Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,364 @@
-import altair as alt
-import numpy as np
-import pandas as pd
 import streamlit as st
+import pandas as pd
+import requests
+import plotly.express as px
+import plotly.graph_objects as go
+from plotly.subplots import make_subplots
+import numpy as np
+import tempfile
+import os
 
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+# 設置頁面配置
+st.set_page_config(
+    page_title="碳排放數據可視化分析",
+    page_icon="🌱",
+    layout="wide",
+    initial_sidebar_state="expanded"
+)
 
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
+# 標題和介紹
+st.title("🌱 碳排放數據可視化分析")
+st.markdown("---")
+st.write("此應用程式分析台灣公司的碳排放數據，包括範疇一和範疇二的排放量。")
 
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
+# 側邊欄設置
+st.sidebar.header("⚙️ 設置選項")
 
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
+# 數據載入功能
+@st.cache_data
+def load_data():
+    """載入並處理碳排放數據"""
+    try:
+        # 顯示載入狀態
+        with st.spinner("正在載入數據..."):
+            url = "https://mopsfin.twse.com.tw/opendata/t187ap46_O_1.csv"
+            response = requests.get(url)
+            
+            # 使用臨時文件
+            with tempfile.NamedTemporaryFile(mode='wb', suffix='.csv', delete=False) as tmp_file:
+                tmp_file.write(response.content)
+                tmp_file_path = tmp_file.name
+            
+            # 讀取CSV文件
+            df = pd.read_csv(tmp_file_path, encoding="utf-8-sig")
+            
+            # 清理臨時文件
+            os.unlink(tmp_file_path)
+            
+            # 數據清理
+            original_shape = df.shape
+            df = df.dropna()
+            
+            # 尋找正確的欄位名稱
+            company_cols = [col for col in df.columns if "公司" in col or "代號" in col or "股票" in col]
+            emission_cols = [col for col in df.columns if "排放" in col]
+            
+            # 自動識別欄位
+            company_col = "公司代號"
+            scope1_col = "範疇一排放量(公噸CO2e)"
+            scope2_col = "範疇二排放量(公噸CO2e)"
+            
+            if company_col not in df.columns and company_cols:
+                company_col = company_cols[0]
+            
+            if scope1_col not in df.columns:
+                scope1_candidates = [col for col in emission_cols if "範疇一" in col or "Scope1" in col]
+                if scope1_candidates:
+                    scope1_col = scope1_candidates[0]
+            
+            if scope2_col not in df.columns:
+                scope2_candidates = [col for col in emission_cols if "範疇二" in col or "Scope2" in col]
+                if scope2_candidates:
+                    scope2_col = scope2_candidates[0]
+            
+            # 轉換數值格式
+            if scope1_col in df.columns:
+                df[scope1_col] = pd.to_numeric(df[scope1_col], errors='coerce')
+            if scope2_col in df.columns:
+                df[scope2_col] = pd.to_numeric(df[scope2_col], errors='coerce')
+            
+            # 移除轉換後的空值
+            available_cols = [col for col in [scope1_col, scope2_col, company_col] if col in df.columns]
+            df = df.dropna(subset=available_cols)
+            
+            return df, original_shape, company_col, scope1_col, scope2_col, company_cols, emission_cols
+    
+    except Exception as e:
+        st.error(f"載入數據時發生錯誤: {str(e)}")
+        return None, None, None, None, None, None, None
 
