Re: [AD] Alleg5 - new driver structure idea

[Bob <ohannessian@xxxxxxxxxx>]

Hein Zelle wrote:

> - how will allegro know which function to call? Does it have to do a
>  string lookup on the driver ID to find the graphics functions, and
>  then a string lookup to find the 'blit' function, for example? I
>  assume not, but I don't see how that would work with the indicated
>  structure. Do you want to have fixed array indices to access a
>  certain driver / function pair?

No, not at all. What's going to happen is that install_graphics() will 
then read through the list of all registered drivers, and look for those 
that have a driver type of "gfx". One it finds them, it will then 
convert them to an AL_GRAPHICS_DRIVER_VTABLE, which is internal to 
Allegro, and is free to change in between versions of Allegro. Allegro 
can then call AL_GRAPHICS_DRIVER_VTABLE->blit() or whatever.

The string conversion is done only once.


> - how bad is the 'internal bloat' compared to what we currently have?

Well, we'd need an extra file (drivers.c), and virtually all of Allegro 
will link to it. It also means that we'll have a lot of string tables in 
Allegro so that the, for example, the graphics driver can recognize the 
blit() function.

There is also the issue of supporting older interfaces. What if we add a 
parameter to blit() for example? We can make a "blit2" entry in the 
driver, but then we also need to support the old "blit" interface.

Finally, I'm not really sure how it'll work for the DOS port, where 
people keep complaining about large executable sizes. Will this system 
really allow us the user to pick and chose the drivers he wants linked 
in, without the #define trickery that was needed before? Or will this 
(somehow) force all of Allegro to be linked in all the time?