app.py

# =============================================================================
# AI DOCUMENT SUMMARIZER - BULLETPROOF HF SPACES VERSION
# This version handles all dependency issues gracefully
# =============================================================================

import gradio as gr
import sys
import time
import warnings
from typing import Tuple, Optional

print("🚀 Starting AI Document Summarizer...")
print(f"Python version: {sys.version}")

# Handle dependencies with proper error messages
dependencies_available = True
error_messages = []

try:
    import torch
    print("✅ PyTorch imported successfully")
    TORCH_AVAILABLE = True
    device_info = f"🖥️ Device: {'GPU' if torch.cuda.is_available() else 'CPU'}"
except ImportError as e:
    print(f"❌ PyTorch import failed: {e}")
    TORCH_AVAILABLE = False
    dependencies_available = False
    error_messages.append("PyTorch not found")

try:
    from transformers import pipeline
    print("✅ Transformers imported successfully")
    TRANSFORMERS_AVAILABLE = True
except ImportError as e:
    print(f"❌ Transformers import failed: {e}")
    TRANSFORMERS_AVAILABLE = False
    dependencies_available = False
    error_messages.append("Transformers not found")

# Suppress warnings if libraries are available
if dependencies_available:
    warnings.filterwarnings("ignore")

class DocumentSummarizer:
    """Robust document summarizer with dependency checking"""
    
    def __init__(self):
        self.models = {}
        self.available = dependencies_available
        
        if not self.available:
            print("❌ Dependencies not available - running in demo mode")
            return
            
        self.configs = {
            "BART": {
                "model_id": "facebook/bart-large-cnn",
                "description": "📰 Great for news and general text"
            },
            "T5-Small": {
                "model_id": "t5-small", 
                "description": "⚡ Fastest processing"
            },
            "DistilBART": {
                "model_id": "sshleifer/distilbart-cnn-12-6",
                "description": "🚀 Lightweight and fast"
            }
        }
        print("✅ Document Summarizer initialized!")
    
    def load_model(self, model_name: str):
        """Load model with comprehensive error handling"""
        if not self.available:
            return None
            
        if model_name not in self.models:
            try:
                config = self.configs.get(model_name, self.configs["T5-Small"])
                print(f"🔄 Loading {model_name}...")
                
                model = pipeline(
                    "summarization",
                    model=config["model_id"],
                    device=-1,  # Force CPU for HF Spaces stability
                    return_tensors="pt"
                )
                
                self.models[model_name] = model
                print(f"✅ {model_name} loaded successfully!")
                return model
                
            except Exception as e:
                print(f"❌ Failed to load {model_name}: {str(e)}")
                return None
        
        return self.models[model_name]
    
    def summarize(self, text: str, model_name: str, max_length: int, min_length: int) -> Tuple[str, str, str, str]:
        """Generate summary or return appropriate error message"""
        
        # Check if dependencies are available
        if not self.available:
            error_msg = f"""
❌ **Dependencies Missing**

The following packages are not installed:
{', '.join(error_messages)}

**To fix this issue:**
1. Make sure your `requirements.txt` contains:
   ```
   torch
   transformers
   gradio
   accelerate
   sentencepiece
   ```

2. Or try the corrected requirements.txt below.

This is likely a Hugging Face Spaces configuration issue.
            """
            return error_msg, "Please check the requirements.txt file", "", ""
        
        # Input validation
        if not text or not text.strip():
            return "⚠️ Please enter some text to summarize.", "", "", ""
        
        text = text.strip()
        word_count = len(text.split())
        
        if word_count < 10:
            return "⚠️ Text too short. Please provide at least 10 words.", "", "", ""
        
        # Load model
        model = self.load_model(model_name)
        if model is None:
            return f"❌ Could not load {model_name} model. Try T5-Small.", "", "", ""
        
        try:
            start_time = time.time()
            
            # Handle T5 special case
            input_text = f"summarize: {text}" if "T5" in model_name else text
            
            result = model(
                input_text,
                max_length=max_length,
                min_length=min_length,
                do_sample=False,
                truncation=True
            )
            
            processing_time = time.time() - start_time
            summary = result[0]['summary_text']
            
            # Calculate metrics
            summary_words = len(summary.split())
            compression_ratio = (summary_words / word_count) * 100
            
            metrics = f"""
📊 **Results:**
- Original: {word_count:,} words
- Summary: {summary_words:,} words  
- Compression: {compression_ratio:.1f}%
- Time: {processing_time:.1f}s
- Model: {model_name}
            """
            
            return summary, metrics, f"{word_count:,}", f"{summary_words:,}"
            
        except Exception as e:
            return f"❌ Error: {str(e)}", "", "", ""

# Initialize summarizer
summarizer = DocumentSummarizer()

# Sample texts
SAMPLES = {
    "Technology News": "Artificial intelligence and machine learning are transforming industries worldwide. From healthcare diagnostics to autonomous vehicles, AI systems are becoming more sophisticated and capable. However, this rapid advancement also raises questions about ethics, job displacement, and the need for proper regulation. Companies are investing billions in AI research while governments work to establish frameworks for responsible AI development.",
    
