Hello everyone,

as some of you may know I have changed job a few weeks back. The process has been rather long since I had to change sponsorship for my Working Visa. The new visa was approved one day before the previous one was to expire. So I started exactly four weeks after I left Front Foot, on August 31st.

I really enjoyed working for Front Foot for the past two years, I learnt a lot from everyone there. I had the chance to work on really interesting projects with them:

• Planet Three Portal for iPhones and Android phones. It won the AIMIA award for the best mobile product of 2010.
• Lotto Results iPhone application (supported by a Rails backend). It was my first iPhone application, and also Front Foot's first.
• Macquarie Dictionary for mobile phones, with interesting XML processing techniques which led me to develop further the nokogiri-happymapper gem.
• Atrium: scalable, multi-channel notification platform. One of the most technically advanced project I've worked on.

So what am I up to now?

I work at Blake Education as a Software Engineer, still working with Ruby on Rails, and other technologies around Ruby. Blake is a publishing company that specialises in education. We run the very successful website readingeggs.com, and we have more projects of course!

Working here is very different from my previous work, the main reason is that most projects I did in the past were just developed for a customer, then shipped. On the contrary readingeggs needs to evolve a lot according to the new needs of the customers.

Being able to re-factor bits of the application, constantly making sure performance is not altered by the addition of features and by the increasing number of customers, and also working on new projects that should re-use as much code as possible...

I really love software engineering, and software architecture, and I really want to be good at it. Of course I already have a decent level, but I feel like I'm going to learn so much at Blake. And it's already started!

I have two more blog posts I want to write about what's been going on lately, so stay tuned!

This is it, you can install my fork of happymapper using the above command! It's now available on Rubygems.org.

It's a replacement to happymapper, so you want to use it using:
require 'happymapper'

Thank you to Matt/GoodGets for using the fork and pushing me to publish it as a gem! Enjoy! :)

It's not built-in the API, so this piece of code will do. It makes your search bar display in black, and with transparency... enjoy! :)

In an iPhone application I'm developing, there are some views that are quite complex to code... and I find them much easier to code in HTML/CSS. Since the Webkit engine available on the iPhone is quite powerful, I find it a much easier way to create complex views...

Now if you want this view to interact with your application it is a bit tricky... but not THAT much. Here is a piece of code I have in my controller... well in my UIWebViewDelegate.

Basically we catch the click on a link, and if the link is of the form file://...#action we then look for the selector action in the controller and trigger it if it exists.

It's THAT simple. :)

I've finally released my first iPhone app under the name Melion Design. This is a photographs portfolio for Andrew Hill. What I want with this app is to demonstrate the engine I've developed to manage your portfolio in your own iPhone application.

The application is free, it's called "The Frog Photos" and you can find it in the app store. Andrew can manage the photos available in the application through a web interface I've developed. It makes it easy to upload, and manage photos. The following features are available:

• Amazon S3 photos storage:  fast download of photos.
• Offline mode: user can still browse your photos when offline.
• Customize the description of each photo.
• Customize the perma-link of each photo.
• Provide a link to your own website.

Don't hesitate to contact me if you're interested in having your own portfolio available in the appstore for everyone to see! :)

Yes, there it is! Version 1.2 has the following additional features:

• Systems - record up to 18 of your favorite winning numbers assisting System Players
• How to win - check what you require to win a division prize for each and every game
• Odds - check the odds of winning each division for each and every game

I'm now working on even more exciting content, but I can't tell you yet what it is!

I just enjoy a lot working on this project. :)

You can find Lotto Result in the app store.

Last November I did a presentation of the Happy Mapper fork I developed. The video is available on Vimeo. Unlike the description says, it is a fork using Nokogiri to replace XML Object, but watching it will clear everything out! It's a lightning talk, so I'm sure I won't waste too much of your time :)

You can also find all the videos from RORO on Vimeo. Enjoy!

Just wrote that little rake file:

I thought that would be useful for some people. It just calls an action from a Rails controller. This action is supposed to be accessed only by the admin user. We also prevent checking for authenticity token. This rake task will be called in a cron job.

I'm proud to announce my first iPhone application I developped at Front Foot has been released!

Winning Numbers is a mobile website adapted to all phones in australia, but you can enjoy it now as a native application for your iPhone under the name Lotto Results.

It provides for now the latest results and dividends for all of the major australian lottery games: Oz Lotto, Powerball, Lotto, Tattslotto, Gold Lotto, and more... You can also save your winning numbers to quickly find them in the latest results, enjoy!

I've updated my version of Happymapper to use Nokogiri. It should be faster. The original Happymapper had already been forked to use Nokogiri by Roland Swingler.

