Re: R: R: R: [AD] Allegro 5 new config routines, alpha 1

[Bob <ohannessian@xxxxxxxxxx>]

Peter Wang wrote:
> On 2002-02-11, Sven Sandberg <svsa1977@xxxxxxxxxx> wrote:
>
> > What makes me want to restrict the use of config variables is that
> > things like current color depth and gfx_capabilities are likely to be
> > used often, in time-critical situations, and from assembler, in which
> > case I think it would be better to keep them in variables or simple
> > inline functions. 

Then they will be available to Allegro internally, but not to applications 
externally. This way, we can change them as we go and not have to worry 
about backward compatibility with cryptic add-ons (see cpu_model and friends).


> > Even if people don't agree with my point of view, I think at least we
> > should agree on where to draw the line.

Which is why I've been trying to get people to comment/complain on the new 
APIs as they are proposed.


> For function arguments, I think a simple starting point would be to
> keep stuff that are "suggestions" in the config system, and use
> arguments for the rest.  Thus, colour depth and update method would be
> passed to `set_gfx_mode', but refresh rate would go in the config
> system.

Color depth needs to be a separate function if you need to switch bitmap 
depths on the fly. So instead of having a dedicated function, we can use the 
config API, which leaves us open for things like color spaces, etc.



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