パスカルの三角形は、最初の行の1から開始して生成されます。後続の行では、数値は、そのすぐ上の左右の2つの数値の合計によって決定されます。

デモンストレーションのために、Pascalの三角形の最初の5行は次のとおりです。

    1
   1 1
  1 2 1
 1 3 3 1
1 4 6 4 1

チャレンジ

入力n（ただし、選択した言語で最も便利な場合）を指定すると、Pascalの三角形の最初のn行が生成されます。nは1から25までの整数であると仮定できます。各行と各番号の間には改行が必要ですが、それ以外に、好きなようにフォーマットできます。

これはcode-golfであるため、最短のソリューションが優先されます。

I / Oの例

> 1
1
> 9
1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1
1 6 15 20 15 6 1
1 7 21 35 35 21 7 1
1 8 28 56 70 56 28 8 1

J、12文字

":@(!{:)\@i.

   i.5
0 1 2 3 4
   {：i.5
4
   （i.5）！{：i.5
1 4 6 4 1
   （！{:) i.5
1 4 6 4 1
   （！{：）\ i.5
10000
1 1 0 0 0
1 2 1 0 0
1 3 3 1 0
1 4 6 4 1
   "：@（！{：）\ i.5
1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
   （ "：@（！{：）\ @ i。）` ''
+ ---------------------------------- +
| +-+ ------------------------------ + |
|| @ | + ------------------------- +-+ ||
|| || +-+ --------------------- + | i。|||
|| ||| \ | + ------------------- + || |||
|| ||| || +-+ --------------- + ||| |||
|| ||| ||| @ | +-+ ---------- + |||| |||
|| ||| ||| || "：| +-+ ------ + ||||| |||
|| ||| ||| || || 2 | +-+-+ ||||||| |||
|| ||| ||| || || ||！| {：|||||||| |||
|| ||| ||| || || | +-+-+ ||||||| |||
|| ||| ||| || | +-+ ------ + |||||| |||
|| ||| ||| | +-+ ---------- + ||||| |||
|| ||| || +-+ --------------- + ||| |||
|| ||| | + ------------------- + || |||
|| || +-+ --------------------- + | |||
|| | + ------------------------- +-+ ||
| +-+ ------------------------------ + |
+ ---------------------------------- +

Python、56バイト

a=[1];exec"print a;a=map(sum,zip([0]+a,a+[0]));"*input()

サンプル使用法：

echo 9 | python filename.py

生産物：

[1]
[1, 1]
[1, 2, 1]
[1, 3, 3, 1]
[1, 4, 6, 4, 1]
[1, 5, 10, 10, 5, 1]
[1, 6, 15, 20, 15, 6, 1]
[1, 7, 21, 35, 35, 21, 7, 1]
[1, 8, 28, 56, 70, 56, 28, 8, 1]

Python、94 91 88 70 63文字

x=[1]
for i in input()*x:
 print x
 x=map(sum,zip([0]+x,x+[0]))

Mathematica：36（41？）

MathematicaにはそのBinomial機能がありますが、それによって面白さがなくなります。私が提案する：

NestList[{0,##}+{##,0}&@@#&,{1},n-1]

上記の行は、次のような不規則な配列をレンダリングします。

{{1}, {1, 1}, {1, 2, 1}, {1, 3, 3, 1}, {1, 4, 6, 4, 1},
 {1, 5, 10, 10, 5, 1}, {1, 6, 15, 20, 15, 6, 1}}

これはMathematicaの基本的な形式であるため、受け入れられると思いましたが、ルールをもう一度読んでみると、受け入れられない可能性があると思います。追加するGrid@と、合計41文字の明確に受け入れられる出力が生成されます。

Grid@NestList[{0,##}+{##,0}&@@#&,{1},n-1]

n = 6：

1                       
1   1                   
1   2   1               
1   3   3   1           
1   4   6   4   1       
1   5   10  10  5   1   
1   6   15  20  15  6   1

C、522

Cの答えを示す自己。より明確にすることができませんでした！余分なキャラクターを見つけるためのボーナスポイント。

#define returns return 0
#define fr for
#define twentyonechexpressis0 0
                                                                                i
                                                                               , x
                                                                              [ 52 ]
                                                                            [ 52] ,j, y
                                                                       ; main (c){fr (;i< c
                                                                    ; i++){ x[i][i]=x[ i][0]= 1
                                                         ; }for(i =2;i<c;i++){for (j=1;j<i;j++){x [i][j] =
                                    1 +x[i][j ]+x[i-1][j-1]+x[i-1] [j]+1-1+1-1+1-1+1-1+1-1+111-11- twentyonechexpressis0 -100-1; }
} ;for(i=0 ;i<c;i++){for(j=0;j<=i;j++){ printf("%3d%c",x[i][j],(1+1+1+1)*(1+1+1+1+1+1+1+1)) ;}putchar(1+1+(1<<1+1)+1+1+1+1+1+111111-111111-1);} /*thiscomment_takes28chars*/ returns; }

