geda-developers team mailing list archive

[Peter Clifton <peter.clifton@xxxxxxxxxxxxxxxxxxxxxxxxx>]

On Mon, 2013-10-21 at 09:19 -0700, Edward Hennessy wrote:
> Peter,
> 
> Thanks for catching this issue.
> 
> I won't be able to check until Tuesday evening. I'm fine with you
> applying the patch, and I'll test more on Tuesday evening.

I've pushed it.. it can always be fixed later if you had different
ideas.

(IE.. is the new page view widget supposed to handle scrolling events
internally? - which might require a little extra refactoring.)

We probably still need to override the GtkScrolledWindow behaviour
whatever we do though.

> I'll give the multiattrib editor a look then too. I'm glad to see
> gschem getting more attention. :)

There are a few unresolved issues with it, mostly:

1. It currently consolidates multiple attributes with the same name
(even within the same symbol).

I need to teach it to list multiple identically named attributes in the
same symbol separately, not as if they came from multiple symbols.

2. It all gets a little muddled when inherited attributes are involved
as well. (I might disable viewing inherited attributes when multiple
objects are selected).

3. I need to consider the usefulness of allowing editing of different
types of inhomogeneous selection sets:

a) Different component types (probably fine)
b) Different object types (complexes, pins, nets, buses)

I'm inclined to leave both for now, but we might consider dropping b) if
it proves of no utility.

We could default to editing (say), symbols, if the selection contains
nets and symbols, and have some kind of precedence rules.

Another option might be to squeeze into the UI a button or option to
restrict what types of object in the selection we are editing.

That might be a nuisance though, depending on how it behaves with
homogeneous selections which don't match the current filter settings.


-- 
Peter Clifton <peter.clifton@xxxxxxxxxxxxxxxxxxxxxxxxx>

Clifton Electronics