Re: EPICS to Python / EPICS to Tcl

[Tim Mooney <mooney@aps.anl.gov>]

The nearest development I know of to this is Miroslaw Dach's embedding 
of Octave
into an ioc, which is described at 
http://epics.web.psi.ch/software/ioc-octave/

I think having the ability to run scripts in an ioc is a good thing, but 
I think it
would be more helpful to have the capability in a record type, than to 
have it
in an asyn driver, because records have links.  I've always thought it 
would be
nice to have a version of the genSub record that would call Python scripts,
instead of compiled code, but I never did anything about it.

Tim Mooney

Timothy Madden wrote:
> Folks
>
> I want to make an IOC that does the following:
>
> Serve PV's that call Asyn, They could BI, BO or any standard record.
> Make an Asyn driver that is linked with  a Python or Tcl interpreter. 
> The IOC written in C, would then call scripts which could be stored as 
> text files on the disk.
>
> A record named "X" would call a script function named "X" and pass its 
> PV value to it. Likewise, the output of the script could be passed 
> back to the PV value. Python and Tcl interpreters provide ways for C 
> to call scripts, pass and retrieve data from scripts. Also, it is not 
> too hard to link  an interpreter into a C program. All that is needed 
> is to write the ASYN-based C functions that call the interpreter. It 
> could be done in Linux or Cygwin.
>
> In this way, one could make an IOC to do whatever one wanted without 
> ever messing a C compiler. The user would write a few scripts, make a 
> st.cmd file, and a db file. In this way, new IOCs could be very 
> quickly developed.
>
> Has anyone already done this?
>
> Tim Madden


--
Tim Mooney (mooney@aps.anl.gov) (630)252-5417
Beamline Controls & Data Acquisition Group (www.aps.anl.gov/aod/bcda)
Advanced Photon Source, Argonne National Lab