#
# Declare three arrays and populate two of them.
#
my @a;
my(@x) = (10, 19, 0, 14, 20, 5, 0, 5, 12, 8, 3, 5);
my(@y) = (21, 20, 1, 15, 8, 18, 16, 18, 0, 1, 11, 18);
#
# Set $_ to a string that looks suspiciously like code.
# (Use ' instead of q{} to make the string stand out better.)
#
$_ = 'map{s@a,g@y;s@a,g@x;}@y;';
#
# Change the letter s to 'unshift', and g to 'pop' + newline
# (Use a more conventional / instead of ; to separate parts of s.)
#
s/s/unshift/g;
s/g/pop\n/g;
# $_ now looks like
# map{unshift@a,pop
# @y;unshift@a,pop
# @x;}@y;
#
# or equivalently
#
# map { unshift(@a, pop @y); unshift(@a, pop @x) } @y;
#
# which if executed, would populate @a with alternating elements
# of @x and @y (@x and @y would be empty, but that's immaterial).
#
# @a = qw(10 21 19 20 0 1 14 15 20 8 5 18 0 16 5 18 12 0 8 1 3 11 5
+18)
#
# Note that with a simple encoding of 1 = a, 2 = b, ...,
# (with 0 = space) this is:
#
# @a = (qw(j u s t), ' ', qw(a n o t h e r)
# , ' ' , qw(p e r l) , ' ' , qw(h a c k e r))
#
# Execute the map, populating @a
eval;
# Now set $_ to this literal string
#
# print join'',(' ','a'..'z')[
#
# followed by all the elements of @a, separated by commas,
# and a closing square bracket.
# The expression argument to join is now an array slice:
# just those elements of (' ', 'a', ... 'z')
# that are indexed by members of @a.
#
# (Using more conventional " instead of qs s and ' instead of q; ;.)
#
$_ = "print join'',(' ','a'..'z')[" . join(',', @a) . '];';
# Create the slice, join the elements and print it.
eval;
Better than caffeine! (Almost ;-) Thanks!
| 