on shutdown, time.sleep(10) is blocking the reconnect thread.
change the _shutdown attribute from bool to an Event, and
use Event.wait instead of time.sleep
Change-Id: Icea6a14ad73df0b3d26ef45806f4c05e6bf18492
assumptions about dispatcher in playground.py are no longer
valid.
- let Dispatcher class in playground inherit from real dispatcher
+ improve log messages
Change-Id: I2a9a9d532dabadc590543660c445c021dd2f2891
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31967
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
unfortunaely, sometimes communication errors happen.
workaround: try several times reading the status
Change-Id: I2788c6c9b4145246cdd51c31b246abffee60f93b
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/32032
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
This reverts commit e67a46cd015c0a1a32d5a4f114b963dd17a7c266.
Reason for revert: required version available from pypi
Change-Id: Ib4f8b0cf62da58e84545511c7521ea93b7ff1342
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31724
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
- when driving a module with <module>(<target>),
keyboard interrupt should send stop()
- make sure keyboard interrupt does not only stop
the current driving, but also skips other code
on the same command line
Change-Id: Ib4d2c4111dc0f23bf07385065766fb9b4a611454
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31926
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Dispatcher.get_module_instance returns None on failure.
If that is the case, the dispatcher should not try to register the
None value as a module.
Change-Id: Ie33b8debc2a829d480d56cafc1eb0ab610181d67
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31713
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
cut of one byte too much in barcode decode
Change-Id: I5f1f8475f197b13af836d685dc6da5a9ee824dc2
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31728
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Some logging calls should not have landed as log.info in the dynamic
modules patch. This fixes that.
Change-Id: I666fc7c9b5c65ddbed1c26ea456becce7870e744
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31707
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Some fixes after the device was tested with socat ptys and NICOS.
Change-Id: I3e9dba2be2547d493c435d1da9844c932a2df4e6
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31662
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Adds a Barcode reader device (for now, only for ANTARES). Not yet
tested with real hardware.
Change-Id: I25f097466be89d152f47b9d05ece8f562e4b34d6
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31412
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
restarting the phytron motor without prior stop leads
to funny behaviour.
- send stop before restart
- stop motor when moving but status not busy
- restart when motor drives the wrong way
+ better status text when stopping
Change-Id: I82cd59297b3c79a354a4eeb5ba03fc65bedf755f
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31929
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
for host name without port, None was used for the port
leading to a confusing error message
- do not call parse_host_port with None as defaultport argument
- improve error message when connection fails
+ fix an error in last line of parse_ipv6_host_and_port
+ fix some issues breaking PEP 8 rules
Change-Id: I437360be96449c164f0080e3c60f1685825d4780
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31911
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
- in SEA, it is not guaranteed that the is_running state is set
before the run command returns. as a consequence, we have to
wait in SeaDrivable.write_target for is_running being set
- syncio has always to be reconnected after asynio
Change-Id: Ia46cff11de86868ce0627faaf6f776282bd7a8f4
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31631
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
the _connect was sometimes started in parallel from
startModules and the first call to doPoll.
remove the first one, and protect the second one
with a lock
Change-Id: I079439e150efd5d005130cef475f6326f933ecbd
add module which scans a connection and registers new devices depending
on the answer.
* change module initialization to demand-based
* move code from server to dispatcher
- remove intermediate step in Attached __get__
TODO:
factor out dispatcher (regards to playground)
discuss factoring out of module creation code from server AND
dispatcher
Change-Id: I7af959b99a84c291c526aac067a4e2bf3cd741d4
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/31470
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
