Implement a plot method for PlotData Objects

[GhislainJ]

• I'm submitting a ...

  • feature request

• What is the current behavior?

One need to call the plot_canvas method defined in the core module of plot_data.

• What is the expected behavior?

plot should be callable from objects directly

• What is the motivation / use case for changing the behavior?

The current implementation is problematic because one need to import plot_data in order to plot locally. By implementing a method for plot_data objects, we could only call object_.plot(), as we know for a fact that we are handling plot data objects and get rid of this import, namely in dessia_common for example. Consider the following dessia common implementation. This change would enable us to get rid of that lousy plot_data import

    def plot(self, selector: Optional[DISPLAY_TYPES] = None, **kwargs):
        """
        Generic plot getting object plot_data views in order to plot locally.

        First, get all settings defined by a 'plot_data_view' decorator.
        If a selector is set, it only plots the desired one.
        If selector is None, plots every plot candidate
        """
        plot_data_settings = self._display_settings_from_type("plot_data", **kwargs)
        if selector is None:
            display_objects = [self._display_from_selector(selector=d.selector, serialize_data=False)
                               for d in plot_data_settings]
            msg = f"Class '{self.full_classname}' does not define any display of type 'plot_data'."
        else:
            display_objects = [self._display_from_selector(selector=d.selector, serialize_data=False)
                               for d in plot_data_settings if d.selector == selector]
            msg = (f"No selector '{selector}' found in class '{self.full_classname}'"
                   f"definition for displays of type 'plot_data'.")
        if not display_objects:
            raise ValueError(msg)

        import plot_data
        for display_object in display_objects:
            plot_data_object = display_object.data
            try:
                # Ideally
                plot_data_object.plot()
            except AttributeError:
                # Supporting plot_data < 0.22.0
                # Removing this and using statement above would remove plot_data requirement
                plot_data.plot_canvas(plot_data_object=plot_data_object, canvas_id='canvas',
                                      width=1400, height=900, debug_mode=False)

• Possible fixes

• Please tell us about your environment:

  • branch:
  • commit:
  • python version:

[GhislainJ]

Actually realized that this is already done in current master and dev version.