import os
import requests
import gradio as gr

# This is the API endpoint for a pre-trained sentiment analysis model.
# This specific model is a DistilBERT model fine-tuned for sentiment analysis.
# The Hugging Face Inference API provides a generous free tier.
# You can find other models here: https://huggingface.co/models?pipeline_tag=text-classification&sort=downloads
API_URL = "https://api-inference.huggingface.co/models/distilbert-base-uncased-finetuned-sst-2-english"

# IMPORTANT: You need a Hugging Face API token to use the Inference API.
# 1. Go to https://huggingface.co/settings/tokens
# 2. Create a new token with "Read" access.
# 3. Copy the token and set it as an environment variable named 'HUGGING_FACE_API_TOKEN'.
#    For example, in your terminal, run:
#    export HUGGING_FACE_API_TOKEN="YOUR_API_TOKEN_HERE"
#    (For Windows, use: set HUGGING_FACE_API_TOKEN="YOUR_API_TOKEN_HERE")
# 4. The script will automatically load this token.
API_TOKEN = os.getenv("HUGGING_FACE_API_TOKEN")

# The headers for the API request, including the authorization token.
headers = {"Authorization": f"Bearer {API_TOKEN}"}

def analyze_sentiment(text):
    """
    Analyzes the sentiment of a given text using the Hugging Face Inference API.

    Args:
        text (str): The input text to analyze.

    Returns:
        str: A formatted string with the sentiment and confidence score,
             or an error message if the API call fails.
    """
    if not API_TOKEN:
        return "ERROR: Hugging Face API token not found. Please set the HUGGING_FACE_API_TOKEN environment variable."

    if not text:
        return "Please enter some text to analyze."

    payload = {"inputs": text}

    try:
        response = requests.post(API_URL, headers=headers, json=payload)
        response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)
        result = response.json()

        # The API response is a list of lists. We'll grab the first item.
        # Example response: [[{'label': 'POSITIVE', 'score': 0.9998782}, {'label': 'NEGATIVE', 'score': 0.00012176}]]
        sentiment_data = result[0]
        
        # Find the sentiment with the highest score
        top_sentiment = max(sentiment_data, key=lambda x: x['score'])
        label = top_sentiment['label']
        score = top_sentiment['score'] * 100  # Convert to percentage

        return f"Sentiment: {label.upper()}\nConfidence: {score:.2f}%"

    except requests.exceptions.RequestException as e:
        # Handle network or API errors
        return f"ERROR: Failed to connect to the API. Check your token and network connection. Details: {e}"
    except Exception as e:
        # Handle other potential errors
        return f"ERROR: An unexpected error occurred. Details: {e}"

# --- Gradio User Interface with Custom Styling ---

# Custom CSS for a cute and aesthetic theme
css = """
body {
    background: linear-gradient(135deg, #f7d9e2, #c7e0ff); /* Soft gradient background */
    font-family: 'Comic Sans MS', 'Arial', sans-serif;
}
.gradio-container {
    background-color: rgba(255, 255, 255, 0.7); /* Semi-transparent white background */
    border-radius: 20px;
    box-shadow: 0 10px 30px rgba(0, 0, 0, 0.1); /* Soft shadow */
    padding: 20px;
}
.gr-button {
    background-color: #ff99cc; /* Pink button */
    border: none;
    color: white;
    font-size: 1.1em;
    border-radius: 15px;
    transition: transform 0.2s ease-in-out;
}
.gr-button:hover {
    background-color: #ff66b2; /* Darker pink on hover */
    transform: scale(1.05); /* Slight grow effect */
}
.label-text {
    font-size: 1.2em;
    font-weight: bold;
    color: #333;
}
.gr-text-box textarea {
    border-radius: 10px;
    border: 1px solid #ccc;
    background-color: #fefefe;
    padding: 10px;
}
"""

# Create the Gradio interface using gr.Blocks for a custom layout
with gr.Blocks(theme=gr.themes.Soft(), css=css) as demo:
    gr.Markdown("# 🌸 The Happy-Go-Lucky Sentiment Analyzer 🌸")
    gr.Markdown("A cute little AI friend that tells you the mood of your text!")
    
    with gr.Row():
        with gr.Column(scale=1):
            gr.Image(
                value="https://placehold.co/200x200/ffb3e6/333333?text=Cute+AI", 
                label="Your AI Friend",
                show_label=True,
                show_download_button=False
            )
        with gr.Column(scale=3):
            input_textbox = gr.Textbox(
                lines=5, 
                label="Tell me something!",
                placeholder="Type your thoughts here, and I'll analyze the sentiment...",
                info="I can tell you if your text is positive or negative."
            )
            analyze_button = gr.Button("💖 Analyze!")
    
    output_label = gr.Label(label="Result")

    # Event listener for the button click
    analyze_button.click(fn=analyze_sentiment, inputs=input_textbox, outputs=output_label)

# Launch the Gradio app
if __name__ == "__main__":
    demo.launch()