...

Cargo.toml

@@ -24,7 +24,18 @@ age = "0.10"
 secrecy = "0.8"
 ed25519-dalek = "2"
 base64 = "0.22"
-tantivy = "0.25.0"
+# Lance vector database dependencies
+lance = "0.33"
+lance-index = "0.33"
+lance-linalg = "0.33"
+# Use Arrow version compatible with Lance 0.33
+arrow = "55.2"
+arrow-array = "55.2"
+arrow-schema = "55.2"
+parquet = "55.2"
+uuid = { version = "1.10", features = ["v4"] }
+reqwest = { version = "0.11", features = ["json"] }
+image = "0.25"
 
 [dev-dependencies]
 redis = { version = "0.24", features = ["aio", "tokio-comp"] }

docs/lance_vector_db.md

@@ -0,0 +1,454 @@
+# Lance Vector Database Operations
+
+HeroDB includes a powerful vector database integration using Lance, enabling high-performance vector storage, search, and multimodal data management. By default, it uses Ollama for local text embeddings, with support for custom external embedding services.
+
+## Overview
+
+The Lance vector database integration provides:
+
+- **High-performance vector storage** using Lance's columnar format
+- **Local Ollama integration** for text embeddings (default, no external dependencies)
+- **Custom embedding service support** for advanced use cases
+- **Text embedding support** (images via custom services)
+- **Vector similarity search** with configurable parameters
+- **Scalable indexing** with IVF_PQ (Inverted File with Product Quantization)
+- **Redis-compatible command interface**
+
+## Architecture
+
+```
+┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
+│   HeroDB        │    │  External        │    │   Lance         │
+│   Redis Server  │◄──►│  Embedding       │    │   Vector Store  │
+│                 │    │  Service         │    │                 │
+└─────────────────┘    └──────────────────┘    └─────────────────┘
+         │                       │                       │
+         │                       │                       │
+    Redis Protocol          HTTP API              Arrow/Parquet
+    Commands                JSON Requests         Columnar Storage
+```
+
+### Key Components
+
+1. **Lance Store**: High-performance columnar vector storage
+2. **Ollama Integration**: Local embedding service (default)
+3. **Custom Embedding Service**: Optional HTTP API for advanced use cases
+4. **Redis Command Interface**: Familiar Redis-style commands
+5. **Arrow Schema**: Flexible schema definition for metadata
+
+## Configuration
+
+### Default Setup (Ollama)
+
+HeroDB uses Ollama by default for text embeddings. No configuration is required if Ollama is running locally:
+
+```bash
+# Install Ollama (if not already installed)
+# Visit: https://ollama.ai
+
+# Pull the embedding model
+ollama pull nomic-embed-text
+
+# Ollama automatically runs on localhost:11434
+# HeroDB will use this by default
+```
+
+**Default Configuration:**
+- **URL**: `http://localhost:11434`
+- **Model**: `nomic-embed-text`
+- **Dimensions**: 768 (for nomic-embed-text)
+
+### Custom Embedding Service (Optional)
+
+To use a custom embedding service instead of Ollama:
+
+```bash
+# Set custom embedding service URL
+redis-cli HSET config:core:aiembed url "http://your-embedding-service:8080/embed"
+
+# Optional: Set authentication if required
+redis-cli HSET config:core:aiembed token "your-api-token"
+```
+
+### Embedding Service API Contracts
+
+#### Ollama API (Default)
+HeroDB calls Ollama using this format:
+
+```bash
+POST http://localhost:11434/api/embeddings
+Content-Type: application/json
+
+{
+  "model": "nomic-embed-text",
+  "prompt": "Your text to embed"
+}
+```
+
+Response:
+```json
+{
+  "embedding": [0.1, 0.2, 0.3, ...]
+}
+```
+
+#### Custom Service API
+Your custom embedding service should accept POST requests with this JSON format:
+
+```json
+{
+  "texts": ["text1", "text2"],           // Optional: array of texts
+  "images": ["base64_image1", "base64_image2"], // Optional: base64 encoded images
+  "model": "your-model-name"             // Optional: model specification
+}
+```
+
+And return responses in this format:
+
+```json
+{
+  "embeddings": [[0.1, 0.2, ...], [0.3, 0.4, ...]],  // Array of embedding vectors
+  "model": "model-name",                               // Model used
+  "usage": {                                          // Optional usage stats
+    "tokens": 100,
+    "requests": 2
+  }
+}
+```
+
+## Commands Reference
+
+### Dataset Management
+
+#### LANCE CREATE
+Create a new vector dataset with specified dimensions and optional schema.
+
+```bash
+LANCE CREATE <dataset> DIM <dimension> [SCHEMA field:type ...]
+```
+
+**Parameters:**
+- `dataset`: Name of the dataset
+- `dimension`: Vector dimension (e.g., 384, 768, 1536)
+- `field:type`: Optional metadata fields (string, int, float, bool)
+
+**Examples:**
+```bash
+# Create a simple dataset for 384-dimensional vectors
+LANCE CREATE documents DIM 384
+
+# Create dataset with metadata schema
+LANCE CREATE products DIM 768 SCHEMA category:string price:float available:bool
+```
+
+#### LANCE LIST
+List all available datasets.
