Boustrophedonise


35

Связанные, но очень разные.

Бустрофедон представляет собой текст , в котором каждая вторая линия письма переворачивается или обратить вспять, с перевернутым буквами.

В этой задаче мы просто перевернем все остальные строки, но оставим действительные символы без изменений. Вы можете выбрать, какие линии повернуть вспять, при условии, что это будут все остальные.

Вы можете взять текст в любом подходящем формате, если вы поддерживаете ноль или более строк печатного ASCII, каждая из которых содержит ноль или более символов.

Примеры:

["Here are some lines","of text for you","to make a","boustrophedon"]:

["Here are some lines","uoy rof txet fo","to make a","nodehportsuob"] or ["senil emos era ereH","of text for you","a ekam ot","boustrophedon"]


["My boustrophedon"]:

["My boustrophedon"] or ["nodehportsuob yM"]

[]:  
[]

["Some text","","More text","","","Last bit of text"]:

["Some text","","More text","","","txet fo tib tsaL"] or ["txet emoS","","txet eroM","","","Last bit of text"]

Не могу понять, должны ли возвращаться и вводиться строки, разделенные текстом, или это может быть файл или список строк.
sergiol


Может ли мой код вести себя непоследовательно, то есть иногда начинать движение с первой строки, а иногда со второй?
Эрик Outgolfer

2
@EriktheOutgolfer Да, я спрашивал об этом ранее и формулировку «Вы можете выбрать, какие строки повернуть вспять, при условии, что это будут все остальные». на самом деле был изменен на то, что сейчас, чтобы сделать его достаточно общим для такого поведения.
Мартин Эндер

1
@totallyhuman Да, согласно ОП.
Адам

Ответы:


20

APL (Dyalog Classic) , 4 байта

⊢∘⌽\

Ввод представляет собой вектор символьных векторов.

это функция, которая обращает вектор (когда is applied monadically).

is " dex " - функция, которая возвращает правильный аргумент. Когда в составе ( ) с другой функцией , fчто заставляет последнее монадическими , как A ⊢∘f Bэквивалентно A ⊢ (f B)и , следовательно f B.

\ is the scan operator. g\A B C ... is the vector A (A g B) (A g (B g C)) ... where g is applied dyadically (infix notation). Substituting ⊢∘⌽ for g it simplifies to:

A (A ⊢∘⌽ B) (A ⊢∘⌽ (B ⊢∘⌽ C)) ...
A (⌽B) (⌽⌽C) ....
A (⌽B) C ....

The reversals at even positions (or odd, depending on how you count) cancel out.

Try it online!


4
That’s literally ]&|.&.>/\ for those who can read J.
FrownyFrog

2
This is really clever.
Erik the Outgolfer

13

Haskell, 26 bytes

zipWith($)l
l=id:reverse:l

Try it online! Usage example: zipWith($)l ["abc","def","ghi"] yields ["abc","fed","ghi"].

Explanation:

l is an infinite list of functions alternating between the identity function and the reverse function.

The main function zips l and the input list with the function application $, that is for an input ["abc", "def", "ghi"] we get [id$"abc", reverse$"def", id$"ghi"].


11

Husk, 4 bytes

z*İ_

Takes and returns a list of strings (the interpreter implicitly joins the result by newlines before printing). The first string is reversed. Try it online!

Explanation

z*İ_  Implicit input.
  İ_  The infinite list [-1,1,-1,1,-1,1..
z     Zip with input
 *    using multiplication.

In Husk, multiplying a string with a number repeats it that many times, also reversing it if the number is negative.


6

JavaScript (ES6), Firefox, 43 bytes

This version abuses the sort algorithm of Firefox. It generates garbage on Chrome and doesn't alter the strings at all on Edge.

a=>a.map((s,i)=>[...s].sort(_=>i&1).join``)

Test cases

Or Try it online! (SpiderMonkey)


JavaScript (ES6), 45 bytes

a=>a.map(s=>(a^=1)?s:[...s].reverse().join``)

Test cases


6

APL (Dyalog Unicode), 10 bytes

⌽¨@{2|⍳≢⍵}

Works both ways:

Try it online! with ⎕IO←1

Try it online! with ⎕IO←0

How it works:

⌽¨@{2|⍳≢⍵}  tacit prefix fn
   {   ≢⍵}  Length of the input
           generate indexes from 1 (or 0 with IO0)
    2|      mod 2; this generates a boolean vector of 0s (falsy) and 1s (truthy)
  @         apply to the truthy indexes...
⌽¨          reverse each element




3

K (oK), 17 14 bytes

Solution:

@[;&2!!#x;|]x:

Try it online!

