import gradio as gr
import os
from groq import Groq
from deep_translator import GoogleTranslator

# Set up Groq API client
client = Groq(
    api_key=os.getenv("GROQ_API_KEY"),  # Ensure you add this key to your environment variables
)

# Function to fetch team and player list from LLM (using Groq API)
def get_team_and_players():
    chat_completion = client.chat.completions.create(
        messages=[
            {"role": "user", "content": "List all MLB teams and their top players."}
        ],
        model="llama-3.3-70b-versatile",  # You can adjust the model here
    )
    
    return chat_completion.choices[0].message.content.strip()

# Function to generate team overview
def get_team_overview(team):
    chat_completion = client.chat.completions.create(
        messages=[
            {"role": "user", "content": f"Provide an overview of the {team} MLB team, including recent performance and standings."}
        ],
        model="llama-3.3-70b-versatile",
    )
    
    return chat_completion.choices[0].message.content.strip()

# Function for season predictions
def predict_season_outcomes(team):
    chat_completion = client.chat.completions.create(
        messages=[
            {"role": "user", "content": f"Predict the potential season outcomes for the {team} based on their current performance."}
        ],
        model="llama-3.3-70b-versatile",
    )
    
    return chat_completion.choices[0].message.content.strip()

# Function for real-time strategy insights
def real_time_tooltips(game_event):
    chat_completion = client.chat.completions.create(
        messages=[
            {"role": "user", "content": f"Explain the strategy behind the following baseball play: {game_event}"}
        ],
        model="llama-3.3-70b-versatile",
    )
    
    return chat_completion.choices[0].message.content.strip()

# Gradio app interface
def create_gradio_interface():
    with gr.Blocks() as demo:
        gr.Markdown("# MLB Fan Engagement App")
        
        with gr.Tab("Team and Player Selection"):
            team_dropdown = gr.Dropdown(choices=[], label="Select Team")
            player_dropdown = gr.Dropdown(choices=[], label="Select Player")
            
            # Update teams and players dynamically using LLM
            def update_teams_and_players():
                team_and_players = get_team_and_players()
                teams = team_and_players.split("\n")  # Assuming each team is on a new line
                team_dropdown.update(choices=teams)
                return team_dropdown, player_dropdown
            
            # Load teams and players when app starts
            update_teams_and_players_button = gr.Button("Load Teams and Players")
            update_teams_and_players_button.click(update_teams_and_players, inputs=[], outputs=[team_dropdown, player_dropdown])
        
        with gr.Tab("Team Overview"):
            team_dropdown_overview = gr.Dropdown(choices=[], label="Select Team")
            overview_output = gr.Textbox(label="Team Overview")
            team_dropdown_overview.change(
                get_team_overview, inputs=team_dropdown_overview, outputs=overview_output
            )
        
        with gr.Tab("Season Predictions"):
            team_dropdown_predictions = gr.Dropdown(choices=[], label="Select Team")
            predictions_output = gr.Textbox(label="Season Predictions")
            team_dropdown_predictions.change(
                predict_season_outcomes, inputs=team_dropdown_predictions, outputs=predictions_output
            )
        
        with gr.Tab("Real-Time Strategy Insights"):
            game_event_input = gr.Textbox(
                label="Describe the game event (e.g., 'Why did the batter bunt in the 8th inning?')"
            )
            strategy_output = gr.Textbox(label="Strategy Explanation")
            game_event_input.submit(real_time_tooltips, inputs=game_event_input, outputs=strategy_output)
        
        demo.launch()

# Run the app
create_gradio_interface()