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
+# 載入數據
+data_result = load_data()
+if data_result[0] is not None:
+    df, original_shape, company_col, scope1_col, scope2_col, company_cols, emission_cols = data_result
+    
+    # 顯示數據基本信息
+    col1, col2, col3 = st.columns(3)
+    with col1:
+        st.metric("原始數據筆數", original_shape[0])
+    with col2:
+        st.metric("處理後數據筆數", df.shape[0])
+    with col3:
+        st.metric("總欄位數", df.shape[1])
+    
+    # 側邊欄控制項
+    st.sidebar.subheader("📊 圖表選項")
+    
+    # 圖表類型選擇
+    chart_types = st.sidebar.multiselect(
+        "選擇要顯示的圖表:",
+        ["旭日圖", "雙層圓餅圖", "散點圖", "綜合旭日圖"],
+        default=["旭日圖", "雙層圓餅圖"]
+    )
+    
+    # 公司數量選擇
+    max_companies = min(30, len(df))
+    num_companies = st.sidebar.slider(
+        "顯示公司數量:",
+        min_value=5,
+        max_value=max_companies,
+        value=min(15, max_companies),
+        step=5
+    )
+    
+    # 顯示數據統計
+    if st.sidebar.checkbox("顯示數據統計", value=True):
+        st.subheader("📈 數據統計摘要")
+        
+        if all(col in df.columns for col in [scope1_col, scope2_col]):
+            col1, col2 = st.columns(2)
+            
+            with col1:
+                st.write("**範疇一排放量統計:**")
+                scope1_stats = df[scope1_col].describe()
+                st.write(f"- 平均值: {scope1_stats['mean']:.2f} 公噸CO2e")
+                st.write(f"- 中位數: {scope1_stats['50%']:.2f} 公噸CO2e")
+                st.write(f"- 最大值: {scope1_stats['max']:.2f} 公噸CO2e")
+                st.write(f"- 最小值: {scope1_stats['min']:.2f} 公噸CO2e")
+            
+            with col2:
+                st.write("**範疇二排放量統計:**")
+                scope2_stats = df[scope2_col].describe()
+                st.write(f"- 平均值: {scope2_stats['mean']:.2f} 公噸CO2e")
+                st.write(f"- 中位數: {scope2_stats['50%']:.2f} 公噸CO2e")
+                st.write(f"- 最大值: {scope2_stats['max']:.2f} 公噸CO2e")
+                st.write(f"- 最小值: {scope2_stats['min']:.2f} 公噸CO2e")
+    
+    # 圖表生成函數
+    def create_sunburst_chart(df, num_companies):
+        """創建旭日圖"""
+        if all(col in df.columns for col in [company_col, scope1_col, scope2_col]):
+            df_top = df.nlargest(num_companies, scope1_col)
+            sunburst_data = []
+            
+            for _, row in df_top.iterrows():
+                company = str(row[company_col])
+                scope1 = row[scope1_col]
+                scope2 = row[scope2_col]
+                
+                sunburst_data.extend([
+                    dict(ids=f"公司-{company}", labels=f"公司 {company}", parents="", values=scope1 + scope2),
+                    dict(ids=f"範疇一-{company}", labels=f"範疇一: {scope1:.0f}", parents=f"公司-{company}", values=scope1),
+                    dict(ids=f"範疇二-{company}", labels=f"範疇二: {scope2:.0f}", parents=f"公司-{company}", values=scope2)
+                ])
+            
+            fig_sunburst = go.Figure(go.Sunburst(
+                ids=[d['ids'] for d in sunburst_data],
+                labels=[d['labels'] for d in sunburst_data],
+                parents=[d['parents'] for d in sunburst_data],
+                values=[d['values'] for d in sunburst_data],
+                branchvalues="total",
+                hovertemplate='<b>%{label}</b><br>排放量: %{value:.0f} 公噸CO2e<extra></extra>',
+                maxdepth=3
+            ))
+            
+            fig_sunburst.update_layout(
+                title=f"碳排放量旭日圖 (前{num_companies}家公司)",
+                font_size=12,
+                height=600
+            )
+            
+            return fig_sunburst
+        return None
+    
+    def create_nested_pie_chart(df, num_companies):
+        """創建雙層圓餅圖"""
+        if all(col in df.columns for col in [company_col, scope1_col, scope2_col]):
+            df_top = df.nlargest(num_companies, scope1_col)
+            
+            fig = make_subplots(
+                rows=1, cols=2,
+                specs=[[{"type": "pie"}, {"type": "pie"}]],
+                subplot_titles=("範疇一排放量", "範疇二排放量")
+            )
+            
+            fig.add_trace(go.Pie(
+                labels=df_top[company_col],
+                values=df_top[scope1_col],
+                name="範疇一",
+                hovertemplate='<b>%{label}</b><br>範疇一排放量: %{value:.0f} 公噸CO2e<br>佔比: %{percent}<extra></extra>',
+                textinfo='label+percent',
+                textposition='auto'
+            ), row=1, col=1)
+            
+            fig.add_trace(go.Pie(
+                labels=df_top[company_col],
+                values=df_top[scope2_col],
+                name="範疇二",
+                hovertemplate='<b>%{label}</b><br>範疇二排放量: %{value:.0f} 公噸CO2e<br>佔比: %{percent}<extra></extra>',
+                textinfo='label+percent',
+                textposition='auto'
+            ), row=1, col=2)
+            
+            fig.update_layout(
+                title_text=f"碳排放量圓餅圖比較 (前{num_companies}家公司)",
+                showlegend=True,
+                height=600
+            )
+            
+            return fig
+        return None
+    
+    def create_scatter_plot(df):
+        """創建散點圖"""
+        if all(col in df.columns for col in [company_col, scope1_col, scope2_col]):
+            fig_scatter = px.scatter(
