pilar@0: from typing import List, Any
pilar@0: 
pilar@0: import xarray as xr
pilar@0: from config import *
pilar@0: import plotly.graph_objects as go
pilar@0: from plotly.subplots import make_subplots
pilar@0: import netCDF4 as nc
pilar@0: from datetime import datetime
pilar@0: import numpy as np
pilar@0: import sys
pilar@0: from os.path import exists
pilar@0: 
pilar@0: 
pilar@0: def read_variable_in_group(nc_file_path, group_name, variable_name):
pilar@0:     try:
pilar@0:         with nc.Dataset(nc_file_path, 'r') as dataset:
pilar@0:             # Find the group
pilar@0:             group = dataset.groups[group_name]
pilar@0: 
pilar@0:             # Access the variable
pilar@0:             try:
pilar@0:                 variable = group.variables[variable_name]
pilar@0:                 dimensions = variable.dimensions
pilar@0:                 # print("Dimensions:", dimensions)
pilar@0:                 # variable_metadata = { variable.name, variable.long_name, variable.units, variable.ndim}
pilar@0: 
pilar@0:                 # variable_metadata.add = variable.long_name
pilar@0:                 variable_metadata = \
pilar@0:                     {'name': variable.name.capitalize(),
pilar@0:                      'long_name': variable.long_name.capitalize(),
pilar@0:                      'units': variable.units,
pilar@0:                      'ndim': variable.ndim}
pilar@0: 
pilar@0:                 # Create an index array for slicing the variable
pilar@0:                 index_array = [slice(None)] * variable.ndim
pilar@0:                 # Access the variable data using the index array
pilar@0:                 variable_data = variable[index_array]
pilar@0: 
pilar@0:                 # Fetch the dimension values
pilar@0:                 dimension_values = {}
pilar@0:                 for dim_name in dimensions:
pilar@0:                     dim_values = group.variables[dim_name][:]
pilar@0:                     dimension_values[dim_name] = dim_values
pilar@0:                     # Adjust the index array for slicing
pilar@0:                     index_array[variable.dimensions.index(dim_name)] = 0
pilar@0: 
pilar@0:                 # Now you have the variable data and its associated dimensions in a dictionary
pilar@0:                 result = {'data': variable_data, 'dimensions': dimension_values, 'metadata': variable_metadata}
pilar@0: 
pilar@0:                 # Print or use the result as needed
pilar@0:                 # print(result)
pilar@0:                 return result
pilar@0: 
pilar@0:             except KeyError:
pilar@0:                 print("The specified variable does not exist in the group.")
pilar@0: 
pilar@0:     except FileNotFoundError:
pilar@0:         print(f"The file '{nc_file_path}' does not exist.")
pilar@0:     except Exception as e:
pilar@0:         print(f"The group {e} does not exist.")
pilar@0: 
pilar@0: 
pilar@0: def read_variable_in_group2(nc_file_path, group_name, variable_name):
pilar@0:     try:
pilar@0:         with nc.Dataset(nc_file_path, 'r') as dataset:
pilar@0:             # Find the group
pilar@0:             group = dataset.groups[group_name]
pilar@0: 
pilar@0:             # Access the variable
pilar@0:             try:
pilar@0:                 variable = group.variables[variable_name]
pilar@0:                 dimensions = variable.dimensions
pilar@0:                 # print("Dimensions:", dimensions)
pilar@0:                 # variable_metadata = { variable.name, variable.long_name, variable.units, variable.ndim}
pilar@0: 
pilar@0:                 # variable_metadata.add = variable.long_name
pilar@0:                 variable_metadata = \
pilar@0:                     {'name': variable.name.capitalize(),
pilar@0:                      'long_name': variable.long_name.capitalize(),
pilar@0:                      'units': variable.units,
pilar@0:                      'ndim': variable.ndim}
pilar@0: 
pilar@0:                 # Create an index array for slicing the variable
pilar@0:                 index_array = [slice(None)] * variable.ndim
pilar@0:                 # Access the variable data using the index array
pilar@0:                 variable_data = variable[index_array]
pilar@0: 
pilar@0:                 # Fetch the dimension values
pilar@0:                 dimension_values = {}
pilar@0:                 for dim_name in dimensions:
pilar@0:                     dim_values = group.variables[dim_name][:]
pilar@0:                     dimension_values[dim_name] = dim_values
pilar@0:                     # Adjust the index array for slicing
pilar@0:                     index_array[variable.dimensions.index(dim_name)] = 0
pilar@0: 
pilar@0:                 # Now you have the variable data and its associated dimensions in a dictionary
pilar@0:                 # result = {'data': variable_data, 'dimensions': dimension_values, 'metadata': variable_metadata}
pilar@0: 
pilar@0:                 # Print or use the result as needed
pilar@0:                 # print(result)
pilar@0:                 return variable_data, dimension_values, variable_metadata
pilar@0: 
pilar@0:             except KeyError:
pilar@0:                 print(f"The variable {variable_name} does not exist in the {group_name} group.")
