- switch to pathlib
- represent multiple confdirs as list of Paths internally, not string
with pathsep
Change-Id: I1418e561641e27cd904af0762be056cd66ee1919
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34464
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
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>
the user would expect that updates are shown in the order
they appeared in the SECoP connection.
For this register_callback needs an additional parameter
to suppress initial callback.
Change-Id: Id1d7605d739b02ec799b3768ae2399558969c381
numpy.finite() does not accept enums -> convert to float first
Change-Id: Ib740800a7410ada38fb48711d4da676af8b64b17
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34608
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
the empty QLabel in widgets for readonly parameters was not hidden
with the group.
Change-Id: I0e5ad49c3b693483a3cabf849e7962925cfcc434
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34607
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
So far, frappy is still supposed to be 3.6 compatible
Change-Id: Ieb56c6500d58cdc1fe2f34b1448b05ba74b0fa6e
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34326
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
overwriting defaults before init() had no effect, as the values were
replaced there.
Now, only unset defaults are updated, keeping the overrides.
Also fix unconditionally taking the values from environment variables.
Change-Id: Idf1c5e2338403e061d20c11ed9c4803d535eb188
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34304
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
The mlzlogger has an extra argument we want to keep using.
To keep compatibility, check if we got the mlzlogger or something else.
Change-Id: I1a4f4e7651370dd71124ab7c3608c6573a585dd9
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34292
Reviewed-by: Georg Brandl <g.brandl@fz-juelich.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
installing a custom logger was not possible, since the RemoteLogHandler
was not installed.
Additionally, search for the RemoteLogHandler recursively upwards, since
not all loggers copy their Hanlders to their children.
Change-Id: If3e19966d9289cacd926648582e9718b7eab279c
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34288
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
leftover error_map from long ago, turned into direct raising of
HardwareError
Change-Id: I8772e20133502bb0a997d5317a74e2bd16b780c6
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/34279
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
- fix a bug then TcpServer can not start dye to address in use
- report errors when restarting interfaces
- increase timeout. the timeout for waiting all interfaces
starting up must be higher than a potential successful
startup of TcpServer, which is currently ~ 10 sec
(might be reduced, but at both places)
Change-Id: I88b967c4baff79fdf94f4c849dd713d2cba6fabc
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33985
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>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
will make setup checking easier, although it does not work for all cases
(from imports for example)
Change-Id: I2e5cf9b427a6f8e6f603d9662cdb9700d5f9ad9a
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33893
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
the error messages for tuple and array swallow all details useful for
debugging. add the original exception for some context
Change-Id: I8e1036bfab10c09d5f9c8af05ac3306af89f54c4
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33850
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
update cache in case of connection errors, as they are not handled
in the rx thread
Change-Id: Icf3377020ec314fcef2982a4e6dc64356f787273
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33744
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
as callbacks trigger by updates while activating may check for
online state
+ remove unused imports
Change-Id: I37df839abf6b7225389b803347234a3d0bc8d799
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33745
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
in case th rx thread crashes, the shutdown callback is called twice
-> improve cleanup code in __rxthread
Change-Id: I0d20aa4304d94b17565b018ad8528d61bbbcbc83
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33614
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
- implement a 'handleError' callback
- SecopClient.handleError is registered as a callback and by default
logs errors up to a limited number of times
- the cache is customized to return an item indicating an undefined value,
helping to avoid follow up errors.
+ frappy.errors: cosmetic fix
+ frappy.client: cosmetic changes
Change-Id: I4614e1d27c7aea3a1a722176c6be55f8563597cd
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33429
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>
buggy updates or errors in callbacks are now converted into
error updates, buggy error updates are skipped.
-> the receive thread should no longer crash
Change-Id: I97e3999db73e64f73dfbc380fac3d7685b6ca31c
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33386
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>
when a reply to a request is not received within 10 s a timeout
error is raised, but any further requests with the same identifier
will be blocked - fix this
Change-Id: Id2fbc8bb6e6ecfd54bba1fa57b62e626e49b54c8
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33385
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
this seems to be broken in change 33266
Change-Id: I4da78f297976daeac0a0709b9c86e6e28fc122bf
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33268
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
triggerPoll must be set on attached io
Change-Id: If3fa1016efa6047371380790f60db246d87b3628
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33269
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
- fix inheritance of command description
- when no stop method is given, then the description should indicate
that stop is a no-op -> add missing doc strings to stop methods
- add test to make sure stop command doc strings are given
when implemented
Change-Id: If891359350e8dcdec39a706841d61d4f8ec8926f
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33266
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Alexander Zaft <a.zaft@fz-juelich.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
on Module, use one single callback list 'paramsCallback' instead of
'valueCallbacks', 'errorCallbacks'. Redesign the mechanism to
avoid most of the closures.
Change-Id: Ie7f68f6bf97ab3f3cd961faa20b0e77730e5b37d
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33118
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
- for the case when the remote module name does not match,
'read', 'change' and 'do' does not work
- a proxy to an IO class has enablePoll == False, but it needs
a triggerPoll for modules relying on it to work
- a proxy on a communicator module has a status even when the
remote does not - this needs 2 fixes
Change-Id: Icd44da4c2984f27ce7147dec633739f9176012ec
Reviewed-on: https://forge.frm2.tum.de/review/c/secop/frappy/+/33168
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
