[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Music stream patch
From: |
Erik Sandberg |
Subject: |
Re: Music stream patch |
Date: |
Tue, 13 Jun 2006 12:49:44 +0100 |
On 6/13/06, Han-Wen Nienhuys <address@hidden> wrote:
patch looks good. One minor issue:
> +
> + assert (infant_event_);
> + SCM infant_scm = infant_event_->get_property ("context");
> + Context *infant = unsmob_context (infant_scm);
>
this looks a bit odd. Isn't there a way to have
send_stream_event()
return a value, which could be infant_event_ or the infant context itself?
I don't see a sensible way: The information would have to go via a
return value of listeners. And listeners are by design passive, so we
shouldn't allow them to return values. Listeners that return values
would basically re-introduce the problem we had previously with
try_music's return value.
In addition, each event can be listened to by any number of listeners,
both in the context tree and in the back-end. I don't see a sensible
way of making send_stream_event combine these return values into one
useful value, and there would be problems with return values that
depend on which back-end is used.
Do you have any concrete ideas?
Erik
- Music stream patch, Erik Sandberg, 2006/06/12
- Re: Music stream patch, Graham Percival, 2006/06/13
- Re: Music stream patch, Han-Wen Nienhuys, 2006/06/13
- Re: Music stream patch,
Erik Sandberg <=
- Message not available
- Message not available
- Re: Music stream patch, Erik Sandberg, 2006/06/26
- Re: Music stream patch, Graham Percival, 2006/06/26
- Re: Music stream patch, Erik Sandberg, 2006/06/26
- Re: Music stream patch, Han-Wen Nienhuys, 2006/06/26
- Re: Music stream patch, Erik Sandberg, 2006/06/27
- Re: Music stream patch, Jan Nieuwenhuizen, 2006/06/27
- Re: Music stream patch, Han-Wen Nienhuys, 2006/06/26