So Play We All News

Ledger

Net Profit: $158.12

Hourly Earnings: $3.24

$19.12 - 2011-09-06 - Rule Violation Payment (Peter)
$19.92 - 2011-08-31 - Rule Violation Payment (Luke)
$28.83 - 2011-08-30 - Rule Violation Payment (Peter)
$10.00 - 2011-08-14 - Rule Violation Payment (Luke)
$30.00 - 2011-08-05 - Rule Violation Payment (Peter)
$10.00 - 2011-07-18 - Rule Violation Payment (Peter)
$10.20 - 2011-07-16 - Rule Violation Payment (Luke)
$20.30 - 2011-07-14 - Rule Violation Payment (Luke)
$29.91 - 2011-06-11 - Rule Violation Payment (Luke)
-$10.16 - 2011-05-18 - Domain Name

A Long Way To Code

published by JGadrow on Sat, 06/04/2011 - 14:22

The Skinny
It's update time again... Well, I got to a point that I at least have something visible online. It's ultra horrible though. I don't even have persistent session storage yet. So, you can verify that the system is able to log you in, but if you reload the page, it forgets who you are again. This is mostly because I had to rush myself just to even get this far and put in lots of sloppy code. So, I've decided that I don't care how far ahead the others get, I am not going to worry about matching their pace.

Really, it was an unrealistic goal that I'm glad to be letting go of. They're using a pre-built framework which means they have options to solve a lot of their issues right out of the box. I, on the other hand, have to implement even my low-level functionality. This means that I'm going to be slower to execute no matter what happens. That's the price that I pay for wanting to use my own implementation. The positive upside is that once I manage to solve these issues, they'll be solved for me in all of my future games and I'll be able to skip this tedium the next time around.

Why would I want to use my own framework as opposed to one that is pre-constructed? Mostly, it's a matter of style. The only framework that I have working experience with is Drupal and, while a completely legitimate and stable system, it's completely designed against the ways that I prefer to work. Even the style guidelines they have in place for how their code is formatted is in direct opposition to how I've always structured my code.

Sure, it slows me down in terms of this competition, however, I will have 100% control over every aspect of the process which is not something that I would have if I were to leverage Drupal to build the game. Additionally, if I were to use Drupal to create a game, there's are several concessions that I would have to make because Drupal was not created with game development in mind. It's possible but that doesn't mean it's the best solution.

Details, Details
So, let's go ahead and talk about what I did today with my four hours. Today, I accomplished the following tasks:

Modify server to allow virtual hosting
Allow public access to the "game"
Create Display class (and corresponding interface)
Create Form class
Create the user_login Form
Display the user_login Form
Add myself as a user
Process submitted user_login Form
Render text if user successfully logged in

Today began simply enough, I just had to edit my server config to create a new level of determining which webroot to serve based upon the domain name that was being requested. After that, I just had to checkout a copy of my framework into the new directory corresponding to http://allabrilyn.com/ and now my game was "public facing." That was my first two points covered. With that part completed in a short amount of time, I started the optimism.

Of course, then came the Display class and interface. Unfortunately, in order to keep pace, I'm being forced to move along at a rate that's not very conducive to thinking things through. So, I started implementation of the Display class a number of times and thought of reasons not to make it work like that in mid-implementation a couple of times. After I while, I realized what was happening but I really wanted to get at least a login form visible today. So, I decided to disregard my feelings on the matter and, like Nike shoes, Just Do It. I came up with a quick and easy implementation that made me cringe but, technically, it works.

Then, I had to begin implementing the Form Display. Of course, after a few moments of thinking it over, I realized that Form is actually a Model as opposed to a View. So, I slammed out some quick implementation for it and added a couple of elements (borrowing a bit from Drupal's Form API for how to structure the table... although Drupal's Forms are in code and not stored as data in a table). Then, I created the SimpleSegment class (an implementation of iSegment which is simply a wrapper element that can be stuffed into a Display for rendering as a string) and implemented an Update method to the Form class that accepted an iDisplay parameter for the addition of the form segment.

After that was completed, I needed a user in the database before I could try and process a form submission. So, I added my details to the database (I did this manually... which means I don't have registration working yet). At this point, I had already sacrificed my values pretty heavily so I just wrote an if/else construct to determine if a submitted form was present and, if so, to process that form to see if the submitted data matched the username and password for a user in the database. If so, a new variable was setup with the user's data contained inside.

At that point, it was a fairly easy task to take the user object and add its data to the display for rendering. I chose to go with something like:

Hello %user%! Your account was created at %time% on %date% and you last logged in at %time% on %date%. Enjoy your stay!

Of course, since you don't know my password, you won't be able to see that message... but I promise you it's there. lol

Conclusions
Circling back to what I stated at the beginning of this post; I managed to get the site "online" but I'm not happy with the results. I'm going to spend a lot of time firming up the low-level functionality of my system before I worry about the actual game itself at this point. There are several systems that I need to have up and running before displaying the game really can be done properly. Sure, I could code a bunch of slop like I did today and just get something up and running... but I don't want to work on just a game, I want to work on my game. Sure, I'll be sacrificing my place in this competition and will probably remain in dead last for a long time... but I'd rather lose the contest and create a game that I'm proud of and that works well than to win... what? An ability to say I kept pace with Harkins and BBrowserGames's progress? That'd be a lie, really, because their code wouldn't be horribly unmaintainable and mine would.

I didn't commit myself to anything today that is 100% able to be thrown away (well, except my sloppy if/else code in the entry point of the application... but even that will likely be maintained... just in other locations of the code) so I didn't really waste my time today. I just took a scenic detour for the sake of showing that I could get stuff up and running... I just choose not to. :P

Comments

Coach Outlet Online

Fri, 12/09/2011 - 20:08

Main menu

You are here

Search form

So Play We All News

User login

Ledger

A Long Way To Code

Tags:

Comments

Pages