Create app.py

app.py

@@ -0,0 +1,160 @@
+import streamlit as st
+import requests
+from bs4 import BeautifulSoup
+import pandas as pd
+import urllib.parse
+from geopy.geocoders import Nominatim
+import plotly.express as px
+
+# Step 1: Fetch hospital URLs and title from the list page
+def get_hospital_urls_and_title(list_url):
+    try:
+        response = requests.get(list_url)
+        response.raise_for_status()
+        soup = BeautifulSoup(response.text, 'html.parser')
+        
+        # Fetch the title of the list page
+        title_element = soup.find('h1', class_='t-intro__title')
+        page_title = title_element.get_text(strip=True) if title_element else 'N/A'
+        
+        # Fetch all hospital URLs
+        hospital_links = soup.find_all('a', class_='m-link')
+        hospital_urls = [urllib.parse.urljoin(list_url, link['href']) for link in hospital_links]
+        
+        return page_title, hospital_urls
+    except Exception as e:
+        st.error(f"Error fetching hospital list: {e}")
+        return 'N/A', []
+
+# Step 2: Fetch hospital details from each hospital's page
+def get_hospital_details(url):
+    try:
+        response = requests.get(url)
+        response.raise_for_status()
+        soup = BeautifulSoup(response.text, 'html.parser')
+        
+        # Fetch the title
+        title_element = soup.find('h1', class_='t-intro__title')
+        title = title_element.get_text(strip=True) if title_element else 'N/A'
+        
+        # Fetch the phone number
+        phone_element = soup.find('a', class_='t-font-large')
+        phone = phone_element.get_text(strip=True) if phone_element else 'N/A'
+        
+        # Fetch the address
+        address_element = soup.find('a', class_='t-font-medium')
+        address = address_element.get_text(strip=True) if address_element else 'N/A'
+        
+        # Fetch the rating
+        rating_element = soup.find('span', class_='t-intro__recommand')
+        rating = rating_element.get_text(strip=True) if rating_element else 'N/A'
+        
+        return {
+            'Title': title,
+            'Phone': phone,
+            'Address': address,
+            'Rating': rating
+        }
+    except Exception as e:
+        st.error(f"Error fetching details from {url}: {e}")
+        return {
+            'Title': 'Error',
+            'Phone': 'Error',
+            'Address': 'Error',
+            'Rating': 'Error'
+        }
+
+# Main function to fetch and display hospital data based on city selection and category
+def fetch_and_display_hospital_data(city_code, category_number):
+    base_url = 'https://www.tw-animal.com'
+    list_url = f'https://www.tw-animal.com/list/pet/{city_code}/{category_number}.html'
+    
+    # Step 1: Get the list page title and hospital URLs
+    page_title, hospital_urls = get_hospital_urls_and_title(list_url)
+    
+    # Step 2: Get details for each hospital
+    hospital_data = []
+    for url in hospital_urls:
+        details = get_hospital_details(url)
+        hospital_data.append(details)
+    
+    # Convert the data to a DataFrame and filter the results
+    df = pd.DataFrame(hospital_data)
+    df_filtered = df.iloc[2:10].reset_index(drop=True)
+    
+    return page_title, df_filtered
+
+# Streamlit interface setup
+def hospital_info_interface(city, category_number):
+    city_codes = {
+        "台北": "07",
+        "新北": "08",
+        "桃園": "09",
+        "台中": "12",
+        "台南": "17",
+        "高雄": "18"
+    }
+    
+    # Validate the category number is within the valid range
+    if not (1 <= category_number <= 1024):
+        st.error("Category number must be between 1 and 1024.")
+        return "Error", pd.DataFrame()
+
+    city_code = city_codes[city]
+    page_title, df = fetch_and_display_hospital_data(city_code, str(category_number))
+    
+    return page_title, df
+
+# Geocode addresses to latitude and longitude
+def geocode_addresses(df):
+    geolocator = Nominatim(user_agent="hospital_locator")
+    latitudes = []
+    longitudes = []
+
+    for address in df['Address']:
+        try:
+            location = geolocator.geocode(address)
+            if location:
+                latitudes.append(location.latitude)
+                longitudes.append(location.longitude)
+            else:
+                latitudes.append(None)
+                longitudes.append(None)
+        except Exception as e:
+            st.error(f"Error geocoding address {address}: {e}")
+            latitudes.append(None)
+            longitudes.append(None)
+
+    df['Latitude'] = latitudes
+    df['Longitude'] = longitudes
+    return df
+
+# Streamlit app setup
+def main():
+    st.title("台灣寵物醫院資料查詢")
+
+    city_list = ["台北", "新北", "桃園", "台中", "台南", "高雄"]
+    city = st.selectbox("選擇縣市", city_list)
+    category_number = st.number_input("輸入分類編號 (1-1024)", min_value=1, max_value=1024, value=1)
+
+    if st.button("查詢"):
+        page_title, df = hospital_info_interface(city, category_number)
+        
+        st.subheader(f"頁面標��: {page_title}")
+        st.dataframe(df)
+        
+        if not df.empty:
+            df = geocode_addresses(df)
+            fig = px.scatter_mapbox(df,
+                                    lat="Latitude",
+                                    lon="Longitude",
+                                    hover_name="Title",
+                                    hover_data=["Phone", "Address", "Rating"],
+                                    zoom=10,
+                                    height=600)
+            fig.update_layout(mapbox_style="open-street-map")
+            fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
+            st.plotly_chart(fig)
+
+if __name__ == "__main__":
+    main()