Either way, isn't [clone] a built in object so can't it take messages _and_ signals on its left inlet like [dac]? I guess maybe it sets up its inlets to be the same as the abstraction which doesn't allow for that.. [haven't looked at the guts].

On Thu, Jun 23, 2016 at 9:52 AM, Miller Puckette <msp@ucsd.edu> wrote:
Good idea ... I have to think about whether this is possible to do.  If
I could ake inlets that take both control and signal, then I could easily
fix clone~ to take control messages like "vis" from inlets of either type.

Also that might be the way for me to fix a long-standing bug - float messages
to signal subpatch inlets don't work.

cheers
Miller

On Thu, Jun 23, 2016 at 09:42:18AM -0700, Alex wrote:
> I'm on pd-0.47-1 binary downloaded from UCSD [mac 64-bit].
>
> I've created a clone of an abstraction that has the left inlet as a signal
> inlet.. from the docs I'd expect I can send a |vis 5 1( message to that
> inlet and display the patch but I can't. If I create another inlet, to the
> right of that inlet, which is a control inlet, I can send the vis message
> to that and my instance does pop up but it does seem to me that the |vis(
> message should always go into the left inlet to keep things consistent?
>
> I attached an example, leftsig_test.pd is the patch to open.
>
> BTW, this clone stuff is really making me wish we could have abstractions
> that take both a signal and control info on a single inlet [like dac and
> some other tilde objects].. I've created a wrapper around clone to play
> with cascading abstractions and it would be nice if I could have my wrapper
> operate the same as I'm assuming clone should? I know there was discussion
> before.. maybe [inlet~] could actually have 2 outlets, a signal outlet and
> a control outlet?
>
> -Alex