[Freeciv-Dev] Re: Introduction and Patch Approval Process

[David Weisgerber <tnt@xxxxxxxxx>]

> > The most ideas I have seen were ahem... I think that the most
> > developpers thought not easy enough as getting sound output is the
> > easiest thing to integrate due having such programs as "play" which is
> > nearly available for every platform. Giving out sound must be an own
> > thread so in fact using "play &" as an command output is the best way
> > because we don't have to mind all the various sound modules and other
> > things.
>
> This will work on Unix-like platforms, but it won't on MS Windows or
> Amiga.  But sound support for a limited set of platforms, with a
> limited method, is better than nothing.  I agree with you that using
> 'play' is a simple way to integrate it.

Maybe we can distribute freeciv with a "play" program for Amiga and Windows.

> > One comment about stability: There is no reason for my patch to be
> > unstable because in fact we don't have to do more than giving the right
> > file to "play".
>
> You have to specify 'the right file' somehow, and hardcoded constants
> in the code aren't good enough.  For example, if every unit has its own
> sound, every modpack (with different units) needs to supply its own set
> of sounds.  So a convention must be implemented for naming the sound
> files.  You also have to supply code (an autoconf test) to detect
> 'play' at compile time and (probably) disable sound if it isn't found.
> There are more details like this.

Oh I didn't think about that but we can integrate such things one step after 
the other.
We don't have to disable sound if play isn't available because the only thing 
it would be some thousands of "command not found" outputs... 
But I'll see what to do.

CU,
David

-- 
Warum ist Windows98 hardwarefreundlicher als Linux?
Es läßt die Resettaste nicht anrosten.