1994 1995 1996 1997 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 1996 1997 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: pyepics not updating pv.enum_strs after connection |
From: | Matt Newville <[email protected]> |
To: | Jameson Graef Rollins <[email protected]> |
Cc: | EPICS tech-talk <[email protected]> |
Date: | Tue, 24 Mar 2015 22:46:59 -0500 |
On Thu, Mar 19 2015, Matt Newville <[email protected]> wrote:
> I think it kind of depends what you're trying to do. FWIW, I have lots of
> GUIs that display Epics variables and get control values in a "lazy
> manner". Generally, one can create a PV, then ask some function to be
> run when the GUI is not busy where you wait-for-connection, get the control
> variables, and add callbacks (in the case where the widgets don't even
> exist when the PV is created-without-connection). Another approach would
> be to use a Timer loop look for newly connected PVs (those for which
> pv.connected was False and is now True) and finish setting up the pvs as
> they connect (including getting ctrl vars).
Hi, Matt. I set up something similar to what you've suggested here and
it seems to work ok. A few seconds after initialization a function gets
executed that periodically checks for newly connected channels and
updates their ctrl attributes. The ENUM records then start to get
filled in with the proper values after the program has been running for
a couple of seconds. Thanks so much for the suggestions.
>> I still don't understand why the enum_strs attribute can't just be
>> updated automatically when the connection is first made, without
>> having to retrieve it manually at some point. It would certainly
>> make things easier from my perspective.
>
> That would be possible. It's not clear to me that every PV always needs
> CTRL variables to be fetched as part of the automatic connection process --
> I might be persuaded to add this as an option,, but am not certain it's
> necessarily always a good idea. It might slow down the connection (I
> have not profiled it carefully), but it's probably acceptable in many
> situations.
The process I describe above to get this working seems pretty hacky to
me, especially given that this is probably a fairly use case (especially
for any GUI apps). It would certainly make things much easier from an
end programmers perspective if it were possible to specify that the
connection callback could also just update the ctrl attributes
automatically.