Re: Sequencer handling of waveforms with dynamic size

["Johnson, Andrew N. via Tech-talk" <tech-talk at aps.anl.gov>]

Unfortunately in version 2.2.9 at least, the current element count value that gets provided by the CA event handler arg's isn't actually saved anywhere by the sequencer code. Looking at the related code I suspect this wouldn't be trivial to fix because when it was written the CA protocol didn't tell clients what the active length of the array was.

 

- Andrew

 

-- 

Complexity comes for free, Simplicity you have to work for.

 

 

From: Tech-talk <tech-talk-bounces at aps.anl.gov> on behalf of Ralph Lange via Tech-talk <tech-talk at aps.anl.gov>
Date: Friday, September 29, 2023 at 7:52 AM
To: EPICS Tech Talk <tech-talk at aps.anl.gov>
Subject: Re: Sequencer handling of waveforms with dynamic size

On Fri, 29 Sept 2023 at 14:46, Ralph Lange <ralph.lange at gmx.de> wrote:

On Fri, 29 Sept 2023 at 13:53, Jure Varlec via Tech-talk <tech-talk at aps.anl.gov> wrote:

Encountering this issue, I'm having a feeling of deja vu, so this might have been discussed before. But I can't find anything.

I found that a sequencer variable assigned to a waveform doesn't seem to know its current size. The pvCount() function always returns the value of NELM (or the array size of the sequencer variable, if smaller), not NORD. Is there some way to obtain the current size of the waveform that I missed? I'm not too keen on creating a separate variable assigned to the NORD field because there's no guarantee that the two variables will be in sync.

 

Isn't https://epics-sequencer.sourceforge.io/sequencer-2-2/Tutorial.html#process-variable-element-count what you're asking for?

 

I mean:

So the pvCount() is just returning the number of elements from the CA metadata? I see.

 

~Ralph