Fix build errors and include document corpus

.gitignore

@@ -1,5 +1,6 @@
 # Data and databases
-data/
+# data/  # ALLOW data/ to be uploaded for HF Spaces
+data/chroma_db/
 .chroma/
 *.db
 *.sqlite3

app.py

@@ -1,30 +1,74 @@
 import os
 import sys
+from pathlib import Path
 
 # Add src to the Python path
+# HF Spaces has the repo content in the root.
+# Our package is in src/iks_rag
 sys.path.insert(0, "src")
 
-from iks_rag.ui.gradio_app import create_interface
-from iks_rag.rag_system import create_rag_system
-
 print("🚀 Initializing IKS RAG System on HuggingFace Spaces...")
 
+# Check environment
+print(f"Python Version: {sys.version}")
+print(f"CWD: {os.getcwd()}")
+
+# Import after path is set
+try:
+    from iks_rag.ui.gradio_app import create_interface
+    from iks_rag.rag_system import create_rag_system
+    print("✅ Successfully imported iks_rag modules")
+except ImportError as e:
+    print(f"❌ ERROR: Failed to import iks_rag modules: {e}")
+    print(f"Python path: {sys.path}")
+    # Create a simple fallback UI if imports fail to show the error on the page
+    import gradio as gr
+    demo = gr.Interface(fn=lambda x: f"Import Error: {e}", inputs="text", outputs="text")
+    demo.launch()
+    sys.exit(1)
+
 # Make sure Gemini key is loaded from space secrets
 api_key = os.environ.get("GOOGLE_API_KEY", "")
 if not api_key:
     print("⚠️ WARNING: GOOGLE_API_KEY not found in environment variables!")
+    print("Please set GOOGLE_API_KEY in the 'Settings > Secrets' section of your Space.")
 
 # Create RAG system using our default config
-# We use the default.yaml which is already configured for Gemini
-rag_system = create_rag_system("configs/rag/default.yaml")
+config_path = "configs/rag/default.yaml"
+if not os.path.exists(config_path):
+    print(f"❌ ERROR: Config file not found at {config_path}")
+else:
+    print(f"✅ Found config at {config_path}")
 
-print("📚 Loading documents (from local ChromaDB)...")
-rag_system.load_documents()
+try:
+    rag_system = create_rag_system(config_path)
+    print("✅ RAG system initialized")
+except Exception as e:
+    print(f"❌ ERROR initializing RAG system: {e}")
+    rag_system = None
 
-stats = rag_system.get_stats()
-print(f"✅ Loaded {stats['documents_loaded']} document chunks")
+# Check data and load
+if rag_system:
+    docs_path = "data/documents"
+    if not os.path.exists(docs_path) or not os.listdir(docs_path):
+        print(f"⚠️ WARNING: No documents found in {docs_path}")
+    else:
+        print(f"📚 Loading documents from {docs_path}...")
+
+    try:
+        rag_system.load_documents()
+        stats = rag_system.get_stats()
+        print(f"✅ Loaded {stats['documents_loaded']} document chunks into ChromaDB")
+    except Exception as e:
+        print(f"❌ ERROR loading documents: {str(e)}")
 
 # Launch the Gradio Interface
-print("🌐 Launching Gradio...")
-demo = create_interface(rag_system)
-demo.launch()
+print("🌐 Launching Gradio Interface...")
+if rag_system:
+    demo = create_interface(rag_system)
+else:
+    import gradio as gr
+    demo = gr.Interface(fn=lambda x: "RAG System failed to initialize", inputs="text", outputs="text")
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -4,10 +4,10 @@ llama-index-embeddings-huggingface>=0.5.0
 llama-index-vector-stores-chroma>=0.4.0
 llama-index-readers-file>=0.1.0
 chromadb>=0.6.0
-sentence-transformers>=3.0.0,<4.0.0
-transformers<4.42.0
-numpy>=1.26.0,<2.0.0
-gradio>=5.0.0
+sentence-transformers>=3.0.0
+transformers>=4.42.0
+numpy>=1.26.0
+gradio>=6.13.0
 google-generativeai>=0.8.0
 pydantic-settings>=2.6.0
 pydantic>=2.9.0

src/iks_rag/ui/gradio_app.py

@@ -4,6 +4,7 @@ Provides a chat interface for students to ask IKS questions.
 """
 
 from pathlib import Path
+from typing import Any
 
 # Load .env FIRST — must happen before iks_rag imports so API keys are available
 try:
@@ -14,10 +15,65 @@ except ImportError:
 
 import gradio as gr
 
-from iks_rag.config import load_config
+from iks_rag.config import load_config, RAGConfig
 from iks_rag.rag_system import RAGSystem
 
