I want to validate some input to check that it is a positive currency value of the format:
0.5, 0.55, 5 (note this one), 5.5, 5.55, 55 etc etc.
The code that I'm using is:
if ($("#gross").val()>0 && !/^\d+?\.?\d?\d$/.test($("#gross").val())) {
alert($("#gross").val() + " is invalid currency");
}
It works for everything except a single digit, eg 5 (and 5.) but does work for 5.5.
What am I doing wrong?
You've forgotten to add a ? at the end, before the $. A better way of doing it would be the following:
/^\d+?\.?\d{0,2}$/
This checks that there are up to two decimal places for the number - if you'd like to check for any amount, you could use something like:
/^(?!\.$)(?:(?!0\d)(\d*)\.?(\d[0-9]*))$/
Note that it's a good idea to explicitly convert your string into a number, and also cache the value of #gross.
var grossVal = $("#gross").val();
if (+grossVal > 0 && !/^\d+?\.?\d{0,2}$/.test(grossVal)) {
alert(grossVal + " is invalid currency");
}
\d and [0-9] are the same. In some flavors \d is equivalent to \p{Nd} (Unicode decimal digit), but I don't know of any flavor where it matches letters. Are you perhaps thinking of \w?
+? will match the fewest possible matches, in this case, 1 digit.
I think you're looking for something like:
/^\d+(\.\d{0,2})?$/
Which would be a series of digits, potentially followed by a decimal and anywhere between 0 to 2 digits.
Consider using alternation to break down a regular expression into the form a|b|c|d.
Then we can use several different forms, let:
a = 0 -- 0
b = [1-9]\d* -- n (non-zero integer), n cannot start with 0
c = 0[.]\d{1,2} -- 0.x or 0.xy
d = [1-9]\d*[.]\d{1,2} -- n.x or n.xy, n (non-zero integer)
This will allow us to reject values like 09 and 1., as they are not covered by any of the individual forms accepted.
5. should be accepted by the regular expression ("eg 5 (and 5.)").