Update app.py

app.py

@@ -1,168 +1,159 @@
+# file_path: app.py
+import streamlit as st
+import requests
+from bs4 import BeautifulSoup
 import pandas as pd
+from google.oauth2.service_account import Credentials
+import gspread
 import plotly.express as px
 import plotly.graph_objects as go
-import numpy as np
 from plotly.subplots import make_subplots
-import streamlit as st
 
-# 讀取數據
+# Google Sheets credentials
+SCOPE = ['https://www.googleapis.com/auth/spreadsheets']
+SERVICE_ACCOUNT_FILE = "realtime-441511-f5708eabdf26.json"
+SPREADSHEET_URL = "https://docs.google.com/spreadsheets/d/1tIsXCbB8P6ZxdnZNnv7S7BBWbbT7lrSjW990zG-vQAA/edit?gid=0#gid=0"
+
+# Streamlit app
+st.title("Booking.com 台南飯店資料爬取與分析")
+st.sidebar.header("功能選擇")
+mode = st.sidebar.selectbox("選擇模式", ["資料爬取", "資料視覺化", "上傳至 Google Sheet"])
+
 @st.cache_data
-def load_data():
-    df = pd.read_csv('booking_hotels_tainan.csv')
-    
-    # 數據預處理
-    def clean_rating(x):
-        if pd.isna(x) or x == '無評分':
-            return 0
-        return float(str(x).replace('分數', '').replace('分', ''))
-    
-    # 清理評分數據
-    df['評分'] = df['評分'].apply(clean_rating)
-    
-    # 創建價格區間
-    df['價格區間'] = pd.qcut(df['價格'], 
-                         q=3, 
-                         labels=['經濟型', '中價位', '高價位'])
-    return df
+def scrape_booking_hotel():
+    url = "https://www.booking.com/searchresults.zh-tw.html"
+    headers = {
+        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
+        'Accept-Language': 'zh-TW,zh;q=0.9,en-US;q=0.8,en;q=0.7',
+    }
+    params = {
+        'ss': '台南',
+        'checkin': '2024-11-16',
+        'checkout': '2024-11-17',
+        'group_adults': '2',
+        'no_rooms': '1',
+        'group_children': '0',
+        'dest_id': '-2637868',
+        'dest_type': 'city'
+    }
+    try:
+        response = requests.get(url, headers=headers, params=params)
+        response.raise_for_status()
+        soup = BeautifulSoup(response.text, 'html.parser')
+        hotels_data = []
+        hotel_cards = soup.find_all('div', {'data-testid': 'property-card'})
+
+        for hotel in hotel_cards:
+            try:
+                name_elem = hotel.find('div', {'data-testid': 'title', 'class': 'f6431b446c'})
+                name = name_elem.text.strip() if name_elem else "無資料"
+                price_elem = hotel.find('span', {
+                    'data-testid': 'price-and-discounted-price',
+                    'class': 'f6431b446c'
+                })
+                price = price_elem.text.strip() if price_elem else "無資料"
+                price = price.replace('TWD', '').replace(' ', '').replace(',', '').strip()
+                rating_container = hotel.find('div', {'class': 'a3b8729ab1'})
+                rating_elem = rating_container.find('div', {'class': 'ac4a7896c7'}) if rating_container else None
+                rating = rating_elem.text.strip() if rating_elem else "無評分"
+                description_elem = hotel.find('div', {'data-testid': 'recommended-units'})
+                if description_elem:
+                    room_type = description_elem.find('h4', {'class': 'abf093bdfe'})
+                    room_type = room_type.text.strip() if room_type else ""
+                    bed_info = description_elem.find('div', {'class': 'abf093bdfe'})
+                    bed_info = bed_info.text.strip() if bed_info else ""
+                    cancellation = description_elem.find('strong', text='可免費取消')
+                    cancellation = "可免費取消" if cancellation else ""
+                    payment = description_elem.find('strong', text='無需訂金')
+                    payment = "無需訂金" if payment else ""
+                    description = f"{room_type} | {bed_info} | {cancellation} | {payment}".strip(' |')
+                else:
+                    description = "無說明"
+                hotels_data.append({
+                    '飯店名稱': name,
+                    '價格': price,
+                    '評分': rating,
+                    '說明': description
+                })
+            except AttributeError:
+                continue
 
-df = load_data()
+        df = pd.DataFrame(hotels_data).drop_duplicates()
+        return df
+    except requests.RequestException:
+        return pd.DataFrame()
 
-def create_price_rating_scatter():
-    """價格vs評分散點圖"""
+def clean_rating(x):
+    if pd.isna(x) or x == '無評分':
+        return 0
+    return float(str(x).replace('分數', '').replace('分', ''))
+
+def create_price_rating_scatter(df):
     fig = px.scatter(
         df,
         x='價格',
         y='評分',
         text='飯店名稱',
-        size='價格',  # 點的大小根據價格變化
-        color='評分',  # 顏色根據評分變化
+        size='價格',
+        color='評分',
         title='台南飯店價��與評分關係圖',
         labels={'價格': '房價 (TWD)', '評分': '評分 (0-10)'}
     )
-    
-    fig.update_traces(
-        textposition='top center',
-        marker=dict(sizeref=2.*max(df['價格'])/(40.**2))
-    )
-    
-    fig.update_layout(
-        height=600,
-        showlegend=True,
-        title_x=0.5,
-        title_font_size=20
-    )
-    
+    fig.update_layout(height=600, title_x=0.5)
     return fig
 
