g+
g+ Communities
Argonne National Laboratory

Experimental Physics and
Industrial Control System

<20122013  2014  Index <20122013  2014 
<== Date ==> <== Thread ==>

Subject: Re: Survey comments
From: Emmanuel Mayssat <emayssat@gmail.com>
To: Andrew Rhyder <Andrew.Rhyder@synchrotron.org.au>
Cc: "So, Sung-Leung" <so@bnl.gov>, "qti-talk@aps.anl.gov" <qti-talk@aps.anl.gov>
Date: Mon, 4 Jun 2012 17:32:20 -0700
 "Unlike SWIG, SIP is specifically designed for bringing together Python and C/C++ and goes to great lengths to make the integration as tight as possible." 
From http://www.riverbankcomputing.co.uk/static/Docs/sip4/introduction.html
PyQt has been wrapped with SIP.
You can get SIP support from the PyQt mailing list.
FYI
--
Emmanuel


On Mon, Jun 4, 2012 at 4:16 PM, Andrew Rhyder <Andrew.Rhyder@synchrotron.org.au> wrote:
Hi Ivan
I am happy with either SIP of SWIG. SWIG is of interest to my colleague Richardo Fernandes who is likely to be doing the work. Any comments on why one should be preferred over the other would be welcomed.
Regarding the Model-View-Controller design pattern, there will always be a natural separation of sorts between the model running on an IOC and a view / controller running on an OPI, but this can be blurred. At the Australian Synchrotron, and I have heard stories from elsewhere, we have ended up with
collections of PVs present in operational IOCs only to manage the look and functionality of MEDM/EDM based GUIs. In an attempt to alleviate this, we have attempted to build the EPICS Qt framework with enough functionality built into the widget set so the required widget interaction can be configured when laying out GUIs with Qt Designer. This leaves us with the opposite danger where the operational model can start to creep into the GUI. In other words, a GUI can be designed that embeds inappropriate operational functionality within the GUI, but at least this is choice that can be avoided. One test to see if the operational model has crept into the GUI it to turn the OPI off. If everything keep going, the OPI is clean!
Regards
Andrew
________________________________________
From: qti-talk-bounces@aps.anl.gov [qti-talk-bounces@aps.anl.gov] On Behalf Of So, Sung-Leung [so@bnl.gov]
Sent: Tuesday, 5 June 2012 2:10 AM
To: qti-talk@aps.anl.gov
Subject: Survey comments

Hi Andrew,

Just a comment on your item 1:

>1.    Default installation should support C++ and python development
>
>We will be working on a python wrapper for EPICS Qt soon - (probably using SWIG?)

The path of least resistance seems to be using SIP and PyQt from:

http://www.riverbankcomputing.co.uk

All Qt-designer widgets can already be used in python code. Python widgets can be used in Qt-designer as plugins :

http://www.riverbankcomputing.co.uk/static/Docs/PyQt4/html/designer.html#writing-qt-designer-plugins

Also, I think working within the tried-and-true Model-View-Controller design pattern is a good choice for Qt and EPICS Qt.

Ivan
so@bnl.gov

<br>This message and any attachments may contain proprietary or confidential information. If you are not the intended recipient or you received the message in error, you must not use, copy or distribute the message. Please notify the sender immediately and destroy the original message. Thank you.


References:
Survey comments So, Sung-Leung
RE: Survey comments Andrew Rhyder

Navigate by Date:
Prev: RE: Survey comments Andrew Rhyder
Next: a MEDM replacement Mezger Anton Christian
Index: <20122013  2014 
Navigate by Thread:
Prev: RE: Survey comments Andrew Rhyder
Next: a MEDM replacement Mezger Anton Christian
Index: <20122013  2014 
ANJ, 05 Jun 2012 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· EPICSv4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·