Odynia.org Blog - Xnyo

Melbourne Public Transport + Google Maps

nospam@example.com (Robert Amos) — Fri, 04 Jul 2008 14:28:35 +1000

While I'm detailing apps that I've played with but never published, there's the beginning of something larger with regards to Melbourne's Public Transport system.

For the moment, its just a google maps version of the network maps with some station/stop info, but you can check it out at http://transport.odynia.org/maps/.

I had planned on going the full stretch and making it the ultimate resource of its type, but my time doesn't seem to be my own lately!

bok

BUBBLES!

nospam@example.com (Robert Amos) — Tue, 22 Aug 2006 09:35:51 +1000

The future.. is Bubbles!

Weeeee

Yes, so I am excited, watch this space!

-bok

Xnyo Development

nospam@example.com (Robert Amos) — Thu, 27 Jul 2006 06:54:58 +1000

Well its official...

There's 116 pages of the Xnyo source printed out in pretty syntax-colours on my desk with a plethora of scribbles, crosses-out, adjustments and notes on there. I've found ALOT of areas that I'm just not satisfied with and want to redo. So I dare say this will end up being a full re-write.

I'll try to make it as backwards compatible as possible, but one thing is for sure, this will be a forward looking rewrite, meaning that unfortunately it won't be PHP4 compatible. I want to start making use of exceptions and better OO practices. Things like autoloading will be incorporated into the plugin system. Exceptions for the errors, __get() and __set() for config options. I'll investigate extra things like PDO and the new filter extension that'll be in 5.2 (which functions remarkably similar to my filter stuff..) and see how they go. I'll keep this blog up to date with development notes as I go.

Should be interesting

-bok

Xnyo, Smarty, AJAX

nospam@example.com (Robert Amos) — Wed, 19 Jul 2006 07:02:47 +1000

Been playing recently with ideas on what I could make Xnyo's AJAX functionality look like. I've started to take a basic look at how other libraries are doing it, so this may evolve in the furture.

So far, I have this:

Click on a comment to hide it. Click here to show all comments.


<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"

    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html>

<head>

    <title>Login</title>

</head>

<style type="text/css">

    #loginbox

    {

        position: absolute;

        top: 200px;

        left: 200px;

        border: 1px solid #000000;

        background-color: #8f9cdd;

        padding: 10px;

        text-align: right;

    }

</style>

<body>

<form name="login">

{effect action="shake" on="login_failure"}

<div id="loginbox">

    <div id="userdiv">Username: <input type="text" name="username" id="username" /></div>

    <div id="passdiv">Password: <input type="password" name="password" id="password" /></div>

    <div id="submitdiv"><button id="login" value="Login" onclick="{post target='xnyo.login'
ids='username,password'}" on_error="login_failure" /></div>

</div>

{/effect}

</form>

</body>

</html>

this is very very early stages yet, just me mainly pushing some ideas onto something more concrete than my brain.

The basic idea here, is that the {effect} function will create an event (or function?) called login_failure that will call the shake effect on the first element found inside it. The {post} function seen in the button element will then post to Pavid (the Xnyo XMLRPC backend) calling the $xnyo->login() method, passing the values of the username and password field. On an error it would trigger the login_failure event, causing the loginbox to shake.

Make sense? Please feel free to leave any comments.

-bok

Thoughts on the next Xnyo

nospam@example.com (Robert Amos) — Mon, 12 Jun 2006 20:22:49 +1000

Howdy,

There's been a lot talk lately about things like Active Record. I too have taken to this concept, and will adapt it into Xnyo 4. I'll be taking the extra step though, and trying to adapt its simplicity and implicitly enable more complex scenarios.

I plan on doing this by extending the existing database specifications section we have, to allow for complex relationships between tables, that will be automatically re-assembled coming into and out of the database. It should be interesting!

The AJAX wave is sweeping across the internet, and there are some killer javascript libraries out there. Several languages have taken the step to incorporate javascript functionality into the language (like ruby with .rjs). I'll be undertaking several approaches to determine what I feel to be the best way to embed ajax functionality into Xnyo. Maybe something like {div id="divone" effect="fadein" when="someobj.onclick"}, I'm not sure though, definitely something I need to think about.

Been thinking alot about backwards compatibility with Xnyo 3. Looking at the minimal install-base for 3, I'm not sure its such a top priority after all. It might be worth redesigning Xnyo from a better OO standpoint, with a "compatibility layer" to run 3.0 applications transparently. Of course if I do this I'll still support 3, not that it needs much of it.

Please, let me know your thoughts, comments and feedback are always welcome!

