About tgext.matplotrender

tgext.matplotrender is a TurboGears2 extension that provides a Rendering Engine for MatplotLib Figures.

Installing

tgext.matplotrender can be installed from pypi:

pip install tgext.matplotrender

should just work for most of the users.

Enabling

To enable tgext.matplotrender add the matplotfig renderer to your configuration:

base_config.renderers.append('matplotfig')

and plug the rendering engine inside your config/app_cfg.py:

import tgext.matplotrender
tgext.matplotrender.plugme(base_config)

Usage

Using tgext.matplotrender is as simple as exposing an action with the matplotfig template and returning the figure itself inside the fig key of action returned dictionary.

Given the following figure:

def _make_fig():
    fig = matplotlib.figure.Figure(figsize=(9, 6))
    fig.Name = "Sinewave"
    ax = fig.add_subplot(111)
    ax.set_xlabel("angle")
    ax.set_ylabel("amplitude")
    t = numpy.arange(0.0, 2.0, 0.01)
    s1 = numpy.sin(2 * numpy.pi * t)
    ax.plot(t, s1, color="k")
    return fig

It can be exposed through TurboGears actions as:

class RootController(TGController):
    @expose('matplotfig')
    def figure(self, *args, **kwargs):
        return dict(fig=_make_fig())

    @expose('matplotfig', render_params=dict(dpi=36))
    def lowres(self, *args, **kwargs):
        return dict(fig=_make_fig())

    @expose('matplotfig')
    def customres(self, *args, **kwargs):
        options = {}
        try:
            options['dpi'] = int(kwargs['dpi'])
        except:
            pass
        return dict(fig=_make_fig(), **options)

Any other value provided in the dictionary will be used as a print_figure argument in matplotlib