pages/5_Past Records.py

import streamlit as st
import pandas as pd
import plotly.express as px
import json
import os
import glob

# Directory paths
data_dir = 'past_temp'
geojson_file = os.path.join(data_dir, 'FavgTS.geojson')

# Load GeoJSON data
def load_geojson():
    with open(geojson_file) as f:
        return json.load(f)

# Create a dictionary to map short forms to long forms
def create_station_map(geojson):
    feature_map = {}
    for feature in geojson['features']:
        short_name = feature['properties']['WeatherStationShortName']
        long_name = feature['properties']['WeatherStationName_en']
        feature_map[short_name] = long_name
    return feature_map

# Load CSV files
def load_csv_files():
    return glob.glob(os.path.join(data_dir, '*.csv'))

# Plot time series
def plot_time_series(df, station_name):
    df['Date'] = pd.to_datetime(df['Date'], format='%Y%m%d', errors='coerce')
    df['Year'] = df['Date'].dt.year
    df['Month'] = df['Date'].dt.month

    fig_all_years = px.line(df, x='Date', y='Value', color='Year',
                            title=f'All-Year Temperature Time Series for {station_name}',
                            labels={'Date': 'Date', 'Value': 'Temperature (°C)', 'Year': 'Year'},
                            line_shape='linear')
    fig_all_years.update_layout(xaxis_title='Date', yaxis_title='Temperature (°C)')

    return fig_all_years

# Plot monthly averages
def plot_monthly_averages(df, station_name):
    df['Date'] = pd.to_datetime(df['Date'], format='%Y%m%d', errors='coerce')
    df['Year'] = df['Date'].dt.year
    df['Month'] = df['Date'].dt.month

    monthly_avg = df.groupby(['Year', 'Month'])['Value'].mean().reset_index()

    fig_monthly_avg = px.line(monthly_avg, x='Month', y='Value', color='Year',
                              title=f'Monthly Average Temperature Time Series for {station_name}',
                              labels={'Month': 'Month', 'Value': 'Average Temperature (°C)', 'Year': 'Year'},
                              line_shape='linear')
    fig_monthly_avg.update_layout(xaxis_title='Month', yaxis_title='Average Temperature (°C)', xaxis_tickformat='%b')

    return fig_monthly_avg

def plot_annual_average(df, station_name):
    annual_avg = df.groupby('Year')['Value'].mean().reset_index()

    fig_annual_avg = px.line(annual_avg, x='Year', y='Value',
                             title=f'Annual Average Temperature Trend for {station_name}',
                             labels={'Year': 'Year', 'Value': 'Average Temperature (°C)'},
                             line_shape='linear')
    fig_annual_avg.update_layout(xaxis_title='Year', yaxis_title='Average Temperature (°C)')

    return fig_annual_avg

# Streamlit app layout
st.set_page_config(layout="wide", page_title="Temperature Time Series")

# Load GeoJSON and create mapping
geojson = load_geojson()
station_map = create_station_map(geojson)

# Load all CSV files
csv_files = load_csv_files()

# Initialize data storage for all CSV files
all_data = []

# Process each CSV file
for file in csv_files:
    try:
        file_name = os.path.basename(file)
        short_form = file_name.split('.')[0]  # Get the file name without extension

        df = pd.read_csv(file)

        if df.shape[1] < 2:
            st.error(f"File {file} does not have the expected number of columns. Skipping.")
            continue

        if df.columns[0] != 'Date':
            df.columns = ['Date', 'Value']

        long_form = station_map.get(short_form, "Unknown Station")
        df['Station'] = long_form
        all_data.append(df)

    except Exception as e:
        st.error(f"Error loading or processing file {file}: {e}")

# Combine all data into a single DataFrame
if all_data:
    combined_df = pd.concat(all_data, ignore_index=True)
    combined_df['Date'] = pd.to_datetime(combined_df['Date'], format='%Y%m%d', errors='coerce')
    combined_df = combined_df.dropna(subset=['Date'])
    combined_df['Year'] = combined_df['Date'].dt.year
    combined_df['Month'] = combined_df['Date'].dt.month

    stations = combined_df['Station'].unique()
    default_station = stations[0] if len(stations) > 0 else None

    if not stations.size:
        st.write("No stations available in the data.")
    else:
        st.subheader('Past Daily Average Temperature Time Series')
        selected_station = st.selectbox("Select a Station", options=stations, index=0)

        station_data = combined_df[combined_df['Station'] == selected_station]

        if not station_data.empty:
            # Create two columns for plots
            col1, col2 = st.columns([2,1.5])

            # Top plot: All-year time series
            with col1:
                fig_all_years = plot_time_series(station_data, selected_station)
                st.plotly_chart(fig_all_years, use_container_width=True)

            # Bottom plot: Monthly average temperatures
            with col2:
                fig_monthly_avg = plot_monthly_averages(station_data, selected_station)
                st.plotly_chart(fig_monthly_avg, use_container_width=True)
        else:
            st.write(f"No data available for the selected station '{selected_station}'.")
else:
    st.write("No data to display.")