Closed cpascual closed 3 years ago
For tips on how to prepare an environment for installation, see http://taurus-scada.org/users/getting_started.html
pip install <tarball_artifact_URL>
pip install https://ci.appveyor.com/api/buildjobs/d5hd5yhe1wtmhecy/artifacts/dist%2Ftaurus-4.7.0a0.tar.gz
pip install taurus_pyqtgraph
taurus --version
python -c "import taurus_pyqtgraph; print(taurus_pyqtgraph.__version__)"
- there is some problem with the polling/tango events. when a client connects to a tango attribute I do not see I start getting "Timeout Error... activating polling" messages every 10s:
(py3qt5) C:\Users\IEUser>taurus form sys/tg_test/1/double_scalar
Dummy-1 INFO 2020-08-03 11:22:01,274 192.168.1.16:10000.sys/tg_test/1.double_scalar: Activating polling. Reason: API_EventTimeout
Dummy-1 INFO 2020-08-03 11:22:11,321 192.168.1.16:10000.sys/tg_test
(...) And so on... repeated every 10s until I close the client
Update: The event issue is now fixed: the problem was with a mismatch in the definition of the TANGO_HOST between the windows machine (in which I had used an IP-based TANGO_HOST
) and the machine running the TangoTest (which used name within alba's domain).
TIP: in order to "fool" the windows machine into properly resolving the TANGO_HOST name (pt168.cells.es:1000
) (while being run on a virtualbox in a laptop outside ALBA's network), I edited the hosts file in the windows machine
Independently of the above, we consistently see the already known problem of Tango crashing when disconnecting. Note that this may be already fixed in cpptango >=9.3.4, but it is not yet available in conda for windows
taurus demo
eval:["FAULT","ON","OFF","ALARM"][randint(4)]
sys/tg_test/1/ampli#rvalue.magnitude
, eval:Q('1mm')#rvalue.units
, eval:10*arange(9)#rvalue[3:4]
)eval:False
as a model for testing)taurus image --demo
taurus --polling-period 333 trend2d --demo
taurus --polling-period 333 trend2d -xt --demo
(known to fail in 4.1.0)taurus --polling-period 333 trend2d -xn --demo
taurus --polling-period 333 trend2d -xe --demo
taurus --polling-period 333 trend2d --demo -b 10
(deactivate auto-scale bottom axis and see that the plot is limited to the
last 10 values )taurus designer
. Check that the taurus widgets are present in the catalogtaurus device sys/tg_test/1
taurus panel
(basically try all features described in the user's guide
taurus form sys/tg_test/1/short_scalar
mm
. Close the form and relaunch. The new units should be used. Change the the write value and check that the orange color is used when in warning values, and that the write widget does not allow to write values out of range.>>{}<<
). Do this in compact and non compact modes.taurus form --config tf.pck
(basically try all features described in the user's guide
taurus gui example01
foogui
) with taurus newgui
(follow the wizard)
taurus newgui
) and selecting the same directory
AttributeError("'NoneType' object has no attribute 'fromXml'")
(#1137)foogui
with pip (using a virtualenv may be a good idea)foogui
using the script that has been installedtaurus icons
. Several tabs with an array of icons should be displayedtaurus plot --ls-alt
(check that it lists "qwt5" and "tpg")taurus plot "eval:Q(rand(333),'mm')" sys/tg_test/1/wave
--use-alt=tpg
to select the tpg implementation)taurus trend --ls-alt
(check that it lists "qwt5" and "tpg")taurus trend "eval:Q(rand(),'mm')" sys/tg_test/1/ampli
(if using py2 and Qt4, you may need to use --use-alt=tpg
to select the tpg implementation)taurus trend -xn "eval:Q(rand(),'mm')" sys/tg_test/1/ampli
(if using py2 and Qt4, you may need to use --use-alt=tpg
to select the tpg implementation)
Done. A few non-RC bugs were found (reported in the issue tracker). The most serious of them is #709 (crashes of tango during exit). While this is certainly annoying , it does not seem to affect the GUIs while running.
This issue is created for reporting the results of manual tests for Jul20 release on debian10 with dependencies installed via conda.
Using a fresh Windows10 machine provisioned with vagrant, using the following Vagrantfile
... and then creating the environment with conda (using the anaconda prompt):