    "Business Update": "The global economy shows mixed signals as markets navigate inflation concerns and supply chain disruptions. Technology stocks have seen volatility while energy sectors gain momentum. Central banks worldwide are adjusting monetary policies to balance growth and inflation targets. Investors remain cautious but optimistic about long-term recovery prospects.",
    
    "Health Research": "Recent studies highlight the importance of preventive healthcare and lifestyle modifications. Research shows that regular exercise, balanced nutrition, and adequate sleep significantly impact long-term health outcomes. Mental health awareness is also growing, with new therapeutic approaches showing promising results in clinical trials."
}

def get_sample(choice):
    return SAMPLES.get(choice, "")

def process_request(text, model, max_len, min_len, sample):
    if sample != "None":
        text = get_sample(sample)
    return summarizer.summarize(text, model, max_len, min_len)

# Create Gradio interface
def create_app():
    with gr.Blocks(
        title="AI Document Summarizer",
        theme=gr.themes.Default(primary_hue="blue")
    ) as app:
        
        gr.Markdown("""
        # 📄 AI Document Summarizer
        ### Intelligent text summarization using transformer models
        """)
        
        # Show status
        if dependencies_available:
            gr.Markdown("✅ **Status:** All dependencies loaded successfully!")
        else:
            gr.Markdown(f"""
            ❌ **Status:** Missing dependencies - {', '.join(error_messages)}
            
            **Fix:** Update your `requirements.txt` with the correct dependencies (see below).
            """)
        
        with gr.Row():
            with gr.Column(scale=2):
                gr.Markdown("## Input")
                
                sample_choice = gr.Dropdown(
                    ["None"] + list(SAMPLES.keys()),
                    label="📋 Try a sample:",
                    value="None"
                )
                
                text_input = gr.Textbox(
                    label="📝 Your text:",
                    placeholder="Enter text to summarize...",
                    lines=8
                )
                
                sample_choice.change(get_sample, sample_choice, text_input)
            
            with gr.Column(scale=1):
                gr.Markdown("## Settings")
                
                model_select = gr.Dropdown(
                    ["BART", "T5-Small", "DistilBART"],
                    label="🤖 Model:",
                    value="T5-Small"
                )
                
                max_length = gr.Slider(50, 200, 100, label="Max length")
                min_length = gr.Slider(20, 80, 30, label="Min length")
        
        submit_btn = gr.Button("🚀 Summarize", variant="primary")
        
        with gr.Row():
            summary_out = gr.Textbox(label="📋 Summary", lines=5)
            metrics_out = gr.Markdown("Metrics will appear here")
        
        with gr.Row():
            orig_count = gr.Textbox(label="Original", interactive=False)
            summ_count = gr.Textbox(label="Summary", interactive=False)
        
        submit_btn.click(
            process_request,
            [text_input, model_select, max_length, min_length, sample_choice],
            [summary_out, metrics_out, orig_count, summ_count]
        )
        
        # Troubleshooting section
        gr.Markdown("""
        ---
        ## 🔧 Troubleshooting
        
        **If you see dependency errors, create these files:**
        
        ### requirements.txt
        ```
        torch
        transformers
        gradio
        accelerate
        sentencepiece
        numpy
        ```
        
        ### README.md
        ```
        ---
        title: AI Document Summarizer
        emoji: 📄
        colorFrom: blue
        colorTo: green
        sdk: gradio
        sdk_version: 4.0.0
        app_file: app.py
        pinned: false
        ---
        ```
        """)
    
    return app

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

# =============================================================================
# DEPLOYMENT FILES - COPY THESE EXACTLY
# =============================================================================

REQUIREMENTS_TXT = """torch
transformers
gradio
accelerate
sentencepiece
numpy"""

README_MD = """---
title: AI Document Summarizer
emoji: 📄
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.0.0
app_file: app.py
pinned: false
---

# AI Document Summarizer

Transform long documents into concise summaries using AI.

## Features
- Multiple transformer models
- Real-time processing
- Sample documents included
- Error handling and troubleshooting

Built for Hugging Face Spaces deployment."""

print(f"""
🚀 BULLETPROOF DEPLOYMENT GUIDE

📁 CREATE EXACTLY THESE 3 FILES:

1️⃣ requirements.txt:
{REQUIREMENTS_TXT}

2️⃣ README.md:
{README_MD}

3️⃣ app.py:
[Copy the entire code above]

🎯 DEPLOYMENT STEPS:
1. Go to hf.co/spaces
2. Create new Space
3. Choose Gradio SDK  
4. Upload these 3 files EXACTLY as shown
5. Wait for build to complete

✅ This version will:
- Show clear error messages if dependencies fail
- Provide troubleshooting instructions
- Work with minimal requirements
- Handle all edge cases gracefully

❌ Common issues fixed:
- Torch import errors
- Transformers import errors  
- Version conflicts
- HF Spaces compatibility
""")