subtraction issue

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: subtraction issue by choroba (Cardinal) on Jan 24, 2017 at 21:47 UTC
See What Every Computer Scientist Should Know About Floating-Point Arithmetic, or try searching for similar terms here: Super Search: floating. ($q=q:Sq=~/;[c](.)(.)/;chr(-\|\|-\|5+lengthSq)`"S\|oS2"`map{chr \|+ord }map{substrSq`S_+\|`\|}3E\|-\|`7**2-3:)=~y+S\|`+$1,++print+eval$q,q,a, [download]	[reply] [d/l]
Re^2: subtraction issue by BillKSmith (Monsignor) on Jan 25, 2017 at 17:14 UTC
For a shorter version, check the FAQ Why am I getting long decimals (eg, 19.9499999999999) instead of the numbers I should be getting (eg, 19.95)? Bill	[reply]
Re: subtraction issue by syphilis (Archbishop) on Jan 24, 2017 at 22:28 UTC
This should result in $finalamount = 0 If you `printf "%a\n", $v_amount;` and `printf "%a\n", $plbamount;` (for which you need to be running perl-5.22.0 or later) you'll get to see the exact value of the 2 variables (in hex). If they turn out to be the same && $finalamount is not zero, then you have a seriously broken perl. Cheers, Rob	[reply] [d/l] [select]
Re^2: subtraction issue by AnomalousMonk (Archbishop) on Jan 24, 2017 at 22:59 UTC
Anonymous: Previous to Perl version 5.22.0, you can use a format like `'%0.20f'`. Give a man a fish: `<%-{-{-{-<`	[reply] [d/l] [select]
Re^3: subtraction issue by syphilis (Archbishop) on Jan 27, 2017 at 09:00 UTC
Previous to Perl version 5.22.0, you can use a format like '%0.20f' Yes, I think that's a good enough replacement for most purposes - and a useful addition to my post. However, it's not guaranteed to give you the exact base 10 value stored in the double - whereas `%a` will give you the exact (base 16) value. Cheers, Rob	[reply] [d/l]
Re^4: subtraction issue by pryrt (Abbot) on Jan 27, 2017 at 16:41 UTC
Re^5: subtraction issue by syphilis (Archbishop) on Jan 28, 2017 at 01:47 UTC
Some notes below your chosen depth have not been shown here
Re: subtraction issue by hippo (Archbishop) on Jan 25, 2017 at 09:07 UTC
Disproven: `#!/usr/bin/perl use strict; use warnings; use Test::More tests => 1; my $v_amount = -683.12; my $plbamt = -683.12; my $finalamount = $v_amount - ($plbamt); # Why have brackets here, OP? is ($finalamount, 0, 'Nothing to see here');` [download] can anyone tell me why this is? Yes, it is because at least one of $v_amount or $plbamt does not have the value which you think it has. See the links the other monks have given you regarding imprecision of floating point numbers in computing.	[reply] [d/l]
Re: subtraction issue by LanX (Saint) on Jan 25, 2017 at 17:46 UTC
> This should result in $finalamount = 0 but it results in 1.13686837721616e-013 can anyone tell me why this is? you most likely have a calculation resulting in a fraction of `0.12` after the point which is `3/25` and `25=5^2` can't be coded without loss in a binary system. Additionally Perl does some DWIM magic to hide minor errors, such that a printed `0.12` is not necessarily `== 0.12` . solution: To calculate dollars "accurately" with 2 decimal places, you need to calculate right from the beginning in cents as integers! So just calculate with integers in the desired accuracy and shift the decimal point afterwards! from Humans have too many fingers Cheers Rolf _{(addicted to the Perl Programming Language and ☆☆☆☆ :) Je suis Charlie!}	[reply] [d/l] [select]