एक फिनिश सिनेमा में सीटें


52

आपको बूलियन मैट्रिक्स के रूप में एक सिनेमा थियेटर का नक्शा दिया गया है: 0 एक मुफ्त सीट का प्रतिनिधित्व करता है, 1 - कब्जा कर लिया। प्रत्येक फिन जो आस- पास के कब्जे वाले स्थान से सबसे दूर ( यूक्लिडियन दूरी ) को चुनता है , या यदि ऐसा कई मौजूद हैं - पंक्ति-प्रमुख क्रम में उनमें से पहला । ऑर्डर सीटों को दिखाने वाले मैट्रिक्स का आउटपुट अंततः में कब्जा कर लिया जाएगा; अर्थात् 0 को 2, 3, 4, आदि से बदलें

// in
0 0 0 0 1
0 0 0 0 0
0 0 0 0 0
0 0 1 1 0
// out
 2  8  3  9  1
10  5 11  6 12
 4 13 14 15  7
16 17  1  1 18

// in
0 0 0 0 0 0 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 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 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 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
// out
  5  43  17  44  45  46  18  47   8  48  49   6  50  19  51   2
 52  24  53  54   1  55  56  25  57  26  58  59  27  60  28  61
 20  62  63  29  64  65   1  66  30  67  68  21  69   9  70  71
 72  73   1  74  31  75  76  77  78   1  79  80  32  81  82  11
 12  83  84   1  85  86  87  13  88  89  90  14  91  92  33  93
 94  34  95  96  97  15  98  99  35 100  36 101 102   1 103  22
104 105  37 106  38 107  39 108 109  16 110  40 111 112  41 113
  4 114 115   7 116  23 117   3 118 119  42 120   1 121 122  10

// in
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 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 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
// out
  2  38 39  26  40   6 41  42  12  43  44   7  45  46  27  47   3
 48  49 15  50  28  51 52  29  53  30  54  55  56  16  57  31  58
 32  59 60  33  61  62 17  63  64  65  18  66  67  68  34  69  35
 70  10 71  72  13  73 74  75   1  76  77  78  11  79  80  14  81
 82  83 36  84  85  86 21  87  88  89  22  90  91  37  92  93  94
 19  95 96  97  23  98 99 100  24 101 102 103  25 104 105 106  20
107 108  4 109 110 111  8 112 113 114   9 115 116 117   5 118 119

I / O प्रारूप आपकी भाषा के लिए स्थापित कोड गोल्फिंग मानदंडों के भीतर लचीला है। आप मान सकते हैं कि इनपुट सही है, आकार का कम से कम 3x3, और इसमें पूरी तरह से समान बूलियन मान शामिल नहीं है। एक समारोह या एक पूरा कार्यक्रम लिखें। प्रति भाषा सबसे छोटा समाधान विजेता माना जाता है; कोई उत्तर स्वीकार नहीं किया जाएगा। मानक खामियों को मना किया जाता है।


6
@ मैं एक असामाजिक व्यक्ति के रूप में, मैं इस बात की पुष्टि कर सकता हूं कि मैं पक्ष में या पीछे की ओर एक सीट के बजाय किसी के पीछे दो सीटें या किसी के पीछे दो सीटें बैठाऊंगा।
पावेल

17
@ आम व्यक्तिगत स्थान की गणना यूक्लिडियन दूरी के साथ की जाती है :)
एंग्स

2
@ पावेल असोसियल, असामाजिक नहीं?
क्रोमैटिक्स

2
@ क्रोमोमिक्स नोप। मैं चाहता हूं कि समाज जल जाए। : पी
पावेल

12
@ पावेल हर्नोसोशल :)
ngn

जवाबों:


11

MATL , 37 बाइट्स

!t~z:Q"@yX:gG&n:!J*w:+X:&-|w/X<&X>(]!

