Re: SNL and CapFast

[Andrew Johnson <anj@ast.cam.ac.uk>]

Rozelle,

On Fri, 28 Jun 96 09:58:29 MDT Rozelle Wright 
<wright@luke.atdiv.lanl.gov> wrote:

> 1.  Could you give us a few more details about what used to work?
> If it is just a matter of putting all of the relevant channels
> in the same symbol so they wind up with related names, perhaps
> it is because the parameter passing scheme has changed between
> edif2sr and e2sr.  I expect you are talking about something
> more complicated. 
> 2.  I don't know of anyone who has made a conversion tool.  It 
> would be really nice to have such a tool. 

What Philip used to do was put a drawing symbol (eg a circle) onto his 
diagram to represent the state machine, and wire the channels the SNL 
program talked to up to control points on this symbol.  Because the circle 
wasn't an EPICS symbol the conversion used to ignore these connections. I 
think the conversion now complains.  (I have to admit to reporting all of 
this this second-hand, as Philip gave up trying to get it working and I 
haven't tried myself; I want to see if we'll be able to use it in future 
designs).

The logical next step is to be able to have the symbol as a hierarchical 
link to the State Transition Diagram which implements the state machine, 
but that would need this conversion tool (which maybe ought to be part of 
e2sr anyway?). I don't believe this would be particularly difficult to 
write for someone who already knows about processing edif.  The difficult 
bit is working out how to draw the STD in a way which allows it to 
be converted into SNL, and allowing people to put normal SNL/C code into 
the diagram too.

- Andrew

  O o ._    Andrew Johnson, Royal Greenwich Observatory
     __\\_	Madingley Road, Cambridge, CB3 0EZ
  __|ooooo|__	Tel +44 (0)1223 374823  Fax 374700
~~\_________/~~	WWW http://www.ast.cam.ac.uk/~anj/