+
+```bash
+LANCE LIST
+```
+
+**Returns:** Array of dataset names
+
+#### LANCE INFO
+Get information about a specific dataset.
+
+```bash
+LANCE INFO <dataset>
+```
+
+**Returns:** Dataset metadata including name, version, row count, and schema
+
+#### LANCE DROP
+Delete a dataset and all its data.
+
+```bash
+LANCE DROP <dataset>
+```
+
+### Data Operations
+
+#### LANCE STORE
+Store multimodal data (text/images) with automatic embedding generation.
+
+```bash
+LANCE STORE <dataset> [TEXT <text>] [IMAGE <base64>] [key value ...]
+```
+
+**Parameters:**
+- `dataset`: Target dataset name
+- `TEXT`: Text content to embed
+- `IMAGE`: Base64-encoded image to embed
+- `key value`: Metadata key-value pairs
+
+**Examples:**
+```bash
+# Store text with metadata
+LANCE STORE documents TEXT "Machine learning is transforming industries" category "AI" author "John Doe"
+
+# Store image with metadata
+LANCE STORE images IMAGE "iVBORw0KGgoAAAANSUhEUgAA..." category "nature" tags "landscape,mountains"
+
+# Store both text and image
+LANCE STORE multimodal TEXT "Beautiful sunset" IMAGE "base64data..." location "California"
+```
+
+**Returns:** Unique ID of the stored item
+
+### Search Operations
+
+#### LANCE SEARCH
+Search using a raw vector.
+
+```bash
+LANCE SEARCH <dataset> VECTOR <vector> K <k> [NPROBES <n>] [REFINE <r>]
+```
+
+**Parameters:**
+- `dataset`: Dataset to search
+- `vector`: Comma-separated vector values (e.g., "0.1,0.2,0.3")
+- `k`: Number of results to return
+- `NPROBES`: Number of partitions to search (optional)
+- `REFINE`: Refine factor for better accuracy (optional)
+
+**Example:**
+```bash
+LANCE SEARCH documents VECTOR "0.1,0.2,0.3,0.4" K 5 NPROBES 10
+```
+
+#### LANCE SEARCH.TEXT
+Search using text query (automatically embedded).
+
+```bash
+LANCE SEARCH.TEXT <dataset> <query_text> K <k> [NPROBES <n>] [REFINE <r>]
+```
+
+**Parameters:**
+- `dataset`: Dataset to search
+- `query_text`: Text query to search for
+- `k`: Number of results to return
+- `NPROBES`: Number of partitions to search (optional)
+- `REFINE`: Refine factor for better accuracy (optional)
+
+**Example:**
+```bash
+LANCE SEARCH.TEXT documents "artificial intelligence applications" K 10 NPROBES 20
+```
+
+**Returns:** Array of results with distance scores and metadata
+
+### Embedding Operations
+
+#### LANCE EMBED.TEXT
+Generate embeddings for text without storing.
+
+```bash
+LANCE EMBED.TEXT <text1> [text2] [text3] ...
+```
+
+**Example:**
+```bash
+LANCE EMBED.TEXT "Hello world" "Machine learning" "Vector database"
+```
+
+**Returns:** Array of embedding vectors
+
+### Index Management
+
+#### LANCE CREATE.INDEX
+Create a vector index for faster search performance.
+
+```bash
+LANCE CREATE.INDEX <dataset> <index_type> [PARTITIONS <n>] [SUBVECTORS <n>]
+```
+
+**Parameters:**
+- `dataset`: Dataset to index
+- `index_type`: Index type (currently supports "IVF_PQ")
+- `PARTITIONS`: Number of partitions (default: 256)
+- `SUBVECTORS`: Number of sub-vectors for PQ (default: 16)
+
+**Example:**
+```bash
+LANCE CREATE.INDEX documents IVF_PQ PARTITIONS 512 SUBVECTORS 32
+```
+
+## Usage Patterns
+
+### 1. Document Search System
+
+```bash
+# Setup
+LANCE CREATE documents DIM 384 SCHEMA title:string content:string category:string
+
+# Store documents
+LANCE STORE documents TEXT "Introduction to machine learning algorithms" title "ML Basics" category "education"
+LANCE STORE documents TEXT "Deep learning neural networks explained" title "Deep Learning" category "education"
+LANCE STORE documents TEXT "Building scalable web applications" title "Web Dev" category "programming"
+
+# Create index for better performance
+LANCE CREATE.INDEX documents IVF_PQ PARTITIONS 256
+
+# Search
+LANCE SEARCH.TEXT documents "neural networks" K 5
+```
+
+### 2. Image Similarity Search
+
+```bash
+# Setup
+LANCE CREATE images DIM 512 SCHEMA filename:string tags:string
+
+# Store images (base64 encoded)
+LANCE STORE images IMAGE "iVBORw0KGgoAAAANSUhEUgAA..." filename "sunset.jpg" tags "nature,landscape"
+LANCE STORE images IMAGE "iVBORw0KGgoAAAANSUhEUgBB..." filename "city.jpg" tags "urban,architecture"
+
+# Search by image
+LANCE STORE temp_search IMAGE "query_image_base64..."
