RE: Calling the streamReload IOC Shell function within C

["Wang, Andrew via Tech-talk" <tech-talk@aps.anl.gov>]

Thank you!

-----Original Message-----
From: Zimoch Dirk (PSI) <dirk.zimoch@psi.ch> 
Sent: Monday, January 24, 2022 1:42 AM
To: Wang, Andrew <wang126@llnl.gov>; EPICS tech-talk <tech-talk@aps.anl.gov>
Subject: AW: Calling the streamReload IOC Shell function within C

Hi Andy,

I am going to add streamReload into devStream.h. You can simply do that for now.
 I just need to check first which Windows decorations are needed...

Dirk

> -----Ursprüngliche Nachricht-----
> Von: Wang, Andrew <wang126@llnl.gov>
> Gesendet: Montag, 24. Januar 2022 09:18
> An: Zimoch Dirk (PSI) <dirk.zimoch@psi.ch>; EPICS tech-talk <tech- 
> talk@aps.anl.gov>
> Betreff: RE: Calling the streamReload IOC Shell function within C
> 
> Hi Dirk,
> 
> Since there isn't a function prototype for streamReload, do you 
> recommend that I manually add it into devStream.h? gcc isn't liking 
> the fact that I am calling to streamReload, because there is no prototype.
> 
> Thanks,
> Andy
> 
> -----Original Message-----
> From: Zimoch Dirk (PSI) <dirk.zimoch@psi.ch>
> Sent: Saturday, January 22, 2022 3:04 AM
> To: EPICS tech-talk <tech-talk@aps.anl.gov>; Wang, Andrew 
> <wang126@llnl.gov>
> Cc: Michael Davidsaver <mdavidsaver@gmail.com>
> Subject: Re: Calling the streamReload IOC Shell function within C
> 
> Hi Andy
> 
> You can call extern „C“ long streamReload(const char* recordname) 
> directly from C/C++. If NULL is passed for recordname, then all 
> StreamDevice records reload. If a record is in the middle of 
> processing, the running protocol is aborted and the record goes to INVALID alarm until the next processing.
> 
> It seems I have forgotten to put the function prototype into 
> devStream.h. Will be done in the next release.
> 
> Dirk
> 
> > Am 22.01.2022 um 06:08 schrieb Michael Davidsaver via Tech-talk 
> > <tech-
> talk@aps.anl.gov>:
> > On 1/21/22 19:39, Wang, Andrew via Tech-talk wrote:
> >> This is a far-fetched question, but I was wondering whether there 
> >> is a
> possibility of executing streamReload, or a similar functionality, 
> within a C function. The reason why I ask is because I saw within this 
> Github link
> (https://urldefense.us/v3/__https://github.com/epics-
> modules/ioczed/blob/master/zzzApp/Db/streamExample.db__;!!G2kpM7uM-
> TzIFchu!gjnSZo93VcnoDkmziao7QLSXtfmZDUg0hZ5hqdkj-
> M2oKQch9GQRdAvkPjUS4spw6a4$
> <https://urldefense.us/v3/__https://github.com/epics-
> modules/ioczed/blob/master/zzzApp/Db/streamExample.db__;!!G2kpM7uM-
> TzIFchu!gjnSZo93VcnoDkmziao7QLSXtfmZDUg0hZ5hqdkj-
> M2oKQch9GQRdAvkPjUS4spw6a4$ >) that the SUB record calls streamReload.
> However, I couldn’t locate the source code for that, so I am not 
> entirely sure how it is used and what its arguments are. Thoughts?
> >
> > The registrar code adding both the function callable from subRecord 
> > as well as iocsh.
> >
> > https://urldefense.us/v3/__https://github.com/paulscherrerinstitute/
> > St
> >
> reamDevice/blob/dfbd308d4602e7cb99d7f60ddf35909a70f41c4f/src/StreamEpi
> > cs.cc*L297-L307__;Iw!!G2kpM7uM-
> TzIFchu!gjnSZo93VcnoDkmziao7QLSXtfmZDUg
> > 0hZ5hqdkj-M2oKQch9GQRdAvkPjUS5UlYp3c$
> >
> > So you could do something like
> >
> >> iocshCmd("streamReload");
> >
> > cf. iocsh.h from Base.
> >
> > Since it is already callable from a subRecord I assume it is 
> > thread-safe.