app.py

#!/usr/bin/env python3
"""
Hugging Face Gradio App for RDF Validation with MCP Server and Anthropic AI

This app serves both as a web interface and can expose MCP server functionality.
Deploy this on Hugging Face Spaces with your Anthropic API key.
"""

import gradio as gr
import os
import json
import sys
import asyncio
import logging
import requests
from typing import Any, Dict, List, Optional
import threading
import time

# CRITICAL: FORCE OVERRIDE ALL ENVIRONMENT VARIABLES THAT COULD INTERFERE
print("🔧 FORCING ENVIRONMENT VARIABLE OVERRIDES...")

# Remove any HF environment variables that could cause URL concatenation
problematic_env_vars = [
    'HF_API_URL',
    'HF_INFERENCE_URL', 
    'HF_ENDPOINT_URL',
    'HF_MODEL',
    'HUGGINGFACE_API_URL',
    'HUGGINGFACE_INFERENCE_URL'
]

for var in problematic_env_vars:
    if var in os.environ:
        old_value = os.environ[var]
        del os.environ[var]
        print(f"🗑️ Removed environment variable: {var} = {old_value}")

print("✅ Environment variables cleaned")

# Add current directory to path
sys.path.append(os.path.dirname(os.path.abspath(__file__)))

# Import our validation logic
try:
    from validator import validate_rdf
    VALIDATOR_AVAILABLE = True
except ImportError:
    VALIDATOR_AVAILABLE = False
    print("⚠️ Warning: validator.py not found. Some features may be limited.")

# Optional: Check if OpenAI and requests are available
try:
    from openai import OpenAI
    OPENAI_AVAILABLE = True
except ImportError:
    OPENAI_AVAILABLE = False
    print("💡 Install 'openai' package for AI-powered corrections: pip install openai")

try:
    import requests
    HF_INFERENCE_AVAILABLE = True
except ImportError:
    HF_INFERENCE_AVAILABLE = False
    print("💡 Install 'requests' package for AI-powered corrections: pip install requests")

# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Configuration - ABSOLUTELY HARDCODED VALUES (NO ENV VARS ALLOWED)
HF_API_KEY = os.getenv('HF_API_KEY', '')  # Only this one env var is allowed
# FORCE HARDCODED VALUES - IGNORE ALL OTHER ENVIRONMENT VARIABLES
HF_ENDPOINT_URL = "https://evxgv66ksxjlfrts.us-east-1.aws.endpoints.huggingface.cloud/v1/"
HF_MODEL = "lmstudio-community/Llama-3.3-70B-Instruct-GGUF"  # Correct model name for your endpoint

print(f"🔐 FORCED hardcoded endpoint: {HF_ENDPOINT_URL}")
print(f"🔐 FORCED hardcoded model: {HF_MODEL}")
print(f"🔑 HF_API_KEY configured: {'Yes' if HF_API_KEY else 'No'}")

# EXTRA PROTECTION: Override any modules that might have cached env vars
import sys
if 'requests' in sys.modules:
    print("🔄 Requests module detected - ensuring no cached env vars")
if 'httpx' in sys.modules:
    print("🔄 HTTPX module detected - ensuring no cached env vars")

# OpenAI client configuration for the endpoint
def get_openai_client():
    """Get configured OpenAI client for HF Inference Endpoint"""
    if not HF_API_KEY:
        print("❌ No HF_API_KEY available for OpenAI client")
        return None
    
    print(f"🔗 Creating OpenAI client with:")
    print(f"   base_url: {HF_ENDPOINT_URL}")
    print(f"   api_key: {'***' + HF_API_KEY[-4:] if len(HF_API_KEY) > 4 else 'HIDDEN'}")
    
    return OpenAI(
        base_url=HF_ENDPOINT_URL,
        api_key=HF_API_KEY,
        timeout=120.0  # Increase timeout for cold starts
    )

# Sample RDF data for examples
SAMPLE_VALID_RDF = '''<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
         xmlns:bf="http://id.loc.gov/ontologies/bibframe/"
         xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">
    
