Basically: an "empty" array can have any size, but an empty list always has size 0.

UPDATE: that should show it...

  DB<1> @a[3..5]=()
 
  DB<2> @b[3..5]=()
 
  DB<3> print scalar @a
6
  DB<4> print ( () = @a,@b )
 
  DB<5> @c= @a,@b
 
  DB<6> print scalar @c
6
  DB<7> @c=(@a,@b)
 
  DB<8> print scalar @c
12  
  DB<9> print ( () = (@a,@b) )
 
  DB<10> push @a,@b
 
  DB<11> print scalar @a
12
  DB<12> x @a
0  undef
1  undef
2  undef
3  undef
4  undef
5  undef
6  undef
7  undef
8  undef
9  undef
10  undef
11  undef
[download]

Line 12 clearly shows, that concatenating two "empty" arrays leads to a larger "empty" array...

Anyway the difference between line 5 and 7 is somehow surprising ...

UPDATE: Aaaaaargh!!! 8)

In line 5, I fell into the scalar comma operator trap again ... it's basically (@c=@a),(@b)

Well, Taulmarill was questioning Ikegamis words 'non-existing array element'. So I'm just pointing out that there is a real difference between non-existing array element and undef'd array element (even if we would wish otherwise to have more consistency).

++ for what you are saying after 'Basically:', an excellent description of the situation

I admit that i might have misunderstood something here .... although i'm not sure what....

Could we from now on agree not to talk about (non)existing array elements, but about elements <= $#array and elements > $#array. That might make things easier for my twisted, sleep deprived mind.

update: also i'd like $[ to stay 0.

++ for what you are saying after 'Basically:', an excellent description of the situation

Actually it should be defined what "empty" means. undef can be an existing element of a list!

  DB<1> p scalar ( ()=(undef,undef) )
2
[download]

Cheers Rolf