import os import glob import pandas as pd import json import numpy as np from abstract_importer import AbstractImporter class JsonImporter(AbstractImporter): def __init__(self, files_path): self.df_samples_list = [] self.df_structure = pd.DataFrame() self.df_variables = pd.DataFrame() super(JsonImporter, self).__init__(files_path) def import_data(self): raw_data = self.read_json_file() self.import_trajectories(raw_data) self.import_structure(raw_data) self.import_variables(raw_data) def import_trajectories(self, raw_data): self.normalize_trajectories(raw_data, 0, 'samples') def import_structure(self, raw_data): self.df_structure = self.one_level_normalizing(raw_data, 0, 'dyn.str') def import_variables(self, raw_data): self.df_variables = self.one_level_normalizing(raw_data, 0, 'variables') def read_json_file(self): try: read_files = glob.glob(os.path.join(self.files_path, "*.json")) for file_name in read_files: with open(file_name) as f: data = json.load(f) return data except ValueError as err: print(err.args) def one_level_normalizing(self, raw_data, indx, variables_key): return pd.json_normalize(raw_data[indx][variables_key]) def normalize_trajectories(self, raw_data, indx, trajectories_key): for sample_indx, sample in enumerate(raw_data[indx][trajectories_key]): self.df_samples_list.append(pd.json_normalize(raw_data[indx][trajectories_key][sample_indx])) def build_list_of_samples_array(self, data_frame): columns_list = [] for column in data_frame: columns_list.append(data_frame[column].to_numpy()) return columns_list def clear_data_frames(self): """ Rimuove tutti i valori contenuti nei data_frames presenti in df_list Parameters: void Returns: void """ for indx in range(len(self.df_samples_list)): #data_frame = data_frame.iloc[0:0] self.df_samples_list[indx] = self.df_samples_list[indx].iloc[0:0] """ij = JsonImporter("../data") ij.import_data() #print(ij.df_samples_list[7]) print(ij.df_structure) print(ij.df_variables) print((ij.build_list_of_samples_array(0)[1].size))"""