Keep elements in sequence that have a letter repeated at least 3 times

Question

Challenge:

Given the input array l with a list of strings, only keep the elements in the sequence that have a letter that's repeated at least 3 times. Like 'lessons' has 3 s letters, so it should be kept in the sequence. But, 'lesson' has only two s letters, so it should be removed.

Notes:

• l will always be a sequence, and its elements will always be strings with only alphabetical characters.

• I am using the example output with Python Lists. You can use any type of sequence in your own language.

Test cases:

['element', 'photoshop', 'good'] -> ['element', 'photoshop']
['happy', 'colorful', 'luggage'] -> ['luggage']
['reference', 'tomorrow', 'today'] -> ['reference', 'tomorrow'] 

This is code-golf, so the shortest code in bytes wins!

Answer 1

Python 3, 51 50 bytes

lambda l:{s for s in l for c in s if s.count(c)>2}

Try it online!

Answer 2

QuadS, 13 bytes

Takes linebreak-separated text

(.).*\1.*\1
%

Try it online!

Search for…

(.) any character (1)
.* zero or more characters
\1 character (1)
.* zero or more characters

and for each match, return:

% the entire line the match occurred on

Answer 3

Japt -f, 7 4 bytes

ü d¤

Try it

ü d¤     :Implicit filter of input array
ü        :Group & sort
  d      :Any truthy (a non-empty string)
   ¤     :  Slice off the first 2 characters

Without flag, 6 bytes

fÈü d¤

Try it

Answer 4

Python 3, 55 bytes

lambda a:filter(lambda x:any(x.count(i)>2for i in x),a)

Returns a filter object. The asterisk is used in the output to unpack the filter object automatically.

Try it online!

Python 2, 55 bytes

lambda a:filter(lambda x:any(x.count(i)>2for i in x),a)

Exactly the same code, just returns a normal Python list instead of a filter object, meaning you don't need to use the asterisk in the output (if that matters).

Try it online!

Answer 5

Nibbles, 6.5 5.5 bytes (11 nibbles)

Edit: -2 bytes by using transpose, inspired by Jonathan Allan's Jelly answer

|$>2`'=~$$

|$              # filter each word int the input by
      =~$       #   group the letters by
         $      #     themselves,
    `'          #   transpose this list of lists,
  >2            #   and remove the first 2 elements
                #   (so if there is at least one letter
                #   present >2 times, the resulting 
                #   list will have nonzero length)

Answer 6

GNU grep -P, 11 bytes

(.).*\1.*\1

P.a. to selecting the language flavour for perl, I'm not counting -P (which sets the language flavour to PCRE from the default BRE).

"List of string" defined as "text file" (POSIX).

Test case typescript:

$ printf '%s\n' \
  'element' 'photoshop' 'good' \
  'happy' 'colorful' 'luggage' \
  'reference' 'tomorrow' 'today' \
| grep -P '(.).*\1.*\1'
element
photoshop
luggage
reference
tomorrow

Answer 7

Python 3, 48 bytes

lambda s:[c for c in s if max(map(c.count,c))>2]

Try it online!

Answer 8

Ruby, 25 bytes

->l{l.grep /(.).*\1.*\1/}

Try it online!

Answer 9

Pip -p, 11 bytes

gFI2<M:^_N_

Try It Online!

Explanation

gFI2<M:^_N_
g            List of command-line arguments
 FI          Filter by this function:
   2<          2 is less than
     M         the maximum of
      :        (force the precedence of unary M to be lower than its rhs)
         N_    the number of occurrences in the argument string of
       ^_      each character in the argument string

Answer 10

JavaScript, 37 bytes

I/O as an array of strings, using GB's Adám's RegEx.

a=>a.filter(s=>/(.).*\1.*\1/.test(s))

Try it online!

44 42 bytes

I/O as an array of character arrays.

a=>a.filter(a=>a.some(c=>(a[c]=-~a[c])>2))

Try it online!

Answer 11

Vyxal, 7 6 bytes

'sĠ'ḢḢ

Try it Online!

Explained

'sĠ'ḢḢ
'      # Filter input where:
 sĠ    #   sorted and grouped by consecutive
   'ḢḢ #   has items where length >= 3

Answer 12

sed -En, 14 bytes

/(.).*\1.*\1/p

Attempt This Online!

Answer 13

x86 .COM, 48 bytes

0100  BA 2F 01 B8 03 0A 89 C7-89 C1 F3 AA CD 21 BE 31
0110  01 8A 4C FF 0F B6 D9 C7-40 01 0A 24 AC 89 C3 FE
0120  0F 74 04 E2 F7 EB D9 B2-31 B4 09 CD 21 EB D1 FF

org 100h
sta:    mov dx, bu
    mov ax, 0x0A03
    mov di, ax
    mov cx, ax
la: rep stosb
    int 21h
    mov si, bu+2
    mov cl, [si-1]
    movzx bx, cl
    mov [bx+si+1], word '$'*256+10
lb: lodsb
    mov bx,ax
    dec byte [bx]
    jz ok
    loop lb
    jmp sta
ok: mov dl, bu+2-$100
    mov ah, 9
    int 21h
    jmp sta
bu: db -1

Unluckily it behaves bad, input is also displayed

Answer 14

Jelly, 6 bytes

ĠZṫɗƇ3

A monadic Link that accepts a list of lists of characters and yields the filtered list.

Try it online!

How?

ĠZṫɗƇ3 - Link: words
     3 - set the right argument to three
    Ƈ  - filter (words) keeping those for which:
   ɗ   -   last three links as a dyad - f(word, 3):
Ġ      -     group indices by value - e.g. "aardvark" -> [[1,2,6],[4],[8],[3,7],[5]]
 Z     -     transpose                                   [[1,4,8,3,5],[2,7],[6]]
  ṫ    -     tail from index (3)                         [[6]]
       -     (an empty list is falsey)                   true (the third "a" gives the 6)

Answer 15

Pip -p, 14 bytes

{FIaM_Na>2}FIg

Try It Online!

