PWC Week 21 - Euler's Constant and Big Numbers
Perl Weekly Challenge, Week 21, Task 1
Write a script to calculate the value of e, also known as Euler’s number and Napier’s constant. Please checkout wiki page for more information.
Ok, so how do we do that?
It is approximately equal to 2.71828, and is the limit of (1 + 1/n)n as n approaches infinity…
They have it to trillions of digits, but we need to do better than five significant digits.
So, we proceed with the simplest thing that could possibly work.
#!/usr/bin/env perl
use strict;
use warnings;
use utf8;
use feature qw{ fc postderef say signatures state switch unicode_strings };
no warnings
qw{ experimental::postderef experimental::smartmatch experimental::signatures };
my $i = $ARGV[0];
$i //= 49;
$i += 0; # force error if $i is non-numeric
my $n = 2**$i;
my $e = find_euler($n);
say <<"END";
i = $i
n = $n
e = $e
END
exit;
sub find_euler ( $n ) {
return ( ( 1 + ( 1 / $n ) )**$n );
}
$ ./euler.pl 48
i = 48
n = 281474976710656
e = 2.71828182845904
$ ./euler.pl 49
i = 49
n = 562949953421312
e = 2.71828182845904
$ ./euler.pl 50
i = 50
n = 1.12589990684262e+15
e = 2.71828182845904
$ ./euler.pl 51
i = 51
n = 2.25179981368525e+15
e = 2.71828182845904
$ ./euler.pl 52
i = 52
n = 4.5035996273705e+15
e = 2.71828182845905
$ ./euler.pl 53
i = 53
n = 9.00719925474099e+15
e = 1
After 249, Perl 5 will not give a representation of the number that’s not rounded into scientific notation. As the wiki page gives us the first 50 digits – 2.71828182845904523536028747135266249 – we see that 48-51 had the first 14 significant digits correct, 252 leaped over, and 253…
I do not have a strong theory about what’s going on with 253.
So, let’s add some complexity.
And by adding some complexity, I mean the slightest bit.
use [bigint](https://metacpan.org/pod/bigint);
#!/usr/bin/env perl
use strict;
use warnings;
use utf8;
use feature qw{ fc postderef say signatures state switch unicode_strings };
no warnings
qw{ experimental::postderef experimental::smartmatch experimental::signatures };
use bigint;
my $i = $ARGV[0];
$i //= 49;
$i += 0; # force error if $i is non-numeric
my $n = 2**$i;
my $e = find_euler($n);
say <<"END";
i = $i
n = $n
e = $e
END
exit;
sub find_euler ( $n ) {
return ( ( 1 + ( 1 / $n ) )**$n );
}
$ ./euler.pl 48
i = 48
n = 281474976710656
e = 1
$ ./euler.pl 49
i = 49
n = 562949953421312
e = 1
$ ./euler.pl 50
i = 50
n = 1125899906842624
e = 1
$ ./euler.pl 51
i = 51
n = 2251799813685248
e = 1
$ ./euler.pl 52
i = 52
n = 4503599627370496
e = 1
Whatever broke 253 before is now breaking everything, and there is no bigfloat to bail us out.
I mean, there is Math::BigInt and it is in Core, and I have looked at it.
And I’ve looked on Perlmonks.
And I’ve looked for blog posts, hoping for a better in for the module than the synopsis.
But, instead, I’m thinking about saying euler.pl 51 without using bigint being good enough, and
