The Midnight Engine – iPhone

So I’ve taken, to porting The Midnight Engine to the iPhone. This is a task that I started 3 years ago. But as many of you loyally Midnight followers will know, time is not a friend to my projects.

I have been playing with getting the engine up and running under AirPlay SDK as this will hopefully give me a little more scope. It kinda goes against my normal philosophy with development, in that an engine must offer just the right amount of assistance to your project but not bog you down with far too many features that you don’t need nor want. There are reasons to consider AirPlay which I won’t go into here, but one of the advantages is that it allows me to keep development in C/C++ which is how TME was written. This is a bonus as I can just port the engine without having to rewrite – well that was the theory.

First I had to scrap my porting attempt from 3 years ago. A because there were an awful lot of errors, and B because I had no idea where I was at.

Creating a new project under XCode and Airplay, I had some 20,000 errors and warnings – mainly errors. Part of this problem was that the codebase had been written to cross-compile across OS’s. And it used to. Not any longer.

It’s been nearly 5 years since I did any proper work on TME, and things have moved on. The compiler didn’t like my code anymore. New keywords had been reserved. Warnings popped up for things that I’d never had warnings for before. eg. Comparing signed and unsigned variables. The Compiler was having different affects on areas of the code that previously compiled fine. The GNU compiler doesn’t appear to be as happy with namespaces as Visual Studio used to be. (that said I can’t get the old project to compile under VS2010 anymore!)

So I had to strip it back. And bit by bit put the code back into the project and painstakingly change everything that needed to be done to make it compile.

After a week – I’m done.

Now I need to do some testing and recoding. First is to get the project loading in the LOM database. This will allow me to wrinkle out some of the OS/Hardware issues in the TME libraries. After that I need to painstakingly test large chunks of the code to make sure they are still doing what I want them to…