+# Then use the returned ID to get embedding and search
+```
+
+### 3. Multimodal Content Management
+
+```bash
+# Setup
+LANCE CREATE content DIM 768 SCHEMA type:string source:string
+
+# Store mixed content
+LANCE STORE content TEXT "Product description for smartphone" type "product" source "catalog"
+LANCE STORE content IMAGE "product_image_base64..." type "product_image" source "catalog"
+
+# Search across all content types
+LANCE SEARCH.TEXT content "smartphone features" K 10
+```
+
+## Performance Considerations
+
+### Vector Dimensions
+- **384**: Good for general text (e.g., sentence-transformers)
+- **768**: Standard for BERT-like models
+- **1536**: OpenAI text-embedding-ada-002
+- **Higher dimensions**: Better accuracy but slower search
+
+### Index Configuration
+- **More partitions**: Better for larger datasets (>100K vectors)
+- **More sub-vectors**: Better compression but slower search
+- **NPROBES**: Higher values = better accuracy, slower search
+
+### Best Practices
+
+1. **Create indexes** for datasets with >1000 vectors
+2. **Use appropriate dimensions** based on your embedding model
+3. **Configure NPROBES** based on accuracy vs speed requirements
+4. **Batch operations** when possible for better performance
+5. **Monitor embedding service** response times and rate limits
+
+## Error Handling
+
+Common error scenarios and solutions:
+
+### Embedding Service Errors
+```bash
+# Error: Embedding service not configured
+ERR Embedding service URL not configured. Set it with: HSET config:core:aiembed url <YOUR_EMBEDDING_SERVICE_URL>
+
+# Error: Service unavailable
+ERR Embedding service returned error 404 Not Found
+```
+
+**Solution:** Ensure embedding service is running and URL is correct.
+
+### Dataset Errors
+```bash
+# Error: Dataset doesn't exist
+ERR Dataset 'mydata' does not exist
+
+# Error: Dimension mismatch
+ERR Vector dimension mismatch: expected 384, got 768
+```
+
+**Solution:** Create dataset first or check vector dimensions.
+
+### Search Errors
+```bash
+# Error: Invalid vector format
+ERR Invalid vector format
+
+# Error: No index available
+ERR No index available for fast search
+```
+
+**Solution:** Check vector format or create an index.
+
+## Integration Examples
+
+### With Python
+```python
+import redis
+import json
+
+r = redis.Redis(host='localhost', port=6379)
+
+# Create dataset
+r.execute_command('LANCE', 'CREATE', 'docs', 'DIM', '384')
+
+# Store document
+result = r.execute_command('LANCE', 'STORE', 'docs', 
+                          'TEXT', 'Machine learning tutorial',
+                          'category', 'education')
+print(f"Stored with ID: {result}")
+
+# Search
+results = r.execute_command('LANCE', 'SEARCH.TEXT', 'docs', 
+                           'machine learning', 'K', '5')
+print(f"Search results: {results}")
+```
+
+### With Node.js
+```javascript
+const redis = require('redis');
+const client = redis.createClient();
+
+// Create dataset
+await client.sendCommand(['LANCE', 'CREATE', 'docs', 'DIM', '384']);
+
+// Store document
+const id = await client.sendCommand(['LANCE', 'STORE', 'docs', 
+                                   'TEXT', 'Deep learning guide',
+                                   'category', 'AI']);
+
+// Search
+const results = await client.sendCommand(['LANCE', 'SEARCH.TEXT', 'docs',
+                                        'deep learning', 'K', '10']);
+```
+
+## Monitoring and Maintenance
+
+### Health Checks
+```bash
+# Check if Lance store is available
+LANCE LIST
+
+# Check dataset health
+LANCE INFO mydataset
+
+# Test embedding service
+LANCE EMBED.TEXT "test"
+```
+
+### Maintenance Operations
+```bash
+# Backup: Use standard Redis backup procedures
+# The Lance data is stored separately in the data directory
+
+# Cleanup: Remove unused datasets
+LANCE DROP old_dataset
+
+# Reindex: Drop and recreate indexes if needed
+LANCE DROP dataset_name
+LANCE CREATE dataset_name DIM 384
+# Re-import data
+LANCE CREATE.INDEX dataset_name IVF_PQ
+```
+
+This integration provides a powerful foundation for building AI-powered applications with vector search capabilities while maintaining the familiar Redis interface.

examples/README.md

@@ -1,6 +1,191 @@
-# HeroDB Tantivy Search Examples
+# HeroDB Examples
 
-This directory contains examples demonstrating HeroDB's full-text search capabilities powered by Tantivy.
+This directory contains examples demonstrating HeroDB's capabilities including full-text search powered by Tantivy and vector database operations using Lance.
