#!/usr/bin/env python
# coding: utf-8

# In[1]:


import xarray as xr
import numpy as np 
import matplotlib.pyplot as plt
import glob
import pandas as pd


# In[2]:


#Define file path to open data
path = r'/work/users/hmo/remoclic/Phuong/PHD/BC_2023/Tan/NC/Geo_*.nc'

#Open and sort the list of file in the previous path 
list_of_files = sorted(glob.glob(path), key = lambda x: x.split('_')[-1])
#list_of_files


# In[3]:


#Write a function to collect "geo" data from the file path, store in a dataset with the columns name is the name of the file
def create_geo_dataframe(file_paths):
    # Create an empty dictionary to store the data of geo
    data_dict = {}

    # Iterate over the file paths
    for file_path in file_paths:
        # Open the netcdf dataset with xarray library 
        dataset = xr.open_dataset(file_path)

        # Extract the "geo" variable
        geo_data = dataset['geo'][0]

        # Get the file name without the directory and extension
        file_name = file_path.split('\\')[-1].split('.')[0]

        # Store the "geo" data in the dictionary with the file name as the key
        data_dict[file_name] = geo_data.values

    # Create the DataFrame from the dictionary
    df = pd.DataFrame(data_dict)

    return df

#Output data storing the geo values
geo_df = create_geo_dataframe(list_of_files)
geo_df


# In[4]:


#print(geo_df.columns)


# In[7]:


# Create a box plot for each column in the DataFrame
fig, ax = plt.subplots(figsize=(10, 6))

#Set the labels
labels = geo_df.columns

#Plot the boxplot for dataframe
bplot1 = ax.boxplot(geo_df,
                    notch = True,
                    vert = True,
                    patch_artist = True, 
                    labels = labels)

#Set the title for figure
ax.set_title('Box plot of '+ str(geo_df.columns[0]).split('_')[0] + ' data')

colors = []
for column in geo_df.columns:
    if 'ORIG' in column:
        color = 'pink'
    elif 'QM' in column:
        color = 'lightgreen'
    else:
        print("Error: This file doesn't contain ORIG or QM.")
        continue
    colors.append(color)

# fill box plots with colors
for patch, color in zip(bplot1['boxes'], colors):
    patch.set_facecolor(color)

# Display the plot
plt.show()

#Save figure
#plt.savefig('E:/thay_Tan/chi_phuong/Box_plot_of_Geo.png')