Re: gEDA-user: two refdes's

[Ales Hvezda <ahvezda@xxxxxxxxxxxxxx>]

> I'm a little surprised by this bug since the code paths for embedding 
> components between gschem and gschlas look identical.  I think something
> is going wrong in the promotion of attributes. 

Ah, I understand what is going wrong here.  The default settings for
the attribute promotion mechanism in libgeda are set to "off" more or 
less.  I'm not too keen on changing these defaults for all programs.

The real cause of this problem is that there isn't symmetry between
the various programs with regard to the rc files.  You can't put 
all the rc keywords into say a system-gafrc (or a local gafc).

I've debating fixing this (for a while now), by changing it so that
you will be able to put any keyword into any *rc file and then all
the programs would read *one* system-gafrc file and bugs like this
wouldn't be an issue.  I won't get rid of the system-programrc files or
the local programrc files since they have their use, but what can go
into them will be come regular and the same across all the programs.
The grossness to this approach is that it pushes more *rc stuff into
libgeda and that isn't so nice.

An alternative approach is to move some of the more common keywords 
from programrc* to system-gafrc (like all the rc keywords that relate to
attribute promotion).  This is certainly the least intrusive change (and
more controlled) and doesn't completely violate the minimal separation
between the programs.  This is probably what I'll end up doing as 
always-promote-attributes is already in system-gafrc. 

-Ales



_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user