    <bf:Work rdf:about="http://example.org/work/1">
        <rdf:type rdf:resource="http://id.loc.gov/ontologies/bibframe/Text"/>
        <bf:title>
            <bf:Title>
                <bf:mainTitle>Sample Monograph Title</bf:mainTitle>
            </bf:Title>
        </bf:title>
        <bf:creator>
            <bf:Agent>
                <rdfs:label>Sample Author</rdfs:label>
            </bf:Agent>
        </bf:creator>
    </bf:Work>
    
</rdf:RDF>'''

SAMPLE_INVALID_RDF = '''<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
    <!-- Missing namespace declarations -->
    <!-- Missing required properties -->
    <bf:Work rdf:about="http://example.org/work/1">
        <bf:title>Incomplete Title</bf:title>
        <!-- Missing rdf:type -->
        <!-- Missing proper title structure -->
    </bf:Work>
</rdf:RDF>'''

# MCP Server Tools (can be used independently)
def validate_rdf_tool(rdf_content: str, template: str = "monograph") -> dict:
    """
    Validate RDF/XML content against SHACL templates.
    
    This tool validates RDF/XML data against predefined SHACL shapes to ensure
    compliance with metadata standards like BIBFRAME. Returns detailed validation
    results with conformance status and specific violation information.
    
    Args:
        rdf_content (str): The RDF/XML content to validate
        template (str): Validation template to use ('monograph' or 'custom')
    
    Returns:
        dict: Validation results with conformance status and detailed feedback
    """
    if not rdf_content:
        return {"error": "No RDF/XML content provided", "conforms": False}
    
    if not VALIDATOR_AVAILABLE:
        return {
            "error": "Validator not available - ensure validator.py is present",
            "conforms": False
        }
    
    try:
        conforms, results_text = validate_rdf(rdf_content.encode('utf-8'), template)
        
        return {
            "conforms": conforms,
            "results": results_text,
            "template": template,
            "status": "✅ Valid RDF" if conforms else "❌ Invalid RDF"
        }
        
    except Exception as e:
        logger.error(f"Validation error: {str(e)}")
        return {
            "error": f"Validation failed: {str(e)}",
            "conforms": False
        }

def get_ai_suggestions(validation_results: str, rdf_content: str) -> str:
    """
    Generate AI-powered fix suggestions for invalid RDF/XML.
    
    This tool analyzes validation results and provides actionable suggestions
    for fixing RDF/XML validation errors using AI or rule-based analysis.
    
    Args:
        validation_results (str): The validation error messages
        rdf_content (str): The original RDF/XML content that failed validation
    
    Returns:
        str: Detailed suggestions for fixing the RDF validation issues
    """
    
    if not OPENAI_AVAILABLE:
        return generate_manual_suggestions(validation_results)
    
    # Get API key dynamically at runtime
    current_api_key = os.getenv('HF_API_KEY', '')
    if not current_api_key:
        return f"""
🔑 **AI suggestions disabled**: Please set your Hugging Face API key as a Secret in your Space settings.

{generate_manual_suggestions(validation_results)}
"""
    
    try:
        # Use OpenAI client with your Hugging Face Inference Endpoint
        print("🔍 Attempting to get OpenAI client for suggestions...")
        client = get_openai_client()
        if not client:
            print("❌ OpenAI client is None for suggestions.")
            return f"""
🔑 **AI suggestions disabled**: HF_API_KEY not configured or client creation failed.

{generate_manual_suggestions(validation_results)}
"""
        print(f"✅ OpenAI client obtained for suggestions. Client timeout: {client.timeout}")
        
        prompt = f"""You are an expert in RDF/XML and SHACL validation. Analyze the following validation results and provide clear, actionable suggestions for fixing the RDF issues.

Validation Results:
{validation_results}

Original RDF (first 1000 chars):
{rdf_content[:1000]}...

Please provide:
1. A clear summary of what's wrong
2. Specific step-by-step instructions to fix each issue
3. Example corrections where applicable
4. Best practices to prevent similar issues

Format your response in a helpful, structured way using markdown."""
        
