Value greater than value issue

[Acrobatic]

Hello,

I know this will be an easy fix--but as of now I'm banging my head
against the wall. I need a fresh perspective from the group to see
what my problem is:

This is a simple accounting application, and the code below is
checking to see if a user's withdraw request is greater than their
available balance:
$withdraw_reque st = $_REQUEST['withdraw_amoun t']; // 543.21
$withdraw_maxim um = $user['available_bala nce']; // 543.21

if($withdraw_re quest $withdraw_maxim um) {
echo "Insufficie nt funds.";
} else {
echo "Processing..." ;
}
In my application, both values are equal (543.21), but
$withdraw_reque st $widthdraw_maxi mum still evaluates to TRUE, and
thus shows "Insufficie nt funds." If the two values are whole numbers,
like "543", then they evaluate the way I expect them to.

I've tried everything I can think of, like

$withdraw_reque st = floatval($_REQU EST['withdraw_amoun t']);
$withdraw_maxim um = floatval($user['available_bala nce']);

as well as doing an "isnumeric" check on both values (they both return
true), but still no luck. If $withdraw_reque st is less than (<) the
$withdraw_maxim um--ie 543.20 < 543.21, the script works fine.

Also, if I hard code the values, the script works fine. Somewhere
between pulling the maximum from the database and getting the
$_REQUEST variable things are getting lost in translation.

Thanks for any advice

[Good Man]

Acrobatic <jb****@gmail.c omwrote in news:7bc09fbf-5b53-4d5d-bfc7-
32**********@d2 1g2000prf.googl egroups.com:

Also, if I hard code the values, the script works fine. Somewhere
between pulling the maximum from the database and getting the
$_REQUEST variable things are getting lost in translation.

Thanks for any advice

my quick suggestion would be to make sure your values really are what you
think they are... echo out to the screen

also, note that "floatval" is a function whereas "float" will evaluate a
string numerically.

so...

<?php

$withdraw_reque st = (float)$_REQUES T['withdraw_amoun t'];
$withdraw_maxim um = (float)$user['available_bala nce'];

echo "comparing $withdraw_reque st and $withdraw_maxim um <br>";

if($withdraw_re quest $withdraw_maxim um) {
echo "Insufficie nt funds.";
} else {
echo "Processing..." ;
}

?>

[Jerry Stuckle]

Acrobatic wrote:

Hello,

I know this will be an easy fix--but as of now I'm banging my head
against the wall. I need a fresh perspective from the group to see
what my problem is:

This is a simple accounting application, and the code below is
checking to see if a user's withdraw request is greater than their
available balance:
$withdraw_reque st = $_REQUEST['withdraw_amoun t']; // 543.21
$withdraw_maxim um = $user['available_bala nce']; // 543.21

if($withdraw_re quest $withdraw_maxim um) {
echo "Insufficie nt funds.";
} else {
echo "Processing..." ;
}
In my application, both values are equal (543.21), but
$withdraw_reque st $widthdraw_maxi mum still evaluates to TRUE, and
thus shows "Insufficie nt funds." If the two values are whole numbers,
like "543", then they evaluate the way I expect them to.

I've tried everything I can think of, like

$withdraw_reque st = floatval($_REQU EST['withdraw_amoun t']);
$withdraw_maxim um = floatval($user['available_bala nce']);

as well as doing an "isnumeric" check on both values (they both return
true), but still no luck. If $withdraw_reque st is less than (<) the
$withdraw_maxim um--ie 543.20 < 543.21, the script works fine.

Also, if I hard code the values, the script works fine. Somewhere
between pulling the maximum from the database and getting the
$_REQUEST variable things are getting lost in translation.

Thanks for any advice

Welcome to the world of floating point numbers - where values are not
generally exact.

See http://www.php.net/manual/en/language.types.float.php for a
discussion on it.

For this particular instance, you should be able to solve the problem with:

$val1=(intval)( (floatval($_REQ UEST['withdraw_amoun t']) + 0.005)* 100);

This gets the float value, adds 0.005 to it then multiplies by 100. It
then gets the integer value of the result. Repeat for the available
balance.

--
=============== ===
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
js*******@attgl obal.net
=============== ===

[Acrobatic]

For this particular instance, you should be able to solve the problem with:

>
$val1=(intval)( (floatval($_REQ UEST['withdraw_amoun t']) + 0.005)* 100);

This gets the float value, adds 0.005 to it then multiplies by 100. It
then gets the integer value of the result. Repeat for the available
balance.

Thanks guys for the advice. Jerry, your workaround did the trick.

[Toby A Inkster]

Acrobatic wrote:

This is a simple accounting application, and the code below is checking
to see if a user's withdraw request is greater than their available
balance:

To build on Jerry's post, floats are a really bad idea for representing
currencies. They're just too woolly. As PHP doesn't include any fixed-
precision decimal data type, the best option is to represent money as an
integer -- using the smallest unit of currency possible (e.g. cents
rather than euros; pence rather than pounds).

You then reformat it as a price in the larger unit of currency only when
it needs to be displayed.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 21:23.]

Sharing Music with Apple iTunes
http://tobyinkster.co.uk/blog/2007/1...tunes-sharing/

[allain]

On Dec 8, 5:51 am, Toby A Inkster <usenet200...@t obyinkster.co.u k>
wrote:

Acrobatic wrote:

This is a simple accounting application, and the code below is checking
to see if a user's withdraw request is greater than their available
balance:

To build on Jerry's post, floats are a really bad idea for representing
currencies. They're just too woolly. As PHP doesn't include any fixed-
precision decimal data type, the best option is to represent money as an
integer -- using the smallest unit of currency possible (e.g. cents
rather than euros; pence rather than pounds).

You then reformat it as a price in the larger unit of currency only when
it needs to be displayed.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 21:23.]

Sharing Music with Apple iTunes
http://tobyinkster.co.uk/blog/2007/1...tunes-sharing/

Geez, that must have been a doozey bug. :)

[Jerry Stuckle]

allain wrote:

On Dec 8, 5:51 am, Toby A Inkster <usenet200...@t obyinkster.co.u k>
wrote:

>Acrobatic wrote:

>>This is a simple accounting application, and the code below is checking
to see if a user's withdraw request is greater than their available
balance:

To build on Jerry's post, floats are a really bad idea for representing
currencies. They're just too woolly. As PHP doesn't include any fixed-
precision decimal data type, the best option is to represent money as an
integer -- using the smallest unit of currency possible (e.g. cents
rather than euros; pence rather than pounds).

You then reformat it as a price in the larger unit of currency only when
it needs to be displayed.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 21:23.]

Sharing Music with Apple iTunes
http://tobyinkster.co.uk/blog/2007/1...tunes-sharing/

Geez, that must have been a doozey bug. :)

Not at all. A very common bug for anyone who's worked with floating
point values for more than six months or so.

--
=============== ===
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
js*******@attgl obal.net
=============== ===