Example:

@[;&2!!#x;|]x:("this is";"my example";"of the";"solution")
("this is"
"elpmaxe ym"
"of the"
"noitulos")

Explanation:

Apply reverse at odd indices of the input list:

@[;&2!!#x;|]x: / the solution
            x: / store input as variable x
@[;      ; ]   / apply @[variable;indices;function] (projection)
          |    / reverse
       #x      / count (length) of x, e.g. 4
      !        / til, !4 => 0 1 2 3
    2!         / mod 2, 0 1 2 3 => 0 1 0 1       
   &           / where true, 0 1 0 1 => 1 3

Notes:

  • switched out &(#x)#0 1 for &2!!#x to save 3 bytes



3

Alumin, 66 bytes

hdqqkiddzhceyhhhhhdaeuzepkrlhcwdqkkrhzpkzerlhcwqopshhhhhdaosyhapzw

Try it online!

FLAG: h
hq
  CONSUME A LINE
  qk
  iddzhceyhhhhhdaeuze
  pk
  rlhcw
  REVERSE STACK CONDITIONALLY
  dqkkrhzpkzerlhcwqops

  OUTPUT A NEWLINE
  hhhhhdao
syhapzw

2
@totallyhuman This is actually my language.
Conor O'Brien


2

R, 85 bytes

for(i in seq(l<-strsplit(readLines(),"")))cat("if"(i%%2,`(`,rev)(l[[i]]),"\n",sep="")

Try it online!

Input from stdin and output to stdout.

Each line must be terminated by a linefeed/carriage return/CRLF, and it prints with a corresponding newline. So, inputs need to have a trailing linefeed.



2

T-SQL, 65 bytes

Our standard input rules allow SQL to input values from a pre-existing table, and since SQL is inherently unordered, the table must have row numbers to preserve the original text order.

I've defined the table with an identity column so we can simply insert lines of text sequentially (not counted toward byte total):

CREATE TABLE t 
    (i int identity(1,1)
    ,a varchar(999))

So to select and reverse alternating rows:

SELECT CASE WHEN i%2=0THEN a
ELSE reverse(a)END
FROM t
ORDER BY i

Note that I can save 11 bytes by excluding the ORDER BY i, and that is likely to return the list in the original order for any reasonable length (it certainly does for the 4-line example). But SQL only guarantees it if you include the ORDER BY, so if we had, say, 10,000 rows, we would definitely need this.


2

Perl 6, 44 bytes

lines.map: ->\a,$b?{a.put;.flip.put with $b}

Try it

lines               # get the input as a list of lines
.map:
-> \a, $b? {        # $b is optional (needed if there is an odd number of lines)
  a.put;            # just print with trailing newline
  .flip.put with $b # if $b is defined, flip it and print with trailing newline
}


1

Actually, 7 bytes

;r'R*♀ƒ

Explanation:

;r'R*♀ƒ
;r       range(len(input))
  'R*    repeat "R" n times for n in range
     ♀ƒ  call each string as Actually code with the corresponding input element as input (reverse each input string a number of times equal to its index)

Try it online!


1

Alice, 13 bytes

M%/RM\
d&\tO/

Try it online!

Input via separate command-line arguments. Reverses the first line (and every other line after that).

Explanation

       At the beginning of each loop iteration there will always be zero
       on top of the stack (potentially as a string, but it will be
       converted to an integer implicitly once we need it).
M      Push the number of remaining command-line arguments, M.
%      Take the zero on top of the stack modulo M. This just gives zero as
       long as there are arguments left, otherwise this terminates the
       program due to the division by zero.
/      Switch to Ordinal mode.
t      Tail. Implicitly converts the zero to a string and splits off the
       last character. The purpose of this is to put an empty string below
       the zero, which increases the stack depth by one.
M      Retrieve the next command-line argument and push it as a string.
/      Switch back to Cardinal mode.
d      Push the stack depth, D.
&\R    Switch back to Ordinal mode and reverse the current line D times.
O      Print the (possibly reversed) line with a trailing linefeed.
\      Switch back to Cardinal mode.
       The instruction pointer loops around and the program starts over
       from the beginning.

1

Standard ML (MLton), 51 bytes

fun$(a::b::r)=a::implode(rev(explode b)):: $r| $e=e

Try it online! Usage example: $ ["abc","def","ghi"] yields ["abc","fed","ghi"].

Explanation:

$ is a function recursing over a list of strings. It takes two strings a and b from the list, keeps the first unchanged and reverses the second by transforming the string into a list of characters (explode), reversing the list (rev), and turning it back into a string (implode).


+1, not enough ML solutions imo
jfh

1

Retina, 18 bytes

{O$^`\G.

*2G`
2A`

Try it online! Explanation: The first stage reverse the first line, then the second stage prints the first two lines, after which the third stage deletes them. The whole program then repeats until there is nothing left. One trailing newline could be removed at a cost of a leading ;.