        # Make API call using OpenAI client
        print(f"🔄 Making SUGGESTION API call to: {HF_ENDPOINT_URL} with model: {HF_MODEL}")
        print(f"🔄 Client base_url: {client.base_url}")
        print("⏳ Attempting client.chat.completions.create() for suggestions...")
        
        chat_completion = client.chat.completions.create(
            model=HF_MODEL,
            messages=[
                {
                    "role": "user",
                    "content": prompt
                }
            ],
            max_tokens=1500,
            temperature=0.7,
            top_p=0.9
        )
        
        print(f"✅ client.chat.completions.create() returned for suggestions. Type: {type(chat_completion)}")
        generated_text = chat_completion.choices[0].message.content
        print("✅ Suggestion API call successful, content extracted.")
        return f"🤖 **AI-Powered Suggestions:**\n\n{generated_text}"
        
    except Exception as e:
        logger.error(f"OpenAI/HF Inference Endpoint error (suggestions): {str(e)}", exc_info=True) # Added exc_info for full traceback
        return f"""
❌ **AI suggestions error**: {str(e)}

{generate_manual_suggestions(validation_results)}
"""

def get_ai_correction(validation_results: str, rdf_content: str) -> str:
    """
    Generate AI-powered corrected RDF/XML based on validation errors.
    
    This tool takes invalid RDF/XML and validation results, then generates
    a corrected version that addresses all identified validation issues.
    
    Args:
        validation_results (str): The validation error messages
        rdf_content (str): The original invalid RDF/XML content
    
    Returns:
        str: Corrected RDF/XML that should pass validation
    """
    
    if not OPENAI_AVAILABLE:
        return generate_manual_correction_hints(validation_results, rdf_content)
    
    # Get API key dynamically at runtime
    current_api_key = os.getenv('HF_API_KEY', '')
    if not current_api_key:
        return f"""<!-- AI correction disabled: Set HF_API_KEY as a Secret in your Space settings -->

{generate_manual_correction_hints(validation_results, rdf_content)}"""
    
    try:
        # Use OpenAI client with your Hugging Face Inference Endpoint
        print("🔍 Attempting to get OpenAI client for correction...")
        client = get_openai_client()
        if not client:
            print("❌ OpenAI client is None for correction.")
            return f"""<!-- AI correction disabled: HF_API_KEY not configured or client creation failed. -->

{generate_manual_correction_hints(validation_results, rdf_content)}"""
        print(f"✅ OpenAI client obtained for correction. Client timeout: {client.timeout}")
        
        prompt = f"""You are an expert in RDF/XML. Fix the following RDF/XML based on the validation errors provided.

Validation Errors:
{validation_results}

Original RDF/XML:
{rdf_content}

Please provide the corrected RDF/XML that addresses all validation issues.
- Return only the corrected XML without additional explanation
- Maintain the original structure as much as possible while fixing errors
- Ensure all namespace declarations are present
- Add any missing required properties
- Fix any syntax or structural issues"""
        
        # Make API call using OpenAI client
        print(f"🔄 Making CORRECTION API call to: {HF_ENDPOINT_URL} with model: {HF_MODEL}")
        print(f"🔄 Client base_url: {client.base_url}")
        print("⏳ Attempting client.chat.completions.create() for correction...")
        
        chat_completion = client.chat.completions.create(
            model=HF_MODEL,
            messages=[
                {
                    "role": "user", 
                    "content": prompt
                }
            ],
            max_tokens=2000,
            temperature=0.3,
            top_p=0.9
        )
        
        print(f"✅ client.chat.completions.create() returned for correction. Type: {type(chat_completion)}")
        corrected_text = chat_completion.choices[0].message.content
        print("✅ Correction API call successful, content extracted.")
        return corrected_text
        
    except Exception as e:
        logger.error(f"OpenAI/HF Inference Endpoint error (correction): {str(e)}", exc_info=True) # Added exc_info for full traceback
        return f"""<!-- AI correction error: {str(e)} -->