+
+## Available Examples
+
+1. **[Tantivy Search Demo](#tantivy-search-demo-bash-script)** - Full-text search capabilities
+2. **[Lance Vector Database Demo](#lance-vector-database-demo-bash-script)** - Vector database and AI operations
+3. **[AGE Encryption Demo](age_bash_demo.sh)** - Cryptographic operations
+4. **[Simple Demo](simple_demo.sh)** - Basic Redis operations
+
+---
+
+## Lance Vector Database Demo (Bash Script)
+
+### Overview
+The `lance_vector_demo.sh` script provides a comprehensive demonstration of HeroDB's vector database capabilities using Lance. It showcases vector storage, similarity search, multimodal data handling, and AI-powered operations with external embedding services.
+
+### Prerequisites
+1. **HeroDB Server**: The server must be running (default port 6379)
+2. **Redis CLI**: The `redis-cli` tool must be installed and available in your PATH
+3. **Embedding Service** (optional): For full functionality, set up an external embedding service
+
+### Running the Demo
+
+#### Step 1: Start HeroDB Server
+```bash
+# From the project root directory
+cargo run -- --dir ./test_data --port 6379
+```
+
+#### Step 2: Run the Demo (in a new terminal)
+```bash
+# From the project root directory
+./examples/lance_vector_demo.sh
+```
+
+### What the Demo Covers
+
+The script demonstrates comprehensive vector database operations:
+
+1. **Dataset Management**
+   - Creating vector datasets with custom dimensions
+   - Defining schemas with metadata fields
+   - Listing and inspecting datasets
+   - Dataset information and statistics
+
+2. **Embedding Operations**
+   - Text embedding generation via external services
+   - Multimodal embedding support (text + images)
+   - Batch embedding operations
+
+3. **Data Storage**
+   - Storing text documents with automatic embedding
+   - Storing images with metadata
+   - Multimodal content storage
+   - Rich metadata support
+
+4. **Vector Search**
+   - Similarity search with raw vectors
+   - Text-based semantic search
+   - Configurable search parameters (K, NPROBES, REFINE)
+   - Cross-modal search capabilities
+
+5. **Index Management**
+   - Creating IVF_PQ indexes for performance
+   - Custom index parameters
+   - Performance optimization
+
+6. **Advanced Features**
+   - Error handling and recovery
+   - Performance testing concepts
+   - Monitoring and maintenance
+   - Cleanup operations
+
+### Key Lance Commands Demonstrated
+
+#### Dataset Management
+```bash
+# Create vector dataset
+LANCE CREATE documents DIM 384
+
+# Create dataset with schema
+LANCE CREATE products DIM 768 SCHEMA category:string price:float available:bool
+
+# List datasets
+LANCE LIST
+
+# Get dataset information
+LANCE INFO documents
+```
+
+#### Data Operations
+```bash
+# Store text with metadata
+LANCE STORE documents TEXT "Machine learning tutorial" category "education" author "John Doe"
+
+# Store image with metadata
+LANCE STORE images IMAGE "base64_encoded_image..." filename "photo.jpg" tags "nature,landscape"
+
+# Store multimodal content
+LANCE STORE content TEXT "Product description" IMAGE "base64_image..." type "product"
+```
+
+#### Search Operations
+```bash
+# Search with raw vector
+LANCE SEARCH documents VECTOR "0.1,0.2,0.3,0.4" K 5
+
+# Semantic text search
+LANCE SEARCH.TEXT documents "artificial intelligence" K 10 NPROBES 20
+
+# Generate embeddings
+LANCE EMBED.TEXT "Hello world" "Machine learning"
+```
+
+#### Index Management
+```bash
+# Create performance index
+LANCE CREATE.INDEX documents IVF_PQ PARTITIONS 256 SUBVECTORS 16
+
+# Drop dataset
+LANCE DROP old_dataset
+```
+
+### Configuration
+
+#### Setting Up Embedding Service
+```bash
+# Configure embedding service URL
+redis-cli HSET config:core:aiembed url "http://your-embedding-service:8080/embed"
+
+# Optional: Set authentication token
+redis-cli HSET config:core:aiembed token "your-api-token"
+```
+
+#### Embedding Service API
+Your embedding service should accept POST requests:
+```json
+{
+  "texts": ["text1", "text2"],
+  "images": ["base64_image1", "base64_image2"],
+  "model": "your-model-name"
+}
+```
+
+And return responses:
+```json
+{
+  "embeddings": [[0.1, 0.2, ...], [0.3, 0.4, ...]],
+  "model": "model-name",
+  "usage": {"tokens": 100, "requests": 2}
+}
+```
+
+### Interactive Features
+
+The demo script includes:
+- **Colored output** for better readability
+- **Step-by-step execution** with explanations
+- **Error handling** demonstrations
+- **Automatic cleanup** options
+- **Performance testing** concepts
+- **Real-world usage** examples
+
+### Use Cases Demonstrated
+
+1. **Document Search System**
+   - Semantic document retrieval
+   - Metadata filtering
+   - Relevance ranking
+
+2. **Image Similarity Search**
+   - Visual content matching
+   - Tag-based filtering
+   - Multimodal queries
+
+3. **Product Recommendations**
+   - Feature-based similarity
+   - Category filtering
+   - Price range queries
+
+4. **Content Management**
+   - Mixed media storage
+   - Cross-modal search
+   - Rich metadata support
+
+---
 
 ## Tantivy Search Demo (Bash Script)
 

examples/lance_vector_demo.sh

@@ -0,0 +1,426 @@
+#!/bin/bash
+
+# Lance Vector Database Demo Script
+# This script demonstrates all Lance vector database operations in HeroDB
+
+set -e  # Exit on any error
+
+# Configuration
+REDIS_HOST="localhost"
+REDIS_PORT="6379"
+REDIS_CLI="redis-cli -h $REDIS_HOST -p $REDIS_PORT"
+
+# Colors for output
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# Helper functions
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+execute_command() {
+    local cmd="$1"
+    local description="$2"
+    
+    echo
+    log_info "Executing: $description"
+    echo "Command: $cmd"
+    
+    if result=$($cmd 2>&1); then
+        log_success "Result: $result"
+    else
+        log_error "Failed: $result"
+        return 1
+    fi
+}
+
+# Check if HeroDB is running
+check_herodb() {
+    log_info "Checking if HeroDB is running..."
