- a new wrapper for a read function is not only to be created when
the a new read function is in the class dict, but also when
it inherited, but not yet wrapped
- a proxy class must not call checkProperties
- removed secop/metaclass.py
- moved code from ModuleMeta to modules.HasAccessibles.__init_subclass__
- reworked properties:
assignment obj.property = value now always allowed
- reworked Parameters and Command to be true descriptors
- Command must now be solely used as decorator
- renamed 'usercommand' to 'Command'
- command methods no longer start with 'do_'
- reworked mechanism to determine accessible order:
the attribute paramOrder, if given, determines order of accessibles
+ fixed some issues makeing the IDE more happy
+ simplified code for StatusType and added a test for it
Change-Id: I8045cf38ee6f4d4862428272df0b12a7c8abaca7
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/25049
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>
- flatten hierarchy (some links do not work when using folders)
- add a tutorial for programming a simple driver
- clean description using inspect.cleandoc
+ fix a bug with 'unit' pseudo property in a Parameter used as override
Change-Id: I31ddba5d516d1ee5e785e28fbd79fca44ed23f5e
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/25000
Tested-by: Jenkins Automated Tests <pedersen+jenkins@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
New Syntax:
- define properties and parameters as class attributes directly
instead of items in class attribute dicts
- define commands with decorator @usercommand(...)
- old syntax is still supported for now
still to do (with decreasing priority):
- turn parameters into descriptors (vs. creating getters/setters)
- migrate all existing code to new syntax
- get rid of or reduce code in metaclasses using __set_name__ and
__init_subclass__ instead, including a fix for allowing py < 3.6
Change-Id: Id47e0f89c506f50c40fa518b01822c6e5bbf4e98
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/24991
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>
- when applying overrides with reorder=True, take ctr from Override,
else copy from the cloned Accesible. This did not work properly
- reworked:
- replaced CountedObj class by object_counter
- accessibles created by a copy or by applying Overrides
do not need fresh counted values
- adjusted tests
Change-Id: Id2fcf1ab1295aa1ea80ea81ae8cd02d36f86e969
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/24926
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>
- flatten hierarchy (some links do not work when using folders)
+ fix a bug with the redorder flag in Override
+ allow removal of parameters
+ clean description using inspect.cleandoc
Change-Id: I3dde4f4cb29c46e8a21014f1fad7aa3ad610a1bf
with handlers, a parameter from the cfg file which is not the
first of parameters with the same handler were not written.
fix: write_<param> method is called for all parameters in
<module>.writeDict even if there is no poll entry.
with this fix, when a parameter has the property initwrite=True,
the write_<param> method is called even when <param>
is not polled and even when <module>.pollerClass is None
Change-Id: I9b397deb5b20709fc4fa7c860c85b251a204c7f6
with handlers, a parameter from the cfg file which is not the
first of parameters with the same handler were not written.
fix: write_<param> method is called for all parameters in
<module>.writeDict even if there is no poll entry.
with this fix, when a parameter has the property initwrite=True,
the write_<param> method is called even when <param>
is not polled and even when <module>.pollerClass is None
Change-Id: I9b397deb5b20709fc4fa7c860c85b251a204c7f6
with handlers, a parameter from the cfg file which is not the
first of parameters with the same handler were not written.
fix: write_<param> method is called for all parameters in
<module>.writeDict even if there is no poll entry.
with this fix, when a parameter has the property initwrite=True,
the write_<param> method is called even when <param>
is not polled and even when <module>.pollerClass is None
Change-Id: I9b397deb5b20709fc4fa7c860c85b251a204c7f6
with handlers, a parameter from the cfg file which is not the
first of parameters with the same handler were not written.
fix: write_<param> method is called for all parameters in
<module>.writeDict even if there is no poll entry.
with this fix, the write_<method> is called even when a parameter
is not polled.
in the previous version, it was not possible to give a ASCII nul
character as end_of_line, because StringType refuses this
- end_of_line might be given as bytes, str or int
- end_of_line might be given as tuple (eol_read, eol_write)
Change-Id: I8b7942320ad3ffe162cdf3a673e113a66a84fb93
- change secop.client.SecopClient to use native types instead of
strings for its setParameter and execCommand methods.
- secop-gui: for now, setParameter accept strings for complex types.
this should be changed to use native type in an other change
- fix bugs in parser.py
+ SecopClient: make visible in an error message that the error
was generated on the SEC node
+ fix a bug when a command is called with 0 as argument
Change-Id: Id87d4678311ef8cf43a25153254d36127e16c6d9
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/23299
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>
motivation: a thread creating a lot of messages like a polling loop
with very short polling frequency or a fast polling connection might
monopolize the output of message over receiving new messages.
In addition, the current design has a latency of 0.3 sec for the
output of asynchronous replies.
Anyway, the output queue is just extending the network output buffer,
which is usally big enough.
- change the name of 'queue_async_reply' to 'send_reply'.
This method anyway was not only used for async replies.
- send_reply is directly sending the reply instead of putting into the
queue. It will slow down the calling thread, if the output buffer
is full, which is desired behaviour.
Change-Id: I305669be2f7c027355b43421432f32be9c166ed4
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/23119
Tested-by: JenkinsCodeReview <bjoern_pedersen@frm2.tum.de>
Reviewed-by: Enrico Faulhaber <enrico.faulhaber@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
- send a heartbeat, if no events for 5 sec.
an interrupted connection (not closed by the other end) may
not be detected for a long time when nothing is sent
+ make the error reply on a non SECoPEror more verbose
e.g. "KeyError('foo')" instead of just "foo"
+ allow cfg file without nodeinterface
+ shorter logger name in HasIodev
Change-Id: I6b1ff23f9bf8c96feb25af44935596437b7d726f
Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/23098
Tested-by: JenkinsCodeReview <bjoern_pedersen@frm2.tum.de>
Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
