Experimental Physics and
Industrial Control System

Michael Davidsaver via Tech-talk <tech-talk at aps.anl.gov> · Tue, 4 Apr 2023 19:52:57 -0700

On 4/4/23 12:50, William F Badgett Jr wrote:

OK, that's a bit disappointing about the Probe.

IMO. network data structure content and layout should be driven by
client needs.  Adherence to this principle is why the EPICS ecosystem
has widely portable semi-generic client applications like CS-Studio
and Archiver Appliance which can be used at any EPICS site without
code changes.

This is a core idea behind the "configuration over compilation" philosophy.

I would suggest some caution when "innovating" with custom meta-data.
While CA enforced this by being difficult to extend, the PVA protocol
gives the ability to transfer semi-arbitrary data structures.  However,
there is an element of "can" vs. "should" to consider.

I'll second what Kay wrote.  There are reasons why the EPICS model is
not to simply replicate an IOC's view of data structures over the network.
From what I have seen, the temptation to simply mirror server side data
structures often arises from an absence of knowledge of client requirements.

To my mind, it is better to stick to the conventional time+alarm+display
meta-data (aka. the NT types), and only _augment_ these after identifying
a specific client application which can and will use some additional
information.

But customizing PV Table and PV Tree would be useful to do, too.  Where would that start?

https://control-system-studio.readthedocs.io/en/latest/app/pvtree/doc/index.html

https://control-system-studio.readthedocs.io/en/latest/preference_properties.html#pvtree

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System