take input and output as strings, or as arrays of digits? \$\endgroup\$ETHproductions– ETHproductions2017年04月29日 16:39:07 +00:00Commented Apr 29, 2017 at 16:39
-
\$\begingroup\$ @ETHproductions Clarified. \$\endgroup\$Linnea Gräf– Linnea Gräf2017年04月29日 16:40:23 +00:00Commented Apr 29, 2017 at 16:40
-
19\$\begingroup\$ A few test cases would be nice. \$\endgroup\$Dennis– Dennis2017年04月29日 18:15:03 +00:00Commented Apr 29, 2017 at 18:15
-
1\$\begingroup\$ Now that 12 answers are already present, all of which performing the same thing, if you are still unclear what is being asked, then it is not the problem of the question. \$\endgroup\$Leaky Nun– Leaky Nun2017年04月30日 02:30:40 +00:00Commented Apr 30, 2017 at 2:30
27 Answers 27
Python 3, (削除) 40 (削除ここまで) 39 bytes
1 byte thanks to Jonathan Allan.
lambda s:sorted(s,key="145926870".find)
-
7\$\begingroup\$ You can drop the
3
, since find will return-1
when an item is not found. \$\endgroup\$Jonathan Allan– Jonathan Allan2017年04月29日 17:51:49 +00:00Commented Apr 29, 2017 at 17:51
05AB1E, (削除) 10 (削除ここまで) (削除) 9 (削除ここまで) 7 bytes
Saved 1 byte thanks to Leaky Nun noting that filtering out duplicates is unnecessary.
Saved 2 bytes thanks to Adnan.
žqRvy†J
Explanation
žq # push pi to 15 decimals (contains all digits but 0)
R # reverse
vy # for each char in pi
†J # move it's occurrences in the input to the front
-
\$\begingroup\$
13žsRvy†J
for 9 bytes \$\endgroup\$Leaky Nun– Leaky Nun2017年04月29日 16:29:52 +00:00Commented Apr 29, 2017 at 16:29 -
\$\begingroup\$ @LeakyNun: Oh yeah, duplicates doesn't matter. Thanks :) \$\endgroup\$Emigna– Emigna2017年04月29日 16:31:21 +00:00Commented Apr 29, 2017 at 16:31
-
3\$\begingroup\$ Can you use
žq
instead of13žs
? \$\endgroup\$Adnan– Adnan2017年04月29日 17:56:28 +00:00Commented Apr 29, 2017 at 17:56 -
\$\begingroup\$ @Adnan It doesn't seem to work. \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年04月29日 18:01:50 +00:00Commented Apr 29, 2017 at 18:01
-
2\$\begingroup\$ @Adnan: Yes of course. I didn't realize there was another pi constant :) \$\endgroup\$Emigna– Emigna2017年04月29日 18:05:17 +00:00Commented Apr 29, 2017 at 18:05
Pyth, (削除) 8 (削除ここまで) 6 bytes
ox+.n0
-1 thanks to Leaky Nun: The input will provide the 0
if it's ever needed.
Trivial -1 thanks to Jakube: Backtick not needed (ah, how did I miss that, HOW?!?).
-
\$\begingroup\$ Woohoo, this even beats 05AB1E! Edit: it doesn't beat 05AB1E, and I don't wanna steal :( \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年04月29日 18:00:24 +00:00Commented Apr 29, 2017 at 18:00
-
3\$\begingroup\$ I found it. You don't need the
0
at the end. If the input has a0
, the0
would be provided by the input; if the input does not have a0
, it won't matter. \$\endgroup\$Leaky Nun– Leaky Nun2017年04月29日 18:16:54 +00:00Commented Apr 29, 2017 at 18:16 -
3\$\begingroup\$ @LeakyNun and you can even save the backtick:
ox+.n0
\$\endgroup\$Jakube– Jakube2017年04月29日 22:14:24 +00:00Commented Apr 29, 2017 at 22:14 -
\$\begingroup\$ OK, disregard the first comment, thanks to LeakyNun and Jakube, I again beat 05AB1E, I hope for good this time. \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年04月30日 08:06:59 +00:00Commented Apr 30, 2017 at 8:06
-
1\$\begingroup\$ That's a beautiful amount of implicit input. \$\endgroup\$izzyg– izzyg2017年05月13日 06:27:53 +00:00Commented May 13, 2017 at 6:27
Jelly, 10 bytes
"ṀSṪw’ṾiμÞ
Takes input as a string of digits.
-3 bytes thanks to @ETHproductions
Explanation
"ṀSṪw’ṾiμÞ
μ - Separate chain into function "ṀSṪw’Ṿi and sort atom Þ.
Þ - Sort the input by
i - Each digit's index in:
"ṀSṪw’ - the literal 3145926870 ...