 
+def format_examples(config: RAGConfig) -> str:
+    """Format example questions.
+    
+    Args:
+        config: RAG configuration
+        
+    Returns:
+        Formatted markdown string
+    """
+    examples = config.ui.examples
+    md = "### Example Questions\n\n"
+    for i, ex in enumerate(examples, 1):
+        md += f"{i}. {ex}\n\n"
+    return md
+
+
+def handle_response(message: str, chat_history: list, rag_system: RAGSystem) -> tuple[str, list]:
+    """Process user message and generate response.
+    
+    Args:
+        message: User input message
+        chat_history: Current chat history
+        rag_system: RAG system instance
+        
+    Returns:
+        Tuple of (empty string, updated chat history)
+    """
+    if not message.strip():
+        return "", chat_history
+
+    # Add user message (Gradio 6.0 message format)
+    chat_history.append({"role": "user", "content": message})
+
+    try:
+        # Query RAG system
+        result = rag_system.query(message)
+        answer = result["answer"]
+
+        # Format sources
+        sources_text = "\n\n**Sources:**\n"
+        for i, source in enumerate(result["sources"][:3], 1):
+            file_name = source["metadata"].get("file_name", "Unknown")
+            sources_text += f"\n{i}. {file_name}"
+
+        full_response = f"{answer}{sources_text}"
+
+        chat_history.append({"role": "assistant", "content": full_response})
+        return "", chat_history
+
+    except Exception as e:
+        error_msg = f"❌ Error: {str(e)}\n\nPlease ensure:\n- API keys are set\n- Documents are loaded"
+        chat_history.append({"role": "assistant", "content": error_msg})
+        return "", chat_history
+
+
 def create_interface(rag_system: RAGSystem) -> gr.Blocks:
     """Create Gradio interface.
 
@@ -83,50 +139,16 @@ def create_interface(rag_system: RAGSystem) -> gr.Blocks:
         # Examples section (hidden by default)
         examples_md = gr.Markdown(visible=False)
 
-        def format_examples():
-            """Format example questions."""
-            examples = config.ui.examples
-            md = "### Example Questions\n\n"
-            for i, ex in enumerate(examples, 1):
-                md += f"{i}. {ex}\n\n"
-            return md
-
         def respond(message: str, chat_history: list):
-            """Process user message and generate response."""
-            if not message.strip():
-                return "", chat_history
-
-            # Add user message (Gradio 6.0 message format)
-            chat_history.append({"role": "user", "content": message})
-
-            try:
-                # Query RAG system
-                result = rag_system.query(message)
-                answer = result["answer"]
-
-                # Format sources
-                sources_text = "\n\n**Sources:**\n"
-                for i, source in enumerate(result["sources"][:3], 1):
-                    file_name = source["metadata"].get("file_name", "Unknown")
-                    sources_text += f"\n{i}. {file_name}"
-
-                full_response = f"{answer}{sources_text}"
-
-                chat_history.append({"role": "assistant", "content": full_response})
-                return "", chat_history
-
-            except Exception as e:
-                error_msg = f"❌ Error: {str(e)}\n\nPlease ensure:\n- Ollama is running (ollama serve)\n- Documents are loaded"
-                chat_history.append({"role": "assistant", "content": error_msg})
-                return "", chat_history
+            return handle_response(message, chat_history, rag_system)
 
         def clear_chat():
             """Clear chat history."""
-            return None
+            return []
 
         def toggle_examples():
             """Toggle examples visibility."""
-            return gr.update(visible=True, value=format_examples())
+            return gr.update(visible=True, value=format_examples(config))
 
         # Event handlers
         submit.click(respond, [msg, chatbot], [msg, chatbot])