pilar@0:     except FileNotFoundError:
pilar@0:         print(f"The file {nc_file_path} does not exist.")
pilar@0:     except Exception as e:
pilar@0:         print(f"The group {e} does not exist.")
pilar@0: 
pilar@0: 
pilar@0: def plot_vertical_profiles_plotly(variables, data, dimensions, y_array, positive_error_array, negative_error_array,
pilar@0:                                   xaxis_label, yaxis_label, title, timeinrows):
pilar@0:     """
pilar@0:     ToDo Complete
pilar@0:     Plot vertical profiles using plotly
pilar@0: 
pilar@0:     Parameters:
pilar@0:     -
pilar@0: 
pilar@0:     Returns:
pilar@0:     - fig: html
pilar@0:     """
pilar@0: 
pilar@0:     # Set number of columns, number of rows, and subtitles if needed
pilar@0:     ncols = len(variables)
pilar@0:     subtitles = None
pilar@0:     if timeinrows is True:
pilar@0:         nrows = len(dimensions['time'])
pilar@0:         if nrows > 1:
pilar@0:             for t in range(nrows):
pilar@0:                 for v in range(ncols):
pilar@0:                     if subtitles is None:
pilar@0:                         subtitles = [str(datetime.fromtimestamp(dimensions['time'][t]))]
pilar@0:                     else:
pilar@0:                         subtitles += [str(datetime.fromtimestamp(dimensions['time'][t]))]
pilar@0:     else:
pilar@0:         nrows = 1
pilar@0:         subtitles = ''
pilar@0: 
pilar@0:     # Create figure
pilar@0:     fig = make_subplots(rows=nrows, cols=ncols,
pilar@0:                         subplot_titles=subtitles,
pilar@0:                         shared_yaxes=True)
pilar@0: 
pilar@0:     # Define colours
pilar@0:     clrs = [['rgb(0,35,255)', 'rgb(0,20,150)', 'rgb(0,170,250)'],  # UV
pilar@0:             ['rgb(90,210,50)', 'rgb(40,90,25)', 'rgb(120,250,80)'],  # VIS
pilar@0:             ['rgb(250,30,30)', 'rgb(200,25,30)', 'rgb(250,125,0)']]  # IR
pilar@0:     clrs = np.array(clrs)
pilar@0: 
pilar@0:     # Plot
pilar@0:     for v in range(len(variables)):
pilar@0:         for t in range(len(dimensions['time'])):
pilar@0:             for w in range(len(dimensions['wavelength'])):
pilar@0:                 tn = str(datetime.fromtimestamp(dimensions['time'][t]))
pilar@0:                 wn = str(dimensions['wavelength'][w])
pilar@0:                 if timeinrows == True:
pilar@0:                     lgnd = wn + ' nm'
pilar@0:                     rown = t + 1
pilar@0:                     colorint = 1
pilar@0:                     colorind = 0
pilar@0:                     subtitle = str(datetime.fromtimestamp(dimensions['time'][t]))
pilar@0:                 else:
pilar@0:                     lgnd = wn + ' nm - ' + tn
pilar@0:                     rown = 1
pilar@0:                     colorind = t
pilar@0:                     colorint = 1 - (0.4 * t)
pilar@0: 
pilar@0:                 # colour definition
pilar@0:                 if abs(dimensions['wavelength'][w] - 1064) < 1:
pilar@0:                     clr = clrs[2, colorind]
pilar@0:                     # clr = 'rgb(235,70,20)'  # 'rgba(215,27,96,0.4)'
pilar@0:                 elif abs(dimensions['wavelength'][w] - 532) < 1:
pilar@0:                     clr = clrs[1, colorind]
