the coding room

I should also mention, just for kicks, that last week I delivered my final paper and thus I finished with my university studies at the University of Piraeus. My studies lasted five years (10 semesters) the last of which were combined with a full-time job. In September I am flying for Hull, UK, for a full-time postgraduate MSc degree in Games Programming[!!!] and I can’t wait.

This experience of a full-time job had many effects on my personal life and my perspective on things, as I came to realize even more, how precious the little time we have is and how we should always make the best out of it.

Working on a full-time basis as a web developer gave me a chance to see how real life is with its tight schedules, deadlines, real-life programming experiences and many other that have to do with the fact that I learned SO F***IN MUCH just working next to people that new what they were doing and why! Things that I never had faced while in the university.

My paper, titled “Design and implementation of a modern and modularized CMS”, introduced a what I think is a flexible new way to look at web sites in terms of file structure, nationalization, modularity and, most importantly, database design.
It tries to see a web page as various nodes (name borrowed from Drupal) that connect with each other, drawing an one-way graph. In reality, these nodes construct the web page, or many other web pages for that matter as they are reusable, in a kind of hierarchy that can be very simple or really complex.
It also tries to have a generic concept of ORM (Object relational mapping), where it tries to map a database table, or a database entity, with a PHP object and make its manipulation ‘fast and furious’.
The design part of the paper is very interesting, as it can be defined as a collection of stuff one has to take into consideration if he ever attempts to create something as vast and massive as a web CMS.

Anyway, It is a very big discussion for a very broad subject. Anyone who cares to leave his notes or thoughts on the subject (a.k.a CMS) is welcome. I am just a little disappointed that I will not be able to get more into these stuff for a very long time.

Once again i fell into the hands of the evil one.

I was implementing this extension to YUI’s extenion that generates nice looking charts. I had everything working on Firefox when I saw that the chart, along with other things, did not generate on IE. I said to myself that it must be my fault, not the YUI or the extensions’ fault. So I went debugging, which you all know is not trivial in IE.

I spent an hour or so just to find out that Yahoo’s connector was not building the query string at all in IE when reading a form with Connector.setForm().
So I took a look inside my template to see what was going on and after a dozen checks It turned out to be the INSANE fact that I was using the word ‘length’ as an id.

I mean, who do I think I am to use ‘length’ as id. I must be crazy.

After a very small search about reserved keywords I found this blog entry.
I do not know how many of you have fallen into this trap but it is common to ignore some ( rare ) things up until the moment the nature of programming makes us face them. And once again, I found something about IE that I didn’t know and once again I wanted to kill it.
But the least I/we can do is raise more anti-IE awareness for the sake of web developers firstly and general web security secondly!! Am I ignorant?

I don’t know about you folks, but I have a little function I call to get an element by its id and it goes something like this

function getObj( pID ) {
    if( util.docRoot.getElementById )
    {
        return util.docRoot.getElementById( pID );
    }
    else if( util.docRoot.all )
    {
        return util.docRoot.all[pID];
    }

    return false;
}

For example, we have two <input> fields as such

<input type=”text” id=”shakeit” name=”shakethis” value=”valueA” />

<input type=”text” id=”shakethis” value=”valueB” />

Well, in Firefox, when I do

var someElement = getObj( ’someID’ );

the someElement contains the element with ID equal to “shakethis”, as expected.
In Internet Explorer though, someElement contains the first occurence of “shakethis”
even if it matches the name of the <input>.

Which, in the above example, means that if we were to print someElement.value this would be equal to “valueB” in Firefox
and “valueA” in IE.

Weird bug or just me being totally ignorant??

Or, as someone would say, is it a bug or a feature??

I was working till seven o’clock yesterday trying to see why the floating box I had created two weeks ago sometimes would not give up and disappear.

The story behind this project

The project is an online jewellery catalog, although you can think of it as an image gallery for simplicity. The goal is to show a pop-up bubble box, when the mouse is over an image, with extra information about that particular product.
Read the rest of this entry »