Experimental Physics and
Industrial Control System

[email protected] · Thu, 14 Nov 2002 18:02:27 -0600 (CST)

I agree whole-heartedly with everything Chris has to say.  I have several  quasi script programs that we use to control experiments--basically push a button here, read a value there, type stuff.  Jave makes a nice cross plateform tool for building UI that have some smarts and some minor number chrunching ability--something way beyond the realm of MEDM.

I could have used TCL/tk or Python, but they don't avoid the CA linkage issue and I've come to really like Java as a development language anyway.    Even though I've linked to CA--through Eric Boucher's JCA--I still have the problem of maintaining and deploying C code with my control apps.

I remember the talks of reverse engineering CA for use in Java--in fact, I seam to remember that some did in fact reverse engineer CA.  I doubt I would use anything reverse engineered like this because of the reasons put forth by Jeff and others, but I got to tell you--it's awfully tempting...

Brian

>On Thu, 2002-11-14 at 17:06, Jeff Hill wrote:
>> 
>> I am certainly no expert on java, but it seems that a "pure java"
>> version of the ca client library would be somewhat at odds with the
>> normal domain of java which from my perspective is rapid prototyping,
>> code component integration, and web development. There may be advantages
>
>  No, there's a lot to be said for Java's cross-platform capabilities.
>Currently, if I want to access CA from java, I need to build the CA
>library for every platform on which my Java app will run.
>
>  This is fairly close to home for me, since I am working on a 
>StripTool-esque Java app.  I had tossed around the idea of implementing
>the CA protocol in Java, but have settled instead on relying upon a
>CORBA <-> CA gateway.
>
>  I think this is one of a whole realm of programs which don't need
>optimal speed to perform efficiently, but which can benefit greatly
>from the rich API java provides.
>
>Chris
>-- 
>Christopher A. Larrieu
>Computer Scientist, Acc. Controls Group
>Thomas Jefferson National Accelerator Facility
>

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System