-def create_price_distribution():
-    """價格分布圖"""
+def create_price_distribution(df):
     fig = go.Figure()
-    
-    # 添加直方圖
     fig.add_trace(go.Histogram(
         x=df['價格'],
         name='價格分布',
         nbinsx=10,
         marker_color='rgb(55, 83, 109)'
     ))
-    
-    # 添加箱型圖
     fig.add_trace(go.Box(
         x=df['價格'],
         name='價格箱型圖',
         marker_color='rgb(26, 118, 255)'
     ))
-    
-    fig.update_layout(
-        title_text='台南飯店價格分布',
-        title_x=0.5,
-        title_font_size=20,
-        xaxis_title='價格 (TWD)',
-        yaxis_title='數量',
-        height=500,
-        bargap=0.2,
-        showlegend=True
-    )
-    
+    fig.update_layout(title_text='台南飯店價格分布', title_x=0.5, height=500)
     return fig
 
-def create_rating_box_by_price_range():
-    """不同價格區間的評分箱型圖"""
-    fig = px.box(
-        df,
-        x='價格區間',
-        y='評分',
-        title='不同價格區間的評分分布',
-        labels={'價格區間': '價格類型', '評分': '評分 (0-10)'},
-        color='價格區間'
-    )
-    
-    fig.update_layout(
-        title_x=0.5,
-        title_font_size=20,
-        height=500,
-        showlegend=False
-    )
-    
-    return fig
+def upload_to_google_sheets(df):
+    creds = Credentials.from_service_account_file(SERVICE_ACCOUNT_FILE, scopes=SCOPE)
+    gs = gspread.authorize(creds)
+    sheet = gs.open_by_url(SPREADSHEET_URL)
+    worksheet = sheet.get_worksheet(0)
+    df1 = df.astype(str)
+    worksheet.update([df1.columns.values.tolist()] + df1.values.tolist())
+    return "資料已成功上傳到 Google Sheet！"
 
-def create_hotel_comparison():
-    """飯店評分與價格比較圖"""
-    fig = make_subplots(specs=[[{"secondary_y": True}]])
-    
-    # 排序數據
-    df_sorted = df.sort_values('評分', ascending=True)
-    
-    # 添加評分柱狀圖
-    fig.add_trace(
-        go.Bar(
-            x=df_sorted['飯店名稱'],
-            y=df_sorted['評分'],
-            name="評分",
-            marker_color='rgb(55, 83, 109)'
-        )
-    )
-    
-    # 添加價格線圖
-    fig.add_trace(
-        go.Scatter(
-            x=df_sorted['飯店名稱'],
-            y=df_sorted['價格'],
-            name="價格",
-            marker_color='rgb(26, 118, 255)'
-        ),
-        secondary_y=True
-    )
-    
-    fig.update_layout(
-        title_text='台南飯店評分與價格比較',
-        title_x=0.5,
-        title_font_size=20,
-        height=700,
-        showlegend=True,
-        xaxis_tickangle=45
-    )
-    
-    fig.update_yaxes(title_text="評分", secondary_y=False)
-    fig.update_yaxes(title_text="價格 (TWD)", secondary_y=True)
-    
-    return fig
+if mode == "資料爬取":
+    st.header("爬取台南飯店資料")
+    if st.button("開始爬取"):
+        df = scrape_booking_hotel()
+        if not df.empty:
+            st.dataframe(df)
+            df.to_csv('booking_hotels_tainan.csv', index=False, encoding='utf-8-sig')
+            st.success("資料爬取成功，已儲存至 booking_hotels_tainan.csv")
+        else:
+            st.error("未能成功爬取資料")
 
-def main():
-    st.set_page_config(page_title="台南飯店分析", layout="wide")
-    st.title("台南飯店分析")
-    
-    scatter_fig = create_price_rating_scatter()
-    dist_fig = create_price_distribution()
-    box_fig = create_rating_box_by_price_range()
-    comparison_fig = create_hotel_comparison()
-    
-    st.plotly_chart(scatter_fig, use_container_width=True)
-    st.plotly_chart(dist_fig, use_container_width=True)
-    st.plotly_chart(box_fig, use_container_width=True)
-    st.plotly_chart(comparison_fig, use_container_width=True)
-    
-    st.write("分析完成！請查看上方產生的互動式視覺化圖表。")
+elif mode == "資料視覺化":
+    st.header("分析與視覺化")
+    try:
+        df = pd.read_csv('booking_hotels_tainan.csv', encoding='utf-8-sig')
+        df['價格'] = pd.to_numeric(df['價格'], errors='coerce')
+        df['評分'] = df['評分'].apply(clean_rating)
+        st.plotly_chart(create_price_rating_scatter(df))
+        st.plotly_chart(create_price_distribution(df))
+    except Exception as e:
+        st.error(f"讀取或分析資料時發生錯誤：{e}")
 
-if __name__ == "__main__":
-    main()
+elif mode == "上傳至 Google Sheet":
+    st.header("上傳資料至 Google Sheet")
+    try:
+        df = pd.read_csv('booking_hotels_tainan.csv', encoding='utf-8-sig')
+        result = upload_to_google_sheets(df)
+        st.success(result)
+    except Exception as e:
+        st.error(f"上傳資料時發生錯誤：{e}")