import gradio as gr
import json
import re
from typing import List, Dict

def split_text_into_sentences(text: str) -> Dict:
    """
    Split text into sentences and return as JSON
    
    Args:
        text (str): Input text paragraph
        
    Returns:
        Dict: JSON response with sentences and metadata
    """
    if not text or not text.strip():
        return {
            "status": "error",
            "message": "Empty input text",
            "sentences": [],
            "count": 0
        }
    
    # Clean the text
    text = text.strip()
    
    # Simple sentence splitting using regex
    # This pattern looks for sentence endings followed by whitespace or end of string
    sentence_pattern = r'(?<=[.!?])\s+(?=[A-Z])'
    
    # Split the text
    sentences = re.split(sentence_pattern, text)
    
    # Clean up sentences (remove extra whitespace)
    sentences = [sentence.strip() for sentence in sentences if sentence.strip()]
    
    # Create response
    response = {
        "status": "success",
        "sentences": sentences,
        "count": len(sentences),
        "original_length": len(text),
        "metadata": {
            "avg_sentence_length": sum(len(s) for s in sentences) / len(sentences) if sentences else 0,
            "longest_sentence": max(len(s) for s in sentences) if sentences else 0,
            "shortest_sentence": min(len(s) for s in sentences) if sentences else 0
        }
    }
    
    return response

def format_json_output(result: Dict) -> str:
    """Format the result as pretty JSON string"""
    return json.dumps(result, indent=2, ensure_ascii=False)

# Create Gradio interface
with gr.Blocks(title="Text to Sentences API") as demo:
    gr.Markdown("# Text to Sentences Splitter API")
    gr.Markdown("Enter a text paragraph and get it split into sentences with JSON output.")
    
    with gr.Row():
        with gr.Column():
            input_text = gr.Textbox(
                label="Input Text",
                placeholder="Enter your text paragraph here...",
                lines=5,
                max_lines=10
            )
            submit_btn = gr.Button("Split into Sentences", variant="primary")
            
        with gr.Column():
            output_json = gr.JSON(
                label="JSON Output",
                show_label=True
            )
            
    # Example inputs
    gr.Examples(
        examples=[
            ["Hello world! How are you today? I hope you're doing well. This is a test sentence."],
            ["The quick brown fox jumps over the lazy dog. Machine learning is fascinating! Natural language processing involves many complex tasks. Text processing is an important skill."],
            ["What is artificial intelligence? AI refers to computer systems that can perform tasks typically requiring human intelligence. These systems can learn, reason, and adapt to new situations."]
        ],
        inputs=input_text,
        outputs=output_json,
        fn=split_text_into_sentences,
        cache_examples=True
    )
    
    # Connect the interface
    submit_btn.click(
        fn=split_text_into_sentences,
        inputs=input_text,
        outputs=output_json
    )
    
    # API documentation
    gr.Markdown("""
    ## API Usage
    
    This app provides both a web interface and API endpoints.
    
    ### Using the API programmatically:
    
    ```python
    import requests
    import json
    
    # Replace with your actual Hugging Face Space URL
    url = "https://your-username-text-splitter.hf.space/api/predict"
    
    payload = {
        "data": ["Your text paragraph here..."]
    }
    
    response = requests.post(url, json=payload)
    result = response.json()
    print(json.dumps(result["data"][0], indent=2))
    ```
    
    ### cURL example:
    ```bash
    curl -X POST https://your-username-text-splitter.hf.space/api/predict \
         -H "Content-Type: application/json" \
         -d '{"data": ["Hello world! How are you? This is a test."]}'
    ```
    
    ### Response format:
    ```json
    {
      "status": "success",
      "sentences": ["Hello world!", "How are you?", "This is a test."],
      "count": 3,
      "original_length": 45,
      "metadata": {
        "avg_sentence_length": 15.0,
        "longest_sentence": 17,
        "shortest_sentence": 12
      }
    }
    ```
    """)

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