If you want to use technique with bar, hist, or scatter
> (which return Patch instances), you'll have to wait a few
> minutes while I update the code, because I haven't added a
> set data method for those classes.
I added some features to CVS to make this work with patches and do
allow the updates to occur flicker free. Below is an simulated
example that uses a bar chart to show network, CPU and memory usage,
updating 4 times per second. If you have trouble with CVS or want a
windows installer of the updated code, let me know
John Hunter
[ I'm reposting this because it didn't seem to get through yesterday.
If this is a duplicate sorry ]
import pygtk
pygtk.require('2.0')
import gtk
import time
from matplotlib.matlab import *
def get_memory():
"Simulate a function that returns system memory"
return 100*(1+sin(0.5*pi*time.time()))
def get_cpu():
"Simulate a function that returns cpu usage"
return 100*(1+sin(0.2*pi*(time.time()-0.25)))
def get_net():
"Simulate a function that returns network bandwidth"
return 100*(1+sin(0.7*pi*(time.time()-0.1)))
def get_stats():
return get_memory(), get_cpu(), get_net()
fig = figure(1)
ax = subplot(111)
ind = arange(1,4)
pm, pc, pn = bar(ind, get_stats())
centers = ind + 0.5*pm.get_width()
pm.set_facecolor('r')
pc.set_facecolor('g')
pn.set_facecolor('b')
ax.set_xlim([0.5,4])
ax.set_xticks(centers)
ax.set_ylim([0,100])
ax.set_xticklabels(['Memory', 'CPU', 'Bandwidth'])
ax.set_ylabel('Percent usage')
ax.set_title('System Monitor')
def updatefig(*args):
m,c,n = get_stats()
pm.set_height(m)
pc.set_height(c)
pn.set_height(n)
ax.set_ylim([0,100])
fig.draw()
return gtk.TRUE
gtk.timeout_add(250, updatefig)
show()
John Hunter