(Inspired by the Keg utility of this challenge)

Given a non-empty input string, e.g. s c 1= e(a"E"), split the input into even-odd chunks.

Example (Feel free to suggest more)

I can only think of this test case, fee free to suggest more.

This input string, when mapped to its code points, yields the list [115, 32, 99, 32, 49, 61, 32, 101, 40, 97, 34, 69, 34, 41]. When applied modulo-2 for every item, this returns [1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1].

In this list let's find the longest possible chunk that is consistent with even and odd code points:

[1, 0, 1, 0, 1], [1, 0, 1, 0, 1, 0, 1, 0, 1]

For the first chunk, this yields [1, 0, 1, 0, 1] because this is the longest chunk that follows the pattern

Odd Even Odd Even Odd Even ...


Even Odd Even Odd Even Odd ...

. Adding another codepoint into [1, 0, 1, 0, 1, 1] breaks the pattern, therefore it is the longest possible even-odd chunk that starts from the beginning of the string.

Using this method, we should split the input into chunks so that this rule applies. Therefore the input becomes (the ; here is simply a separator; this can be any separator that is not an empty string, Including the string itself):

s c 1;= e(a"E")

However, returning a list of strings is also permitted.


  • This is so the shortest solution wins. Let it be known that flags don't count towards being in the pattern. They also don't count towards byte count in this challenge.

  • The input will only be in ASCII, and the mapping will always be in ASCII (as far as I can tell most golflangs use a superset of ASCII).

Answering some of the comments

  • You may output strings as lists of codepoints.

  • "Any separator" includes the input string itself.

  • You may insert other characters like the MATL answer, such as alphanumeric characters.

  • You may not use integers as input instead of ASCII. Doing that will trivialize the challenge.

  • 2
    \$\begingroup\$ Some more test cases would be helpful. Also, can the input be empty? \$\endgroup\$ – Luis Mendo Dec 22 '19 at 13:56
  • 2
    \$\begingroup\$ May we insert other characters / strings like the MATL answer? \$\endgroup\$ – Jonathan Allan Dec 22 '19 at 15:02
  • 4
    \$\begingroup\$ May we simply return a list of strings, like the Japt answer used to, and the 05AB1E answer does? \$\endgroup\$ – Jonathan Allan Dec 22 '19 at 15:04
  • 5
    \$\begingroup\$ Can we use integers as input instead of ASCII? It seems needlessly complex to require ASCII as input when we only care about their code points. \$\endgroup\$ – Post Rock Garf Hunter Dec 22 '19 at 16:50
  • 1
    \$\begingroup\$ ...Can output be a list of lists of code-points? \$\endgroup\$ – Jonathan Allan Dec 23 '19 at 13:17

11 Answers 11


05AB1E, 12 6 4 5 bytes


Input as a string, output as a 2D list of ASCII codepoint integers.

-6 bytes thanks to @Grimmy.
+1 byte now that the allowed I/O rules are finally settled..

Try it online or try it online with output as a list of strings instead.


Ç      # Transform the (implicit) input-string to a list of codepoint integers
 .¬    # Split this list of integers at:
   +   #  Sum the two codepoint at both sides of the potential split
    È  #  And check whether it is odd
       # (after which the resulting 2D integer list is output implicitly)
  • 2
    \$\begingroup\$ 7: Ç¥È0šÅ¡, 6: .¬JÇOÈ \$\endgroup\$ – Grimmy Dec 22 '19 at 14:55
  • \$\begingroup\$ The output has more than just the string joined by a separator string. If this is allowed then a function returning a list of strings should be allowed. I have inquired. \$\endgroup\$ – Jonathan Allan Dec 22 '19 at 15:07
  • \$\begingroup\$ @Grimmy I wasn't even aware we had a builtin (and it worked like that - I had to add debug-lines to see it was checking between a pair of characters). Thanks! \$\endgroup\$ – Kevin Cruijssen Dec 22 '19 at 18:31
  • 2
    \$\begingroup\$ @JonathanAllan Ah, missed that part about the delimiter. I assumed a flexible output was allowed by default, since it wasn't mentioned (I read past it). I'll await clarification from OP, but I've added the three trailing characters required to fix it in my answer. \$\endgroup\$ – Kevin Cruijssen Dec 22 '19 at 18:34
  • \$\begingroup\$ @JonathanAllan OP has clarified that an output as list or using the input as delimiter is allowed. \$\endgroup\$ – Kevin Cruijssen Dec 23 '19 at 11:11

Japt -R, 9 bytes

óÈcv ¦Ycv

Try it

óÈcv ¦Ycv     :Implicit input of string
ó             :Partition
 È            :Between characters X & Y where
  c           :Charcode of X
   v          :Parity
     ¦        :Is not equal to
      Ycv     :Parity of charcode of Y
              :Implicit output, joined by newlines
  • \$\begingroup\$ The output has more than just the string joined by a separator string. If this is allowed then a function returning a list of strings should be allowed. I have inquired. \$\endgroup\$ – Jonathan Allan Dec 22 '19 at 14:52
  • \$\begingroup\$ @JonathanAllan: Updated. \$\endgroup\$ – Shaggy Dec 22 '19 at 15:02
  • 2
    \$\begingroup\$ Oh ha ha Japt's flags to the rescue. \$\endgroup\$ – Jonathan Allan Dec 22 '19 at 15:03

Python 2, 72 71 bytes

lambda s:''.join(x+(ord(x)-~ord(y))%2*';'for x,y in zip(s,s[1:]))+s[-1]

Try it online!

