On Mon, Oct 13, 2008 at 01:20:20AM +0200, Carlos Nieves Ónega wrote:
> El sáb, 11-10-2008 a las 23:42 -0700, [EMAIL PROTECTED] escribió:
> [snip]
> > - For compound objects, calculate a special bounding box that excludes  
> > pins and text in the computation.  The user can select a compound  
> > object by either clicking within this special bounding box, or on the  
> > pins and text.
> [snip]
> 
> If you go down this way, take a look at gschem/src/g_hook.c , functions
> custom_world_get_single_object_bounds and
> custom_world_get_object_list_bounds.
> 
> You can tell them to exclude some attributes, or even some object types
> when calculing the bounding box.
> 
> I think it is only currently used by the attribute auto placing code.

Attribute auto placing code? This is news to me, maybe I have disabled
it by mistake, but for me attributes have a tendency to accumulate on
top of each other, which certainly saves space but somehow impairs
readability :-)

Speaking of this, there is one feature that I'd like to see, namely to
have two different snapping grids, the normal one for moving components
and so on, and a finer one automatically chosen when moving a single
attribute. I don't know easy it would be to implement or what the
criterion would be to select the fine grid (two possibilities at least:
either the selection is a single attached attribute, or the selection does
not contain any pin). 

I find myself often switching the grid between component placement 
and attribute placement and this is a bit irritating. Hey even OrCAD
under DOS had such a feature 15 years ago that the grid was finer
when moving (single) attributes instead of a whole component.

With some help (on where I have to look) I might even try to code this
feature next month.

        Regards,
        Gabriel


_______________________________________________
geda-dev mailing list
[email protected]
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev

Reply via email to