{generate_manual_correction_hints(validation_results, rdf_content)}"""

def generate_manual_suggestions(validation_results: str) -> str:
    """Generate rule-based suggestions when AI is not available"""
    suggestions = []
    
    if "Constraint Violation" in validation_results:
        suggestions.append("• Fix SHACL constraint violations by ensuring required properties are present")
    
    if "Missing property" in validation_results or "missing" in validation_results.lower():
        suggestions.append("• Add missing required properties (check template requirements)")
    
    if "datatype" in validation_results.lower():
        suggestions.append("• Correct data type mismatches (ensure proper literal types)")
    
    if "namespace" in validation_results.lower() or "prefix" in validation_results.lower():
        suggestions.append("• Add missing namespace declarations at the top of your RDF")
    
    if "XML" in validation_results or "syntax" in validation_results.lower():
        suggestions.append("• Fix XML syntax errors (check for unclosed tags, invalid characters)")
    
    if not suggestions:
        suggestions.append("• Review detailed validation results for specific issues")
        suggestions.append("• Ensure your RDF follows the selected template requirements")
    
    suggestions_text = "\n".join(suggestions)
    
    return f"""
📋 **Manual Analysis:**

{suggestions_text}

💡 **General Tips:**
• Check namespace declarations at the top of your RDF
• Ensure all required properties are present
• Verify data types match expected formats
• Make sure XML structure is well-formed

🔧 **Common Fixes:**
• Add missing namespace prefixes
• Include required properties like rdf:type
• Fix malformed URIs or literals
• Ensure proper XML syntax
"""

def generate_manual_correction_hints(validation_results: str, rdf_content: str) -> str:
    """Generate manual correction hints when AI is not available"""
    return f"""<!-- Manual correction hints based on validation results -->
<!-- Set HF_API_KEY as a Secret in your Space settings for AI-powered corrections -->

{rdf_content}

<!-- 
VALIDATION ISSUES FOUND:
{validation_results[:500]}...

MANUAL CORRECTION STEPS:
1. Add missing namespace declarations
2. Include required properties (rdf:type, etc.)
3. Fix XML syntax errors
4. Ensure proper URI formats
5. Validate data types
-->"""

def validate_rdf_interface(rdf_content: str, template: str, use_ai: bool = True):
    """Main validation function for Gradio interface"""
    if not rdf_content.strip():
        return "❌ Error", "No RDF/XML data provided", "", ""
    
    # Validate RDF
    result = validate_rdf_tool(rdf_content, template)
    
    if "error" in result:
        return f"❌ Error: {result['error']}", "", "", ""
    
    status = result["status"]
    results_text = result["results"]
    
    if result["conforms"]:
        suggestions = "✅ No issues found! Your RDF/XML is valid according to the selected template."
        corrected_rdf = "<!-- Already valid - no corrections needed -->\n" + rdf_content
    else:
        if use_ai:
            suggestions = get_ai_suggestions(results_text, rdf_content)
            corrected_rdf = get_ai_correction(results_text, rdf_content)
        else:
            suggestions = generate_manual_suggestions(results_text)
            corrected_rdf = generate_manual_correction_hints(results_text, rdf_content)
    
    return status, results_text, suggestions, corrected_rdf

def get_rdf_examples(example_type: str = "valid") -> str:
    """
    Retrieve example RDF/XML snippets for testing and learning.
    
    This tool provides sample RDF/XML content that can be used to test
    the validation system or learn proper RDF structure.
    
    Args:
        example_type (str): Type of example ('valid', 'invalid', or 'bibframe')
    
    Returns:
        str: RDF/XML example content
    """
    examples = {
        "valid": SAMPLE_VALID_RDF,
        "invalid": SAMPLE_INVALID_RDF,
        "bibframe": '''<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
         xmlns:bf="http://id.loc.gov/ontologies/bibframe/"
         xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">
    
    <bf:Instance rdf:about="http://example.org/instance/1">
        <rdf:type rdf:resource="http://id.loc.gov/ontologies/bibframe/Print"/>
        <bf:instanceOf rdf:resource="http://example.org/work/1"/>
        <bf:title>
            <bf:Title>
                <bf:mainTitle>Example Book Title</bf:mainTitle>
            </bf:Title>
        </bf:title>
        <bf:provisionActivity>
            <bf:Publication>
                <bf:date>2024</bf:date>
                <bf:place>
                    <bf:Place>
                        <rdfs:label>New York</rdfs:label>
                    </bf:Place>
                </bf:place>
            </bf:Publication>
        </bf:provisionActivity>
    </bf:Instance>
    
