ErrorException: bcmul(): bcmath function argument is not well-formed

[ryanrapini]

In PHP 7.4 I believe this becomes a warning.

https://www.php.net/manual/en/migration74.incompatible.php

BCMath Arbitrary Precision Mathematics ¶

BCMath functions will now warn if a non well-formed number is passed, such as "32foo". The argument will be interpreted as zero, as before.

Code to reproduce:

$converter = UnitConverter::binary();
$converted_size = $converter->convert('1000')->from('mg')->to('g');

Result:

ErrorException: bcmul(): bcmath function argument is not well-formed

/vendor/jordanbrauer/unit-converter/src/Calculator/BinaryCalculator.php:59
/vendor/jordanbrauer/unit-converter/src/Calculator/Formula/UnitConversionFormula.php:35
/vendor/jordanbrauer/unit-converter/src/Calculator/AbstractCalculator.php:162
/vendor/jordanbrauer/unit-converter/src/UnitConverter.php:359
/vendor/jordanbrauer/unit-converter/src/UnitConverter.php:314
/src/Units.php:15
/vendor/laravel/framework/src/Illuminate/Support/Facades/Facade.php:261
/tests/Unit/Units.php:236

Additionally, I tried using the standard UnitConverter::default() but I get weird things like:

$converted_size = $converter->convert('1')->from('mg')->to('g');
echo($converted_size);
0.0

Any advice?

[jordanbrauer]

In PHP 7.4 I believe this becomes a warning.

https://www.php.net/manual/en/migration74.incompatible.php
...

@ryanrapini interesting, thanks for the heads up on this; I will take a peak at this later today and see what I can do!

Additionally, I tried using the standard UnitConverter::default() but I get weird things like:

This will depend on what you're converting. Since you are converting a smaller unit to a larger unit, you will wind up with a decimal, so you will to need to pass the convert method a precision value for rounding. In your case, if you were to pass it a 3 you would get a 0.001 as your result (1 gram is 1,000 milligrams), like so

$converted_size = $converter->convert('1', 3)->from('mg')->to('g');
echo($converted_size);
0.001

Hopefully that helps! 🙂

P.S. FWIW, the default precision is 2 (see the AbstractCalculator::DEFAULT_PRECISION constant)

[ryanrapini]

Thanks, this is great help. I dug in a little deeper to the issue before I had to stop working on it - the issue is coming because mul() is being passed these arguments:

  0 => "1"
  1 => "1.0E-6"

the 1.0E-6 is the issue, of course.

I believe you can force php to output "0.000006" (which is what that function wants) with, for example,

echo sprintf('%.8f', floatval('-1.0E-5'));//force 8 decimal places

(from here)

but I'm not sure this is the right way to go about this,

Hope this helps.

[ryanrapini]

P.S: your documentation appears to have weird compile errors, FYI

https://jordanbrauer.github.io/unit-converter/classes/UnitConverter.UnitConverterInterface.html

[jordanbrauer]

Thanks, this is great help.

Yeah, not a problem! Thanks for the heads up RE: documentation being out of whack.

mul() is being passed these arguments:

0 => "1"
1 => "1.0E-6"

What conversion are you performing when this happens? E.g., value and to/from units

UnitConverter::default()
    ->convert($n)
    ->from($x)
    ->to($y);

You can also get a small debug log that might show more of what's going on with the numbers/formula

$converter->getConversionLog();

[jordanbrauer]

@ryanrapini The issue should be rectified in #163 . If you would like to take advantage of it right away, you can target dev-master in your composer.json, otherwise it will be in the next release!

Here is some samples to show new behaviour working as expected.

$bc = new BinaryCalculator(13); // high precision binary calculator

dump(
    $bc->add('2.3562e-007', '1.8281e-009'), # (string) 0.0000002374481
    $bc->mul('1', '1.0E-6'),                # (string) 0.000001
);

[ryanrapini]

Thanks for your hard work!