gEDA-dev: Add pin wierdness in gschem

[Carlos Nieves Ónega]

El mié, 13-12-2006 a las 02:02 +0100, Carlos Nieves Ónega escribió:
> El sáb, 09-12-2006 a las 20:50 -0500, Ales Hvezda escribió:
[snip]
> > I've been testing CVS (HEAD) and I noticed two things:
> > 
> > 1) Adding a pin to the schematic causes some of the attributes to be placed
> >    near the origin instead of near the pin.  This is new behavior.  Does 
> >    this happen for anybody else?

This has to do with the attribute autoplacement. The algorithm is
becoming clever, so I needed to add a new function to get the object
type to libgeda.
It returns the object->type converted into a scheme string.

Now the backend needs to know the OBJ_* definitions which are in
libgeda/include/o_type.h . By now, I included the definitions in the
backend, however I think they should be in libgeda.
If no one objects, I will move them into libgeda...

> > 2) There have been at least two reports of gschem crashing when adding pins:
> > 
> > http://sourceforge.net/tracker/index.php?func=detail&aid=1568155&group_id=161080&atid=818426
> "ERROR: /undefined in ISOLatin1Extended"? 
> I think you mean:
> http://sourceforge.net/tracker/index.php?func=detail&aid=1599582&group_id=161080&atid=818426
> 
> I think I have a fix for this one. I would like to know how it happens.
> Is the submitter subscribed to this list?

I committed into CVS a fix for this bug, improving error detection in
g_add_attrib.

Regards,

Carlos