EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  <19961997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  <19961997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Re: access to array elements
From: [email protected] (William Lupton)
To: [email protected]
Date: Wed, 7 Aug 96 11:48:28 HST
Dear all,

  Just to summarize, there have been several replies (all basically
positive and all but one sent via tech-talk).

Nick Rees said:
> My only comment is that I see this as something which may be better
> handled at the server end, rather than the client end

  When I made the original suggestion, I _was_ thinking of some
client-side convention that would not even impact the CA library, so
xxx[1] would still have read the entire array and then just retain the
second element.

  However, I agree with Nick that it _does_ seem more like a server side
function. Nick suggests that he is thinking in terms of handling the
entire "xxx.YYY[n-m]" string on the server side. The trouble with this
is that CA needs the "xxx.YYY" bit for name resolution (and so would
have to know about the suffix syntax), whereas database processing
handles the "YYY[n-m]" bit.

Jeff Hill said:
> I agree that this would be quite useful (it has been suggested before).
> The database and CA client API's will need modifications so that the
> first element (in addition to the count can be specified).

  Jeff is envisaging a CA client extension to allow access to any
contiguous set of elements.

Jeff continued:
> With the ability to fetch any element out of the array it should be
> reasonably easy to parse the "[ element ]" out of the PV name string and
> use this as an offset off of what is specified from the client API
> (so this can be transparent to existing code).

  I think this is referring to code above CA which would take something
like "xxx[2-4]" and parse it into name="xxx", offset=2, count=3, i.e.
this is still referring to a client-side facility but assuming the CA
extension to specify the offset?

  So are there opinions on how far to go here?

1) pure client convention (perhaps using provided library routines)

2) (1) plus CA extension to support offset

3) handle entirely on server side

4) other

  William


Navigate by Date:
Prev: Re: access to array elements Jeff Hill
Next: Initialization of analog output records Jens Bergl
Index: 1994  1995  <19961997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: access to array elements Jeff Hill
Next: Dev/Driver for Dynapower and Inverpower Power Supplies Sebastien Fabbro
Index: 1994  1995  <19961997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·