+    if ! $REDIS_CLI ping > /dev/null 2>&1; then
+        log_error "HeroDB is not running. Please start it first:"
+        echo "  cargo run -- --dir ./test_data --port $REDIS_PORT"
+        exit 1
+    fi
+    log_success "HeroDB is running"
+}
+
+# Setup embedding service configuration
+setup_embedding_service() {
+    log_info "Setting up embedding service configuration..."
+    
+    # Note: This is a mock URL for demonstration
+    # In production, replace with your actual embedding service
+    execute_command \
+        "$REDIS_CLI HSET config:core:aiembed url 'http://localhost:8080/embed'" \
+        "Configure embedding service URL"
+    
+    # Optional: Set authentication token
+    # execute_command \
+    #     "$REDIS_CLI HSET config:core:aiembed token 'your-api-token'" \
+    #     "Configure embedding service token"
+    
+    log_warning "Note: Embedding service at http://localhost:8080/embed is not running."
+    log_warning "Some operations will fail, but this demonstrates the command structure."
+}
+
+# Dataset Management Operations
+demo_dataset_management() {
+    echo
+    echo "=========================================="
+    echo "         DATASET MANAGEMENT DEMO"
+    echo "=========================================="
+    
+    # List datasets (should be empty initially)
+    execute_command \
+        "$REDIS_CLI LANCE LIST" \
+        "List all datasets (initially empty)"
+    
+    # Create a simple dataset
+    execute_command \
+        "$REDIS_CLI LANCE CREATE documents DIM 384" \
+        "Create a simple document dataset with 384 dimensions"
+    
+    # Create a dataset with schema
+    execute_command \
+        "$REDIS_CLI LANCE CREATE products DIM 768 SCHEMA category:string price:float available:bool description:string" \
+        "Create products dataset with custom schema"
+    
+    # Create an image dataset
+    execute_command \
+        "$REDIS_CLI LANCE CREATE images DIM 512 SCHEMA filename:string tags:string width:int height:int" \
+        "Create images dataset for multimodal content"
+    
+    # List datasets again
+    execute_command \
+        "$REDIS_CLI LANCE LIST" \
+        "List all datasets (should show 3 datasets)"
+    
+    # Get info about datasets
+    execute_command \
+        "$REDIS_CLI LANCE INFO documents" \
+        "Get information about documents dataset"
+    
+    execute_command \
+        "$REDIS_CLI LANCE INFO products" \
+        "Get information about products dataset"
+}
+
+# Embedding Operations
+demo_embedding_operations() {
+    echo
+    echo "=========================================="
+    echo "         EMBEDDING OPERATIONS DEMO"
+    echo "=========================================="
+    
+    log_warning "The following operations will fail because no embedding service is running."
+    log_warning "This demonstrates the command structure and error handling."
+    
+    # Try to embed text (will fail without embedding service)
+    execute_command \
+        "$REDIS_CLI LANCE EMBED.TEXT 'Hello world'" \
+        "Generate embedding for single text" || true
+    
+    # Try to embed multiple texts
+    execute_command \
+        "$REDIS_CLI LANCE EMBED.TEXT 'Machine learning' 'Artificial intelligence' 'Deep learning'" \
+        "Generate embeddings for multiple texts" || true
+}
+
+# Data Storage Operations
+demo_data_storage() {
+    echo
+    echo "=========================================="
+    echo "         DATA STORAGE DEMO"
+    echo "=========================================="
+    
+    log_warning "Storage operations will fail without embedding service, but show command structure."
