app.py

import os
import streamlit as st
import pandas as pd
import requests
import time

# ✅ Safely get the API key from the environment or fallback to empty string
API_KEY = os.getenv("SERPAPI_API_KEY", "")

# ✅ Ensure API_KEY is available
if not API_KEY:
    raise ValueError("API key is missing. Please set the SERPAPI_API_KEY environment variable.")

# Function to get collaboration data using SERPAPI
def get_collaboration_data(competitors, target="Virgin Media"):
    results = []
    
    for comp in competitors:
        query = f"{comp} {target} partnership OR collaboration"
        params = {
            "q": query,
            "api_key": API_KEY,
            "engine": "google",  # You can switch to other engines if needed
            "num": 5  # Number of results to return per query
        }

        print(f"Searching: {query}")  # Log the query for tracking
        try:
            # Make a request to the SERP API
            response = requests.get("https://serpapi.com/search", params=params)
            response.raise_for_status()  # Check for request errors (e.g., 4xx, 5xx)
            
            # Parse the JSON response
            data = response.json()

            # Extract the organic results
            for item in data.get("organic_results", []):
                results.append({
                    "Competitor": comp,
                    "Title": item.get("title", "No title available"),
                    "Link": item.get("link", "No link available"),
                    "Snippet": item.get("snippet", "No snippet available")
                })
        except requests.exceptions.RequestException as e:
            print(f"Error fetching data for {comp}: {e}")
        except ValueError as e:
            print(f"Error parsing JSON response for {comp}: {e}")
        
        time.sleep(1)  # ✅ Respect API rate limits

    # Return results as a DataFrame
    return pd.DataFrame(results)

# Set the page configuration
st.set_page_config(page_title="Virgin Media Collaboration Finder", layout="centered")

# Page title
st.title("🕵️ Virgin Media - Competitor Collaboration Finder")
st.markdown("Use this tool to find which competitors have worked with **Virgin Media**.")

# URL to the competitors CSV on Hugging Face
csv_url = "https://huggingface.co/spaces/Satyam0077/anapan_ai_Competitor_Task/resolve/main/competitors.csv"

# Load competitor list from Hugging Face URL
try:
    competitors_df = pd.read_csv(csv_url)
    competitors = competitors_df['Company'].dropna().tolist()
except Exception as e:
    st.error(f"❌ Failed to load the CSV: {e}")
    st.stop()

# Run search when button is clicked
if st.button("🔍 Run Collaboration Search"):
    with st.spinner("Searching... This may take a few moments..."):
        result_df = get_collaboration_data(competitors)  # This will now work as the function is defined
        if not result_df.empty:
            # Save results to the "results" folder
            os.makedirs("results", exist_ok=True)
            result_csv_path = os.path.join("results", "virgin_collab_results.csv")
            result_df.to_csv(result_csv_path, index=False)
            st.success("✅ Search Completed!")
            st.dataframe(result_df)
        else:
            st.warning("⚠️ No collaboration data found. Nothing was saved.")

# Show saved results when button is clicked
if st.button("📂 Show Saved Results"):
    try:
        saved_df = pd.read_csv("results/virgin_collab_results.csv")
        if saved_df.empty:
            st.warning("⚠️ The saved file is empty. Try running the search again.")
        else:
            st.success("✅ Loaded saved results.")
            st.dataframe(saved_df)
    except FileNotFoundError:
        st.warning("⚠️ No saved results found. Please run the search first.")
    except pd.errors.EmptyDataError:
        st.warning("⚠️ The saved file exists but is empty or invalid.")