pilar@0:                 elif abs(dimensions['wavelength'][w] - 355) < 1:
pilar@0:                     clr = clrs[0, colorind]
pilar@0:                 else:
pilar@0:                     clr = 'rgb(0,0,0)'
pilar@0: 
pilar@0:                 data_values = data[variables[v]][w, t, :]
pilar@0:                 error_values = positive_error_array[variables[v]][w, t, :]
pilar@0:                 vertical_levels = y_array["data"][t, :]
pilar@0: 
pilar@0:                 # Add trace
pilar@0:                 fig.add_trace(go.Scatter(
pilar@0:                     x=data_values,
pilar@0:                     y=vertical_levels,
pilar@0:                     error_x=dict(array=error_values, width=0, thickness=0.1),
pilar@0:                     # ToDo Include distinction between positive and negative errors
pilar@0:                     mode='lines',
pilar@0:                     line=dict(width=2, color=clr),
pilar@0:                     name=lgnd),
pilar@0:                     row=rown, col=v + 1,
pilar@0:                 )
pilar@0: 
pilar@0:                 # ToDo Check formatting options (showlegend, mapbox)
pilar@0:                 fig.update_xaxes({"title": xaxis_label[v]}, row=rown, col=v + 1)
pilar@0:                 fig.update_xaxes({"mirror": True, "ticks": 'outside', "showline": True, "color": "black"}, row=rown,
pilar@0:                                  col=v + 1)
pilar@0: 
pilar@0:                 fig.update_xaxes(ticks="outside")
pilar@0:                 fig.update_yaxes(ticks="outside", row=rown, col=v + 1)  # ToDo check tickformat (i.e. tickformat=":.2e")
pilar@0:                 if v == 0:
pilar@0:                     fig.update_yaxes({"title": yaxis_label}, row=rown, col=v + 1)
pilar@0: 
pilar@0:     # Set axis labels and title
pilar@0:     fig.update_layout(
pilar@0:         yaxis_title=yaxis_label,
pilar@0:         title=title,
pilar@0:         template="seaborn",  # You can change the template as per your preference
pilar@0:         # ['ggplot2', 'seaborn', 'simple_white', 'plotly', 'plotly_white', 'plotly_dark', 'presentation',
pilar@0:         #  'xgridoff', 'ygridoff', 'gridon', 'none']
pilar@0:         height=max(600, 450 * rown), width=min(max(400 * len(variables), 750), 1400),
pilar@0:     )
pilar@0: 
pilar@0:     return fig
pilar@0: 
pilar@0: 
pilar@0: # Files for testing
pilar@0: # netcdf_file = 'hpb_012_0000598_201810172100_201810172300_20181017oh00_ELDAmwl_v0.0.1.nc'      # Backscatter and extinction, no error
pilar@0: # netcdf_file = 'hpb_012_0000627_202308240200_202308240400_20230824hpb0200_ELDAmwl_v0.0.1.nc'     # Extinction with error
pilar@0: 
pilar@0: 
pilar@0: try:
pilar@0:     netcdf_file = sys.argv[1]
pilar@0: except:
pilar@0:     print('Syntax is incorrect, you need to pass the filename of the file to be plotted.\n'
pilar@0:           'Example: ./eldamwl_plot.py hpb_012_0000598_201810172100_201810172300_20181017oh00_ELDAmwl_v0.0.1.nc')
pilar@0:     exit()
pilar@0: 
pilar@0: netcdf_file_path = netcdf_path + netcdf_file
pilar@0: 
pilar@0: file_exists = exists(netcdf_file_path)
pilar@0: if not file_exists:
pilar@0:     print('The file that you provided does not exist.')
