Re: AAI/AAO database information

[Andrew Johnson <anj@aps.anl.gov>]

Till Straumann wrote:
> Andrew Johnson wrote:
>
> > The aai and aao records are a bit of an anachronism; they're not 
> > really used very much, 
>
> FYI. WE do use them at SSRL/SLAC.

Interesting; presumably not asynchronously, since on examining the 
source code it appears that any attempt to use either record type with 
an asynchronous device support will cause the record's process() routine 
to return immediately if it finds PACT true on entry.

> > there's no soft or test device support included in Base, and they're 
> > not documented anywhere either.
>
> The source is good enough.

There still no soft or test device support for them in base though.

> >   I think they were contributed by one site a long time ago, and we 
> > just haven't got around to removing them yet (they'll be gone if R3.15 
> > is ever released).
>
> Has there been a discussion about this? We would be in quite a bit of 
> trouble if these are removed.

No, I admit there has been no online discussion about this.  However you 
wouldn't be in any trouble if we do remove them since you can easily 
pull the R3.14.x versions into a new support application, build that 
alongside base and add it to your IOC app's configure/RELEASE file.

If you think that's likely to be a lot of work, you should know that 
there will be some even more significant changes in the R3.15 tree which 
will require some conversion of applications.  However 3.15 isn't on 
track for release anytime soon, and I'm not going to remove the aai/aao 
records from any future 3.14.x releases.

- Andrew
--
There is considerable overlap between the intelligence of the smartest
bears and the dumbest tourists -- Yosemite National Park Ranger