Golfscript（21文字）

~]({0\{.@+\}/;1].p}*;

説明が要求されたため：

# Stack contains 'n'
~](
# Stack: [] n
{
    # prev_row is [\binom{i,0} ... \binom{i,i}]
    # We loop to generate almost all of the next row as
    #     [(\binom{i,-1} + \binom{i,0}) ... (\binom{i,i-1} + \binom{i,i})]
    # \binom{i,-1} is, of course, 0
    # Stack: prev_row
    0\
    # Stack: 0 prev_row
    {
        # Stack: ... \binom{i,j-1} \binom{i,j}
        .@+\
        # Stack: ... (\binom{i,j-1} + \binom{i,j}) \binom{i,j}
    }/
    # Stack: \binom{i+1,0} ... \binom{i+1,i} \binom{i,i}
    # unless it's the first time round, when we still have 0
    # so we need to pop and then push a 1 for \binom{i+1,i+1}
    ;1]
    # next_row
    .p
}*
# final_row
;

ハスケル、94 92

f=[1]:[zipWith(+)(0:x)x++[1]|x<-f]
main=readLn>>=mapM_(putStrLn.unwords.map show).(`take`f)

出力：

1
1 1
1 2 1
1 3 3 1
1 4 6 4 1

各数字の間にスペースを印刷しない71文字のバージョン：

f=[1]:[zipWith(+)(0:x)x++[1]|x<-f]
main=readLn>>=mapM_ print.(`take`f)

出力：

[1]
[1,1]
[1,2,1]
[1,3,3,1]

Scala、81 78 72 70文字

81文字：最初の試行、Pythonバージョンから恥知らずにコピーされた:)

var x=Seq(1)
for(i<-1 to args(0).toInt){println(x)
x=(0+:x,x:+0).zipped.map(_+_)}

スクリプトとして実行するか、REPLで直接実行します。

驚くほど読みやすく慣用的なもので70文字にカットします。

Seq.iterate(Seq(1),readInt)(a=>(0+:a,a:+0).zipped.map(_+_))map println

または、まったく異なる方法で72 70文字：

0 to(readInt-1)map(i=>println(0 to i map(1 to i combinations(_)size)))

ルビー：51 49 46文字

（45文字のコード+ 1文字のコマンドラインオプション）

p=[];$_.to_i.times{n=0;p p.map!{|i|n+n=i}<<1}

おかげで：

• jsvnm値の切り替えの代替案を提案する（2文字）
• GB以前の改善後に未使用の変数を見つけるための（4文字）

サンプル実行：

bash-4.4$ ruby -ne 'p=[];$_.to_i.times{n=0;p p.map!{|i|n+n=i}<<1}' <<< 1
[1]

bash-4.4$ ruby -ne 'p=[];$_.to_i.times{n=0;p p.map!{|i|n+n=i}<<1}' <<< 9
[1]
[1, 1]
[1, 2, 1]
[1, 3, 3, 1]
[1, 4, 6, 4, 1]
[1, 5, 10, 10, 5, 1]
[1, 6, 15, 20, 15, 6, 1]
[1, 7, 21, 35, 35, 21, 7, 1]
[1, 8, 28, 56, 70, 56, 28, 8, 1]

オンラインでお試しください！

JavaScript（90 85 83 81）

for(n=prompt(o=i='');i++<n;o+='\n')for(s=j=1;j<=i;s=s*(i-j)/j++)o+=s+' ';alert(o)

デモ：http : //jsfiddle.net/tcRCS/3/

注：数値は組み込みの整数データ型をオーバーフローし、浮動小数点数になるため、約n> 30の間は実際にはうまく機能しません。

編集1：ステートメントwhileへの変換forと組み合わせにより5文字を削除

編集2：s=文を内部に移動しfor、2文字を保存します

編集3：s=1,j=1初期化子を結合しs=j=1て2文字を保存する

R、39文字

Rはこのタスクに最適なツールのようです:-)

x=1;for(i in 1:n)x=c(print(x),0)+c(0,x)

in Q (25 characters/20 with shorter version)

t:{(x-1) (p:{0+':x,0})\1}

Shorter

t:{(x-1){0+':x,0}\1}

Sample usage:

q)t 4
1
1 1
1 2 1
1 3 3 1

awk - 73 chars

fairly straightforward implementation:

{for(i=0;i<$1;++i)for(j=i;j>=0;)printf"%d%c",Y[j]+=i?Y[j-1]:1,j--?32:10}

sample run:

% awk -f pascal.awk <<<10
1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1
1 6 15 20 15 6 1
1 7 21 35 35 21 7 1
1 8 28 56 70 56 28 8 1
1 9 36 84 126 126 84 36 9 1

Perl, 52, 49 characters

Edit: using say instead of print

map{@_=(1,map$_[$_-1]+$_[$_],1..@_);say"@_"}1..<>

Perl, 47 54 characters

$p=1;map{print"@{[split//,$p]}\n";$p*=11}1..<>

It takes a number from the command line, but doesn't perform any error checks.