pilar@0:     exit()
pilar@0: 
pilar@0: # Groups of data in the NetCDF file (low and high resolution)
pilar@0: groups = ['lowres_products', 'highres_products']
pilar@0: groups_names = ['Low resolution', 'High resolution']
pilar@0: 
pilar@0: # Variables that we want to plot
pilar@0: variable = ['backscatter', 'extinction', 'lidarratio', 'volumedepolarization', 'particledepolarization']    # ToDo Test particledepolarization when available
pilar@0: # Errors associated to the variables
pilar@0: variable_error = ['error_backscatter', 'error_extinction', 'error_lidarratio',
pilar@0:                   'positive_systematic_error_volumedepolarization', 'error_particledepolarization']
pilar@0: variable_negative_error_volumedepol = 'negative_systematic_error_volumedepolarization'
pilar@0: # Other variables:
pilar@0: #   cloud_mask (time, level)
pilar@0: 
pilar@0: 
pilar@0: for g in range(len(groups)):
pilar@0:     # Read the altitude variable
pilar@0:     altitude = read_variable_in_group(netcdf_file_path, groups[g], 'altitude')
pilar@0: 
pilar@0:     if altitude['metadata']['units'] == 'm':
pilar@0:         altitude['data'] = altitude['data'] / 1000.0
pilar@0:         altitude['metadata']['units'] = 'km'
pilar@0: 
pilar@0:     # Initialize variables
pilar@0:     data = None
pilar@0:     variables_with_data = None
pilar@0:     variables_axis = None
pilar@0: 
pilar@0:     # Read the data
pilar@0:     for v in range(len(variable)):
pilar@0:         # Initialize var
pilar@0:         var = None
pilar@0: 
pilar@0:         try:
pilar@0:             var, dim, met = read_variable_in_group2(netcdf_file_path, groups[g], variable[v])
pilar@0:         except TypeError:
pilar@0:             print(f"The variable {variable[v]} is not present. Cannot unpack non-iterable NoneType object")
pilar@0: 
pilar@0:         if var is not None:  # If the variable that we're trying to read exists
pilar@0:             if var.max() != var.min():  # If there is data in the array
pilar@0:                 if variables_with_data is None:
pilar@0:                     variables_with_data = [variable[v]]
pilar@0:                     variables_axis = [met['long_name'] + ' [ ' + met['units'] + ' ]']
pilar@0:                 else:
pilar@0:                     variables_with_data += [variable[v]]
pilar@0:                     variables_axis += [met['long_name'] + ' [ ' + met['units'] + ' ]']
pilar@0: 
pilar@0:                 var_error, dim_error, met_error = read_variable_in_group2(netcdf_file_path, groups[g],
pilar@0:                                                                           variable_error[v])
pilar@0:                 if variable[v] == 'volumedepolarization':
pilar@0:                     var_error_negative, dim, met = read_variable_in_group2(netcdf_file_path, groups[g],
pilar@0:                                                                            variable_negative_error_volumedepol)
pilar@0:                 else:
pilar@0:                     var_error_negative = var_error
pilar@0: 
pilar@0:                 # Add read data to the data variable to have it all grouped
pilar@0:                 if data is None:
pilar@0:                     # Start populating data and data_error
pilar@0:                     data = {variable[v]: var}
pilar@0:                     data_error_positive = {variable[v]: var_error}
pilar@0:                     data_error_negative = {variable[v]: var_error_negative}
pilar@0:                 else:
pilar@0:                     # Add more data to data and data_error
pilar@0:                     data[variable[v]] = var
pilar@0:                     data_error_positive[variable[v]] = var_error
pilar@0:                     data_error_negative[variable[v]] = var_error_negative
pilar@0:             else:
pilar@0:                 print(f'There is no data for {variable[v]} in the {groups_names[g]} group.')