+    
+    # Store text documents
+    execute_command \
+        "$REDIS_CLI LANCE STORE documents TEXT 'Introduction to machine learning algorithms and their applications in modern AI systems' category 'education' author 'John Doe' difficulty 'beginner'" \
+        "Store a document with text and metadata" || true
+    
+    execute_command \
+        "$REDIS_CLI LANCE STORE documents TEXT 'Deep learning neural networks for computer vision tasks' category 'research' author 'Jane Smith' difficulty 'advanced'" \
+        "Store another document" || true
+    
+    # Store product information
+    execute_command \
+        "$REDIS_CLI LANCE STORE products TEXT 'High-performance laptop with 16GB RAM and SSD storage' category 'electronics' price '1299.99' available 'true'" \
+        "Store product with text description" || true
+    
+    # Store image with metadata (using placeholder base64)
+    execute_command \
+        "$REDIS_CLI LANCE STORE images IMAGE 'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8/5+hHgAHggJ/PchI7wAAAABJRU5ErkJggg==' filename 'sample.png' tags 'test,demo' width '1' height '1'" \
+        "Store image with metadata (1x1 pixel PNG)" || true
+    
+    # Store multimodal content
+    execute_command \
+        "$REDIS_CLI LANCE STORE images TEXT 'Beautiful sunset over mountains' IMAGE 'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8/5+hHgAHggJ/PchI7wAAAABJRU5ErkJggg==' filename 'sunset.png' tags 'nature,landscape' location 'California'" \
+        "Store multimodal content (text + image)" || true
+}
+
+# Search Operations
+demo_search_operations() {
+    echo
+    echo "=========================================="
+    echo "         SEARCH OPERATIONS DEMO"
+    echo "=========================================="
+    
+    log_warning "Search operations will fail without data, but show command structure."
+    
+    # Search with raw vector
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH documents VECTOR '0.1,0.2,0.3,0.4,0.5' K 5" \
+        "Search with raw vector (5 results)" || true
+    
+    # Search with vector and parameters
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH documents VECTOR '0.1,0.2,0.3,0.4,0.5' K 10 NPROBES 20 REFINE 2" \
+        "Search with vector and advanced parameters" || true
+    
+    # Text-based search
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH.TEXT documents 'machine learning algorithms' K 5" \
+        "Search using text query" || true
+    
+    # Text search with parameters
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH.TEXT products 'laptop computer' K 3 NPROBES 10" \
+        "Search products using text with parameters" || true
+    
+    # Search in image dataset
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH.TEXT images 'sunset landscape' K 5" \
+        "Search images using text description" || true
+}
+
+# Index Management Operations
+demo_index_management() {
+    echo
+    echo "=========================================="
+    echo "         INDEX MANAGEMENT DEMO"
+    echo "=========================================="
+    
+    # Create indexes for better search performance
+    execute_command \
+        "$REDIS_CLI LANCE CREATE.INDEX documents IVF_PQ" \
+        "Create default IVF_PQ index for documents"
+    
+    execute_command \
+        "$REDIS_CLI LANCE CREATE.INDEX products IVF_PQ PARTITIONS 512 SUBVECTORS 32" \
+        "Create IVF_PQ index with custom parameters for products"
+    
+    execute_command \
+        "$REDIS_CLI LANCE CREATE.INDEX images IVF_PQ PARTITIONS 256 SUBVECTORS 16" \
+        "Create IVF_PQ index for images dataset"
+    
+    log_success "Indexes created successfully"
+}
+
+# Advanced Usage Examples
+demo_advanced_usage() {
+    echo
+    echo "=========================================="
+    echo "         ADVANCED USAGE EXAMPLES"
+    echo "=========================================="
+    
+    # Create a specialized dataset for semantic search
+    execute_command \
+        "$REDIS_CLI LANCE CREATE semantic_search DIM 1536 SCHEMA title:string content:string url:string timestamp:string source:string" \
+        "Create dataset for semantic search with rich metadata"
+    
+    # Demonstrate batch operations concept
+    log_info "Batch operations example (would store multiple items):"
+    echo "  for doc in documents:"
+    echo "    LANCE STORE semantic_search TEXT \"\$doc_content\" title \"\$title\" url \"\$url\""
+    
+    # Show monitoring commands
+    log_info "Monitoring and maintenance commands:"
+    execute_command \
+        "$REDIS_CLI LANCE LIST" \
+        "List all datasets for monitoring"
+    
+    # Show dataset statistics
+    for dataset in documents products images semantic_search; do
+        execute_command \
+            "$REDIS_CLI LANCE INFO $dataset" \
+            "Get statistics for $dataset" || true
+    done
+}
+
+# Cleanup Operations
+demo_cleanup() {
+    echo
+    echo "=========================================="
+    echo "         CLEANUP OPERATIONS DEMO"
+    echo "=========================================="
+    
+    log_info "Demonstrating cleanup operations..."
+    
+    # Drop individual datasets
+    execute_command \
+        "$REDIS_CLI LANCE DROP semantic_search" \
+        "Drop semantic_search dataset"
+    
+    # List remaining datasets
+    execute_command \
+        "$REDIS_CLI LANCE LIST" \
+        "List remaining datasets"
+    
+    # Ask user if they want to clean up all test data
+    echo
+    read -p "Do you want to clean up all test datasets? (y/N): " -n 1 -r
+    echo
+    if [[ $REPLY =~ ^[Yy]$ ]]; then
+        execute_command \
+            "$REDIS_CLI LANCE DROP documents" \
+            "Drop documents dataset"
+        
+        execute_command \
+            "$REDIS_CLI LANCE DROP products" \
+            "Drop products dataset"
+        
+        execute_command \
+            "$REDIS_CLI LANCE DROP images" \
+            "Drop images dataset"
+        
+        execute_command \
+            "$REDIS_CLI LANCE LIST" \
+            "Verify all datasets are cleaned up"
+        
+        log_success "All test datasets cleaned up"
+    else
+        log_info "Keeping test datasets for further experimentation"
+    fi
+}
+
+# Error Handling Demo
+demo_error_handling() {
+    echo
+    echo "=========================================="
+    echo "         ERROR HANDLING DEMO"
+    echo "=========================================="
+    
+    log_info "Demonstrating various error conditions..."