+                df,
+                x=scope1_col,
+                y=scope2_col,
+                hover_data=[company_col],
+                title="範疇一 vs 範疇二排放量散點圖",
+                labels={
+                    scope1_col: "範疇一排放量 (公噸CO2e)",
+                    scope2_col: "範疇二排放量 (公噸CO2e)"
+                },
+                hover_name=company_col
+            )
+            
+            fig_scatter.update_layout(height=600)
+            return fig_scatter
+        return None
+    
+    def create_comprehensive_sunburst(df, num_companies):
+        """創建綜合旭日圖"""
+        if all(col in df.columns for col in [company_col, scope1_col, scope2_col]):
+            df_copy = df.copy()
+            df_copy['total_emission'] = df_copy[scope1_col] + df_copy[scope2_col]
+            df_copy['emission_level'] = pd.cut(df_copy['total_emission'], 
+                                            bins=[0, 1000, 5000, 20000, float('inf')], 
+                                            labels=['低排放(<1K)', '中排放(1K-5K)', '高排放(5K-20K)', '超高排放(>20K)'])
+            
+            sunburst_data = []
+            companies_per_level = max(1, num_companies // 4)
+            
+            for level in df_copy['emission_level'].unique():
+                if pd.isna(level):
+                    continue
+                level_companies = df_copy[df_copy['emission_level'] == level].nlargest(companies_per_level, 'total_emission')
+                
+                for _, row in level_companies.iterrows():
+                    company = str(row[company_col])
+                    scope1 = row[scope1_col]
+                    scope2 = row[scope2_col]
+                    total = scope1 + scope2
+                    
+                    sunburst_data.extend([
+                        dict(ids=str(level), labels=str(level), parents="", values=total),
+                        dict(ids=f"{level}-{company}", labels=f"{company}", parents=str(level), values=total),
+                        dict(ids=f"{level}-{company}-範疇一", labels=f"範疇一({scope1:.0f})", 
+                             parents=f"{level}-{company}", values=scope1),
+                        dict(ids=f"{level}-{company}-範疇二", labels=f"範疇二({scope2:.0f})", 
+                             parents=f"{level}-{company}", values=scope2)
+                    ])
+            
+            fig_comprehensive = go.Figure(go.Sunburst(
+                ids=[d['ids'] for d in sunburst_data],
+                labels=[d['labels'] for d in sunburst_data],
+                parents=[d['parents'] for d in sunburst_data],
+                values=[d['values'] for d in sunburst_data],
+                branchvalues="total",
+                hovertemplate='<b>%{label}</b><br>排放量: %{value:.0f} 公噸CO2e<extra></extra>',
+                maxdepth=4
+            ))
+            
+            fig_comprehensive.update_layout(
+                title="分級碳排放量旭日圖",
+                font_size=10,
+                height=700
+            )
+            
+            return fig_comprehensive
+        return None
+    
+    # 顯示選中的圖表
+    st.subheader("📊 互動式圖表")
+    
+    if "旭日圖" in chart_types:
+        st.write("### 🌞 旭日圖")
+        fig1 = create_sunburst_chart(df, num_companies)
+        if fig1:
+            st.plotly_chart(fig1, use_container_width=True)
+        else:
+            st.error("無法創建旭日圖，缺少必要欄位")
+    
+    if "雙層圓餅圖" in chart_types:
+        st.write("### 🥧 雙層圓餅圖")
+        fig2 = create_nested_pie_chart(df, num_companies)
+        if fig2:
+            st.plotly_chart(fig2, use_container_width=True)
+        else:
+            st.error("無法創建圓餅圖，缺少必要欄位")
+    
+    if "散點圖" in chart_types:
+        st.write("### 📈 散點圖")
+        fig3 = create_scatter_plot(df)
+        if fig3:
+            st.plotly_chart(fig3, use_container_width=True)
+        else:
+            st.error("無法創建散點圖，缺少必要欄位")
+    
+    if "綜合旭日圖" in chart_types:
+        st.write("### 🌟 綜合旭日圖")
+        fig4 = create_comprehensive_sunburst(df, num_companies)
+        if fig4:
+            st.plotly_chart(fig4, use_container_width=True)
+        else:
+            st.error("無法創建綜合旭日圖，缺少必要欄位")
+    
+    # 顯示原始數據
+    if st.sidebar.checkbox("顯示原始數據"):
+        st.subheader("📋 原始數據預覽")
+        st.dataframe(df.head(100), use_container_width=True)
+    
+    # 數據下載功能
+    if st.sidebar.button("下載處理後數據"):
+        csv = df.to_csv(index=False, encoding='utf-8-sig')
+        st.sidebar.download_button(
+            label="💾 下載 CSV 文件",
+            data=csv,
+            file_name="carbon_emission_data.csv",
+            mime="text/csv"
+        )
+    
+    # 偵錯信息
+    if st.sidebar.checkbox("顯示偵錯信息"):
+        st.subheader("🔧 偵錯信息")
+        st.write("**識別的欄位:**")
+        st.write(f"- 公司欄位: {company_col}")
+        st.write(f"- 範疇一欄位: {scope1_col}")
+        st.write(f"- 範疇二欄位: {scope2_col}")
+        st.write("**所有可用欄位:**")
+        st.write(df.columns.tolist())
 
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
+else:
+    st.error("無法載入數據，請檢查網路連接或數據源。")
 
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+# 頁面底部信息
+st.markdown("---")
+st.markdown(
+    """
+    **數據來源:** 台灣證券交易所公開資訊觀測站  
+    **更新時間:** 根據數據源自動更新  
+    **製作:** Streamlit 碳排放數據分析應用
+    """
+)