Accessing Data Frame From Geom about ggpy HOT 5 CLOSED

class geom_boxplot(geom):
    VALID_AES = ['y', 'lower','middle','upper','x','ymax','ymin','alpha',
                 'colour','color','fill','linetype','shape','size','weight']


    def plot_layer(self, layer):
        layer = {k: v for k, v in layer.items() if k in self.VALID_AES}
        layer.update(self.manual_aes)

        # Option 1 (No Pandas)

        # Boxplot takes in an array or sequence of vectors
        # The goal is to group the 'y' values by grouped 'x'
        # Then take the transpose and give it to plt.boxplot(**layer)
        # If we do not have a y value, no need to change anything
        #        if "y" in layer:
        # list(set([...])) just gets unique elements
        #           unique = list(set(layer["x"]))            
        # Group y values in len(x) arrays, having values y.x
        # Store it back in x as a transpose.            
        #          pass

        # plt.boxplot(**layer)
        # Option 2 With Pandas and DataFrame.boxplot()
        # Need access to the DataFrame passed to ggplot(..)
        df = DataFrame({'y': layer['y']})
        if 'x' in layer:
            layer['by'] = 'x'
            df['x'] = layer['x']
            del layer['x']
        if 'y' in layer:
            layer['column'] = 'y'
            del layer['y']
        df.boxplot(**layer)