+    
+    # Try to access non-existent dataset
+    execute_command \
+        "$REDIS_CLI LANCE INFO nonexistent_dataset" \
+        "Try to get info for non-existent dataset" || true
+    
+    # Try to search non-existent dataset
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH nonexistent_dataset VECTOR '0.1,0.2' K 5" \
+        "Try to search non-existent dataset" || true
+    
+    # Try to drop non-existent dataset
+    execute_command \
+        "$REDIS_CLI LANCE DROP nonexistent_dataset" \
+        "Try to drop non-existent dataset" || true
+    
+    # Try invalid vector format
+    execute_command \
+        "$REDIS_CLI LANCE SEARCH documents VECTOR 'invalid,vector,format' K 5" \
+        "Try search with invalid vector format" || true
+    
+    log_info "Error handling demonstration complete"
+}
+
+# Performance Testing Demo
+demo_performance_testing() {
+    echo
+    echo "=========================================="
+    echo "         PERFORMANCE TESTING DEMO"
+    echo "=========================================="
+    
+    log_info "Creating performance test dataset..."
+    execute_command \
+        "$REDIS_CLI LANCE CREATE perf_test DIM 128 SCHEMA batch_id:string item_id:string" \
+        "Create performance test dataset"
+    
+    log_info "Performance testing would involve:"
+    echo "  1. Bulk loading thousands of vectors"
+    echo "  2. Creating indexes with different parameters"
+    echo "  3. Measuring search latency with various K values"
+    echo "  4. Testing different NPROBES settings"
+    echo "  5. Monitoring memory usage"
+    
+    log_info "Example performance test commands:"
+    echo "  # Test search speed with different parameters"
+    echo "  time redis-cli LANCE SEARCH.TEXT perf_test 'query' K 10"
+    echo "  time redis-cli LANCE SEARCH.TEXT perf_test 'query' K 10 NPROBES 50"
+    echo "  time redis-cli LANCE SEARCH.TEXT perf_test 'query' K 100 NPROBES 100"
+    
+    # Clean up performance test dataset
+    execute_command \
+        "$REDIS_CLI LANCE DROP perf_test" \
+        "Clean up performance test dataset"
+}
+
+# Main execution
+main() {
+    echo "=========================================="
+    echo "    LANCE VECTOR DATABASE DEMO SCRIPT"
+    echo "=========================================="
+    echo
+    echo "This script demonstrates all Lance vector database operations."
+    echo "Note: Some operations will fail without a running embedding service."
+    echo "This is expected and demonstrates error handling."
+    echo
+    
+    # Check prerequisites
+    check_herodb
+    
+    # Setup
+    setup_embedding_service
+    
+    # Run demos
+    demo_dataset_management
+    demo_embedding_operations
+    demo_data_storage
+    demo_search_operations
+    demo_index_management
+    demo_advanced_usage
+    demo_error_handling
+    demo_performance_testing
+    
+    # Cleanup
+    demo_cleanup
+    
+    echo
+    echo "=========================================="
+    echo "         DEMO COMPLETE"
+    echo "=========================================="
+    echo
+    log_success "Lance vector database demo completed successfully!"
+    echo
+    echo "Next steps:"
+    echo "1. Set up a real embedding service (OpenAI, Hugging Face, etc.)"
+    echo "2. Update the embedding service URL configuration"
+    echo "3. Try storing and searching real data"
+    echo "4. Experiment with different vector dimensions and index parameters"
+    echo "5. Build your AI-powered application!"
+    echo
+    echo "For more information, see docs/lance_vector_db.md"
+}
+
+# Run the demo
+main "$@"

examples/tantivy_search_demo.sh

@@ -1,239 +0,0 @@
-#!/bin/bash
-
-# HeroDB Tantivy Search Demo
-# This script demonstrates full-text search capabilities using Redis commands
-# HeroDB server should be running on port 6381
-
-set -e  # Exit on any error
-
-# Configuration
-REDIS_HOST="localhost"
-REDIS_PORT="6382"
-REDIS_CLI="redis-cli -h $REDIS_HOST -p $REDIS_PORT"
-
-# Start the herodb server in the background
-echo "Starting herodb server..."
-cargo run -p herodb -- --dir /tmp/herodbtest --port ${REDIS_PORT} --debug &
-SERVER_PID=$!
-echo
-sleep 2 # Give the server a moment to start
-
-# Colors for output
-RED='\033[0;31m'
-GREEN='\033[0;32m'
-BLUE='\033[0;34m'
-YELLOW='\033[1;33m'
-NC='\033[0m' # No Color
-
-# Function to print colored output
-print_header() {
-    echo -e "${BLUE}=== $1 ===${NC}"
-}
-
-print_success() {
-    echo -e "${GREEN}✓ $1${NC}"
-}
-
-print_info() {
-    echo -e "${YELLOW}ℹ $1${NC}"
-}
-
-print_error() {
-    echo -e "${RED}✗ $1${NC}"
-}
-
-# Function to check if HeroDB is running
-check_herodb() {
-    print_info "Checking if HeroDB is running on port $REDIS_PORT..."
