Subject: Re: [linux-audio-dev] LAAGA
From: Richard C. Burnett (burnett_AT_tality.com)
Date: Wed Jun 13 2001 - 17:08:33 EEST
Ok, so in LAAGA being the same type of interface as LADSPA, then if you
added new properties, then the server/engine would have to be modified to
give you these new properties, not the API.
However, isn't this really bad design? First, it opens the door to having
to know about the engine more than you really need too. Second, it means
that the API information is really incomplete, because each engine has its
own set of properties it could make. Third, it can be abused really
easily, when someone makes an engine that they start doing lots of
property stuff that shouldn't be there.
I think that this generic property interface while providing for changes
in needed information, its more of a hack for not fully thinking out what
you need from the API. Its going to be a pain when different engines have
different properties and they don't all line up. I still think they
should be hardcoded, ie get_samplerate() type functions.
Rick
On Wed, 13 Jun 2001, Richard Guenther wrote:
> On Tue, 12 Jun 2001, Rick Burnett wrote:
>
> >
> > I guess I do not understand *why* we would need some sort of generic
> > property mechanism. If it was found out that additional properties
> > were needed to be queried or changed in the future, you will still
> > have to put in the appropriate 'glue' code between the audio engine
> > and the audio drivers, right? So if this is the case, then there is
> > no reason to have a generic mechanism for setting and controlling
> > properties, my reasons being:
>
> [lots of points deleted]
>
> You want to mix apps with "different" properties. Just have the
> properties identified by a string (which defines their type, too)
> and have them point to a memory region. Keep a list of properties
> at the originating port (if you go for my LAAGA approach, otherwise
> stick it somewhere else). Any app can now query the properties it
> knows and handle them, simply ignoring the others. No versioning
> problem at all (just the problem of coosing unique identifiers).
>
> Richard.
>
> --
> Richard Guenther <richard.guenther_AT_uni-tuebingen.de>
> WWW: http://www.tat.physik.uni-tuebingen.de/~rguenth/
> The GLAME Project: http://www.glame.de/
>
>
+------------------------+-----------------------+
| T a l i t y | +------+ |
+------------------------+ +----+-+ | |
| Richard Burnett | +-+ | |
| Senior Design Engineer +---+ +----+ |
| burnett_AT_tality.com | | |
| | | |
| Phone: 919.380.3014 | |
| Fax: 919.380.3903 | | |
+------------------------------------------------+
This archive was generated by hypermail 2b28 : Wed Jun 13 2001 - 18:29:27 EEST