Just realized it only works up to n=4. It was some old code I had on my hd.

This works though:

map{@a=(1,map$a[$_-1]+=$a[$_],1..@a);print"@a\n"}a..n

n has to be input into the script though, or it would be one character more.

Keg, 40 bytes

1®n1¿1.
,(|(©n|:©n$@MCƒℤ. ,⑨)©n⑨®n_01.
,

Explained

Perl, 77 Chars

$o[0]=1;for(1..<>){$"=" ";for(1..$_){$n[$_]=$o[$_]+$o[$_-1]}@o=@n;print"@o
"}

Example input

5

Example output

 1
 1 1
 1 2 1
 1 3 3 1
 1 4 6 4 1

C, 132 127 characters

c[25][25],n,i,j;main(){for(scanf("%d",&n);i<n;i++)for(j=0;j<=i;j++)printf("%d%c",c[i][j]=j?c[i-1][j-1]+c[i-1][j]:1,i-j?32:10);}

Pascal: 216 192 characters

(Not a real competitor, just an honorific presence.)

var p:array[0..1,0..25]of LongInt;i,j,n,u:Word;begin
Read(n);u:=0;for i:=1to n do begin
p[1,1]:=1;for j:=1to i do begin
p[u,j]:=p[1-u,j-1]+p[1-u,j];Write(p[u,j],' ')end;u:=1-u;Writeln
end
end.

Sample run:

bash-4.2$ fpc pascal.pas 
/usr/bin/ld: warning: link.res contains output sections; did you forget -T?

bash-4.2$ ./pascal <<< 1
1 

bash-4.2$ ./pascal <<< 9
1 
1 1 
1 2 1 
1 3 3 1 
1 4 6 4 1 
1 5 10 10 5 1 
1 6 15 20 15 6 1 
1 7 21 35 35 21 7 1 
1 8 28 56 70 56 28 8 1 

MATL, 10 bytes

Language created after this challenge

1iq:"tTTY+

Try it online!

1       % Push a 1. This will be the first row
iq:     % Take input n. Generate range [1,2,...,n-1]
"       % For each (that is, repeat n-1 times)
  t     %   Duplicate latest row
  TT    %   Push [1 1]
  Y+    %   Convolve latest row with [1 1] to produce next row
        % Implicitly end for each
        % Implicitly display stack contents

D 134 128 chars

import std.stdio;void main(){int n,m;int[]l,k=[0,1];readf("%d",&n);foreach(i;0..n){writeln(l=k~0);k=[];foreach(e;l)k~=m+(m=e);}}

output for 9 is

>9
[0, 1, 0]
[0, 1, 1, 0]
[0, 1, 2, 1, 0]
[0, 1, 3, 3, 1, 0]
[0, 1, 4, 6, 4, 1, 0]
[0, 1, 5, 10, 10, 5, 1, 0]
[0, 1, 6, 15, 20, 15, 6, 1, 0]
[0, 1, 7, 21, 35, 35, 21, 7, 1, 0]
[0, 1, 8, 28, 56, 70, 56, 28, 8, 1, 0]

taking full advantage of "you may format it however you like"; there is a space between each number and a linebreak

edit repositioned the assignment to l to shave of some chars

Scala, 131 characters

object P extends App{var x=List(1)
while(x.size<=args(0).toInt){println(x.mkString(" "))
x=(0+:x:+0).sliding(2).map(_.sum).toList}}

Takes the input from the command line.

Output for n=10:

1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1
1 6 15 20 15 6 1
1 7 21 35 35 21 7 1
1 8 28 56 70 56 28 8 1
1 9 36 84 126 126 84 36 9 1

F♯ - 203 characters

My first attempt at a round of code golf, and first attempt at functional programming. There is probably some obvious way to shorten it I haven't quite figured out yet. It complies in VS2010s F♯ compiler (which has the effect of running #light by default unlike earlier versions), and also works in the F♯ interpreter. Accepts input via stdin. Wish there was a better way for the input/output though! Lots of characters!

open System
let rec C r m =if r=0||m<=0||m>=r then 1 else C(r-1)m+C(r-1)(m-1)
for j = 0 to Convert.ToInt32(Console.ReadLine ()) do (
 [0..j]|>List.map(C j)|>List.iter(fun k->printf "%i " k)
 printf "\n")

Why is there no accepted answer to this question?

