Il percorso dello gnu


23

Golf un programma o una funzione che dà il nth posizione del gnu che parte da piazza 1 su un infinito scacchiera che è numerato in una spirale piazza in senso antiorario, dove gli gnu sempre visite il numero più basso quadrato si può raggiungere che non ha ancora visitato.

Ispirazione: The Trapped Knight e OEIS A316667 .

Modifica: questa sequenza è ora sull'OEIS come A323763 .

Il codice può produrre il nth di posizione, le prime n posizioni, o generare la sequenza prendendo alcun input.

Sentiti libero di dare la sua posizione dopo (o fino a) n salti invece, ma in tal caso, indicalo chiaramente nella tua risposta e assicurati che un input di n=0 produca 1(o [1]se appropriato).

Questo è , quindi l'obiettivo è produrre codice funzionante nel minor numero di byte possibile nella lingua scelta.

Nota: lo gnu rimane intrappolato (molto simile al cavaliere fa il suo 2016th posizione, piazza 2084 , e il cammello fa il suo 3723rd , quadrato 7081 ) al suo 12899744968esimo posizione sulla piazza 12851850258 . Il comportamento del codice potrebbe essere indefinito per n maggiore di questo. (Grazie a Deadcode per il codice C ++ che l'ha trovato!)

Dettaglio

La scheda è simile alla seguente e continua indefinitamente:

101 100  99  98  97  96  95  94  93  92  91
102  65  64  63  62  61  60  59  58  57  90
103  66  37  36  35  34  33  32  31  56  89
104  67  38  17  16  15  14  13  30  55  88
105  68  39  18   5   4   3  12  29  54  87
106  69  40  19   6   1   2  11  28  53  86
107  70  41  20   7   8   9  10  27  52  85
108  71  42  21  22  23  24  25  26  51  84
109  72  43  44  45  46  47  48  49  50  83
110  73  74  75  76  77  78  79  80  81  82
111 112 113 114 115 116 117 118 119 120 121

Uno gnu è un pezzo degli scacchi fata "gnu" - un pezzo degli scacchi non standard che può muoversi sia come cavaliere (una (1,2) -foglia) sia come cammello (una (1,3) -foglia).
Come tale, potrebbe spostarsi in una di queste posizioni dalla sua posizione iniziale di 1 :

  .   .   .   .   .   .   .   .   .   .   .
  .   .   .   .  35   .  33   .   .   .   .
  .   .   .   .  16   .  14   .   .   .   .
  .   .  39  18   .   .   .  12  29   .   .
  .   .   .   .   .  (1)  .   .   .   .   .
  .   .  41  20   .   .   .  10  27   .   .
  .   .   .   .  22   .  24   .   .   .   .
  .   .   .   .  45   .  47   .   .   .   .
  .   .   .   .   .   .   .   .   .   .   .

Il più basso di questi è 10 e lei non ha ancora visitato quel quadrato, quindi 10 è il secondo termine della sequenza.

Successivamente potrebbe spostarsi da 10 in una di queste posizioni:

  .   .   .   .   .   .   .   .   .   .   .
  .   .   .   .   .   .  14   .  30   .   .
  .   .   .   .   .   .   3   .  29   .   .
  .   .   .   .   6   1   .   .   .  53  86
  .   .   .   .   .   .   . (10)  .   .   .
  .   .   .   .  22  23   .   .   .  51  84
  .   .   .   .   .   .  47   .  49   .   .
  .   .   .   .   .   .  78   .  80   .   .
  .   .   .   .   .   .   .   .   .   .   .

Tuttavia, ha già visitato piazza 1 quindi la sua terza posizione è piazza3 , la più bassa che non ha ancora visitato.


I primi 100 termini del percorso dello gnu sono:

1, 10, 3, 6, 9, 4, 7, 2, 5, 8, 11, 14, 18, 15, 12, 16, 19, 22, 41, 17, 33, 30, 34, 13, 27, 23, 20, 24, 44, 40, 21, 39, 36, 60, 31, 53, 26, 46, 25, 28, 32, 29, 51, 47, 75, 42, 45, 71, 74, 70, 38, 35, 59, 56, 86, 50, 78, 49, 52, 80, 83, 79, 115, 73, 107, 67, 64, 68, 37, 61, 93, 55, 58, 54, 84, 48, 76, 43, 69, 103, 63, 66, 62, 94, 57, 87, 125, 82, 118, 77, 113, 72, 106, 148, 65, 97, 137, 91, 129, 85

I primi 11 salti sono mosse da cavaliere, quindi i primi 12 termini coincidono con A316667 .


I commenti non sono per una discussione estesa; questa conversazione è stata spostata in chat .
Mego

Risposte:


21

JavaScript (Node.js) ,  191 ... 166  164 byte

Salvato 2 byte grazie a @grimy .