1

Wolfram Language (Mathematica), 33 bytes

Fold[StringReverse@*Append,{},#]&

Try it online!

How it works

StringReverse@*Append, when given a list of strings and another string as input, adds the string to the end of the list and then reverses all of the strings.

Folding the input with respect to the above means we:

  • Reverse the first line.
  • Add the second line to the end and reverse both of them.
  • Add the third line to the end and reverse all three.
  • Add the fourth line to the end and reverse all four.
  • And so on, until we run out of lines.

Each line gets reversed one time fewer than the previous line, so the lines alternate direction.


1

CJam, 11 bytes

{2/Waf.%:~}

Try it online! (CJam array literals use spaces to separate elements)

Explanation:

{              Begin block, stack: ["Here are some lines" "of text for you" "to make a" "boustrophedon"]
 2/            Group by 2:         [["Here are some lines" "of text for you"] ["to make a" "boustrophedon"]]
   W           Push -1:            [["Here are some lines" "of text for you"] ["to make a" "boustrophedon"]] -1
    a          Wrap in array:      [["Here are some lines" "of text for you"] ["to make a" "boustrophedon"]] [-1]
     f.%       Vectorized zipped array reverse (black magic):
                                   [["senil emos era ereH" "of text for you"] ["a ekam ot" "boustrophedon"]]
        :~     Flatten:            ["senil emos era ereH" "of text for you" "a ekam ot" "boustrophedon"]
          }

Explanation for the Waf.% "black magic" part:

  • W is a variable preinitialized to -1. a wraps an element in an array, so Wa is [-1].
  • % pops a number n and an array a and takes every nth element of the array. When n is negative, it also reverses it, meaning that W% reverses an array.
  • . followed by a binary operation applies that operation to corresponding elements of an array, so [1 2 3] [4 5 6] .+ is [5 7 9]. If one array is longer than the other, the elements are kept without modification, meaning that Wa.% reverses the first element of an array.
  • f followed by a binary operation will take an element from the stack and then act like {<that element> <that operation>}%, that is, go through each element in the array, push its element, push the element first popped from the stack, run the operation, and then collect the results back into an array. This means that Wa.f% reverses the first element of every element in the array.


1

Swift, 90 85 82 72 bytes

-10 bytes thanks to @Mr.Xcoder

func f(a:[String]){print(a.reduce([]){$0.map{"\($0.reversed())"}+‌​[$1]})}

You can use print and drop the return type declaration: func f(a:[String]){print(a.reduce([]){$0.map{"\($0.reversed())"}+[$1]})}
Mr. Xcoder

1

Ruby, 19 + 2 = 21 bytes

+2 bytes for -nl flags.

$.%2<1&&$_.reverse!

Try it online!

Explanation

Practically identical to the Perl 5 answer, though I hadn’t seen that one when I wrote this.

With whitespace, the code looks like this:

$. % 2 < 1 && $_.reverse!

The -p option makes Ruby effectively wrap your script in a loop like this:

while gets
  # ...
  puts $_
end

The special variable $_ contains the last line read by gets, and $. contains the line number.

The -l enables automatic line ending processing, which automatically calls chop! on each input line, which removes the the \n before we reverse it.


1

GNU sed, 31 + 1 = 32 bytes

+1 byte for -r flag.

G
:
s/(.)(.*\n)/\2\1/
t
s/.//
N

Try it online!

Explanation

G                   # Append a newline and contents of the (empty) hold space
:
  s/(.)(.*\n)/\2\1/   # Move the first character to after the newline
  t                   # If we made the above substitution, branch to :
s/.//               # Delete the first character (now the newline)
N                   # Append a newline and the next line of input

1

Charcoal, 9 bytes

EN⎇﹪ι²⮌SS

Try it online! Link is to verbose version of code. Note: Charcoal doesn't know the length of the list, so I've added it as an extra element. Explanation:

 N          First value as a number
E           Map over implicit range
    ι       Current index
     ²      Literal 2
   ﹪        Modulo
  ⎇         Ternary
       S    Next string value
      ⮌     Reverse
        S   Next string value
            Implicitly print array, one element per line.

1

Befunge-93, 48 bytes

 <~,#_|#*-+92:+1:
#^_@  >:#,_"#"40g!*40p91+,~:1+

Try It Online

Prints first line in reverse. Has a trailing newline.

Basically, it works by alternating between printing as it gets input and storing the input on the stack. When it reaches a newline or end of input, it prints out the stack, prints a newline, and modifies the character at 0,4 to be either a # or a no-op to change the mode. If it was the end of input, end the program

Используя наш сайт, вы подтверждаете, что прочитали и поняли нашу Политику в отношении файлов cookie и Политику конфиденциальности.
Licensed under cc by-sa 3.0 with attribution required.