EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  <20022003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  <20022003  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: VDCT groups
From: John Maclean <[email protected]>
To: Luedeke Andreas <[email protected]>
Cc: Matthias Clausen <[email protected]>, EPICS tech-talk <[email protected]>
Date: Tue, 03 Dec 2002 10:00:13 -0600
Luedekes' message makes me think that now would be a good time for a
status update on the progress of VDCT development. For those at the
recent collaboration meeting, please forgive the repetition.

As you may recall, I've been acting as the contact point between the
community and Cosylab, the VDCT developers. I asked any one interested
to try VDCT and send me feedback on problems and requirements for new
features. Well, I got a lot of feedback. From this we were able to draw
up a list of requirements that was split into two work packages, A and
B. Package A was funded by the APS and is complete. Package B was funded
by DIAMOND and is nearing completion. 

I think that when package B is finished VDCT will be a very good tool
for database development. We are now at the stage of fixing bugs rather
than implementing new features with this package so I hope that the
final release for this package will be available within a few weeks. 

Although hierarchy is implemented in package B, it is somewhat difficult
to use. Among the things I would like to put into the next work package
is an improvement in the way hierarchy is done. Last week we had a
meeting at the APS with Matej Sekoranja (the VDCT author), Rozelle
Wright (LANL), Andrew Johnson (APS) and myself. We devised an improved
specification for hierarchy implementation. This will give everything
available in CAPFAST and more.

I've tried to keep discussions of VDCT requirements to those who have
expressed an interest and to keep the noise off of tech-talk. I intend
to get a VDCT status web page up when I get the chance. This would be
the place to find what's happening with VDCT, what new requirements are
being mooted etc.

John.

Luedeke Andreas wrote:
> 
> Maybe tech-talk is a good place to discuss extensions of the VisualDCT.
>
>
> Currently I'm using the VisualDCT to analyse complex databases.
> Those Databases were not written with VisualDCT.
> The features required here are:
> 
>    * displaying records
>    * showing internal interconnections (Links) between the records
>    * grouping of the records according to parts of their names
>      (it splits the channel name at a predefined character according
>      to our naming convention into a "device" and a "property" and
>      displays all records of one device in a "group")
> 
> These features allow us to analyse databases or templates of dozens or hundreds
> of records with rich interconnections. We can use these features because:
> 
>    * Grouping is done only by the record names, no extra information is required.
>    * We just need to drag the records to appropriate positions to
>      get a full view of the information- and process flow in the database
>      due to the "rubber band" link display.
> 
> We have an (old) "wish list" of what would be nice extensions of VDCT for this purpose:
> 
>    * Show links between groups (display a group as a "meta-record")
>    * Mark external links with a label "External:" and
>    * display external links also in the "group" displays.
>    * Provide an "auto-placement" function, that tries to place all records
>         + to eliminate the overlap of records
>         + to minimize the crossing of links (auto-routing)
>    * Visualization of the process chain (an EPICS database simulator)
> 
> The implementation of those features would only make sense if a larger
> community uses the VisualDCT, since quite some manpower would be needed.
> 
> A second purpose for the VisualDCT is to be a programming environment for
> EPICS databases. The requirements here are partly different:
> 
>    * Syntax checking (already implemented)
>    * Handling of substitution files and check of template expansion.
>      (All substitution variables provided? Duplicated channels generated?
>      Expanded links resolved?...)
>    * Tool for database documentation (already many good features provided)
> 
> The idea of hierarchical databases falls into this category, too.
> I think this should be an "official" extension to the syntax of
> databases and templates, too. I.e. dbLoadTemplate should
> be extended or replaced to load databases with this syntax.
> But even before that, it's a great feature for VisualDCT.
> 
> Graphical database configuration is a very basic feature of EPICS.
> I guess everyone could add more ideas to the discussion.
> Even if we do not have the manpower to implement it :-)
> 
> Andreas
> 
> Matthias Clausen wrote:
> 
> > Grouping and hierarchies would be one of the most important reasons to use VDCT.
> > We even need the transparent forwarding of macro substitutions - which is disabled at the moment.
> >
> > We should collect the requirements for the next release...
> >
> > Matthias
> >
> > John Maclean wrote:
> >
> > > We are drawing up requirements for future versions of VDCT.
> > >
> > > Based on discussions with a number of people it appears the the grouping
> > > facility is not used but is often a source of confusion.
> > >
> > > So the question is, should we remove grouping from VDCT? If we don't
> > > remove it then it will certainly be turned off by default.
> > >
> > > Note that groups are different from the hierarchy scheme that is now
> > > being implemented. See the user manual at
> > > http://www.cosylab.com/visualdct/builds/VisualDCT/2.3.1238/doc/MAN-VisualDCT_Users_Manual.html
> > > if you need more information on groups or hierarchies.
> > >
> > > If no one objects we'll remove grouping.
> > >
> > > John Maclean.
> 
> --
> Dr. Andreas Luedeke, SLS Operations Manager, Tel:+41-56-3104002
> mailto:[email protected], http://people.web.psi.ch/luedeke

References:
VDCT groups John Maclean
Re: VDCT groups Matthias Clausen
Re: VDCT groups Luedeke Andreas

Navigate by Date:
Prev: Re: VDCT groups Luedeke Andreas
Next: Re: VDCT groups John Maclean
Index: 1994  1995  1996  1997  1998  1999  2000  2001  <20022003  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: VDCT groups John Maclean
Next: device support routines for RS232 Lifang Zheng
Index: 1994  1995  1996  1997  1998  1999  2000  2001  <20022003  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 ·