from smolagents import CodeAgent,DuckDuckGoSearchTool, HfApiModel,load_tool,tool
import datetime
import requests
import pytz
import yaml
from tools.final_answer import FinalAnswerTool

from Gradio_UI import GradioUI

# My first tool !
@tool
def coin_predictor_tool()-> str: #it's import to specify the return type
    #Keep this format for the description / args / args description but feel free to modify the tool
!pip install -q llama-index requests

# Step 2: Define the Agent Tool
from llama_index.core.tools import FunctionTool
import requests
from datetime import datetime

def coin_predictor_tool() -> str:
    """
    Predicts Bitcoin's price for today (March 10, 2025) based on available data or trends.
    Returns a formatted string with the predicted price and reasoning.
    """
    # Simulate current date
    today = datetime(2025, 3, 10)
    current_date_str = today.strftime('%Y-%m-%d')

    # Placeholder for real data (replace with API call in practice)
    # Example: Fetch last 7 days of BTC prices from an API
    try:
        # Hypothetical API call (e.g., CoinGecko)
        url = "https://api.coingecko.com/api/v3/coins/bitcoin/market_chart"
        params = {
            "vs_currency": "usd",
            "days": "7",  # Last 7 days
            "interval": "daily"
        }
        response = requests.get(url, params=params, timeout=10)
        response.raise_for_status()
        data = response.json()
        
        # Extract prices (simulated here; replace with real data)
        prices = data.get("prices", [])  # [[timestamp, price], ...]
        if not prices:
            raise ValueError("No price data available")
        
        # Simple prediction: Average of last 7 days + trend adjustment
        recent_prices = [price[1] for price in prices[-7:]]  # Last 7 days' closing prices
        avg_price = sum(recent_prices) / len(recent_prices)
        
        # Assume a trend (e.g., based on last day's change)
        last_change = recent_prices[-1] - recent_prices[-2]
        trend_factor = 1 + (last_change / recent_prices[-2])  # % change applied
        predicted_price = avg_price * trend_factor

    except Exception as e:
        # Fallback simulation if API fails or for demo purposes
        print(f"API error: {e}. Using simulated data.")
        # Simulated data based on recent trends (e.g., from your provided context)
        avg_price = 91981  # From Web ID 5, March 10, 2025 price
        trend_factor = 1.0418  # +4.18% from Web ID 7's 24h change
        predicted_price = avg_price * trend_factor

    # Format the output
    output = (
        f"Bitcoin Price Prediction for {current_date_str}:\n"
        f"Predicted Price: ${predicted_price:,.2f} USD\n"
        f"Reasoning: Based on a 7-day average of ${avg_price:,.2f} with a "
        f"{(trend_factor-1)*100:.2f}% trend adjustment from recent data."
    )
    return output

# Create the tool
bitcoin_price_tool = FunctionTool.from_defaults(
    fn=coin_predictor_tool,
    name="bitcoin_price_predictor",
    description="Predicts Bitcoin's price for today (March 10, 2025) using recent trends or API data."
)

# Step 3: Test the tool standalone
print(bitcoin_price_tool())


@tool
def get_current_time_in_timezone(timezone: str) -> str:
    """A tool that fetches the current local time in a specified timezone.
    Args:
        timezone: A string representing a valid timezone (e.g., 'America/New_York').
    """
    try:
        # Create timezone object
        tz = pytz.timezone(timezone)
        # Get current time in that timezone
        local_time = datetime.datetime.now(tz).strftime("%Y-%m-%d %H:%M:%S")
        return f"The current local time in {timezone} is: {local_time}"
    except Exception as e:
        return f"Error fetching time for timezone '{timezone}': {str(e)}"


final_answer = FinalAnswerTool()

# If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
# model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud' 

model = HfApiModel(
max_tokens=2096,
temperature=0.5,
model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
custom_role_conversions=None,
)


# Import tool from Hub
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)

with open("prompts.yaml", 'r') as stream:
    prompt_templates = yaml.safe_load(stream)
    
agent = CodeAgent(
    model=model,
    tools=[final_answer], ## add your tools here (don't remove final answer)
    max_steps=6,
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name=None,
    description=None,
    prompt_templates=prompt_templates
)


GradioUI(agent).launch()