Experimental Physics and
Industrial Control System

"Jeff Hill" <[email protected]> · Thu, 13 Sep 2007 17:23:41 -0600

All,

Here is a latest revision of how DBR types might be mapped to DataAccess
property space. There are two points to bear in mind here. First, this must
map to preexisting DBR_XXXX functionality. Second, this needs to be the core
elements of an interface safely carrying us forward in the future - serving
as a scaffolding on which to hang all future PV properties. Therefore the
initial names and the depth of the hierarchy might be quite important. 
I am not at this time defining the properties that do not map to preexisting
DBR_XXXX functionality.

I am open to any suggestions on how this might better be arranged and or
named. In particular, the alarm acknowledgement part may still need some
work.

PS: I am aware of and have been referring to the following wiki page. My
primary divergence is a tendency towards more hierarchy in what I have
defined below. The increased hierarchy seems to be desirable from my
perspective because.

1) The implementation of a catalog indexing a property container can be
quite efficient even without a hash table - if there are relatively few
properties in each bucket in the hierarchy. It does appear at this time that
CA will make a regular and heavy use of random access to a property
container using the property identifier as an index. 

2) The property names can be in most situations single words, and that
appears to be a cosmetically hygienic approach when property names start to
appear in the source code.

3) It's easier to include sufficient hierarchy up front, and relatively
harder to add hierarchy later when more parameters start showing up.

http://www.aps.anl.gov/epics/wiki/index.php/V4_Standard_Properties_and_Event
s

// DBR Property Hierarchy
// ----------------------
// value
//   time
//   precision
//     decimalDigits
//   units
//   limits
//     control
//       upper
//       lower
//     display
//       upper
//       lower
//     alarm
//       major
//         upper
//         lower
//       minor
//         upper
//         lower
//   alarm 
//     condition
//       severity
//       status
//     acknowledgement
//       pending (is a transient alarm pending acknowledgement)
//       severity (the highest unacknowledged alarm severity)
//   stateSet
//     labels
//   class
//

Thanks in advance for any comments.

Jeff
______________________________________________________
Jeffrey O. Hill           Internet     [email protected]
LANL MS H820              Voice        505 665 1831
Los Alamos NM 87545 USA   FAX          505 665 5107

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System