Re: BSC spec query

["g8kmh" <g8kmh@xxxxxxxxxxx>]

> > Also there's a typo on page 8, para 2 which I think should be
> > xAPBSC.query and then another para should be added for the case
where
> > an xAPBSC.cmd results in no change of status.
> >
> Can you confirm exactly what/where this is. I read para 2 and I
think it
> is correct as written (as a response to a xAPBSC.cmd message).  I
agree
> that we could do with clarification on which response to send when
a
> xAPBSC.cmd is sent that doesn't actually result in a state change.
It
> should be a xAPBSC.info.
>

Yep, that's the section. I'd suggest something like:

A xAPBSC.info must be returned by a device supporting BSC whenever a
xAPBSC.query message
is received querying the status of that endpoint.

A xAPBSC.info must be returned by a device supporting BSC whenever a
xAPBSC.cmd message
is received which results in no change to the state of the device.

>
> > In the plug in I'm testing, a DMX interface, how addressing works
in
> > real life is going to be pretty important.
> >
> I'm looking forward to seeing this app and mixing my C-Bus and DMX
> stuff..  dug out the serial interface last week in anticipation...

Well the breadboard code is working against a single (hardwired) sub-
instance.. hence the focus on addressing.

> > Interestingly, the BSC spec says that all devices must respond to
> > target=*.*.> which isn't the quite the same as >:>
> >
> We had the '>' discussion a bit after the BSC spec was written when
we
> came a cross a  couple of real life constructs that needed the
> interpretation. The application of > to match everything or nothing
is
> not technically quite right but it works .    Note that *.*.> would
only
> address the whole application and not any endpoints whereas >:>
or
even
> *.*.>:> would address the whole application and every endpoint
and
> lastly that >.*.> would only address endpoints (and not the main
> application).
>

..more testing cases to do then!

Lehane