</rdf:RDF>'''
    }
    
    return examples.get(example_type, examples["valid"])

# Create Gradio Interface
def create_interface():
    """Create the main Gradio interface"""
    
    # Check API key status dynamically
    current_api_key = os.getenv('HF_API_KEY', '')
    api_status = "🔑 AI features enabled" if (OPENAI_AVAILABLE and current_api_key) else "⚠️ AI features disabled (set HF_API_KEY)"
    
    with gr.Blocks(
        title="RDF Validation Server with AI",
        theme=gr.themes.Soft(),
        css="""
        .status-box { 
            font-weight: bold; 
            padding: 10px; 
            border-radius: 5px; 
        }
        .header-text {
            text-align: center;
            padding: 20px;
        }
        """
    ) as demo:
        
        # Header
        debug_info = f"""
        Debug Info:
        - OPENAI_AVAILABLE: {OPENAI_AVAILABLE}
        - HF_INFERENCE_AVAILABLE: {HF_INFERENCE_AVAILABLE}
        - HF_API_KEY set: {'Yes' if current_api_key else 'No'}
        - HF_API_KEY length: {len(current_api_key) if current_api_key else 0}
        - HF_ENDPOINT_URL: {HF_ENDPOINT_URL}
        - HF_MODEL: {HF_MODEL}
        """
        
        gr.HTML(f"""
        <div class="header-text">
            <h1>🔍 RDF Validation Server with AI</h1>
            <p>Validate RDF/XML against SHACL schemas with AI-powered suggestions and corrections</p>
            <p><strong>Status:</strong> {api_status}</p>
            <details><summary>Debug Info</summary><pre>{debug_info}</pre></details>
        </div>
        """)
        
        # Main interface
        with gr.Row():
            with gr.Column(scale=1):
                gr.Markdown("### 📝 Input")
                
                rdf_input = gr.Textbox(
                    label="RDF/XML Content",
                    placeholder="Paste your RDF/XML content here...",
                    lines=15,
                    show_copy_button=True
                )
                
                with gr.Row():
                    template_dropdown = gr.Dropdown(
                        label="Validation Template",
                        choices=["monograph", "custom"],
                        value="monograph",
                        info="Select the SHACL template to validate against"
                    )
                    
                    use_ai_checkbox = gr.Checkbox(
                        label="Use AI Features",
                        value=True,
                        info="Enable AI-powered suggestions and corrections"
                    )
                
                validate_btn = gr.Button("🔍 Validate RDF", variant="primary", size="lg")
        
        # Results section
        with gr.Row():
            with gr.Column():
                gr.Markdown("### 📊 Results")
                
                status_output = gr.Textbox(
                    label="Validation Status",
                    interactive=False,
                    lines=1,
                    elem_classes=["status-box"]
                )
                
                results_output = gr.Textbox(
                    label="Detailed Validation Results",
                    interactive=False,
                    lines=8,
                    show_copy_button=True
                )
                
                suggestions_output = gr.Textbox(
                    label="💡 Fix Suggestions",
                    interactive=False,
                    lines=8,
                    show_copy_button=True
                )
        
        # Corrected RDF section
        with gr.Row():
            with gr.Column():
                gr.Markdown("### 🛠️ AI-Generated Corrections")
                
                corrected_output = gr.Textbox(
                    label="Corrected RDF/XML",
                    interactive=False,
                    lines=15,
                    show_copy_button=True,
                    placeholder="Corrected RDF will appear here after validation..."
                )
        
        # Examples and controls
        with gr.Row():
            gr.Markdown("### 📚 Examples & Tools")
            
        with gr.Row():
            example1_btn = gr.Button("✅ Valid RDF Example", variant="secondary")
            example2_btn = gr.Button("❌ Invalid RDF Example", variant="secondary")
            example3_btn = gr.Button("📖 BibFrame Example", variant="secondary")
            clear_btn = gr.Button("🗑️ Clear All", variant="stop")
        
