# Author:       David Tschan
# Institute:    Paul-Scherrer Institute, Laboratory for Atmospheric Chemistry, Multiphase Chemistry Group, 
#               Forschungsstrasse 111, 5232 Villigen PSI, Switzerland 
# Role:         Master Student
# License:      n/a
# Date:         2026

### ----------------------------------------------------------------------------------------------------------------------------------- ###
### SUBMODULE DOCSTRING

"""The docstring for this submodule
"""

### ----------------------------------------------------------------------------------------------------------------------------------- ###
### IMPORTS AND CONFIGURATIONS

import pandas as pd
import numpy as np
import matplotlib 
import matplotlib.pyplot as plt

### ----------------------------------------------------------------------------------------------------------------------------------- ###
### Helper.py

def quick_plot_setup(fontsize = 8):

    plt.style.use(r'C:\Users\tschan_d\Documents\code\.plotstyle\quick.mplstyle')
    matplotlib.rcParams.update({'font.size': fontsize})

    return ".pdf"

def final_plot_setup(fontsize = 8):

    matplotlib.use('pgf')
    plt.style.use(r'C:\Users\tschan_d\Documents\code\.plotstyle\final.mplstyle')

    matplotlib.rcParams.update({'text.usetex':      True,
                                'pgf.rcfonts':      False,
                                'font.family':      'serif',
                                'font.size':        fontsize,
                                'legend.fontsize':  fontsize,
                                'xtick.labelsize':  fontsize,
                                'ytick.labelsize':  fontsize,
                                'pgf.texsystem':    r'lualatex',
                                'pgf.preamble':     r'\usepackage[fontsize = ' + str(fontsize) + r']{fontsize}' + 
                                                    r'\usepackage{unicode-math}' + 
                                                    r'\usepackage{accents}' + 
                                                    r'\usepackage{fontspec}' + 
                                                    r'\usepackage[T1]{fontenc}' +
                                                    r'\usepackage{unicode-math}' +
                                                    r'\setmainfont{STIX Two Text}' +
                                                    r'\setmathfont{STIX Two Math}' + 
                                                    r'\usepackage{siunitx}' + 
                                                    r'\usepackage{mhchem}' + 
                                                    r'\newcommand{\der}[1]{\ensuremath{\textnormal{d}\,{#1}}}' + 
                                                    r'\DeclareSIUnit\ppb{ppb}' + 
                                                    r'\DeclareSIUnit\ppm{ppm}'})

    # Exporting via the pgf backend produces some font issues, therefore outputting via pdf is most consistent...
    return ".pdf"

class DATFile():

    """The docstring for this class
        
        Description: Bla bla bla
    
        Attributes
        ----------
        word: str
            A word
    
        Methods
        -------
        __init__: HelloWorld
            The class constructor
        """

    def __init__(self, file, data_name = 'value'):

        self.file = file
        self.data_name = data_name
        self.data = None

    def read(self):

        dates = []
        times = []
        values = []

        with open(self.file, 'r', errors='ignore') as f:
            for i, line in enumerate(f.readlines()):

                if i >= 3:
                    
                    items = line.split()
                    dates.append(items[0])
                    times.append(items[1])
                    values.append(float(items[2]))

        pre_datetimes = [date + ' ' + time for date, time in zip(dates, times)]
        datetimes = pd.to_datetime(pre_datetimes, format=r'%d.%m.%Y %H:%M:%S')
        self.data = pd.DataFrame({'timestamp': datetimes, self.data_name: values})

    def exclude_outliers():
            
        pass

def compute_R2(y_true, y_est):

    residuals = y_est - y_true
    distances = y_true - np.mean(y_true)

    return 1 - np.sum(np.power(residuals, 2)) / np.sum(np.power(distances, 2))