Ṿ - transformed into the list 3,1,4,5,9,2,6,8,7,0
-
\$\begingroup\$ I think
3145926870
can be represented as a 4-digit base-250 string (meaning it takes up 6 bytes instead of 10), but I'm not sure how to compress it as such. \$\endgroup\$ETHproductions– ETHproductions2017年04月29日 16:46:23 +00:00Commented Apr 29, 2017 at 16:46 -
\$\begingroup\$ Does Jelly not have a builtin for pi? \$\endgroup\$math junkie– math junkie2017年04月29日 16:46:48 +00:00Commented Apr 29, 2017 at 16:46
-
\$\begingroup\$ @mathjunkie but Jelly is not very efficient on string manipulation \$\endgroup\$Leaky Nun– Leaky Nun2017年04月29日 16:47:12 +00:00Commented Apr 29, 2017 at 16:47
-
\$\begingroup\$ @mathjunkie Yes, but the manipulations to the list take too many bytes \$\endgroup\$fireflame241– fireflame2412017年04月29日 16:47:43 +00:00Commented Apr 29, 2017 at 16:47
-
Japt, (削除) 10 (削除ここまで) 9 bytes
8 bytes of code, +1 for the -P
flag.
–!bMP+U
Try it online! Takes input as a string.
Explanation
–!bMP+'0 // Implicit input
¬ // Split the input into chars.
ñ // Sort each char in the resulting list by
!b // its index in
MP+U // Math.PI + the input.
-P // Join the result back into a single string.
// Implicit: output result of last expression
JavaScript (ES6), 54 bytes
f=
s=>[...s].sort((a,b)=>k[a]-k[b],k=`9150236874`).join``
<input oninput=o.textContent=f(this.value)><pre id=o>
Uses strings for I/O.
Jelly, (削除) 8 (削除ここまで) 7 bytes
-1 byte thanks to Dennis (use any existing 0
in the input, clever.)
ØP;ṾiμÞ
How?
ØP;ṾiμÞ - Main link: string s (char list)
μÞ - sort the characters, c, of s by:
i - first index of c in:
ØP - pi yield: 3.141592653589793
; - concatenate with left: [3.141592653589793, c]
Ṿ - un-evaluate: "3.141592653589793,c" (a char list with the digit character c)
if any c is 0 ^ it will then be to the right of all others
-
\$\begingroup\$ ...and there I was searching for squares -
3820009
(sqrt of14592468760081
) is still3
digits in base250
. \$\endgroup\$Jonathan Allan– Jonathan Allan2017年04月29日 18:25:28 +00:00Commented Apr 29, 2017 at 18:25 -
\$\begingroup\$ The
Ṿ
in your explanation is misplaced. \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年04月30日 09:56:33 +00:00Commented Apr 30, 2017 at 9:56
CJam, (削除) 15 (削除ここまで) (削除) 12 (削除ここまで) (削除) 10 (削除ここまで) 8 bytes
r{P`#c}$
-3: Use a string based on the P
pi variable instead of a literal.
-2: Decided I don't need to uniquify at all, since finding an index takes the first occurrence anyways.
-2: Thanks to jimmy23013 for an interesting approach using x mod 65536.
Explanation:
r{P`#c}$ e# Takes an input token r e# Take the integer as a string {P`#c} e# Sorting key: P e# Push P (defaults to 3.141592653589793) ` e# Convert to string representation # e# Find char's index in the string we made e# A '.' will never be found in an integer, but it doesn't matter, since the shifting preserves ideal sorting. e# A '0' will be indexed as -1. c e# Convert index to char e# This first calculates index % 65536, and then converts to char. We need this because otherwise 0 would be indexed as -1, i.e. smallest index. e# We don't need to convert back to integer, since we can use lexicographical sorting. $ e# Sort with key
-
1\$\begingroup\$ Yay, beats MATL :) \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年04月29日 17:53:42 +00:00Commented Apr 29, 2017 at 17:53
-
\$\begingroup\$ tio.run/nexus/cjam#@19UHZCgnFyr8v@/oYGRsYmpmbmFJQA \$\endgroup\$jimmy23013– jimmy230132017年05月01日 19:02:06 +00:00Commented May 1, 2017 at 19:02
-
\$\begingroup\$ @jimmy23013 Wow that's clever. It's almost like there's a builtin for int(x)%65536, and
ci
would even convert back to integer. \$\endgroup\$Erik the Outgolfer– Erik the Outgolfer2017年05月02日 11:12:30 +00:00Commented May 2, 2017 at 11:12
C (clang), (削除) 85 (削除ここまで) 73 bytes