-bok

Upcoming Xnyo features

nospam@example.com (Robert Amos) — Fri, 02 Jun 2006 03:52:12 +1000

Hi all,

I'm throwing up the list of possible features for Xnyo 4.0. Alot of these you'll find in other frameworks, but then there are a lot of things in other frameworks I'm not fond of. One of the things I hear most often about Xnyo is from people who have similar-minded development styles to myself, which means Xnyo suits them perfectly. What we'll be trying to do with 4.0 is evolve along that same philospohy, but add features designed to make it even faster to work, without losing the flexibility thats inherit in Xnyo. I want you to be able to work quickly, securely and proactively, but at the same time I don't want to tie you to a naming/storage scheme of my own choosing.

The list of possible new features stands at:

Active Record

The industry is seemingly raving with this design pattern. While it is mainly useful for simple projects, incorporating it into our existing database abstraction layer should allow us to scale it up to complex projects as well.

Revamped Debug Console

The debug console came into life relatively late in the development of Xnyo 3.0. As such it was never properly designed into the structure of Xnyo, so hacks were necessary to keep from being a performance drain on production systems. I would like to design Xnyo to incorporate this more fully, with performance in mind, as well as extend and enhance its capabilities. And a facelift probably wouldn't go astray

Web 2.1

With the seeming rise and rise of Web 2.0 applications I'd like to bring Xnyo up to Web 2.1 (beat that!) and incorporate alot more dynamic features, as well as AJAX and the rest of the bandwagon. Though I have something special in mind here

That's all I have so far, I'll keep updating this entry with more details as they emerge. If you have anything you'd like to see added, please feel free to post a comment!

Cheers,

-bok

PS. For those that have asked, you can see what the current debug console looks like on Xny by appending ?xnyo_debug=true to the end of the main page of the Xnyo website. No direct link for spam reasons

History of Xnyo and future directions

nospam@example.com (Robert Amos) — Tue, 23 May 2006 17:49:35 +1000

This might be a long story, but I've received several questions recently as to the future of Xnyo, so I'll try to address those here.

The early days: Library

I live in a small city (population 40k or so) in central NSW, Australia. Its safe to say that there is very little in the way of software development done here, and even less PHP. Initially, I was tasked with rebuilding the Dubbo South High School website and intranet (see here and here) which is now known as the Dubbo College South Campus. As a student there I was quite excited by the prospect, and set to work ernestly. As my first web project I built up a basic set of functions that made life easier as I learnt. This evolved into the first release, simply called "Library" this was never made public.

The middle ages: Onyx 2.0

Shortly afterwards I discovered the joys of Object-Oriented development and took a stab at reinventing the Library using these programming techniques, which resulted in the complete rewrite known as Onyx, which was released under a BSD license back in May 2003 after a 6 month beta release cycle. Public interest in Onyx was minimal, but that was alright as it was doing the job it was designed for.

With the discussion and development direction of PHP5, as well as a programming job I held down in South Africa building ontop of Onyx in several high-stress projects I became aware that it could only scale so far. I'd learnt alot and my techniques had improved considerably. Though never formally trained in the "proper" programming techniques I took a keen interest in security and improving my coding styles. I was always able to be creative in my problem solving, leading me to develop things not previously thought of in PHP.

Present Day: Xnyo 3.0

This lead to another complete re-write with security as a primary concern, as well as adapability, scalibility, and all those other -bility words. This is where Xnyo stands today, mostly where I left it, after having to return to Dubbo to my family. I got a local job fixing PCs, doing network admin/design, and Microsoft CRM work. Unfortunately none of this work involved PHP, and therefore no Xnyo. Work piled on work and my free time wavered. Xnyo stagnated, and aside from bug fixes from a few users, not much happened.

The future: Xnyo 4.0

The good news is though, is that I plan on moving to Melbourne shortly and returning to PHP work, my first love. This should allow me to pick up development of Xnyo where I left off. This means another rewrite! And no I'm not changing the name this time. I've spent a bit of time experimenting with Ruby on Rails, and am very impressed. PHP is still my preferred language though, but I'd like to bring over some of the concepts of RoR into Xnyo, in order to make me even more productive! Also, theres that whole new world of AJAX and Web 2.0 and as much as I don't believe the hype I still don't want to miss the boat

As Xnyo 3 grew and ideas flourished useful things were added in (like the debug console) that the core wasn't designed to take. So another rebuild is necessary, but naturally it will maintain backwards-compatibility. Such is the nature of Xnyo that old Onyx 2.0b1 apps can run on Xnyo 3.0 with very little modification (other than the name change, obviously).

So sometime soon, in the midst of organising to move, finding a new job (see plug) and moving itself I plan on writing up a basic roadmap for the future of Xnyo 3.x and Xnyo 4.x. Ok maybe after I move then..

Watch this space!

-Rob

Plug: If you're a company in Melbourne looking to hire a PHP developer who also happens to be an excellent unix admin, MCSA, and Mac OS X admin email me at blog [at] odynia.org