Update app.py

app.py

@@ -2,21 +2,32 @@ import streamlit as st
 import pandas as pd
 import requests
 import plotly.express as px
-import plotly.graph_objects as go
 import io
 
-# 設置全局主題 - 使用更豐富的彩色主題
-theme = px.colors.qualitative.Bold
+# 設置全局主題
+THEME = px.colors.qualitative.Bold
 
-# Function to download and load CSV data
-def download_and_load_csv(url):
+# 下載並載入 CSV 數據
+@st.cache_data
+def download_and_clean_csv(url):
     response = requests.get(url)
     response.encoding = 'utf-8'
     df = pd.read_csv(io.StringIO(response.text), encoding='utf-8')
-    df = df.fillna(0)  # 填充 NaN 為 0
+    
+    # 資料清理
+    df = df.dropna().drop_duplicates()  # 移除缺失值與重複數據
+    df.columns = df.columns.str.strip()  # 去除欄位名稱前後空格
+    df = df.applymap(lambda x: x.strip() if isinstance(x, str) else x)  # 去除字串內的空格
+    
+    # 嘗試轉換數值欄位
+    num_cols = ["範疇一排放量(噸CO2e)", "範疇二排放量(噸CO2e)", "範疇三排放量(噸CO2e)"]
+    for col in num_cols:
+        if col in df.columns:
+            df[col] = pd.to_numeric(df[col], errors='coerce').fillna(0)
+    
     return df
 
-# 美化函數
+# 美化圖表
 def beautify_chart(fig):
     fig.update_layout(
         font_family="Arial",
@@ -31,87 +42,29 @@ def beautify_chart(fig):
     fig.update_yaxes(showline=True, linewidth=2, linecolor='lightgray', gridcolor='lightgray')
     return fig
 
-# 通用函數來生成並顯示圖表
-def generate_plots(df, df_name, selected_columns):
-    selected_columns = [col for col in selected_columns if col in df.columns]  # 確保欄位存在
-    
-    if not selected_columns:
-        st.write(f"{df_name} 中沒有找到可用的數值列來繪製圖表")
-        return
-
-    # 移除所有選擇欄位值皆為 0 的行
-    df = df[(df[selected_columns] != 0).any(axis=1)]
-    
-    if df.empty:
-        st.write(f"{df_name} 中的選擇欄位數據全為 0 或缺失，無法繪製圖表。")
-        return
-    
-    with st.expander(f"顯示/隱藏 {df_name} 圖表"):
-        st.subheader(f"{df_name} 線圖")
-        fig_line = px.line(df, x="公司名稱", y=selected_columns, title=f"{df_name} 線圖", color_discrete_sequence=theme)
-        fig_line = beautify_chart(fig_line)
-        st.plotly_chart(fig_line)
-
-        st.subheader(f"{df_name} 餅圖")
-        total_emissions = df.groupby("公司名稱")[selected_columns].sum().reset_index()
-        total_emissions = total_emissions.melt(id_vars=["公司名稱"], value_vars=selected_columns, var_name="Emission Type", value_name="Total Emissions")
-        fig_pie = px.pie(total_emissions, values='Total Emissions', names='公司名稱', 
-                         title=f"{df_name} 各公司排放量的餅圖", 
-                         color_discrete_sequence=theme, hole=0.3)
-        fig_pie.update_traces(textposition='inside', textinfo='percent+label')
-        fig_pie = beautify_chart(fig_pie)
-        st.plotly_chart(fig_pie)
-
-        if len(selected_columns) >= 2:
-            st.subheader(f"{df_name} 散點圖")
-            fig_scatter = px.scatter(df, x="公司名稱", y=selected_columns, 
-                                     title=f"{df_name} 散點圖", color_discrete_sequence=theme)
-            fig_scatter = beautify_chart(fig_scatter)
-            st.plotly_chart(fig_scatter)
-
-# URLs for the CSV files
+# 下載、清理並合併數據
 urls = [
     "https://mopsfin.twse.com.tw/opendata/t187ap46_L_1.csv",
     "https://mopsfin.twse.com.tw/opendata/t187ap46_O_2.csv",
     "https://mopsfin.twse.com.tw/opendata/t187ap46_L_6.csv"
 ]
-
-# Download, load, and clean data from each CSV file
-dfs = [download_and_load_csv(url) for url in urls]
-
-# Combine all DataFrames into one
+dfs = [download_and_clean_csv(url) for url in urls]
 combined_df = pd.concat(dfs, ignore_index=True)
 
-# 確保所有 NaN 值填充為 0
-combined_df = combined_df.fillna(0)
-
-# 移除所有排放量皆為 0 的行
-combined_df = combined_df[(combined_df["範疇一排放量(噸CO2e)"] != 0) | 
-                          (combined_df["範疇二排放量(噸CO2e)"] != 0) | 
-                          (combined_df["範疇三排放量(噸CO2e)"] != 0)]
-
-# Set up the Streamlit app
+# 設定 Streamlit 介面
 st.title("台灣企業ESG數據分析與揭露")
 
-# Display the combined DataFrame
-st.subheader("爬取的資料預覽")
+# 顯示清理後的數據
+st.subheader("清理後的數據預覽")
 st.dataframe(combined_df)
 
-# Specify the columns related to emissions
-emission_columns = [
-    "範疇一排放量(噸CO2e)", 
-    "範疇二排放量(噸CO2e)", 
-    "範疇三排放量(噸CO2e)"
-]
-
-# Create buttons for each emission category
-selected_columns = []
-for column in emission_columns:
-    if st.button(f"顯示 {column}"):
-        selected_columns.append(column)
+# 可視化數據
+emission_columns = ["範疇一排放量(噸CO2e)", "範疇二排放量(噸CO2e)", "範疇三排放量(噸CO2e)"]
+selected_columns = [col for col in emission_columns if col in combined_df.columns]
 
-# Generate plots for the selected emission categories
 if selected_columns:
-    generate_plots(combined_df, "Combined Data", selected_columns)
+    st.subheader("排放量折線圖")
+    fig_line = px.line(combined_df, x="公司名稱", y=selected_columns, title="企業排放量趨勢", color_discrete_sequence=THEME)
+    st.plotly_chart(beautify_chart(fig_line))
 else:
-    st.write("請選擇至少一個排放類別來顯示圖表。")
+    st.write("未找到有效的排放數據")