case_study.py

from pickle import Pickler, Unpickler
from typing import Optional

from cells.config import CFG
from cells.processing.cell import CellsStore, DistMatrix, ImageSize
from cells.processing.border_distance import calculate_mutual_distances
from cells.processing.img_processing import read_cells


class CaseStudy:
    def __init__(self, imageFileName: str, imageSize: ImageSize):
        self.imageFile = imageFileName
        self.imageSize = imageSize
        p_from = imageFileName.rfind('/') + 1
        p_end = imageFileName.rfind('.')
        self.pickleFile = f"pickle_files/{imageFileName[p_from:p_end]}.dat"

        self.cells: CellsStore = None
        self.distances_gt: DistMatrix = None
        self.distances: DistMatrix = None

    def calculate_cells(self):
        print(f"Calculating pre-processing data for {self.imageFile}")
        self.cells = read_cells(self.imageFile)

    def calculate_opt_distances(self):
        if self.cells is None:
            self.calculate_cells()
        old_precise, CFG["precise"] = CFG["precise"], False
        self.distances = calculate_mutual_distances(self.cells)
        CFG["precise"] = old_precise

    def calculate_gt_distances(self):
        if self.cells is None:
            self.calculate_cells()
        old_precise, CFG["precise"] = CFG["precise"], True
        self.distances_gt = calculate_mutual_distances(self.cells)
        CFG["precise"] = old_precise


def store_case_study(case_study: CaseStudy):
    print(f"Removing Numba data from CellsStore of {case_study.pickleFile}")
    for cell in case_study.cells.values():
        cell.surface_pixels_numba = None
    #
    print(f"Storing case study into {case_study.pickleFile}")
    with open(case_study.pickleFile, "wb") as file:
        Pickler(file).dump(case_study)


def load_case_study(reference_case_study: CaseStudy) -> Optional[CaseStudy]:
    print(f"Loading case study from {reference_case_study.pickleFile}")
    with open(reference_case_study.pickleFile, "rb") as file:
        cls: CaseStudy = Unpickler(file).load()
        cls.imageSize = reference_case_study.imageSize
        return cls