im.ralphholzmann

jQuery Required Form Validation

Sat, 24 Jan 2009 10:38:00 -0600

Recently I was asked to create a web form for my internship that had about 30 text input fields. All fields were required, and they asked that form have a validation check. So, instead of writing 30 individual checks for each input, I wrote a some jQuery.

First of all, your required form text inputs must look like this:

Notice how the required text inputs have the class “required”. Here is the form validation function that gets called when the form tries to submit:

function validate() {

	// Initialize empty string for errors
	var errors = "";

	// Trim whitespace before and after every input
	$('input:text').each(function(){
		$(this).val($(this).val().trim());
	});



	// Check if all required inputs are empty
	$('.required').each(function(){
		if($(this).val() == "") {
			errors += "" + $(this).attr("title") + " cannot be blank";
  		}
	});
	
	// If there are no errors, submit the form, otherwise display errors.
	if(errors == "") {
		return true;  	
	} else {
  		$('#errors').html("The following errors occured: " + errors + "").slideDown();
  		return false;
	}
}

Which uses this prototype function for Strings I call trim. It trims the whitespace from the beginning and end of any String.

String.prototype.trim = function () {
	return this.replace(/^\s+|\s+$/g,"");
}

The first line initializes an errors String to hold any errors that may occur

The next function takes the value of each input (of type text), trims the whitespace from it, and then replaces it with the trimmed version. This stops the user from simply entering a few spaces in an input to circumvent required fields.

The next function selects all fields with the class “required”, then for each input, it checks if its value is empty. If it is, it grabs the title of that input and uses it in the error message.

Finally, if there are no errors, the function returns true and the form is submitted. Otherwise, the errors are displayed and the form is not submitted.

This nifty little function not only saved me a ton of time, but also a lot of lines of tedious coding. I hope this helps save you some time too!

Click here for a working demo

What I learned about C++ STL vector::erase

Fri, 24 Oct 2008 10:09:00 -0500

In file structures this week we had an assignment to implement LZ77 compression using a custom bit by bit input/output class. This resulted in having to read/write individual bits, using boolean vectors. I would pull individual bits off of the front of the vector, decode them, and then delete them from the vector using erase like this:

someBooleanVector.erase(someBooleanVector.begin(), someBooleanVector.begin() + length);

This worked well for very small text files, but took FOREVER on larger text files (like the hamlet text file that was included with the assignment). After doing a ton of testing, I narrowed it down to the .erase function calls. Then, after googling some information about vectors, I learned that .erase doesn’t simply erase the first x amount of elements. What it does is clear the first x amount of elements, and then shifts all the remaining elements down to the front of the vector, then edits the vector.size() accordingly.

This was costing my program a ton of runtime. A program that was supposed to take less than a second to run, was taking 17 seconds to run. The reason was the shifting.

For example, at capacity, my boolean vector would have 32,768 elements. If I deleted one element from the vector, it would have to perform 32,767 shifts, costing my program a ton of runtime.

I simply fixed this by completely reversing the way I constructed the boolean vector, then using the much more efficient .back() and .pop_back(), which require no shifting, to access and delete records. This sped up my program from 17 seconds to half a second.

I hope reading this helps someone out there who may have runtime issues with vectors.

Have a good weekend.

Portfolio

Tue, 07 Oct 2008 16:40:30 -0500

I added a link to my portfolio of web design projects to the left. Check it out and tell me what you think

ralphholzmann@gmail.com

~Ralph

Finding P and Q in an RSA encryption

Fri, 26 Sep 2008 09:07:00 -0500

Our last assignment in File Structures dealt with decrypting messages ecrypted with RSA and a Viginere cypher. Our professor stressed that we should come up with the most efficient way of finding primes p and q given n, because that would require the most time and he would take points off for decrypters that take more than 10 seconds to run. Considering the assignment was due 5 minutes ago, I’m going to post my algorithm for finding P and Q which I’m pretty proud of because it runs really fast.

// Recursively tries to find prime numbers,
// p and q, given n, starting each factor
// at the square root of n. Values p and q are
// passed by reference so I wouldn’t have to
// return an array or anything
void findPQ(unsigned long n, int up, int down, unsigned long& p, unsigned long& q) {

bool foundUp, foundDown = false;

// Finds lower prime from root
while (!foundDown){
if (!isPrime(down)) {
down—;
} else {
foundDown = true;
}
}

// Finds upper prime from root
while (!foundUp){
if (!isPrime(up)) {
up++;
} else {
foundUp = true;
}
}

// Checks primes p and q against n
if( down > 0 && up < 65535) {
if ((up * down) == n) {
p = up;
q = down;
} else if ((up * down) < n) {
findPQ(n, (up + 2), down, p, q);
} else {
findPQ(n, up, (down - 2), p, q);
}
}
}

My decrypter can decrypt Hamlet in less than a second!

Have a good weekend.

~Ralph

Redesign (sort of)

Tue, 23 Sep 2008 11:53:57 -0500

I just got done recoding my blog, and this is my first post! Luckily, this time I integrated my design with Tumblr, so it will be a lot easier to make posts. There are still some bugs to work out, which I’ll be pounding out in a couple days, but let me know what you think!

~Ralph