matplotlib-users Mailing List for matplotlib

On Fri, Nov 13, 2009 at 10:39 AM, Gregor Thalhammer
<Gre...@i-...> wrote:
> 1) I would like to stay independent of the backend, therefore I would
> prefer a approach like in idle_and_timeout
> http://matplotlib.sourceforge.net/examples/event_handling/idle_and_timeout.html
> However, with my default backend WXAgg I do not continuously idle
> events, but only once e.g. after each mouse movement. Can someone give
> me more information about the intended behaviour, current status and
> support of idle events?
I worked on this for a while and thought I had wx working but maybe I
missed something or maybe there is a wx version issue. The real
stumbling block was tk since it did not have a native idle and I was
stuck trying to make it work with threading. In and of itself, this
was OK, but it broke CTRL-C in ipython due to problems with cross
thread signal handling. So I put the project on the back burner.
Perhaps we can borrow some knowledge from the ipython developers who
have recently made good progress here in various GUIs backends.
>
> 2) I want to limit the lifetime of my connection object, that takes care
> of the communication with the microscope stage, so that the connection
> is closed if I close the matplotlib figure. I could not find a
> 'close_event'. To achieve this I could embed a matplotlib figure into a
> wx frame, but this is somewhat tediuos. For the wxagg backend, does
> somebody know a more simple solution? Do others also feel the need to
> add a close_event? In case, I could try to take care of the wx backend.
Adding a close event at the mpl level should be fairly
straight-forward, and shouldn't require much if any additional GUI
programming since we already handle the window destroy event across
backends for pylab figure management.
JDH

Taro Sato-4 wrote:
> 
> Hello. I'm trying to embed mpl in an wxpython app which is intended
> to run repeatedly within a single script. (It's a workaround for not
> being able to use pyplot.show() more than once in a script.)
> 
> All seems to work as intended when I don't attach any key press events
> (here to just destroy the app) to canvas, but when I do, as in the
> script below, for some reason I keep getting errors that appear to
> indicate that some remnant of the zombie of the previous app instance
> may be sticking around when they shouldn't exist. Both the script in
> question and the (typical) error message are shown below in this
> message.
> 
> 
I think I figured out what causes wx._core.PyDeadObjectError above.
In backend_bases.py, LocationEvent class has the attribute lastevent
(initially set to None) which keeps the Event instance of a previous event. 
LocationEvent._update_enter_leave() method processes axes_leave_event for
lastevent if not None, and then processes axes_enter_event for the current
event.
When an application window first gets created, lastevent is set to None. 
When the application is destroyed and a new instance of the application is
created, lastevent needs to be reset to None, which apparently is not
happening. The reference to the previous event which occurred within the
destroyed application is being accessed via lastevent, hence the
PyDeadObjectError.
I'm not sure what the best timing is to rest lastevent (I'm not thoroughly
familiar with mpl source codes). But modifying
LocationEvent._update_enter_leave() such that the PyDeadObjectError
exception raised by accessing lastevent solves the specific issue in the
original post. 
Should I file a bug report on this?
Cheers,
Taro
-- 
View this message in context: http://old.nabble.com/Strange-event-errors-with-WXAgg-%2B-wx-tp26312977p26355887.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On Nov 14, 2009, at 17:40 , Jae-Joon Lee wrote:
What backend are you using?
ah! it seems to work with the tkagg backend, but not the wxagg backend.
I am using ipython with the -pylab flag.
I remember having some problem with the tk backend (perhaps when I 
was doing some wx embedding), but I can't recall what it was. is 
there a reason the wx backend doesn't work as I expect?
on a related note, is there a way in ipython to specify the backend 
for the -pylab flag, overriding the one specified in matplotlibrc?
		thanks,
			bb
There was a similar report that the "draw" method of the qt4agg
backend does not update the screen immediately but wait until it gets
idle.
See if using other backends makes any change.
import numpy
import matplotlib.pyplot as plt
h,=plt.plot([1,2,3])
plt.ylabel('some numbers')
plt.show()
for i in range(100):
 h.set_ydata(numpy.random.rand(3)*2+1)
 plt.draw()
-- 
Brian Blais
bb...@br...
http://web.bryant.edu/~bblais