> > ruby -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~ /^1$|^(11+?)\1+$/' 42

> This is pretty cool.  Particularly as a simple benchmark for regex
> backtracking.
> It highlites some of the performance issues of backtracking in ruby1.8
> and 1.9.
> And it also highlites issues with Perl's backtracking.

> Speed issues:
> > time ruby1.9 -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~ /^1$|^(11+?)\1+$/' 22441
> Prime

> real    0m2.111s
> user    0m2.084s
> sys     0m0.024s

> > time perl -wle 'print "Prime" if (1 x shift) !~ /^1$|^(11+?)\1+$/' 22441
> Prime

> real    0m0.230s
> user    0m0.208s
> sys     0m0.020s

> But... on larger numbers...

> > ruby1.9 -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~ /^1$|^(11+?)\1+$/' 104729
> Prime

> > perl -wle 'print "Prime" if (1 x shift) !~ /^1$|^(11+?)\1+$/' 104729
> Segmentation fault

> > perl -v
> This is perl, v5.8.8 built for x86_64-linux-gnu-thread-multi

> ruby -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~
> /^1$|^(11+?)\1+$/' [number]

> I cannot take credit for this one -- it originally came from Perl. The
> website listed below gives the full explanation of how the regexp
> works.

> http://montreal.pm.org/tech/neil_kandalgaonkar.shtml

> I highly recommend reading this if you want to flex your regexp  
> muscles today.

> ruby -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~
> /^1$|^(11+?)\1+$/' [number]

> I cannot take credit for this one -- it originally came from Perl. The
> website listed below gives the full explanation of how theregexp
> works.

> http://montreal.pm.org/tech/neil_kandalgaonkar.shtml

> I highly recommend reading this if you want to flex yourregexpmuscles today.

> Blessings,
> TwP

> ruby -wle 'puts "Prime" unless ("1" * ARGV[0].to_i) =~ /^1$|^(11+?)\1+$/' 42

> It may be the answer to life, the universe, and everything, but it
> certainly isn'tprime!