इसे ऑनलाइन आज़माएं! या सभी परीक्षण मामलों को सत्यापित करें । आप ASCII कला में भरे सिनेमा को भी देखना चाहेंगे ।

व्याख्या

!t        % Implicit input: M×N matrix of zeros and ones. Transpose and duplicate.
          % The transpose is needed because MATL uses column-major (not row-major)
          % order. It will be undone at the end
~z        % Number of zeros, say Z
:Q        % Range, add 1 element-wise: gives the array [2, 3, ..., Z+1]. These are
          % the new values that will be written into the matrix
"         % For each k in that array
  @       %   Push k. Will be written in a position to be determined
  y       %   Duplicate from below: pushes a copy of the current matrix, that has
          %   values up to k-1 already written in
  X:      %   Linearize into an (R*C)×1 vector, in column-major order
  g       %   Convert to logical: this replaces non-zero values by 1 
  G&n     %   Push input size as two separate numbers: M, N
  :!      %   Range, transpose: gives the column vector [1; 2; ...; N]
  J*      %   Multiply by imaginary unit, 1j, element-wise
  w:      %   Swap, range: gives the row vector [1, 2, ..., M]
  +       %   Add, with broadcast. Gives an N×M complex matrix defining a grid of
          %   coordinates: [1+1j, ..., M+1j; 2+1j, ... 2+1j; ...; N+1j, ..., N+Mj]
  X:      %   Linearize into an (M*N)×1 vector, in column-major order
  &-|     %   (M*N)×(M*N) matrix of absolute differences. This gives all distances
          %   between seats. Rows of this matrix represent currently used seats,
          %   and columns correspond to potential new positions
  w/      %   Swap, divide with broadcast. This divides the rows representing
          %   occupied seats by 1, and those with unocuppied seats by 0. So the
          %   latter rows are set to infinity, which effectively removes them for
          %   the subsequent minimization
  X<      %   Mimimum of each column: this gives the minimum distance to currently
          %   occupied seats for each potential new seat
  &X>     %   Argument maximum: gives the index of the first maximizing value
  (       %   Write value k at that position, using linear indexing
]         % End
!         % Transpose. Implicit display

11

जावास्क्रिप्ट (ईएस 6), 156 137 बाइट्स

@ L4m2 के लिए धन्यवाद 18 बाइट्स सहेजे गए

यह काफी है map()...

f=(a,n=1)=>a.map(B=(r,y)=>r.map((_,x)=>a.map(b=q=>q.map(v=>b=b<(d=X*X--+Y*Y)|!v?b:d,X=x)&Y--,Y=y)|v|b<=B||(R=r,C=x,B=b)))|B?f(a,R[C]=++n):a

इसे ऑनलाइन आज़माएं!

टिप्पणी की गई

f = (a, n = 1) =>               // a = input array; n = seat counter
  a.map(B =                     // initialize B to a non-numeric value
    (r, y) =>                   // for each row r at position y in a[]:
    r.map((_, x) =>             //   for each target seat at position x in r[]:
      a.map(b =                 //     initialize b to a non-numeric value
        q =>                    //     for each row q in a[]:
        q.map(v =>              //       for each reference seat v in q[]:
          b = b < (             //         if b is less than d, defined as
            d = X * X-- + Y * Y //           the square of the Euclidean distance
          ) | !v ?              //           or the reference seat is empty
            b                   //             let b unchanged
          :                     //           else:
            d,                  //             update b to d
          X = x                 //         start with X = x
        ) & Y--,                //       end of q.map(); decrement Y
        Y = y                   //       start with Y = y
      ) |                       //     end of inner a.map()
      b <= B ||                 //     unless b is less than or equal to B,
      (R = r, C = x, B = b)     //     update B to b and save this position in (R, C)
    )                           //   end of r.map()
  ) | B ?                       // end of outer a.map(); if B was updated:
    f(a, R[C] = ++n)            //   update the best target seat and do a recursive call
  :                             // else:
    a                           //   stop recursion and return a[]


b=b<(d=X*X--+Y*Y)|!v?b:d
l4m2

v|b<=B v|अनावश्यक cuz है अगर vतबb=0
l4m2

3

हास्केल , 216 213 185 184 बाइट्स

import Data.Array
m a=[snd$maximum a|a/=[]]
f k=k//m[(r,((a,b),maximum(elems k)+1::Int))|s<-[assocs k],((a,b),0)<-s,r<-[minimum[(x-a)^2+(y-b)^2|((x,y),i)<-s,i>0]]]
(until=<<((==)=<<))f

एक सरणी के रूप में इनपुट लेता है। इनपुट और आउटपुट रिवर्स ऑर्डर में हैं। फिक्की के लिए फिक्स्ड पॉइंट मैजिक का श्रेय लियाकोनी को दिया गया

इसे ऑनलाइन आज़माएं!


