Experimental Physics and
Industrial Control System

Tim Mooney <[email protected]> · Tue, 27 Oct 2009 09:54:25 -0500

haquin wrote:

Hi Tim,

Thank you for your proposal but:

I looked in the genSub record code and I think that what I do should

work ...

After deeper investigations, it turns out that when I initialize the

INPH field type with FTH=STRING or CHAR it is then assigned a CA_LINK

type ((*plinkin).type)==CA_LINK instead of CONSTANT ...

That's why the Constant string is not copied at init.

When FTH=LONG then ((*plinkin).type)==CONSTANT

How can I force it to be a CONSTANT

I think the only way to make the link type CONSTANT is to make the
link value entirely numeric.  As I read the code, FTH should have no
effect on the link type, because the EPICS code that makes the decision
doesn't look at anything other than the link value.

Tim Mooney a écrit :

haquin wrote:

Hi,

I initialize the INPH (with macro substitution) of a gensub record

with a string,

but it is never transferred to the corresponding H variable ...

in the IOC shell I use the commands:
epics> dbgf MyGensub.INPH
       DBR_STRING:          MyString NPP NMS

epics> dbgf MyGensub.H
       DBR_STRING:

The behaviour is different if I use LONG instead of STRING ...

Can someone tell me how to do retreive MyString in MyGensub.H ?

You can use a stringout record to hold the string, like so:

record(stringout, "myStringInit") {
   field(VAL, "$(MYSTRINGMACRO)")
   field(UDF, "0")
}
record(aSub, "myASub") {
   ....
   field(INPH, "myStringInit")
   ....
}

Then, dbLoadRecords("myDatabase.db", "MYSTRINGMACRO=MyString")

--
Tim Mooney ([email protected]) (630)252-5417
Beamline Controls & Data Acquisition Group
Advanced Photon Source, Argonne National Lab.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System