if (a)
{
  b();
}
else
{
  if (c)
  {
    d();
  }
  else
  {
    e();
  }
}

if a: b() else if c: d() else e()

if (a) b(); else if (c) d(); else e();

if (b>0) CallSomeFunction() else i = i + 1; for(int t=i;i<100;i++) callOther(i); while(true) { bla-bla-bla}

p += q;

p.x := p.x + q.x; p.y := p.y + q.y; p.z := p.z + q.z;

p = Vx*cos(t) + Vy*sin(t); // кружок-с

Set(p, Add(Mul(Vx,cos(t), Mul(Vy,sin(t)))); // <- пока писал уже запутался со скобками

char*f="char*f=%c%s%c;main(){printf(f,34,f,34,10);}%c";
main(){printf(f,34,f,34,10);}

Q

P>А в обероне есть классы?

pointer of abstract record

Т>char*f="char*f=%c%s%c;main(){printf(f,34,f,34,10);}%c";
Т>main(){printf(f,34,f,34,10);}
Т>

 def l = '\\';
 def q = '\"';
 def b = '\$';
 def n = '\n';

 def introspect(s)
 {
  System.Console.WriteLine($" def l = '$l$l';$n def q = '$l$q';$n def b = '$l$b';$n def n = '$(l)n';$n$n def introspect(s)$n {$n  System.Console.WriteLine($b$q$s$q);$n }$n$n introspect($q$s$q);");
 }

 introspect(" def l = '$l$l';$n def q = '$l$q';$n def b = '$l$b';$n def n = '$(l)n';$n$n def introspect(s)$n {$n  System.Console.WriteLine($b$q$s$q);$n }$n$n introspect($q$s$q);");

n[i]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=Z/I)*I)/J)*J)/K)*K)/L)*L)/Y)*Y);

K>

У тех, кто таким образом вспомнит Будду, Дхарму и Сангху, монахи,
Не будет ни страха, ни столбняка, ни мороза по коже.

    a ? b() : c ? d() : e();

n[i]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=(Z-=(n[i++]=Z/I)*I)/J)*J)/K)*K)/L)*L)/Y)*Y);

({:[x<4;,2;r,1_&{@[x#1;y*!:'-_-x%y;:;0]}[x]r:_f@-_-_sqrt x]}@_1e6)10000

+/f@&~(f1e6>+/-2#){x,+/-2#x}/1)!2

fib=: {."1@(]`(({: , +/)@])@.(> {^<_))&1 2 +/ (#~ -.@(2&|)) fib 1e6

1000000([:+/[:x:]#~>)p:i.100000

+/ i.@x:&.(p:^:_1) 1e6

+/((#":!100x)#10)#:!100x

print [a for a in range(2,1e6) if 0 not in [a%b for b in range(2,1+pow(a,0.5))]][10001]

LCR>Кстати, мне — всё понятно.

print reduce(lambda x,y:int(x)+int(y),str(reduce(lambda x,y:x*y,range(1,101),1)))

-- Поздравляю, ваша жена только-что родила.
— Малчика?
— Нет.
— А кого?

When the human mind is engaged in thought of an explicit nature, the speech center of the brain is involved. This means that we are verbalizing our thoughts even though we may not actually be speaking. When we are thinking we are necessarily using language or a notation of some sort to accomplish this verbalization. Hence it follows that notation, particularly the scientific notation of the field of science is a critical tool for creative thinking.

    truth_table1 =: #&2 #: [: i. 2&^
    truth_table2 =: (] # 2:) #: [: i. 2: ^ ]
    truth_table3 =: monad : '((] y.) # (2: y.)) #: (i. (2: y.) ^ (] y.))'
    truth_table4 =: monad : '(y. # 2) #: (i. 2 ^ y.)'
    truth_table1 3
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1

gen_elem  = (#);  // сгенерировать вектор y y .. y с кол-вом элементов x
gen_vect  = (i.); // сгенерировать вектор 0..(y-1)
antibase  = (#:); // конвертировать y в вектор используя базу из x
right     =  (]);  // взять правый аргумент
two       = (2:); // возвращает двойку для любого аргумента
power     = (^);  // возведение в степень

// тип type - потому что может быть n-мерный массив, необязательно число
type truth_table(type args)
{
    return antibase(gen_elem(right(args), 2), gen_vect(power(2, right(args)));
}

// выкинул функцию right, потому что в Java принято 
type truth_table_simplified(type x, type y)
{
    return antibase(gen_elem(y, 2), gen_vect(power(2, y)));
}

b := for initial       i := 0;
       bvalue := a[0];
     while (i < n - 1) repeat
       i := old i + 1;
       bvalue := old bvalue + a[i];
     returns
       array of bvalue
     end for;

>> antibase = (#; // конвертировать y в вектор используя базу из x

  2 2 2 2 #: 13
1 1 0 1      NB. бинарное представление

  5 4 3 2 #: 13
0 2 0 1      NB. 13 - это 201 в (5,4,3,2)-ичной системе счисления

  24 60 60 #: 1300
0 21 40      NB. 1300 секунд - это 0 часов, 21 минута и 40 секунд

[we expect kdb to be 10-100 times faster and scale better than other DBMS's.]

I'm curious how KDB would stack up against these numbers and whether Kx has any plans to post TPC scores.

IBM's new world record is 2733 queries per hour (QphH), with a system cost of $347/QphH.
— Microsoft's old world record was 1699 QphH at $161/QphH.
— Microsoft achieved its score by using eight PIII 700MHz Intel processors on a single server; IBM used 16 PIII 700MHz processors spread across a four-node cluster.
-----
kdb would do about 14000 QphH with 16 p3 700MHZ
>and whether Kx has any plans to post TPC scores.
it would be fun but tpc-h says vertical partitioning is not allowed.
-----
(Why not send it in with a question about why arbitrary rules should be allowed to hinder progress...?)
-----
They freely acknowledge this problem, and "address" it by stating categorically that any changes would result a different benchmark (not "tpc-h").

[The implicit assumption is that vertical partitioning requires a lot of work and that they're not really excluding any meaningful database implementations with this rule. But implicit assumptions are hard to address in explicit conversation.]

An important part of TPC is how fast you can back out of incomplete failed transactions.

KDB is designed to eliminate all work on failed transactions -- that's one of the reasons it's fast.

So far, no one has exhibited much interest in building a superstructure around KDB to benchmark doing pointless work then undoing it.
Until that's done, KDB would not be suitable for TPC.