Restituisce il N ° periodo.

n=>(g=(x,y)=>n--?g(Buffer('QPNP1O?O@242Q3C3').map(m=c=>g[i=4*((x+=c%6-2)*x>(y+=c%7-2)*y?x:y)**2,i-=(x>y||-1)*(i**.5+x+y)]|i>m||(H=x,V=y,m=i))&&H,V,g[m]=1):m+1)(1,2)

Provalo online! oppure Vedi una versione formattata

Come?

Indici a spirale

Per convertire le coordinate (x,y) nell'indice a spirale I , per prima cosa calcoliamo il livello L con:

L=max(|x|,|y|)

Che dà:

3210+1+2+333333333232222231321112303210123+13211123+23222223+33333333

Quindi calcoliamo la posizione P nel layer con:

P={2L+x+yif x>y(2L+x+y)if xy

Che dà:

3210+1+2+330123456210123471210125803210369+143234710+254567811+36789101112

L'indice finale io è dato da:

I=4L2P

NB: la formula precedente fornisce una spirale con indice 0.

Nel codice JS, in realtà calcoliamo subito 4L2 con:

i = 4 * (x * x > y * y ? x : y) ** 2

E quindi sottrarre P con:

i -= (x > y || -1) * (i ** 0.5 + x + y)

Mosse dello gnu

(x,y)

321x+1+2+3391128101761213y+1541415+220+331

We walk through them by applying 16 pairs of signed values (dx,dy). Each pair is encoded as a single ASCII character.

 ID | char. | ASCII code | c%6-2 | c%7-2 | cumulated
----+-------+------------+-------+-------+-----------
  0 |  'Q'  |     81     |   +1  |   +2  |  (+1,+2)
  1 |  'P'  |     80     |    0  |   +1  |  (+1,+3)
  2 |  'N'  |     78     |   -2  |   -1  |  (-1,+2)
  3 |  'P'  |     80     |    0  |   +1  |  (-1,+3)
  4 |  '1'  |     49     |   -1  |   -2  |  (-2,+1)
  5 |  'O'  |     79     |   -1  |    0  |  (-3,+1)
  6 |  '?'  |     63     |   +1  |   -2  |  (-2,-1)
  7 |  'O'  |     79     |   -1  |    0  |  (-3,-1)
  8 |  '@'  |     64     |   +2  |   -1  |  (-1,-2)
  9 |  '2'  |     50     |    0  |   -1  |  (-1,-3)
 10 |  '4'  |     52     |   +2  |   +1  |  (+1,-2)
 11 |  '2'  |     50     |    0  |   -1  |  (+1,-3)
 12 |  'Q'  |     81     |   +1  |   +2  |  (+2,-1)
 13 |  '3'  |     51     |   +1  |    0  |  (+3,-1)
 14 |  'C'  |     67     |   -1  |   +2  |  (+2,+1)
 15 |  '3'  |     51     |   +1  |    0  |  (+3,+1)

We keep track of the minimum encountered value in m and of the coordinates of the corresponding cell in (H,V).

Once the best candidate has been found, we mark it as visited by setting a flag in the object g, which is also our main recursive function.

On the first iteration, we start with x=1 and y=2. This ensures that the first selected cell is (0,0) and that it's the first cell to be marked as visited.


3
So much golfing, can't wait for the rundown of how all the magic works!
Jonathan Allan

did you have to use Buffer to force each character to be interpreted as a single byte?
Jonah

1
@Jonah Although it's been deprecated, the Buffer constructor still accepts a string. So, yes, this is a rather cheap way to convert it to a list of bytes -- as opposed to [..."string"].map(c=>do_something_with(c.charCodeAt())).
Arnauld

1
-2 bytes on the coordinate encoding: TIO
Grimmy

@Grimy Nicely done!
Arnauld

8

Coconut, 337 276 bytes

import math
def g((x,y))=
 A=abs(abs(x)-abs(y))+abs(x)+abs(y)
 int(A**2+math.copysign(A+x-y,.5-x-y)+1)
def f():
 p=x,y=0,0;s={p};z=[2,3,1,1]*2
 while 1:yield g(p);p=x,y=min(((a+x,b+y)for a,b in zip((1,1,2,-2,-1,-1,3,-3)*2,z+[-v for v in z])if(a+x,b+y)not in s),key=g);s.add(p)

Returns a generator of values. Could probably be golfed more. (Especially the sequence of difference tuples.) Spiral algorithm taken from this math.se answer.

Try it online!


1
for a,b in ( -> for a,b in( (maybe you can golf the delta tuple of tuples itself too)
Jonathan Allan

1
No need for q and a zip is shorter for the tuples: 306 bytes may still be golfable of course
Jonathan Allan

1
...how about this for 284? EDIT... this for 278
Jonathan Allan

1
FWIW, that math.se answer has x and y swapped and both negative relative to the coordinate system in this challenge (where positive x is right and y is up). Not that it'd make any difference due to the symmetries, but still.
Deadcode

1
0.5->.5 for another byte save; A**2->A*A for one more.
Jonathan Allan

8

05AB1E, 77 65 58 57 52 bytes

Xˆ0UF3D(Ÿ0KãʒÄ1¢}εX+}Dε·nàDtyÆ+yO·<.±*->}D¯KßDˆkèU}¯

-6 bytes thanks to @Arnauld by using a port of his formula.