*z=L"9150236874"-48;s(*a,*b){return z[*a]-z[*b];}f(*t,n){qsort(t,n,4,s);}
-
\$\begingroup\$ @ceilingcat, wow, it took my small mind forever to figure out what you did there!! \$\endgroup\$jdt– jdt2021年09月24日 16:08:00 +00:00Commented Sep 24, 2021 at 16:08
-
-
-
PHP, 71 Bytes
The regex solution is shorter
for(;~$c=_3145926870[$i++];)echo str_repeat($c,substr_count($argn,$c));
or
for(;~$c=_3145926870[$i++];)echo str_pad("",substr_count($argn,$c),$c);
PHP, 78 Bytes
for(;~$c=$argn[$i++];)$j[strpos("3145926870",$c)].=$c;ksort($j);echo join($j);
PHP, 112 Bytes
$a=str_split($argn);usort($a,function($x,$y){return strpos($d="3145926870",$x)<=>strpos($d,$y);});echo join($a);
-
\$\begingroup\$ I've added a 69 byte solution. Maybe we can get it down to 66 byte together ;) \$\endgroup\$Christoph– Christoph2017年05月02日 14:13:11 +00:00Commented May 2, 2017 at 14:13
Ruby, 50 bytes
n=gets;"3145926870".each_char{|c|$><<c*n.count(c)}
C (gcc), 78 bytes
f(char*s){for(char*d="3145926870",*p;*d;d++)for(p=s;*p;)*p++-*d||putchar(*d);}
MATL, 14 bytes
YP99Y$uj!y=sY"
Explanation with an example
The symbol ;
is used as row separator in matrices. So [1 2 3]
is a row vector, [1; 2; 3]
is a column vector, and [1 2; 3 4]
is a square matrix. The latter can also be represented, for clarity, as
[1 2;
3 4]
Consider input 2325
as an example.
YP % Push approximation of pi as a double (predefined literal)
% 3.14159265358979
99Y$ % Variable-precision arithmetic with 99 digits. Gives a string.
% The input 3.14159265358979 is recognized as representing pi
% STACK: '3.141592653589793238462 ··· 707'
u % Unique entries, keeping order of their first appearance
% STACK: '3.145926870'
j % Input line as a string
% STACK: '3.145926870', '2352'
! % Transpose
% STACK: '3.145926870', ['2'; '3';'5'; '2']
y % Duplicate the second-top element in the stack
% STACK: '3.145926870', ['2'; '3';'5'; '2'], '3.145926870'
= % Test for equality, with broadcast. This gives a matrix with
% all pairwise comparisons)
% STACK: '3.145926870', [0 0 0 0 0 0 1 0 0 0 0;
% 1 0 0 0 0 0 0 0 0 0 0;
% 0 0 0 0 1 0 0 0 0 0 0;
% 0 0 0 0 0 0 1 0 0 0 0]
s % Sum of each column
% STACK: '3.145926870', [1 0 0 0 1 0 2 0 0 0 0]
Y" % Run-length decoding. Implicitly display
% STACK: '3522'
C# Interactive, (削除) 37 (削除ここまで) 36 Bytes
i.OrderBy(c=>"145926870".IndexOf(c))
Actually you have to execute this in the C# interactive for proper results, but I guess this is what you meant with exit status. The variable i actually is the input variable (it can be for example a string), so it's basically the method parameter.
I think the code itself is pretty straight forward.
-
\$\begingroup\$ Where's the
3
? \$\endgroup\$Paul– Paul2017年05月01日 13:44:44 +00:00Commented May 1, 2017 at 13:44 -
1\$\begingroup\$ @Paul it's not neccessary, as it returns -1 if the element isn't found. \$\endgroup\$MetaColon– MetaColon2017年05月01日 13:47:19 +00:00Commented May 1, 2017 at 13:47
-
\$\begingroup\$ This is just a snippet of code though, I'm pretty sure even in interactive you have to specify why
i
is somewhere so it can be taken as input. Also if you're saying C# you have to includeusing System.Linq;
into the byte count. However, if this is Interactive you should specify the language as C# Interactive not solely C#. \$\endgroup\$TheLethalCoder– TheLethalCoder2017年05月18日 13:13:34 +00:00Commented May 18, 2017 at 13:13 -
\$\begingroup\$ @TheLethalCoder I updated it to C# Interactive. The using isn't neccesaary in the interactive, as it's included automatically. \$\endgroup\$MetaColon– MetaColon2017年05月18日 13:16:08 +00:00Commented May 18, 2017 at 13:16
PHP, (削除) 66 (削除ここまで) 65 bytes
Saved 1 byte thanks to Titus.
while(~$d=_3145926870[++$i])echo preg_filter("/[^$d]/",'',$argn);
APL (Dyalog Unicode), 13 bytes
{⍵[⍵⍋⍨14⍕○しろまる1]}
○しろまる1
Pi times 1.
