Module gonioanalysis.tkgui.plotting

class RecordingPlotter:
    '''For plotting a single image folder and its data.

    Uses tk_steroids.matplotlib.CanvasPlotter to plot on matplotlib figures.
    
    Attributes
    -----------
    core
        Reference to the Core instance
    selected_recording
        The recording currently plotted at the plotter
    i_repeat
        None if show all traces, otherwise the index of repeat to be shown
    N_repeats
        The total number of repeats in the image folder.

    '''

    def __init__(self, core):
        '''
        core    An instance of the Core class in core.py
        '''

        self.core = core
        
        # Keeps internally track of the current recording on plot
        self.selected_recording = None

        self.colorbar = None
        self.roi_rectangles = []
        
        self.N_repeats = 0
        self.i_repeat = None


    def _check_recording(self, skip_datafetch=False):
        '''
        Check from core if the selected has changed.
        '''
        selected_recording = self.core.selected_recording

        if self.selected_recording != selected_recording:
            self.i_repeat = None
        
        self.selected_recording = selected_recording

        if not skip_datafetch:
            if self.core.analyser.is_measured():
                self.movement_data = self.core.analyser.get_movements_from_folder(selected_recording)
                self.N_repeats = len(next(iter(self.movement_data.values())))
                pass
            else:
                self.movement_data = {}
                self.N_repeats = 0
        


    def magnitude(self, ax, **kwargs):
        '''
        Plot a displacement over time of the current specimen/recording.
        '''
        self._check_recording(skip_datafetch=True)
        
        ax, self.magnitudes, self.N_repeats = plot_1d_magnitude(self.core.analyser,
                self.selected_recording,
                i_repeat=self.i_repeat,
                label='EYE-repIREPEAT',
                ax=ax,
                **kwargs)


    def vectormap(self, ax, **kwargs):

        self.N_repeats = 0
        
        ax, self.vectors = plot_3d_vectormap(self.core.analyser,
                ax=ax,
                **kwargs)



    def xy(self, ax, **kwargs):
        '''
        Plot (x, y) where time is encoded by color.
        '''
        self._check_recording()
        
        ax, self.xys = plot_xy_trajectory([self.core.analyser],
                {self.core.analyser.name: [self.selected_recording]},
                i_repeat=self.i_repeat,
                ax=ax,
                **kwargs)

    
    def ROI(self, canvas_plotter):
        '''
        Plot specimen/recording image, and the ROIs and imaging parameters on top of it.
        '''
        self._check_recording(skip_datafetch=True)
        
        fig, ax = canvas_plotter.get_figax()
        
        # Plot the image
        image_fns = self.core.analyser.list_images(self.selected_recording, absolute_path=True)
        self.N_repeats = len(image_fns)

        if self.i_repeat:
            i_frame = self.i_repeat
        else:
            i_frame = 0
        
        image_fn = image_fns[i_frame]
        self.image = tifffile.TiffFile(image_fn).asarray(key=0)

        canvas_plotter.imshow(self.image, cmap='gray', slider=True)


        # Draw ROI rectangles
        for old_roi in self.roi_rectangles:
            try:
                old_roi.remove()
            except NotImplementedError:
                # This error occurs when something goes from from here on before
                # the ax.add_patch(roi) line. We can just ignore this as the
                # ROI has not been ever added to the ax
                continue
        self.roi_rectangles = []
        
        for roi in self.core.analyser.get_rois(self.selected_recording):
            patch = matplotlib.patches.Rectangle((roi[0], roi[1]), roi[2], roi[3],
                    fill=False, edgecolor='White')
            self.roi_rectangles.append(patch)
        
        for roi in self.roi_rectangles:
            ax.add_patch(roi)
        

        # Text about imaging parameters
        imaging_params = self.core.analyser.get_imaging_parameters(self.selected_recording)
        if imaging_params:
            text = '\n'.join(['{}: {}'.format(setting, value) for setting, value in imaging_params.items()])
        else:
            text = 'Unable to fetch imaging parameters'

        try:
            self.roi_text.set_text(text)
        except AttributeError:
            self.roi_text = ax.text(0,1, text, ha='left', va='top', fontsize='small', 
                    transform=ax.transAxes)