Module gonioanalysis.image_tools

class ImageShower:
    '''
    From  cascade training, a generalized image shower
    '''
    
    def __init__(self, fig, ax):
        
        self.fig = fig
        self.ax = ax
        
        self.fns = None

        self.buttonActions = []
        self.image_brightness = 0
        self.image_maxval = 1
        self.cid = self.fig.canvas.mpl_connect('key_press_event', self.callbackButtonPressed)
        
        self.pupils = -1
        
        self.title = ''

        self.fig.show()


    def callbackButtonPressed(self, event):
        '''
        A callback function connecting to matplotlib's event manager.
        '''
        
        # Navigating between the images
        if event.key == 'z':
            self.image_maxval -= 0.3
            self.updateImage(strong=True)
        
        elif event.key == 'x':
            self.image_maxval += 0.3
            self.updateImage(strong=True)
        
        elif event.key == 'a':
            self.image_brightness += 50
            self.updateImage(strong=True)
        elif event.key == 'c':
            self.image_brightness += -50
            self.updateImage(strong=True)
        
        for button, action in self.buttonActions:
            if event.key == button:
                print(event.key)
                action()


    def setImages(self, image_fns):
        '''
        Set the images that ImageShower shows.
        '''
        self.fns = image_fns
        self.cache = {}

    def setImage(self, i):
        fn = self.fns[i]
        try:
            self.image = self.cache[fn]
        except KeyError:
            self.image = tifffile.imread(fn) 
        
        self.updateImage()
        
    def cacheImage(self, i):
        '''
        Loads image to cache for faster showup. 
        '''
        fn = self.fns[i]
        # Uncomment to take cahcing in use
        #self.cache[fn] = tifffile.imread(fn)

    def setTitle(self, title):
        self.title = title

    def updateImage(self, strong=False):
        capvals = (0, np.mean(self.image) *self.image_maxval)
        self.ax.clear() 
        self.ax.set_title(self.title)
        self.ax.imshow(self.image-self.image_brightness,cmap='gist_gray', interpolation='nearest', vmin=capvals[0], vmax=capvals[1])
        
        if not strong:
            self.fig.canvas.draw()
        else:
            self.fig.show()
    
    def close(self):
        plt.close(self.fig)