Nokogiri is a ruby gem to parse XML and HTML using XPath or CSS selectors. It's written mostly in C, and is very fast. I intend to use this version of Happymapper in an upcoming project with intensive parsing!

One thing that as made web applications much easier to use in the last few years is the integration of Ajax into web pages. Before that, each click of the user would block the application until the next page loads... not really a great user experience.

Now when you want to trigger a long task in the background, you can use Javascript to trigger asynchronous requests... that is just great!

Well in standard applications design, it is also very handy to be able to do such thing, and we usually do it using threads. The iPhone platform obviously provides everything you need to perform that, and that's pretty simple! Just awesome! Here is how it works...

The documentation you want to read is the NSOperation Class it provides the tools to trigger a task in the background. And as the documentation says, you will most of the time want to manage a queue of operations rather than just a single operation. So the best thing to do (in my opinion) is to create a Manager (singleton) which owns a queue of operations: NSOperationQueue Class.

This way you can add operations to the queue, and they will all be treated in order. A good example of this kind of manager is the ConnectionManager of the ObjectiveResource library.

You can use it this way:

- (void)fileDownloaded: (NSData *)data {
              // Do stuff here
          }
          - (void)downloadFile: (NSUrl *)url {
              // Download file here
          
              [self performSelectorOnMainThread: @selector(fileDownloaded:)
                                     withObject: downloadedData
                                  waitUntilDone: YES];
          }

And somewhere in your code trigger the asynchronous download:

[[ConnectionManager sharedInstance] runJob: @selector(downloadFile:)
                                            onTarget: self
                                        withArgument: url];

That's it! So when you call runJob on the connection manager, it queues the job in the operation queue... so if you need to download ten files, they will be processed asynchronously one-by-one while the user is still using your app, without interruption.

Hope this was helpful!

Routing using Rails is quite simple, you define routes and where they point at. The order in which routes are declared defines their priority. Very simple.

Now Rails also provides conditions on routes according to the HTTP method used, and using the request_routing plugin we can use more conditions. We specially use it for the domain, or sub domain used for example. The Frontfoot website and the Frontfoot mobile site are the same rails application, they just use different routing according to the domain used.

map.connect '/:action',  :controller => 'mobile',
                                   :conditions => { :domain => /\.mobi/i }
          map.connect '/:action',  :controller => 'main'

That works well if the domain has only one TLD like frontfoot.mobi, but if the domain has two TLDs it doesn't work. In this case we do the test on frontfoot.mobi so it works well.

Now what if we have two domains like for example frontfoot.com.au and ffmobile.com.au. Well that doesn't work. The :domain option will test the domain against 'com.au' we could then use sub-domain to get 'frontfoot' but in that case if we use 'www' it won't work because the :subdomain option tests only the first sub-domain.

So here is the solution! I have written a fork of the plugin to add a :domain_2 option which just gets the domain considering it has 2 TLDs. And there you go you can write:

map.connect '/:action',  :controller => 'mobile',
                                   :conditions => { :domain_2 => /ffmobile\.com\.au/i }
          map.connect '/:action',  :controller => 'main'

Enjoy! It's tested and on my github. :)

Part of my job consists in getting a lot of data from different sources. Whether it's a SOAP or REST webservice, a TCP Socket, or simply files on a FTP, most of this data is in XML format.

There are different libraries to parse XML in Ruby. We used to do with XML Object, but the resulting code is really difficult to read and maintain. We prefer Happy Mapper.

The resulting code is very readable and easy to maintain... and the library can be used in most cases. It's also quite fast since it uses XML Object for the actual parsing. Examples are in the source tree, and the source itself is quite short and easy to understand.

Enjoy easy and happy XML parsing! :)

• The brand new Apple Guide for Application Submission
• AdHoc distribution building

When you are compiling in Xcode make sure you use the good building profile and it refers correctly your entitlements for AdHoc distribution, also make sure to use the good provisioning profiles for Distribution and AdHoc distribution.

Now all my workmates have a version of the application on their iPhone, and the application is being reviewed by Apple. Tell you more when it's available!

Well hello there!

Let me present myself, I'm Damien, software engineer for Front Foot Media Solutions in Sydney Australia.

Lately I've been working a lot on mobile platforms, whether it's for mobile websites with a strong back-end like Three Mobile's iPhone landing page: Planet 3, or for iPhone native applications.

I really enjoy developing for the iPhone, we focus a lot on ergonomics, providing the good features and making them easy and enjoyable for the user!

I also work on my own projects, and some free lancing as well for really small projects aside from work, which is what Melion Design is about! So don't hesitate to contact me if you want a good front-end and back-end developer for your web app.