1 byte thx to Jonathan Allan


Charcoal, 19 bytes


Try it online! Link is to verbose version of code. Explanation:

 θ                  Input string
⭆                   Map over characters and join
    ¶               Literal newline
   ×                Repeated by
      κ             Current index
     ∧              Logical And
       ¬            Logical Not
           ι        Current character
          ℅         Ordinal
         ⁺          Plus
              θ     Input string
             §      Indexed by
                κ   Current index
               ⊖    Decremented
            ℅       Ordinal
        ﹪           Modulo
                 ²  Literal 2
  ⁺                 Concatenated with
                  ι Current character
                    Implicitly print

Jelly 6 bytes

(7 if we must return a list of strings - add .)


A monadic Link accepting a list of integers which yields a list of lists of integers.

Try it online!


ḂI¬Żœṗj⁷ - Main Link: list of characters (A)
Ḃ        - least significant bit (vectorises)
 I       - incremental differences
  ¬      - logical NOT (vectorises)
   Ż     - prepend a zero
    œṗ   - at truthy indices (of left) partition (right=A)

Original challenge 7 bytes


A full program using 2 as the separator.

Try it online!


żOḂMḊƲƝ - Main Link: list of characters, S    e.g. ['4','5','7','4','0']
                                                   (from program argument '45740')
      Ɲ - for neighbours of S:                     ['4','5'] ['5','7'] ['7','4'] ['4','0']
     Ʋ  -   last four links as a monad:
 O      -     to ordinal (vectorises)              [52,53]   [53,55]   [55,52]   [52,48]
  Ḃ     -     least significant bit (vectorises)   [0,1]     [1,1]     [1,0]     [0,0]
   M    -     maximal indices                      [2]       [1,2]     [1]       [1,2]
    Ḋ   -     dequeue                              []        [2]       []        [2]
        - }                                        [[],[2],[],[2]]
ż       - (S) zip with (that)                      [['4',[]],['5',[2]],['7',[]],['4',[2]],['0']]
        - implicit, smashing print                 4527420

C (clang), 49 bytes


Try it online!

  • 1
    \$\begingroup\$ Very nice idea with the least significant bit sum! Applause!! :-) \$\endgroup\$ – Noodle9 Dec 24 '19 at 10:39

J, 18 14 bytes


Try it online!

-4 bytes thanks to Bubbler

  • 2|] remainders mod 2
  • 2=/\ consecutive pairs of those, are the equal?
  • 1, start it off with a group
  • <;.1~ cut into groups using the first element as a delimiter, ie, starting a new group whenever we consecutive items are equal, odd odd or even even
  • 1
    \$\begingroup\$ 14 bytes. Cut works better in this case. \$\endgroup\$ – Bubbler Dec 23 '19 at 6:37

C (clang), 90 ... 67 66 bytes


Try it online!

Saved 9 10 11 bytes thanks to @ceilingcat!!!


MATL, 10 bytes


Outputs each character on a different line. The separator is 1.

Try it online!


"      % Input: string (implicit). For each
  I    %   Push contents of clipboard I, initially 3
  @    %   Push current character
  2\   %   Modulo 2 of (code point) of that character. Gives 0 or 1
  XI   %   Copy result into clipboard I
  =    %   Equal? This compares the current and previous contents of
       %   clipboard I. Gives true or false
  f    %   Find. This outputs indices of true entries. Gives 1 or []
  @    %   Push current character
       % End (implicit)
       % Display stack (implicit), bottom to top. Each entry is displayed
       % on a different line. [] is not shown and doesn't start a new line

JavaScript (Node.js), 94 97 95 bytes

+1 byte for allowing more than 1 cut. Had to completely rewrite my code but had a lot of help from tsh's answer


Try it online!

  • 1
    \$\begingroup\$ This only work if the input should be split into two strings. Wont work for "aaa", "abcdef". \$\endgroup\$ – tsh Dec 24 '19 at 2:49
  • \$\begingroup\$ @tsh thanks for pointing it out, I made the wrong assumption that was the only expected input. \$\endgroup\$ – Joost K Dec 24 '19 at 10:22

JavaScript (Node.js), 53 bytes


Try it online!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.