Weird math problem, PHP thinks a calculation is a String

[eluser]Fabdrol[/eluser]
Hi guys, I've got a problem with PHP.
Some how, I keep getting an error with this code:

Code:

$ANS = ($L1 - ­(( 1.0­ ‐ $G1 ) * ( $L1 + $M1 + $A1 + 1.0 ))) / ( $G1 * (2.0 * $K1 + $M1 ));

The error reads:

Code:

Parse error: syntax error, unexpected T_STRING

Which is weird, since it's not a String, neither is one of the variables. To be sure I've added + 0.0 to each variable, to make sure they're floats.

Does anybody know what to do with this?

Fabian

[eluser]mdpauley[/eluser]
Can you provide example data for the formula so I can try it?

[eluser]mdpauley[/eluser]
I just tried the forumla with some fake data and it ran fine for me on 5.2.13.

[eluser]diego6616[/eluser]
Be aware that PHP it's saying it can't PARSE the file. It means that somewere there is a syntax error. Check your code ahead and behind the formula, since the formula with fake values apparently run prefectly alone.

[eluser]Fabdrol[/eluser]
Yeah, I know. However, visually, there's no issue, so I figured it had to be something else.. In a bright moment I decided to rewrite the formula, and that turned out to do the trick. Apparently there was something gone wrong when copying it from the Excel file I used to design the formula..

[eluser]intractve[/eluser]
May not be applicable here, but just putting this here so people looking for it may find it.

PHP generally fails at basic math using regular operands like + - etc. with floating point numbers (1.222 + 1.435) etc.

For Eg:

Code:

var_dump(0.7 + 0.1 == 0.8); //outputs bool(false)

Excerpt from php.net

Floating point numbers have limited precision. Although it depends on the system, PHP typically uses the IEEE 754 double precision format, which will give a maximum relative error due to rounding in the order of 1.11e-16. Non elementary arithmetic operations may give larger errors, and, of course, error progragation must be considered when several operations are compounded.
...
So never trust floating number results to the last digit, and never compare floating point numbers for equality. If higher precision is necessary, the arbitrary precision math functions and gmp functions are available.

Solution is to use bcmath - a precision math extension installed in most php installations.

Hope this is useful to someone.

[eluser]Fabdrol[/eluser]
Thanks George.

However, for most applications, the 'default' precision is okay, you just don't compare them. In stead, define a 'ceil' value and a 'floor' value, and check if the outcome is between them.
Of course, if you need the precision, you could use the precision math functions.

[eluser]Eric Cope[/eluser]
Usually when I get that error, I missed a $ for a variable. Had one of your variables been a string, it would have converted it to 0.0 and proceeded with the math.