VBA - 249 chars

Sub t(n)
ReDim a(1 To n,1 To n*2)
a(1,n)=1:y=vbCr:z=" ":d=z & 1 & z & y:For b=2 To n:For c=1 To n*2:x=a(b-1,c)
If c>1 Then a(b,c)=a(b-1,c-1)+x
If c<n*2 Then a(b,c)=a(b-1,c+1)+x
d=IIf(a(b,c)<>0,d & z & a(b,c) & z,d):Next:d=d & y:Next:MsgBox d
End Sub

postscript - 59 chars (63 if you count -dn= to get the number of rows in)

[1]n{dup ==[0 3 2 roll{dup 3 2 roll add exch}forall]}repeat

run with

gs -q -dn=10 -dBATCH pascal.ps 

to get

[1]
[1 1]
[1 2 1]
[1 3 3 1]
[1 4 6 4 1]
[1 5 10 10 5 1]
[1 6 15 20 15 6 1]
[1 7 21 35 35 21 7 1]
[1 8 28 56 70 56 28 8 1]
[1 9 36 84 126 126 84 36 9 1]

Mathematica 35 chars

Here is the dull and lazy way of slicing Pascal's triangle:

Table[n~Binomial~k,{n,0,5},{k,0,n}]

(* out *)
{{1}, {1, 1}, {1, 2, 1}, {1, 3, 3, 1}, {1, 4, 6, 4, 1}, {1, 5, 10, 10,5, 1}}

APL, 19 15 characters

A bit late to the party, perhaps?

{⍪{⍵!⍨⍳⍵+1}¨⍳⍵}

It doesn't beat the J entry, though.

This assumes that the index origin (⎕IO) is set to 0. Unfortunately, with an index origin of 1, we need 25 18 characters:

{⍪{⍵!⍨0,⍳⍵}¨1-⍨⍳⍵}

There are two ⍨s in the code to express my frustration.

Demo:

      {⍪{⍵!⍨⍳⍵+1}¨⍳⍵}5
1
1 1
1 2 1
1 3 3 1
1 4 6 4 1

Explanations

Short version:

• ⍳⍵ (with an index origin of 0) produces an array of the numbers from 0 to ⍵-1 inclusive, where ⍵ is the right argument to the function.
• ⍳⍵+1 generates all numbers from 0 to ⍵
• {⍵!⍨⍳⍵+1} generates ⍵ choose k for every element k in ⍳⍵+1. The ⍨ (commute) operator swaps the arguments to a function around, such that the right hand argument becomes the left, and vice versa.
• {⍵!⍨⍳⍵+1}¨⍳⍵ passes each element in ⍳⍵ using the ¨ (each) operator. The result is a one dimensional array containing the first ⍵ rows of the Pascal's Triangle.
• The one argument form of ⍪ takes a one dimensional vector, and makes it a column rather than a row. Each row of the triangle is put on its own line.

Long answer:

• Virtually the same as the other version, except that 1-⍨ is placed before an ⍳ to replicate an index origin of 0.
• 0,⍳⍵ with an index origin of 1 replicates ⍳⍵+1 with an index origin of 0.

Maple, 46

seq(print(seq(binomial(i,k),k=0..i)),i=0..n-1)

Usage:

> f:=n->seq(print(seq(binomial(i,k),k=0..i)),i=0..n-1);
> f(3)
    1
   1 1
  1 2 1

VBA, 162 142 102 80 bytes

_{^{Saved 22 bytes thanks to Taylor Scott.}}

_{This is an old question now but I saw a shorter solution for VBA.}

[B2].Resize([A1],[A1])="=IF(COLUMN()>ROW(),"""",IF(ROW()=2,1,IFERROR(A1+B1,1)))"

This is meant to be run in the immediate window. Input is in cell A1 of the active worksheet. Output is in the active worksheet starting at B2 and using however many cells are required based on the input. The COLUMN()>ROW() check keeps the top right of the triangle blank. The ROW()=2 check makes the first value 1 to initiate the triangle. I could have shifted the output down and dropped this check, but it introduces a lot of extraneous output before the actual triangle and I didn't feel that it was in the spirit of the challenge.

I originally posted a much more complicated method that calculated every value based on its row and column. All this method does, though, is to use in-cell formulas. I start at B2 so I can reference the row above it without #REF! errors. Then, it copies and pastes the same formula over a block of cells n wide and n tall. The input and output for n=25 looks like this:

05AB1E, 14 bytes

FN©>F®Ne})ˆ}¯»

Try it online!

1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1