app.py

import os
import requests
import gradio as gr
from langchain_groq import ChatGroq
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

def get_weather_and_insight(city):
    # Fetching weather data
    url = f'http://api.openweathermap.org/data/2.5/weather?q={city}&appid={os.getenv("OPENWEATHER_API_KEY")}&units=metric'
    res = requests.get(url)
    data = res.json()

    if res.status_code != 200:
        return "Failed to retrieve weather data. Please check the city name and try again."

    humidity = data.get('main', {}).get('humidity')
    temp = data.get('main', {}).get('temp')
    description = data.get('weather', [{}])[0].get('description')

    # Generating a creative weather summary using LangChain LLM
    groq_api_key = os.getenv("GROQ_API_KEY")
    system = f"You are a local weather reporter bringing the latest update for {city}. Right now, it's \n🌡️ {temp}°C, degrees Celsius with about 💧 {humidity}% percent humidity. The weather of the city can be best described as {description} with a short fun fact."
    human = "{text}"
    prompt = ChatPromptTemplate.from_messages(
        [("system", system), ("human", human)]
    )
    chat = ChatGroq(api_key=groq_api_key, model_name="llama3-70b-8192")
    chain = prompt | chat | StrOutputParser()
    output = chain.invoke({"text": city})

    # Adding emojis for visual flair
    weather_emoji = "☀️" if "clear" in description else "☁️" if "cloud" in description else "🌧️" if "rain" in description else "❄️" if "snow" in description else "🌫️"
    
    # Creating the final response string
    final_response = f"🌍 {city.upper()} Weather Insight: {output} {weather_emoji}"

    return final_response

# Gradio Interface
iface = gr.Interface(
    fn=get_weather_and_insight,
    inputs=gr.Textbox(label="Enter City Name", placeholder="Type city here..."),
    outputs=gr.Textbox(label="Weather Report and Insight"),
    title="WeatherAssistantApp",
    description="Enter a city name to get a detailed weather report with an AI-generated insight."
)

iface.launch()