pilar@0: 
pilar@0:     dimensions = dim
pilar@0: 
pilar@0:     # Ask user if he/she wants to plot different temporal profiles in different rows
pilar@0:     if g == 0:
pilar@0:         # Initialize
pilar@0:         timeinplots = False
pilar@0:         timeinrows = False
pilar@0:         # Ask user how he/she wants the plot
pilar@0:         if len(dimensions['time']) > 1:
pilar@0:             answer = input(
pilar@0:                 f"There are {len(dimensions['time'])} temporal profiles, do you want to plot them on different plots [y/n]? ")
pilar@0:             if answer[0:1].lower() == 'y':
pilar@0:                 timeinplots = True
pilar@0:             if timeinplots == False:
pilar@0:                 answer = input(
pilar@0:                     f"There are {len(dimensions['time'])} temporal profiles, do you want to plot them on different rows [y/n]? ")
pilar@0:                 if answer[0:1].lower() == 'y':
pilar@0:                     timeinrows = True
pilar@0: 
pilar@0:     if variables_with_data is not None:
pilar@0:         timeiter = dimensions['time']
pilar@0:         if timeinplots == True:
pilar@0:             dimensions_t = dimensions
pilar@0:             for t in range(len(dimensions['time'])):
pilar@0:                 dimensions_t['time'] = [timeiter[t]]
pilar@0:                 for vr in range(len(variables_with_data)):
pilar@0:                     if vr == 0:
pilar@0:                         data_t = {variables_with_data[vr]: data[variables_with_data[vr]][:, t:t + 1, :]}
pilar@0:                     else:
pilar@0:                         data_t[variables_with_data[vr]] = data[variables_with_data[vr]][:, t:t + 1, :]
pilar@0: 
pilar@0:                 fig = plot_vertical_profiles_plotly(
pilar@0:                     variables=variables_with_data,
pilar@0:                     data=data_t,
pilar@0:                     dimensions=dimensions_t,
pilar@0:                     y_array=altitude,
pilar@0:                     positive_error_array=data_error_positive,
pilar@0:                     negative_error_array=data_error_negative,
pilar@0:                     xaxis_label=variables_axis,
pilar@0:                     yaxis_label=altitude['metadata']['long_name'] + ' [' + altitude['metadata']['units'] + ']',
pilar@0:                     title=netcdf_file[0:3].upper() + " - " + str(
pilar@0:                         datetime.fromtimestamp(dimensions_t['time'][0]).strftime(
pilar@0:                             '%d/%m/%Y %H:%M:%S')) + " - ELDA MWL " +
pilar@0:                           groups_names[g],
pilar@0:                     timeinrows=True
pilar@0:                 )
pilar@0: 
pilar@0:                 # Show the plot (in Jupyter Notebook or Jupyter Lab)
pilar@0:                 fig.show()
pilar@0: 
pilar@0:                 # Save the plot
pilar@0:                 filename = "ELDAmwl_" + netcdf_file[0:3].upper() + "_" + datetime.fromtimestamp(
pilar@0:                     dimensions['time'][0]).strftime('%Y%m%d-%H%M%S') + "_" + groups[g] + ".html"
pilar@0:                 fig.write_html(html_path + filename)
pilar@0: 
pilar@0:         else:
pilar@0:             # Plotting in one plot (even if there are different time profiles)
pilar@0:             fig = plot_vertical_profiles_plotly(
pilar@0:                 variables_with_data,
pilar@0:                 data,
pilar@0:                 dimensions,
pilar@0:                 altitude,
pilar@0:                 data_error_positive,
pilar@0:                 data_error_negative,
pilar@0:                 xaxis_label=variables_axis,
pilar@0:                 yaxis_label=altitude['metadata']['long_name'] + ' [' + altitude['metadata']['units'] + ']',
pilar@0:                 title=netcdf_file[0:3].upper() + " - " + str(
pilar@0:                     datetime.fromtimestamp(dimensions['time'][0]).strftime('%d/%m/%Y')) + " - ELDA MWL " + groups_names[
pilar@0:                           g],
pilar@0:                 timeinrows=timeinrows
pilar@0:             )
pilar@0: 
pilar@0:             # Show the plot (in Jupyter Notebook or Jupyter Lab)
pilar@0:             fig.show()
pilar@0: 
pilar@0:             # Save the plot
pilar@0:             if timeinrows == True:
pilar@0:                 filename = "ELDAmwl_" + netcdf_file[0:3].upper() + "_" + datetime.fromtimestamp(
pilar@0:                     dimensions['time'][0]).strftime('%Y%m%d%H%M%S') + '-' + datetime.fromtimestamp(
pilar@0:                     dimensions['time'][-1]).strftime('%Y%m%d%H%M%S') + "_" + groups[g] + "_timeinrows.html"
pilar@0:             else:
pilar@0:                 filename = "ELDAmwl_" + netcdf_file[0:3].upper() + "_" + datetime.fromtimestamp(
pilar@0:                     dimensions['time'][0]).strftime('%Y%m%d%H%M%S') + '-' + datetime.fromtimestamp(
pilar@0:                     dimensions['time'][-1]).strftime('%Y%m%d%H%M%S') + "_" + groups[g] + ".html"
pilar@0: 
pilar@0:             fig.write_html(html_path + filename)
pilar@0: 
pilar@0:     else:
pilar@0:         print(f'There is no data to be plotted in the {groups_names[g]} group.')