Re: gEDA-user: Bug or feature?

[sdb@xxxxxxxxxx (Stuart Brorson)]

> 
> Hi Stuart,
> 
> 
> >Yeah, I like John's suggestion to use the RC files also.  My little
> >wrinkle is that I prefer to put the variable into gafrc instead of
> >gnetlistrc.  Gafrc is read by all programs, and is (in principle)
> >being phased-in while the individual RC files are being phased out.  
> 
> 
> 	I don't know if I will be able to completely get rid of the
> program specific rc files ever.  This is because I don't think that
> some program specific things belong in the gafrc.  Plus, the last time
> I looked at moving all the keywords into one rc file (gafrc), I wasn't
> particularly pleased with putting a whole bunch of completely gschem
> specific (say) things into libgeda.

That's fine.  I tried to hedge my statements since I know you think
this way.  I think that over time the ${HOME} and local project dir
RC files can become gafrc, while the global system-*rc files can
remain specific to each program (which can also read in the
system-gafrc file too . . .)  What do you think about that?

My aim is simply to reduce confusion amongst newbies.

> 	However, if other programs will use $SPICE_MODEL_DIR, then I'm
> all for putting it in the gafrc, otherwise it should go into gnetlistrc.

It can go into either since both the [gschem|gnetlist|gattrib]rc and
gafrc files in all three locations are read, but I will make it so it
will go into gafrc by default.  

Stuart