[Halld-online] epics2et

[David Lawrence]

Hi All,

  I have finally tracked down the bug that was preventing the EPICS values from being inserted
into the evio data file and have committed the changes to svn. I have also added a DEPICSvalue
class to the DAQ plugin. The values can now be accessed in the normal “JANA” way via:

#include <DAQ/DEPICSvalue.h>
vector<const DEPICSvalue*> epicsvalues;
loop->Get(epicsvalues);

Note that EPICS events will have no detector-based data in them. The two event types are mutually
exclusive. At this point, the “last encountered” values are not maintained anywhere so if you want to use
the most recent EPICS value when processing an event, your code will need to remember it.
There are some potential gotchas with race conditions when using multiple threads so those
details will need to be worked out later.

Please note that the program that inserts these into the data stream is currently commented 
out (it is in the “run_go” script so feel free to uncomment it Sergey if you get to it first). We
should be able to activate it tomorrow at the latest.

The configuration file used to determine which variables are inserted and how often can be found 
in svn here:

https://halldsvn.jlab.org/repos/trunk/online/daq/config/monitoring/epics2et.conf

If you have values that should go in here and are not listed, let me know and I will add them.
Keep in mind though that this should NOT become a duplicate of the EPICS Mya archive.
Only those few values that would be really convenient to have in the data stream should be
put there. Several of them are recorded infrequently (every 15 minutes) so that should be
OK.

Regards,
-David