I have a program that has 2 very simple threads:
I also have a matplotlib animation running in my main(). It has a scope class that is the example from matplotlib.
When the program starts to run, it shows the plot and everything is OK. The problem is that as soon as the user enters a key, the program crashes and python exits with a fatal error.
The ui thread has nothing to do with matplotlib and scope class. If I delete the code that creates the plots, the ui thread has no problem and program runs smoothly. I also notice matplotlib on my system uses tkinter for creating windows.
Do you have any hints or experiences with why the matplotlib animation causes problem? Can't a thread be used with a matplotlib plot?
I am running this in a command line window in Windows7 with Python 2.7.
matplotlib version : 2.0.2
Tkinter version : 8.5
Error:
Fatal Python error: GC object already tracked
This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
Or this error:
TclStackFree: incorrect freePtr. Call out of sequence?
This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
Code:
import threading
import serial
from matplotlib.lines import Line2D
import matplotlib.pyplot as plt
import matplotlib.animation as animation
class listner(threading.Thread):
def __init__(self,port):
threading.Thread.__init__(self)
self.sport=None
self.is_running=True
self.init_port(port)
def run(self):
print ' Reading from Port'
while self.is_running:
try:
self.sport.read(1)
except:
print 'Error reading port'
def init_port(self,port):
print '1'
if self.sport==None or not self.sport.is_open :
try:
self.sport = serial.Serial(port,115200)
self.sport.timeout = 1
self.sport.reset_input_buffer()
self.sport.reset_output_buffer()
self.port_open=True
except:
print " Port error Listener Initing\n",self.port_open,'\n',self.sport
else:
pass
def process(self):
pass
class ui(threading.Thread):
def __init__(self):
threading.Thread.__init__(self)
self.running = True
def run(self):
print 'Starting UI:\n'
while self.running:
print ' Enter input ''S'':\n'
user = raw_input()
def main(port):
listner_thread = None
try:
listner_thread = listner(port)
listner_thread.start();
except:
print "Listener Thread Failed To Start"
return
ui_thread=None
try:
ui_thread = ui()
ui_thread.start()
except:
print "UI Thread Failed To Start"
return
run_charts()
def run_charts():
fig, (ax1, ax2) = plt.subplots(2, 1)
scope1 = Scope(ax1)
ani1 = animation.FuncAnimation(fig, scope1.update, emit_ch1, interval=10,blit=True)
scope2 = Scope(ax2)
ani2 = animation.FuncAnimation(fig, scope2.update, emit_ch2, interval=10,blit=True)
plt.show()
def emit_ch1():
yield 0.001
def emit_ch2():
yield -0.001
class Scope(object):
def __init__(self, ax, maxt=2, dt=0.02):
self.ax = ax
self.dt = dt
self.maxt = maxt
self.tdata = [0]
self.ydata = [0]
self.line = Line2D(self.tdata, self.ydata)
self.ax.add_line(self.line)
self.ax.set_ylim(-.009, 0.009)
self.ax.set_xlim(0, self.maxt)
def update(self, y):
t = self.tdata[-1] + self.dt
self.tdata.append(t)
self.ydata.append(y)
self.line.set_data(self.tdata, self.ydata)
return self.line,
if __name__ == '__main__':
main('COM11')
The first error, "Fatal Python Error:GC object Already Tracked" was closed in 2013 with a status of "CLOSED WONTFIX" See the bug report on Bugzilla
It seems the interim fix when it was raised again in 2015 [in relation to dask] was to use a single thread using this code:
import dask
dask.set_options(get=dask.async.get_sync)
but that the problem actually lay with dataframe.read_csv problems.
The problem was eventually solved in a later version of pandas. If you upgrade your version of matplotlib, the problem will most likely be resolved also with a similar fix.
Hope this helps
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With