Re: [Matplotlib-users] timer objects in macosx backend

2013-07-17 Thread Michiel de Hoon
Hi Justin, The .stop() method was indeed never implemented for Timer objects in the MacOSX backend. I am not sure if a .stop() method is really needed, because deleting the timer has the same effect as stopping the timer. Is there some reason you prefer >>> t.stop() instead of >>> del t ? Best,

Re: [Matplotlib-users] Install problems, OSX 10.6.8, Python 3.3.1

2013-07-17 Thread Scott Lasley
I am able to import matplotlib.pyplt under python 3.3.2 from www.python.org in OS X 10.6.8. matplotlib 1.2.1, libfreetype and libpng were built from source using gcc 4.2.1 from Xcode 3.2.6. Are you using python from macports or from www.python.org? You might try using python from the installer

Re: [Matplotlib-users] numpy masked array plot - isolated single unmasked value is hidden

2013-07-17 Thread Michael Droettboom
You could use a single pixel for a marker (','), I guess. But as you say, you need at least two points for a line segment. Mike On 07/17/2013 10:45 AM, Gregorio Bastardo wrote: > Hi, > > The following example demonstrates the problem, value 5 could not be > seen w/o marker: > > data = np.arange

[Matplotlib-users] timer objects in macosx backend

2013-07-17 Thread Justin Lazear
Hi all, I'm using a timer object to interact with the MPL event loop on my OS X laptop. However, it seems to be missing a few key methods that are making using it a little difficult. In particular, I can't find a way to stop the timer from sending events: $ ipython --pylab In [1]

[Matplotlib-users] numpy masked array plot - isolated single unmasked value is hidden

2013-07-17 Thread Gregorio Bastardo
Hi, The following example demonstrates the problem, value 5 could not be seen w/o marker: data = np.arange(10) mask = [0,0,0,1,1,0,1,0,0,0] x = np.ma.masked_array(data, mask) plot(x) plot(x, '+') In my datasets, isolated unmasked values are rare, but placing a marker to spot them makes the whole

Re: [Matplotlib-users] cbook.report_memory IOError

2013-07-17 Thread Benjamin Root
I have a vague recollection of a similar problem faced by @dopplershift when he was updating the animation module to pipe a stream to a mencoder/ffmpeg process on certain Macs. Maybe this is the same problem? On Wed, Jul 17, 2013 at 9:18 AM, Michael Droettboom wrote: > To debug, it might be he

Re: [Matplotlib-users] cbook.report_memory IOError

2013-07-17 Thread Michael Droettboom
To debug, it might be helpful to try ps -p PID -o rss,vsz (where PID is the process id of an interesting process) and see what happens. Mike On 07/17/2013 02:05 AM, Eric Firing wrote: > On 2013/07/16 5:50 PM, K.-Michael Aye wrote: >> Hi! >> >> I have just run an old code that I believe was wor

Re: [Matplotlib-users] Install problems, OSX 10.6.8, Python 3.3.1

2013-07-17 Thread Michael Droettboom
Do any Mac experts have any ideas? One thing that may help us is to fire this up in gdb and get a traceback. 1) Run "gdb python" (or "gdb python3"). 2) At the gdb prompt, type "run" 3) At the Python prompt, type "import matplotlib.pyplot" 4) Python should crash, then type "bt" to get a backtrace

Re: [Matplotlib-users] [Patch] Patch for fontmanager crash

2013-07-17 Thread Michael Droettboom
This patch doesn't make a whole lot of sense to me. get_name should work just fine if self._family is None, and indeed it does in my own testing: ``` from matplotlib import font_manager f = font_manager.FontProperties(None) print f._family print f.get_family() print f.get_name() ``` So I'd mu

Re: [Matplotlib-users] Graph ticks label missing !

2013-07-17 Thread Michael Droettboom
Can you please provide a completely standalone example? The following code has undefined variables etc. Mike On 07/08/2013 10:49 AM, Nicolas Mailhot wrote: >subplot.set_xticks([dd.monthstart(m) for m in months]) >subplot.set_xticks([w.gmticks() for w in weeks], minor=True) >subplot.

Re: [Matplotlib-users] Smooth animations

2013-07-17 Thread Michael Droettboom
There is nothing available to push points -- matplotlib uses Numpy arrays internally for the data, and they can not be (efficiently) resized. I would try implementing this before assuming it's too slow. Mike On 07/05/2013 12:02 PM, v0idnull wrote: Yes, but this is where I am failing. I don't