Re: pyepics callback function problem

[Janko Kolar <janko@i-tech.si>]

Hello

Your suggestion solved the problem.

Thank you for your quick response.

                               Janko
> Hi Janko,
>
>   
> > I hope this is a right place to ask a question about pyepics.
> >     
> I think it is.
>
>   
> > I wrote a simple program to monitor their value:
> >
> > import sys
> > import time
> > from epics import PV
> >
> > def callback_function(pvname=None, value=None, char_value=None,
> > type=None,units=None, enum_strs=None, **kw):
> >     print "------Printout from callback ------"
> >     fmt = ' pvname \t = %s \n value \t\t = %s \n\
> > char_value \t = %s \n type \t\t = %s \n\
> > units \t\t = %s \n enum_strs \t = %s \n'
> >     print fmt % (pvname, value, char_value, type, units, enum_strs)
> >     print "----------------------------------\n"
> >
> > pll = PV('LUCA:ENV:ENV_SC_PLL_MONITOR')
> >
> > vent = PV('LUCA:ENV:ENV_FRONT_VENT_ACT_MONITOR')
> >
> > pll.add_callback(callback_function)
> > pll.run_callbacks()
> >
> > vent.add_callback(callback_function)
> > vent.run_callbacks()
> >
> > while True:
> >     time.sleep(0.001)
> >
> >     
> Sorry for the trouble.
>
> The issue is that not all of the "extra control fields" (including the
> names of the enumeration states, units, and precision) have been
> determined when you do PV.run_callbacks().  That is only the "native
> form" of the PV
>
> The easiest fix is to explicitly make the control fields be looked up,
> either with PV.get_ctrlvars() or PV.get(as_string=True) which also
> makes sure that the "char_value" is well formed.  That is, I think you
> want
>
>    pll.add_callback(callback_function)
>    pll.get(as_string=True)
>    pll.run_callbacks()
>
>    vent.add_callback(callback_function)
>    vent.get(as_string=True)
>    vent.run_callbacks()
>
>    while True:
>        time.sleep(0.001)
>
> Perhaps an option should be added to PV.add_callback() to make sure
> the control fields are initialized.  Either that, or this should be
> better documented.
>
> --Matt Newville <newville at cars.uchicago.edu> 630-252-0431
>
>