14⍕
convert to a string with 14 decimal digits. This includes all unique digits except 0.
⍵⍋⍨
the indices that would sort the input string ⍵
according to order given by the previous string.
⍵[...]
index with those indices into the input to sort it.
05AB1E, (削除) 5 (削除ここまで) 6 bytes
Had to realise that 0
is not present in the standard length pi constant.
Σтžsyk
Σтžsyk
Σ Sort by the result of code
тžs Push 100 digits of pi
yk Index of digit in pi
-
\$\begingroup\$ Too bad you needed that zero for this method. It should be optimal for this language at least. Can't ask for more than that :) \$\endgroup\$Emigna– Emigna2017年05月18日 13:24:34 +00:00Commented May 18, 2017 at 13:24
Java 7, 110 bytes
String c(String s){String r="";for(char i:"3145926870".toCharArray())r+=s.replaceAll("[^"+i+"]","");return r;}
Explanation:
String c(String s){ // Method with String parameter and String return-type
String r=""; // Result String
for(char i:"3145926870".toCharArray()) // Loop over the characters of "3145926870"
r+=s.replaceAll("[^"+i+"]",""); // Append the result-String with all the occurrences of the current character
// End of loop (implicit / single-line body)
return r; // Return the result-String
} // End of method
Test code:
class M{
static String c(String s){String r="";for(char i:"3145926870".toCharArray())r+=s.replaceAll("[^"+i+"]","");return r;}
public static void main(String[] a){
System.out.println(c("12345678908395817288391"));
}
}
Output:
33311145599922688888770
Clojure, 38 bytes
#(sort-by(zipmap"3145926870"(range))%)
Input in string, returns a sequence of characters. zipmap
creates a "dictionary" object, which can be used in a function context as well.
(f "1234")
(3円 1円 4円 2円)
If input digits were guaranteed to be unique then you could simply do #(filter(set %)"3145926870")
.
PHP, (削除) 69 (削除ここまで) 68
for(;(~$d=$argn[$j++])||~$c=_3145926870[$i+++$j=0];)$c==$d&&print$d;
Still beaten by preg_filter but I thought it was quite nice itself. Maybe someone can golf off some bytes.
-
\$\begingroup\$
$c!=$d?:print$d
as alternative for$c==$d&&print$d
I only see in the moment \$\endgroup\$Jörg Hülsermann– Jörg Hülsermann2017年05月02日 14:21:12 +00:00Commented May 2, 2017 at 14:21 -
1\$\begingroup\$
_3145926870
instead of `"3145926870" save 1 Byte \$\endgroup\$Jörg Hülsermann– Jörg Hülsermann2017年05月02日 14:26:18 +00:00Commented May 2, 2017 at 14:26 -
\$\begingroup\$
for(;(~$d=$argn[$j++])?:~$c=_3145926870[++$i+$j=0];$c!=$d?:print$d);
is also a working alternative \$\endgroup\$Jörg Hülsermann– Jörg Hülsermann2017年05月02日 14:57:14 +00:00Commented May 2, 2017 at 14:57
Pip, 10 bytes
PI.0@?_SKa
Explanation
SKa Sort the command-line argument a by this key function:
@?_ Index of first occurrence in
PI.0 Pi (3.141592653589793) concatenated with 0
If the 0
isn't added, PI@?0
returns nil, which behaves... strangely... when used as a sorting key.
-
\$\begingroup\$ Try it Online! for 6 bytes \$\endgroup\$2023年06月18日 00:43:00 +00:00Commented Jun 18, 2023 at 0:43
Perl 6, 34 bytes
*.comb.sort:{3145926870.index: $_}
*\ # WhateverCode lambda (this is the parameter)
.comb # split into digits
.sort: { # sort by
3145926870.index: $_ # its index in this number
}
-
\$\begingroup\$ you can save 1 byte by dropping '3' from the lookup. \$\endgroup\$jdt– jdt2021年08月13日 23:34:13 +00:00Commented Aug 13, 2021 at 23:34
k, 19 bytes
{x@<"3145926870"?x}
Explanation:
{ } /function(x)
"3145926870"?x /for each x: "3145926870".index(x)
< /get indices with which to sort
x@ /sort x by those indices
TI-Basic, 100 bytes
Input Str1
seq(expr(sub(Str1,I,1)),I,1,length(Str1→A
seq(1+sum(not(cumSum(Ans(I)={3,1,4,5,9,2,6,8,7}))),I,1,dim(Ans→B
SortD(ʟB,ʟA
sum(seq(10^(I-1)ʟA(I),I,1,dim(ʟA
Takes input as a string. Outputs a number that is stored in Ans
and is displayed at the end.