napari_utility_functions

### OPEN EVERYTHING IN NAPARI's ipython.

def view_img(url_image):
    """
	This function takes as input the ome.zarr path.
	ex.  view_img("D:/hybiss_experiment/section.ome.zarr")
	will display directly in the napari's image window
	"""
    viewer.open(url_image, plugin="napari-ome-zarr")
	
	
def load_spots(url_spots):
    """
	This function takes as input the decoded spots path:
	ex.  df = load_spots("D:/hybiss_experiment/decoded/decoded_spots.csv")
	It returns a pandas DataFrame, so assign to variable (df).
	"""
    import pandas as pd
	df = pd.read_csv(url_spots)
	df = df.dropna()
	return df
	
	
def plot_ts(df, t_name, size, color):
    """
	This function takes the df created with the function load_spots,
	t_name is the gene name as written in the HybIss codebook (df.target)
	size is the desired size of the points
	color is the desired color of the points
	It plots directly to the napari's image window
	"""
    t_coords = df[df.target == t_name][["y", "x"]]
	viewer.add_points(t_coords, size=size, face_color=color, edge_color="white")
	
	
def export_pickled(outname):
    """
	This function takes as input the name of a file in which to put
	the pickled dictionary {"anatomical_acronym": coordinates}
	It saves in a .pkl file in the working directory.
	"""
    import pickle as pkl
	import os
    # Delete any layer but the ANNOTATIONS.
    data_dict = {}
	for i in viewer.layers:
	    data_dict[i.name] = i.data
    # for windows
	with open(f"{outname}.pkl", "wb") as f:
	    pkl.dump(data_dict, f)


def read_pickled(filename):
    """
	This function takes as input a .pkl file and 
	returns the dictionary {"anatomical_acronym": coordinates}
	"""
    import pickle as pkl
	with open(filename, "rb") as f:
	    polygons = pkl.loads(f.read())
	return polygons



def load_shapes(filename):
    """
	This function takes as input the .pkl file,
	reads it and creates the polygon layers in napari's image window.
	"""
    polygons = read_pickled(filename)
	for i in polygons:
	    viewer.add_shapes(
		    polygons[i],
			shape_type="polygon",
			name=f"{i}")