Doc #76287 [Com]: bcmod returns incorrect values

Edit report at https://bugs.php.net/bug.php?id=76287&edit=1 ID: 76287 Comment by: kacper dot rowinski at gmail dot com Reported by: kacper dot rowinski at gmail dot com Summary: bcmod returns incorrect values Status: Analyzed Type: Documentation Problem Package: BC math related Operating System: Linux debian PHP Version: 7.2.5 Assigned To: cmb Block user comment: N Private report: N New Comment: Tx for info I hope this will be soon added to documentation as it's little confusing. Previous Comments: ------------------------------------------------------------------------ [2018-04-30 13:54:06] cmb@php.net bcmod() doesn't use floor() but rather truncates towards zero, which is also defined this way for POSIX fmod(), so that the result always has the same sign as the dividend. Therefore, this is not a bug, but rather a documentation issue. ------------------------------------------------------------------------ [2018-04-30 12:36:40] kacper dot rowinski at gmail dot com Prove https://www.wolframalpha.com/input/?i=0.00000000099999+mod+-0.00056 ------------------------------------------------------------------------ [2018-04-30 12:33:45] kacper dot rowinski at gmail dot com Description: ------------ Using bcmod on float return incorrect value Test script: --------------- <?php $leftOperand = '0.00000000099999'; $modulus = '-0.00056'; $scale =14; // incorrect : echo bcmod($leftOperand,$modulus, $scale); // this how it should work // mod(a, b) = a - b * floor(a/b) /* echo bcsub( $leftOperand, bcmul( $modulus, floor(bcdiv($leftOperand, $modulus, $scale)), $scale ), $scale ); */ Expected result: ---------------- -0.00055999900001 Actual result: -------------- 0.00000000099999 ------------------------------------------------------------------------ -- Edit this bug report at https://bugs.php.net/bug.php?id=76287&edit=1