        # Event handlers
        validate_btn.click(
            fn=validate_rdf_interface,
            inputs=[rdf_input, template_dropdown, use_ai_checkbox],
            outputs=[status_output, results_output, suggestions_output, corrected_output]
        )
        
        # Auto-validate on input change (debounced)
        rdf_input.change(
            fn=validate_rdf_interface,
            inputs=[rdf_input, template_dropdown, use_ai_checkbox],
            outputs=[status_output, results_output, suggestions_output, corrected_output]
        )
        
        # Example buttons
        example1_btn.click(
            lambda: get_rdf_examples("valid"),
            outputs=[rdf_input]
        )
        
        example2_btn.click(
            lambda: get_rdf_examples("invalid"),
            outputs=[rdf_input]
        )
        
        example3_btn.click(
            lambda: get_rdf_examples("bibframe"),
            outputs=[rdf_input]
        )
        
        clear_btn.click(
            lambda: ("", "", "", "", ""),
            outputs=[rdf_input, status_output, results_output, suggestions_output, corrected_output]
        )
        
        # Footer with instructions
        gr.Markdown("""
        ---
        ### 🚀 **Deployment Instructions for Hugging Face Spaces:**
        
        1. **Create a new Space** on [Hugging Face](https://huggingface.co/spaces)
        2. **Set up your Hugging Face Inference Endpoint** and get the endpoint URL
        3. **Set your tokens** in Space settings (use Secrets for security):
           - Go to Settings → Repository secrets  
           - Add: `HF_API_KEY` = `your_huggingface_api_key_here`
           - Endpoint is now hardcoded to your specific Inference Endpoint
        4. **Upload these files** to your Space repository
        5. **Install requirements**: The Space will auto-install from `requirements.txt`
        
        ### 🔧 **MCP Server Mode:**
        This app functions as both a web interface AND an MCP server for Claude Desktop and other MCP clients.
        
        **Available MCP Tools (via SSE):**
        - `validate_rdf_tool`: Validate RDF/XML against SHACL shapes
        - `get_ai_suggestions`: Get AI-powered fix suggestions  
        - `get_ai_correction`: Generate corrected RDF/XML
        - `get_rdf_examples`: Retrieve example RDF snippets
        
        **MCP Connection:** 
        1. When deployed on Hugging Face Spaces, the MCP server is available at:
           `https://your-space-id.hf.space/gradio_api/mcp/sse`
        2. Use this URL in Claude Desktop's MCP configuration
        3. The app automatically exposes functions with proper docstrings as MCP tools
        
        ### 💡 **Features:**
        - ✅ Real-time RDF/XML validation against SHACL schemas
        - 🤖 AI-powered error suggestions and corrections (with HF Inference Endpoint)
        - 📚 Built-in examples and templates
        - 🔄 Auto-validation as you type
        - 📋 Copy results with one click
        
        **Note:** AI features require a valid Hugging Face API key (HF_API_KEY) set as a Secret. Manual suggestions are provided as fallback.
        """)
    
    return demo

# Launch configuration
if __name__ == "__main__":
    # Force verify environment is clean
    print("🔍 FINAL CHECK: Verifying problematic environment variables are removed...")
    for var in problematic_env_vars:
        if var in os.environ:
            print(f"⚠️ WARNING: {var} still exists! Value: {os.environ[var]}")
            del os.environ[var]
            print(f"🗑️ FORCE REMOVED: {var}")
        else:
            print(f"✅ {var} confirmed not in environment")
    
    demo = create_interface()
    
    # Configuration for different environments
    port = int(os.getenv('PORT', 7860))  # Hugging Face uses PORT env variable
    
    demo.launch(
        server_name="0.0.0.0",      # Important for external hosting
        server_port=port,           # Use environment PORT or default to 7860
        share=False,                # Don't create gradio.live links in production
        show_error=True,            # Show errors in the interface
        show_api=True,              # Enable API endpoints
        allowed_paths=["."]         # Allow serving files from current directory
    )