Experimental Physics and
Industrial Control System

Gerrit Kühn via Tech-talk <tech-talk at aps.anl.gov> · Fri, 3 Nov 2023 11:12:22 +0100

Hello,

I have a (probably very simple, but I fail to see the answer) question on
records with constant input:
What I'd like to have is a stringin record that is initialised with a
constant value that can later be changed with caput. From the
documentation on the stringin record I get 

"The INP field determines where the string input record gets its string.
It can be a database or channel access link, or a constant. If constant,
the VAL field is initialized with the constant and can be changed via
dbPuts."

which sounds exactly like what I need.

However, when trying a record like

record (stringin, "$(P):$(T)-$(B)_$(R)_$(D)_$(UN)_UID")
{  field (DESC, "Device Identifier for $(U)")
   field (DTYP, "Soft Channel")
   field (INP, "test")
}

this does not work for me. I don't see any error messages, but I get an
empty record:

---
epics@epics-appl:~$ caget A0:PEM-3405_106_TF1_DO1_UID
A0:PEM-3405_106_TF1_DO1_UID    
---

What is even more strange (I think) is that I get an error message when
trying to write a value to this record:

---
epics@epics-appl:~$ caput A0:PEM-3405_106_TF1_DO1_UID
test
Old : A0:PEM-3405_106_TF1_DO1_UID    
CA.Client.Exception...............................................
    Warning: "Channel write request failed"
    Context: "New : A0:PEM-3405_106_TF1_DO1_UID    test
op=1, channel=A0:PEM-3405_106_TF1_DO1_UID, type=DBR_STRING, count=1,
ctx="A0:PEM-3405_106_TF1_DO1_UID"" Source File: ../oldChannelNotify.cpp
line 159 Current Time: Fri Nov 03 2023 11:04:49.636763495
..................................................................
---

The write appears to succeed anyway, though, the value is updated.

It looks like I can get the desired behaviour when using the VAL field
instead of INP in the record definition. While this is fine by me, it
feels like I misunderstand how the constant INP field is supposed to work
as described in the manual above. Could anyone shed a light on this mor me?

cu
  Gerrit

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System