{FIaM_Na>2}FIg
{         }FIg   Keep words that return truthy resuts for ...
   aM_Na>2       Does the letter appear more than two times?
 FI              Filter out falsy results (return either "1" or "")

Answer 16

Retina 0.8.2, 13 bytes

G`(.).*\1.*\1

Try it online! Takes newline-separated words. Explanation: G is Retina's "grep" operation, which just keeps matching lines, according to the obvious regular expression.

Answer 17

Charcoal, 12 bytes

ＷＳ¿⊙ι›№ικ²⟦ι

Try it online! Link is to verbose version of code. Takes newline-terminated words. Explanation:

ＷＳ

Loop through the words.

¿⊙ι›№ικ²

If any character appears more than twice, then...

⟦ι

... output the word on its own line.

9 bytes by taking the input in JSON format:

Φθ⊙ι›№ιλ²

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

 θ          Input array
Φ           Filtered where
   ι        Current word
  ⊙         Any character satisfies
     №      Count of
       λ    Current character
      ι     In current word
    ›       Is greater than
        ²   Literal integer `2`
            Implicitly output each match on its own line

Answer 18

Jelly, 7 bytes

ĠẈ>2ẸµƇ

Try it online!

How it works

ĠẈ>2ẸµƇ - Main link. Takes a list of words L on the left
     µƇ - Filter L by the following:
Ġ       -   Group the indices of identical letters
 Ẉ      -   Length of each group
  >2    -   Greater than 2?
    Ẹ   -   Any true?

Answer 19

C (gcc), 95 90 85 bytes

• -5 thanks to l4m2
• -5 thanks to ceilingcat and c--

Lowercase strings only.

Iterates through each string and prints it as soon as it encounters 3 of a letter, skipping the string otherwise. Rather than clear the counters each time, I simply create brand new counters.

f(int**s){for(char*t,*i;t=*s;*t&&puts(*s),s++)for(i=calloc(99,9);*t&&++i[*t]<3;t++);}

Try it online!

Answer 20

Python 3, 54¹ 53 50 bytes

To get the ball rollin'...

-1 thanks to a fix pointed out by xnor
-3 bytes because sets are also fine

lambda n:{i for i in n for y in i if i.count(y)>2}

_{1: crossed out 54 still looks like 54 :(}

Answer 21

Java 8, 100 bytes

l->{l.removeIf(s->s.chars().noneMatch(c->s.length()-s.replace(""+(char)c,"").length()>2));return l;}

Try it online!

Answer 22

APL (Dyalog Extended), 11 bytes

Anonymous tacit prefix function taking a list of strings

⊢/⍨3∊¨3⌊⍧⍨¨

Try it online!

⊢ the argument…

/⍨ filtered by…

3∊¨ whether 3 is a member of each,…

3⌊ where 3 caps the values of…

⍧⍨¨ the self-counts for each

Answer 23

Pyth, 7 bytes

.-#+GGQ

Try it online!

Explanation

  #        # filter each element of
      Q    # eval(input())
.-         # on the bag-wise subtraction of
   +GG     # two alphabets concatenated

Answer 24

Factor + math.unicode, 43 bytes

[ [ histogram values [ 2 > ] ∃ ] filter ]

Attempt This Online!

• [ ... ] filter Select sequences whose
• histogram histograms
• values [ ... ] ∃ contain any values
• 2 > greater than two

Answer 25

Haskell, 54 bytes

import Data.List
r=filter(any((>2).length).group.sort)

Try it online!

Answer 26

05AB1E, 7 6 bytes

ʒD¢à3@

I/O as a list of lists of characters.

Try it online.

Old 7 bytes answers with I/O as a list of strings:

ʒ{Åγà3@

Try it online.

{Åγà could alternatively be D.M¢ for the same byte-count:
Try it online.

Explanation:

ʒ       # Filter the (implicit) input-list of lists of characters:
 D      #  Duplicate the current list of characters
  ¢     #  Pop both, and count how many times each character occurs in the list
   à    #  Pop and leave the maximum
    3@  #  Check whether this maximum is >= 3
        # (after which the filtered list of lists of characters is output implicitly)

ʒ       # Filter the (implicit) input-list of strings:
 {      #  Sort the characters in the string
  Åγ    #  Run-length encode it; pushing a list of characters and lengths separated to
        #  the stack
    à   #  Pop the list of lengths and leave its maximum
     3@ #  Check whether this maximum is >= 3
        # (after which the filtered list of strings is output implicitly)

 D      #  Duplicate the current string
  .M    #  Pop and push its most frequently occurring character
    ¢   #  Pop both, and count how many times this most frequent character occurs in the
        #  string

Answer 27

QB45, 122 bytes

for t=1to n:redim p(26):for i=1to len(a$(t)):v=asc(mid$(a$(t),i,1))-96:p(v)=p(v)+1:if p(v)>2then ?a$(t):exit for
next:next

Answer 28

PowerShell Core, 30 bytes

$args|?{$_-match'(.).*\1.*\1'}

Try it online!

Or 40 bytes without a regex: $args|?{($_|% t*y|group|%{$_|% c*})-ge3}

Answer 29

Arturo, 62 35 bytes

$=>[select&=>[tally&|some?=>[&>2]]]

Try it

-27 due to 0.9.82 introducing tally.

Answer 30

Julia 1.7, 37 bytes

!l=l[l.|>i->any(count.([i...],i).>2)]

Try it online!