1
180 बाइट्स के साथuntil((==)=<<f)f
ovs

3

पायथन 2 , 200 187 बाइट्स

a=input()
z=len(a[0]);P=[divmod(i,z)for i in range(len(a)*z)];i=2
while 0in sum(a,[]):t,y,x=max((min((u-U)**2+(v-V)**2for V,U in P if a[V][U]),-v,-u)for v,u in P);a[-y][-x]=i;i+=1
print a

इसे ऑनलाइन आज़माएं!

-13 बाइट्स टीके से एक टिप के लिए नहीं है कि कोशिकाओं 0 के लिए अनावश्यक जाँच को हटाकर चार्ल्स


मेरे पास लगभग एक ही समाधान है, लेकिन पायथन 3, एक फ़ंक्शन, और 194 बाइट्स
ऐसा नहीं है कि चार्ल्स

मेरा पोस्टिंग के बजाय, मुख्य बचत यह है कि मैं ,v,uजनरेटर के अंत में जोड़ता हूं max, और आपको ऐसा करने की आवश्यकता नहीं है if a[v][u]<1क्योंकि वे 0अधिकतम होंगे और इसलिए अधिकतम नहीं। तो मेरी लाइन मूल रूप से है*_,y,x=max((min(...),-v,-u,v,u)for v,u in P)
नहीं कि चार्ल्स

आश्चर्यजनक समान कोड, हालांकि। वाह।
चार्ल्स

ईमानदारी से, मुझे यकीन नहीं है कि *,v,uचरित्रों को जोड़ना आपके लिए एक बचत --है। :)
नहीं कि चार्ल्स

@Not चार्ल्स कि: अच्छी, पूरी तरह से याद किया है कि if a[v][u]<1अनावश्यक है (के बाद से गैर शून्य कोशिकाओं एक होगा min()की 0)।
चास ब्राउन






1

क्लोजर, 247 बाइट्स

#(let[R(range(count %))C(range(count(% 0)))](loop[M % s 2](if-let[c(ffirst(sort-by last(for[x R y C :when(=((M x)y)0)][[x y](-(nth(sort(for[i R j C :when(>((M i)j)0)](+(*(- i x)(- i x))(*(- j y)(- j y)))))0))])))](recur(assoc-in M c s)(inc s))M)))

इनपुट एक vec-of-vecs है M, जिसे बाय में संशोधित किया loopजाता है assoc-in। जब कोई खाली स्थान नहीं मिलता है ( if-letतब) परिणाम वापस आ जाता है।


1

जेली , 35 33 30 29 बाइट्स

ZJæịþJFạþx¥F«/MḢṬ×FṀ‘Ɗo@FṁµÐL

इसे ऑनलाइन आज़माएं!

बदला ×ı+साथ æị(जटिल गठबंधन), एक नई युग्म आधारित पर j.जम्मू से, एक बाइट की बचत।

यहाँ TIO के लिए एक अधिक कुशल संस्करण है। इसे ऑनलाइन आज़माएं!

व्याख्या

ZJæịþJFạþx¥F«/MḢṬ×FṀ‘Ɗo@FṁµÐL  Input: matrix M
Z                              Transpose
 J                             Enumerate indices - Get [1 .. # columns]
     J                         Enumerate indices - Get [1 .. # rows]
  æịþ                          Outer product using complex combine
                                 (multiply RHS by 1j and add to LHS)
      F                        Flatten
           F                   Flatten input
          ¥                    Dyadic chain
         x                       Times - Repeat each of LHS by each of RHS
       ạþ                        Outer product using absolute difference
            «/                 Reduce by minimum
              M                Indices of maximal values
               Ḣ               Head
                Ṭ              Untruth - Return a Boolean array with 1's at the indices
                 ×             Times
                     Ɗ         Monadic chain
                  F              Flatten input
                   Ṁ             Maximum
                    ‘            Increment
                      o@       Logical OR
                        F      Flatten input
                         ṁ     Mold - Reshape to match the input
                          µÐL  Repeat until result converges

हमारी साइट का प्रयोग करके, आप स्वीकार करते हैं कि आपने हमारी Cookie Policy और निजता नीति को पढ़ और समझा लिया है।
Licensed under cc by-sa 3.0 with attribution required.