Re: [AD] Shawn's thoughts about new Allegro

[Bob <ohannessian@xxxxxxxxxx>]

Stepan Roh wrote:

[snip] 
> I totally agree with Shawn that callback based input is not good. I did
> callback-based event processing library in the past and it was a pain and
> hard to debug. Message queues are too high-level. Polling system is
> probably the most reasonable solution.


I'd like to extend al_readkey() to support all KEY_*, not just a few of 
them, and also add timestamps to them if possible. The current Allegro input 
system is very prone to frme rate variations. A sudden drop in frame rate 
could mean keys get dropped, controls don't respond as well, etc.
For SUUTB, we built an input system over Allegro's to try and deal with 
this, but it's not very nice (callbacks, timers and polling all in one - yuk).

Also, al_readkey() can return an object instead of a number.
Something like:

struct KEY {
    int ascii;
    int scancode;
    long time_stamp;
};


This will mean that Allegro will need to manage an event queue of some sort.
I'm open to other suggestions.


--
- Robert J Ohannessian
"Microsoft code is probably O(n^20)" (my CS prof)
http://pages.infinit.net/voidstar/