Sorry, the replay-all is not my default. In that case it is easier for me to monkey patch just the mainloop to invoke Gtk.main everytime.
But again, even if we are talking about interactive (ipython), why is it unbalanced? I mean calls Gtk.main vs Gtk.main_quit? Federico On Fri, Oct 11, 2013 at 3:28 PM, Thomas A Caswell <tcasw...@uchicago.edu> wrote: > Please keep all emails in-band > > I was commenting that the issue you are having with getting easy-to-use > pre-built figures in a non-interactive program without dragging pyplot in is > the same as what I think the root of 2503 is and the re-factor I proposed to > make your life easier would also help that case (I think). > > The gui backends were (as I understand it, someone please correct me if I am > wrong) built to play nice with ipython to put together a MATLAB-like > interface. The fact that you can then embed the gui-framework dependent > bits of matplotlib in other gui applications is a nice side-effect, but not > one of the initial design goals. This is why the `figure_manager` code is > (too-)tightly coupled to _pylab_helpers. > > See the examples at http://matplotlib.org/examples/user_interfaces/ making > a window with a canvas + tool bar is pretty easy. > > A quick and dirty solution might be to just monkey patch the > figure_manager.destroy function when your app starts up to remove the check > that shuts down the main loop. > > Tom > > > On Fri, Oct 11, 2013 at 1:40 PM, Federico Ariza <ariza.feder...@gmail.com> > wrote: >> >> Sorry I don't get it. >> Are you suggesting to explain this little predicament in the PR to >> give another point of view? or >> You want me to check the PR and try to use the solutions proposed there? >> >> Federico >> >> On Fri, Oct 11, 2013 at 1:32 PM, Thomas A Caswell <tcasw...@uchicago.edu> >> wrote: >> > embedding vs launching is a distinction without a difference, you are >> > integrating matplotlib with your own gui application. >> > >> > That said, it would be nice to re-factor the figure_manager classes so >> > they >> > they make no reference to `Gcf` or anything associated with pylab and >> > could >> > be easily re-used. >> > >> > I think that would also help with the issues in >> > https://github.com/matplotlib/matplotlib/pull/2503 >> > >> > Tom >> > >> > >> > On Fri, Oct 11, 2013 at 12:14 PM, Federico Ariza >> > <ariza.feder...@gmail.com> >> > wrote: >> >> >> >> Again >> >> >> >> In the example the plotting is inside the callback (just for >> >> simplicity), but in reality, the plotting is in another class, >> >> somewhere else that can be called standalone to produce the plots. >> >> >> >> Federico >> >> >> >> On Fri, Oct 11, 2013 at 1:12 PM, Federico Ariza >> >> <ariza.feder...@gmail.com> wrote: >> >> > I am not embedding, just launching, as the example shows. >> >> > >> >> > Federico >> >> > >> >> > On Fri, Oct 11, 2013 at 1:11 PM, Thomas A Caswell >> >> > <tcasw...@uchicago.edu> wrote: >> >> >> If you are embedding matplotlib, do not import `pyplot`. `pyplot` >> >> >> sets >> >> >> up a >> >> >> bunch of gui-magic (tm) in the background (as you found in >> >> >> `figure_manager`). >> >> >> >> >> >> Tom >> >> >> >> >> >> >> >> >> On Fri, Oct 11, 2013 at 11:57 AM, Federico Ariza >> >> >> <ariza.feder...@gmail.com> >> >> >> wrote: >> >> >>> >> >> >>> Hello everybody >> >> >>> >> >> >>> Working on one GTK3 app, that calls matplotlib to plot some >> >> >>> figures, I >> >> >>> found that closing all the figures from matplotlib kills my app >> >> >>> also. >> >> >>> The problem.... >> >> >>> >> >> >>> Gtk.main() is called only if there is no previous invocation, in my >> >> >>> case, my Gtk3 app invokes main, so the mainloop won't call it >> >> >>> again. >> >> >>> >> >> >>> #in backend_gtk3.py >> >> >>> # >> >> >>> class Show(ShowBase): >> >> >>> def mainloop(self): >> >> >>> if Gtk.main_level() == 0: >> >> >>> Gtk.main() >> >> >>> >> >> >>> But in the "destroy" method of the figure manager calls >> >> >>> Gtk.main_quit >> >> >>> everytime that there are no more figures >> >> >>> >> >> >>> #in backend_gtk3.py inside destroy method of FigureManagerGTK3 >> >> >>> # >> >> >>> if Gcf.get_num_fig_managers()==0 and \ >> >> >>> not matplotlib.is_interactive() and \ >> >> >>> Gtk.main_level() >= 1: >> >> >>> Gtk.main_quit() >> >> >>> >> >> >>> >> >> >>> So basically we are not calling Gtk.main but we are Gtk.calling >> >> >>> main_quit. >> >> >>> Isn't it more natural to call Gtk.main the same amount of times >> >> >>> that >> >> >>> we are going to call Gtk.main_quit? >> >> >>> >> >> >>> Adding matplotlib.rcParams['interactive'] = True doesn't help >> >> >>> >> >> >>> Here is my little testing code >> >> >>> >> >> >>> ############################## >> >> >>> #file myapp.py >> >> >>> >> >> >>> import matplotlib >> >> >>> matplotlib.rcParams['interactive'] = True >> >> >>> matplotlib.use('GTK3AGG') >> >> >>> import matplotlib.pyplot as plt >> >> >>> >> >> >>> from gi.repository import Gtk >> >> >>> >> >> >>> class MyWindow(Gtk.Window): >> >> >>> >> >> >>> def __init__(self): >> >> >>> Gtk.Window.__init__(self, title="Hello World") >> >> >>> >> >> >>> self.button = Gtk.Button(label="Click Here") >> >> >>> self.button.connect("clicked", self.on_button_clicked) >> >> >>> self.add(self.button) >> >> >>> >> >> >>> def on_button_clicked(self, widget): >> >> >>> fig = plt.figure() >> >> >>> ax = fig.add_subplot(111) >> >> >>> ax.plot([1,2,3]) >> >> >>> plt.show() >> >> >>> >> >> >>> win = MyWindow() >> >> >>> win.connect("delete-event", Gtk.main_quit) >> >> >>> win.show_all() >> >> >>> Gtk.main() >> >> >>> ######################### >> >> >>> >> >> >>> I know this is related to interactive mode, but running from >> >> >>> console >> >> >>> >>> python myapp.py >> >> >>> reproduces the problem >> >> >>> >> >> >>> Why hasattr(sys, 'ps1') is False? if I am running it from console? >> >> >>> how >> >> >>> do I change this? >> >> >>> >> >> >>> >> >> >>> Thanks >> >> >>> Federico >> >> >>> >> >> >>> P.S. Does anybody had the time to check my PR for >> >> >>> multi-figure-manager? >> >> >>> https://github.com/matplotlib/matplotlib/pull/2465 >> >> >>> >> >> >>> -- >> >> >>> Y yo que culpa tengo de que ellas se crean todo lo que yo les digo? >> >> >>> >> >> >>> -- Antonio Alducin -- >> >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> ------------------------------------------------------------------------------ >> >> >>> October Webinars: Code for Performance >> >> >>> Free Intel webinars can help you accelerate application >> >> >>> performance. >> >> >>> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the >> >> >>> most >> >> >>> from >> >> >>> the latest Intel processors and coprocessors. See abstracts and >> >> >>> register > >> >> >>> >> >> >>> >> >> >>> >> >> >>> http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk >> >> >>> _______________________________________________ >> >> >>> Matplotlib-devel mailing list >> >> >>> Matplotlib-devel@lists.sourceforge.net >> >> >>> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> -- >> >> >> Thomas A Caswell >> >> >> PhD Candidate University of Chicago >> >> >> Nagel and Gardel labs >> >> >> tcasw...@uchicago.edu >> >> >> jfi.uchicago.edu/~tcaswell >> >> >> o: 773.702.7204 >> >> > >> >> > >> >> > >> >> > -- >> >> > Y yo que culpa tengo de que ellas se crean todo lo que yo les digo? >> >> > >> >> > -- Antonio Alducin -- >> >> >> >> >> >> >> >> -- >> >> Y yo que culpa tengo de que ellas se crean todo lo que yo les digo? >> >> >> >> -- Antonio Alducin -- >> > >> > >> > >> > >> > -- >> > Thomas A Caswell >> > PhD Candidate University of Chicago >> > Nagel and Gardel labs >> > tcasw...@uchicago.edu >> > jfi.uchicago.edu/~tcaswell >> > o: 773.702.7204 >> >> >> >> -- >> Y yo que culpa tengo de que ellas se crean todo lo que yo les digo? >> >> -- Antonio Alducin -- > > > > > -- > Thomas A Caswell > PhD Candidate University of Chicago > Nagel and Gardel labs > tcasw...@uchicago.edu > jfi.uchicago.edu/~tcaswell > o: 773.702.7204 -- Y yo que culpa tengo de que ellas se crean todo lo que yo les digo? -- Antonio Alducin -- ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk _______________________________________________ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
