We have started to have a quick look at SDL1.3 specifically for the iPhone. We
are curious how the main loop is handled and it seems different then all the
code samples we have seen using an NSTimer path or a separate thread (which
seems overkill)?
Could anyone give a rough overview how this works currently in SDL1.3 for the
iPhone as it seems interesting? We have been digging through the code but we
are no MacOS experts is why 
Hello,
In order to retain strict compatibility with other SDL programs and in
order to not require the use of any Objective-C, SDL for iPhone had to
take some extreme steps in terms of control flow.
In a normal iPhone application, the main() function calls
UIApplicationMain. After this, control flow isn’t really in the hands
of your code, it’s in the hands of the Cocoa Touch library. After
this point, your application mainly responds to events caught by the
main event loop, and then returns control to the main event loop. But
doing things this way isn’t just the orthodox way of doing things on
the iPhone platform – if you don’t do it this way your app will not
function properly.
But this model of control flow is very much incompatible with the way
things are done in a normal SDL application. A normal SDL application
retains control throughout the entire program execution. It typically
contains a main() function with a for() loop that loops until the user
decides to quit.
So how to reconcile these two approaches? The way I chose to do it
was to place the main() function in the SDL library itself. This main
function calls UIApplicationMain, as is necessary. Your main function
will be renamed SDL_main(), or something to this effect, and called
later by SDL. So how do we return control to the Cocoa Touch Library
periodically (as is necessary)? We do this whenever you draw to the
screen. Control then returns back to where you were, and you’ll never
even notice it left 
The end result is that a programmer gets to program in the way that
SDL applications are normally made (as one big for loop), but the
control flow follows the pattern that the user interface libraries on
iPhone expect and require in order to function correctly.
It’s an approach that is not without it’s flaws (getting the SDL
application to respond to a quit event correctly required a horrible
hack), but one that provides compatibility across platforms and
generally works just fine.
I hope that answers your question,
Quoting WB <warrick_buchanan at hotmail.com>:> We have started to have a quick look at SDL1.3 specifically for the
iPhone. We
are curious how the main loop is handled and it seems different then all the
code samples we have seen using an NSTimer path or a separate thread (which
seems overkill)?Could anyone give a rough overview how this works currently in SDL1.3 for the
iPhone as it seems interesting? We have been digging through the code but we
are no MacOS experts is why
SDL mailing list
SDL at lists.libsdl.org
http://lists.libsdl.org/listinfo.cgi/sdl-libsdl.org
–
Holmes Futrell
@Holmes_Futrell