import os
import subprocess
import stat
import pandas as pd
import torch
import numpy as np



def Mazda(init_samples):
    ##########################################
    # Scaling
    ##########################################
    
    # Define the path to your Excel file
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/Info_Mazda_CdMOBP_edited.xlsx'
    
    # Read the Excel file into a DataFrame
    dataframe = pd.read_excel(file_path, sheet_name='Explain_DV_and_Const.')
    
    # Display the DataFrame to ensure it has been read correctly
    bounds = dataframe.values[1:, 1:3]
    bounds_tensor = torch.tensor(bounds, dtype=torch.float32)
    # print(bounds_tensor.shape)
    
    range_bounds = bounds_tensor[:,1] - bounds_tensor[:,0]
    
    scaled_samples = init_samples * range_bounds + bounds_tensor[:,0]
    # print(scaled_samples)
    
    # Convert the torch tensor to a numpy array
    data_numpy_back = scaled_samples.numpy()
    
    # Create a pandas DataFrame from the numpy array
    dataframe_back = pd.DataFrame(data_numpy_back)
    
    # Write the DataFrame to a text file with space-separated values
    output_file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_vars_eval.txt'
    
    dataframe_back.to_csv(output_file_path, sep='\t', header=False, index=False)
    #####################
    #####################
    
    
    #####################
    # Run Bash file
    #####################
    
    # Change the current working directory
    os.chdir('/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2')
    
    # Get the current permissions of the file
    current_permissions = os.stat(os.getcwd()).st_mode
    
    # Add execute permissions for the owner, group, and others
    new_permissions = current_permissions | stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
    
    # Apply the new permissions
    os.chmod(os.getcwd(), new_permissions)
    
    # Script name
    script_name = 'run.sh'
    
    # Run the bash script in the background
    process = subprocess.Popen(['bash', script_name], stdout=subprocess.PIPE, stderr=subprocess.PIPE, start_new_session=True)
    process.wait()
    
    # Optional: capture the output and error messages
    stdout, stderr = process.communicate()
    
    os.chdir('/home/turbo/rosenyu/Bank_High_DIM/')
    # print(os.getcwd())
    #####################
    #####################
    
    
    #####################
    # Read in objective and constraints
    #####################
    
    # Read the data from the file into a pandas DataFrame
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_objs_eval.txt'
    objs_dataframe = pd.read_csv(file_path, delim_whitespace=True, header=None)
    
    # Convert the DataFrame to a numpy array
    objs_data_numpy = objs_dataframe.values
    
    # Convert the numpy array to a torch tensor
    objs_data_tensor = torch.tensor(objs_data_numpy, dtype=torch.float32)
    objs_data_tensor = objs_data_tensor[:,0].reshape(objs_data_tensor.shape[0],1)
    
    
    
    # Read the data from the file into a pandas DataFrame
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_cons_eval.txt'
    cons_dataframe = pd.read_csv(file_path, delim_whitespace=True, header=None)
    
    # Convert the DataFrame to a numpy array
    cons_data_numpy = cons_dataframe.values
    
    # Convert the numpy array to a torch tensor
    cons_data_tensor = torch.tensor(cons_data_numpy, dtype=torch.float32)
    
    
    # print(objs_data_tensor)
    # print(cons_data_tensor)
    return cons_data_tensor, -objs_data_tensor







def Mazda_softpen(init_samples):
    ##########################################
    # Scaling
    ##########################################
    
    # Define the path to your Excel file
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/Info_Mazda_CdMOBP_edited.xlsx'
    
    # Read the Excel file into a DataFrame
    dataframe = pd.read_excel(file_path, sheet_name='Explain_DV_and_Const.')
    
    # Display the DataFrame to ensure it has been read correctly
    bounds = dataframe.values[1:, 1:3]
    bounds_tensor = torch.tensor(bounds, dtype=torch.float32)
    # print(bounds_tensor.shape)
    
    range_bounds = bounds_tensor[:,1] - bounds_tensor[:,0]
    
    scaled_samples = init_samples * range_bounds + bounds_tensor[:,0]
    # print(scaled_samples)
    
    # Convert the torch tensor to a numpy array
    data_numpy_back = scaled_samples.numpy()
    
    # Create a pandas DataFrame from the numpy array
    dataframe_back = pd.DataFrame(data_numpy_back)
    
    # Write the DataFrame to a text file with space-separated values
    output_file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_vars_eval.txt'
    
    dataframe_back.to_csv(output_file_path, sep='\t', header=False, index=False)
    #####################
    #####################
    
    
    #####################
    # Run Bash file
    #####################
    
    # Change the current working directory
    os.chdir('/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2')
    
    # Get the current permissions of the file
    current_permissions = os.stat(os.getcwd()).st_mode
    
    # Add execute permissions for the owner, group, and others
    new_permissions = current_permissions | stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
    
    # Apply the new permissions
    os.chmod(os.getcwd(), new_permissions)
    
    # Script name
    script_name = 'run.sh'
    
    # Run the bash script in the background
    process = subprocess.Popen(['bash', script_name], stdout=subprocess.PIPE, stderr=subprocess.PIPE, start_new_session=True)
    process.wait()
    
    # Optional: capture the output and error messages
    stdout, stderr = process.communicate()
    
    os.chdir('/home/turbo/rosenyu/Bank_High_DIM/')
    # print(os.getcwd())
    #####################
    #####################
    
    
    #####################
    # Read in objective and constraints
    #####################
    
    # Read the data from the file into a pandas DataFrame
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_objs_eval.txt'
    objs_dataframe = pd.read_csv(file_path, delim_whitespace=True, header=None)
    
    # Convert the DataFrame to a numpy array
    objs_data_numpy = objs_dataframe.values
    
    # Convert the numpy array to a torch tensor
    objs_data_tensor = torch.tensor(objs_data_numpy, dtype=torch.float32)
    objs_data_tensor = objs_data_tensor[:,0].reshape(objs_data_tensor.shape[0],1)
    
    
    
    # Read the data from the file into a pandas DataFrame
    file_path = '/home/turbo/rosenyu/Bank_High_DIM/Mazda_CdMOBP/Mazda_CdMOBP/rosen_sample_t2/pop_cons_eval.txt'
    cons_dataframe = pd.read_csv(file_path, delim_whitespace=True, header=None)
    
    # Convert the DataFrame to a numpy array
    cons_data_numpy = cons_dataframe.values
    
    # Convert the numpy array to a torch tensor
    cons_data_tensor = torch.tensor(cons_data_numpy, dtype=torch.float32)


    
    cost = cons_data_tensor
    cost[cost<0] = 0
    cost = cost.sum(dim=1).reshape(cost.shape[0], 1)
    objs_data_tensor = objs_data_tensor + cost

    
    # print(objs_data_tensor)
    # print(cons_data_tensor)
    return cons_data_tensor, -objs_data_tensor