Math-Expression-Evaluator
view release on metacpan or search on metacpan
optimization nearly doubles the execution speed. The compiled and the
optimized-and-then-compiled versions are both much faster.
With this example expression the optimization prior to compilation pays
off if you evaluate it more than 1000 times. But even if you call it
"10**5" times the optimized and compiled version is only 3% faster than
the directly compiled one (mostly due to perl's overhead for method
calls).
So to summarize you should compile your expresions, and if you have
really many iterations it might pay off to optimize it first (or to
write your program in C instead ;-).
BUGS AND LIMITATIONS
* Modulo operator produces an unnecessary big AST, making it
relatively slow
INTERNALS
The AST can be accessed as "$obj-"{ast}>. Its structure is described in
Math::Expression::Evaluator::Parser (or you can use Data::Dumper to
figure it out for yourself). Note that the exact form of the AST is
benchmark.pl view on Meta::CPAN
use strict;
use warnings;
use lib 'lib';
use Carp qw(confess);
use Benchmark qw(cmpthese);
use Math::Expression::Evaluator;
use Data::Dumper;
my $statement = '2 + a + 5 + (3+4)';
my $iterations = $ARGV[0] || 200;
sub with_optimize {
my $m = Math::Expression::Evaluator->new($statement);
$m->optimize;
for (1..$iterations){
$m->val({a => $_});
}
}
sub no_optimize {
my $m = Math::Expression::Evaluator->new($statement);
for (1..$iterations){
$m->val({a => $_});
}
}
sub compiled {
my $m = Math::Expression::Evaluator->new($statement);
my $c = $m->compiled();
for (1..$iterations){
$c->({a => $_});
}
}
sub opt_compiled {
my $m = Math::Expression::Evaluator->new($statement);
$m->optimize();
my $c = $m->compiled();
for (1..$iterations){
$c->({a => $_});
}
}
my %tests = (
optimize => \&with_optimize,
no_optimize => \&no_optimize,
compiled => \&compiled,
opt_compiled => \&opt_compiled,
lib/Math/Expression/Evaluator.pm view on Meta::CPAN
As you can see, the non-optimized version is painfully slow, optimization
nearly doubles the execution speed. The compiled and the
optimized-and-then-compiled versions are both much faster.
With this example expression the optimization prior to compilation pays off
if you evaluate it more than 1000 times. But even if you call it C<10**5>
times the optimized and compiled version is only 3% faster than the directly
compiled one (mostly due to perl's overhead for method calls).
So to summarize you should compile your expresions, and if you have really
many iterations it might pay off to optimize it first (or to write your
program in C instead ;-).
=head1 BUGS AND LIMITATIONS
=over 4
=item *
Modulo operator produces an unnecessary big AST, making it relatively slow
( run in 2.819 seconds using v1.01-cache-2.11-cpan-71847e10f99 )