Outputs the first n+1 values as a list (of decimals).

Try it online (the ï in the footer removes the .0 to make the output more compact, but feel free to remove it to see the actual result).

Code explanation:

Xˆ             # Put integer 1 in the global_array (global_array is empty by default)
0U             # Set variable `X` to 0 (`X` is 1 by default)
F              # Loop the (implicit) input amount of times:
 3D          #  Push the list in the range [-3,3]: [-3,-2,-1,0,1,2,3]
     0K        #  Remove the 0: [-3,-2,-1,1,2,3]
       ã       #  Cartesian product with itself, creating each possible pair: [[3,3],[3,2],[3,1],[3,-1],[3,-2],[3,-3],[2,3],[2,2],[2,1],[2,-1],[2,-2],[2,-3],[1,3],[1,2],[1,1],[1,-1],[1,-2],[1,-3],[-1,3],[-1,2],[-1,1],[-1,-1],[-1,-2],[-1,-3],[-2,3],[-2,2],[-2,1],[-2,-1],[-2,-2],[-2,-3],[-3,3],[-3,2],[-3,1],[-3,-1],[-3,-2],[-3,-3]]
        ʒ   }  #  Filter this list of pairs by:
         Ä     #   Where the absolute values of the pair
          1¢   #   Contains exactly one 1
               #  (We now have the following pairs left: [[3,1],[3,-1],[2,1],[2,-1],[1,3],[1,2],[1,-2],[1,-3],[-1,3],[-1,2],[-1,-2],[-1,-3],[-2,1],[-2,-1],[-3,1],[-3,-1]])
 εX+}          #  Add the variable `X` (previous coordinate) to each item in the list
 D             #  Duplicate this list of coordinates
  ε            #  Map each `x,y`-coordinate to:
   ·           #   Double both the `x` and `y` in the coordinate
    n          #   Then take the square of each
     à         #   And then pop and push the maximum of the two
   Dt          #   Duplicate this maximum, and take its square-root
     yÆ        #   Calculate `x-y`
       +       #   And add it to the square-root
   yO          #   Calculate `x+y`
     ·         #   Double it
      <        #   Decrease it by 1
             #   And pop and push its signum (-1 if < 0; 0 if 0; 1 if > 0)
   *           #   Multiply these two together
    -          #   And subtract it from the duplicated maximum
   >           #   And finally increase it by 1 to make it 1-based instead of 0-based
  }D           #  After the map: Duplicate that list with values
    ¯K         #  Remove all values that are already present in the global_array
      ß        #  Pop the list of (remaining) values and push the minimum
       Dˆ      #  Duplicate this minimum, and pop and add the copy to the global_array
         k     #  Then get its index in the complete list of values
          è    #  And use that index to get the corresponding coordinate
           U   #  Pop and store this coordinate in variable `X` for the next iteration
             # After the outer loop: push the global_array (which is output implicitly)

General explanation:

We hold all results (and therefore values we've already encountered) in the global_array, which is initially started as [1].
We hold the current x,y-coordinate in variable X, which is initially [0,0].

The list of coordinates we can reach based on the current x,y-coordinate are:

[[x+3,y+1], [x+3,y-1], [x+2,y+1], [x+2,y-1], [x+1,y+3], [x+1,y+2], [x+1,y-2], [x+1,y-3], [x-1,y+3], [x-1,y+2], [x-1,y-2], [x-1,y-3], [x-2,y+1], [x-2,y-1], [x-3,y+1], [x-3,y-1]]

The list I mention in the code explanation above holds these values we can jump to, after which the current x,y (stored in variable X) is added.

Then it will calculate the spiral values based on these x,y-coordinates. It does this by using the following formula for a given x,y-coordinate:

T=max((2x)2,(2y)2)
R=T(xy+T)signum((x+y)21)+1

Which is the same formula @Arnauld is using in his answer, but written differently to make use of 05AB1E's builtins for double, square, -1, +1, etc.

(If you want to see just this spiral part of the code in action: Try it online.)

After we've got all the values we can reach for the given x,y-coordinate, we remove all values that are already present in the global_array, and we then get the minimum of the (remaining) values.
This minimum is then added to the global_array, and variable X is replaced with the x,y-coordinate of this minimum.

After we've looped the input amount of times, the program will output this global_array as result.


1
FWIW, here is a port of my own formula to convert the coordinates into spiral indices. It's 5 bytes shorter but yields floats. (I don't know if this is a problem or not.)
Arnauld

(Note that y in your code is y in mine.)
Arnauld

@Arnauld Thanks, that saves 5 additional bytes. :) EDIT: Which you already mentioned in your first comment. ;p
Kevin Cruijssen
Utilizzando il nostro sito, riconosci di aver letto e compreso le nostre Informativa sui cookie e Informativa sulla privacy.
Licensed under cc by-sa 3.0 with attribution required.