Building An Extensible Form Validator Class - Running On Empty
( Page 6 of 12 )

Let's now begin constructing the validation routines themselves.

Here's the first one, which performs a very basic test - it checks to see whether the corresponding variable contains a value or not.

<?php
class FormValidator
{
	// snip

	//
	// methods (public)
	//

	// check whether input is empty
	function isEmpty($field, $msg)
	{
		$value = $this->_getValue($field);
		if (trim($value) == "")
		{
			$this->_errorList[] = array("field" => $field,
"value" => $value, "msg" => $msg);
			return false;
		}
		else
		{
			return true;
		}
	}

	// snip
}
?>

This is fairly simple, but worth examining in detail, since all subsequent methods will follow a similar structure.

The isEmpty() public method is called with two arguments: the name of the form variable to be tested, and the error message to be displayed if it is found to be empty. This method internally calls the _getValue() method to obtain the value of the named form variable; it then trim()s this value and checks to see if it is empty.

If the variable is empty, a new element is added to the private class variable $_errorList (an array) to represent the error. This new element is itself a three-element associative array, with the keys "field", "value" and "msg", representing the form variable name, the current value of that variable, and the corresponding error message respectively.

If the variable is not empty, it will pass the test and the method will simply return true; this is handy in case you want to wrap the method call in a conditional test in your form processing script.

The reason for passing the form variable's name to the class method, rather than using its value directly, should also now be clear. When I pass the variable name to the method as a string, it can be used to access the corresponding form variable via the _getValue() method, and also becomes available within the class scope. This allows me to populate the $_errorList array with both the form variable name and its value; this additional information can come in handy when debugging long and complex forms.

If, instead, I had used the form variable's value directly, I would have certainly been able to record the incorrect form value in the $_errorList array, but would have no way to record the corresponding form variable name.

This might sound overly complicated, but it's actually fairly simple. In order to understand it better, try altering the class methods to accept form data directly, and you will see that it becomes much harder to record the name of the offending form variable(s) in the $_errorList array.

Finally, in case you're wondering, the $this prefix provides a convenient way to access variables and functions which are "local" to the class.

Let's move on to a couple of other, similar functions:

<?php
class FormValidator
{
	// snip

	// check whether input is a string
	function isString($field, $msg)
	{
		$value = $this->_getValue($field);
		if(!is_string($value))
		{
			$this->_errorList[] = array("field" => $field,
"value" => $value, "msg" => $msg);
			return false;
		}
		else
		{
			return true;
		}
	}

	// check whether input is a number
	function isNumber($field, $msg)
	{
		$value = $this->_getValue($field);
		if(!is_numeric($value))
		{
			$this->_errorList[] = array("field" => $field,
"value" => $value, "msg" => $msg);
			return false;
		}
		else
		{
			return true;
		}
	}
}

?>

These are very similar, simply using PHP's built-in variable functions to determine whether the data passed to them is numeric or character data.

>>> More PHP Articles          >>> More By icarus, (c) Melonfire