-    if ! $REDIS_CLI ping > /dev/null 2>&1; then
-        print_error "HeroDB is not running on port $REDIS_PORT"
-        print_info "Please start HeroDB with: cargo run -- --port $REDIS_PORT"
-        exit 1
-    fi
-    print_success "HeroDB is running and responding"
-}
-
-# Function to execute Redis command with error handling
-execute_cmd() {
-    local cmd="$1"
-    local description="$2"
-    
-    echo -e "${YELLOW}Command:${NC} $cmd"
-    if result=$($REDIS_CLI $cmd 2>&1); then
-        echo -e "${GREEN}Result:${NC} $result"
-        return 0
-    else
-        print_error "Failed: $description"
-        echo "Error: $result"
-        return 1
-    fi
-}
-
-# Function to pause for readability
-pause() {
-    echo
-    read -p "Press Enter to continue..."
-    echo
-}
-
-# Main demo function
-main() {
-    clear
-    print_header "HeroDB Tantivy Search Demonstration"
-    echo "This demo shows full-text search capabilities using Redis commands"
-    echo "HeroDB runs on port $REDIS_PORT (instead of Redis default 6379)"
-    echo
-
-    # Check if HeroDB is running
-    check_herodb
-    echo
-
-    print_header "Step 1: Create Search Index"
-    print_info "Creating a product catalog search index with various field types"
-    
-    # Create search index with schema
-    execute_cmd "FT.CREATE product_catalog SCHEMA title TEXT description TEXT category TAG price NUMERIC rating NUMERIC location GEO" \
-                "Creating search index"
-    
-    print_success "Search index 'product_catalog' created successfully"
-    pause
-
-    print_header "Step 2: Add Sample Products"
-    print_info "Adding sample products to demonstrate different search scenarios"
-    
-    # Add sample products using FT.ADD
-    execute_cmd "FT.ADD product_catalog product:1 1.0 title 'Wireless Bluetooth Headphones' description 'Premium noise-canceling headphones with 30-hour battery life' category 'electronics,audio' price 299.99 rating 4.5 location '-122.4194,37.7749'" "Adding product 1"
-    execute_cmd "FT.ADD product_catalog product:2 1.0 title 'Organic Coffee Beans' description 'Single-origin Ethiopian coffee beans, medium roast' category 'food,beverages,organic' price 24.99 rating 4.8 location '-74.0060,40.7128'" "Adding product 2"
-    execute_cmd "FT.ADD product_catalog product:3 1.0 title 'Yoga Mat Premium' description 'Eco-friendly yoga mat with superior grip and cushioning' category 'fitness,wellness,eco-friendly' price 89.99 rating 4.3 location '-118.2437,34.0522'" "Adding product 3"
-    execute_cmd "FT.ADD product_catalog product:4 1.0 title 'Smart Home Speaker' description 'Voice-controlled smart speaker with AI assistant' category 'electronics,smart-home' price 149.99 rating 4.2 location '-87.6298,41.8781'" "Adding product 4"
-    execute_cmd "FT.ADD product_catalog product:5 1.0 title 'Organic Green Tea' description 'Premium organic green tea leaves from Japan' category 'food,beverages,organic,tea' price 18.99 rating 4.7 location '139.6503,35.6762'" "Adding product 5"
-    execute_cmd "FT.ADD product_catalog product:6 1.0 title 'Wireless Gaming Mouse' description 'High-precision gaming mouse with RGB lighting' category 'electronics,gaming' price 79.99 rating 4.4 location '-122.3321,47.6062'" "Adding product 6"
-    execute_cmd "FT.ADD product_catalog product:7 1.0 title 'Comfortable meditation cushion for mindfulness practice' description 'Meditation cushion with premium materials' category 'wellness,meditation' price 45.99 rating 4.6 location '-122.4194,37.7749'" "Adding product 7"
-    execute_cmd "FT.ADD product_catalog product:8 1.0 title 'Bluetooth Earbuds' description 'True wireless earbuds with active noise cancellation' category 'electronics,audio' price 199.99 rating 4.1 location '-74.0060,40.7128'" "Adding product 8"
-    
-    print_success "Added 8 products to the index"
-    pause
-
-    print_header "Step 3: Basic Text Search"
-    print_info "Searching for 'wireless' products"
-    
-    execute_cmd "FT.SEARCH product_catalog wireless" "Basic text search"
-    pause
-
-    print_header "Step 4: Search with Filters"
-    print_info "Searching for 'organic' products"
-    
-    execute_cmd "FT.SEARCH product_catalog organic" "Filtered search"
-    pause
-
-    print_header "Step 5: Numeric Range Search"
-    print_info "Searching for 'premium' products"
-    
-    execute_cmd "FT.SEARCH product_catalog premium" "Text search"
-    pause
-
-    print_header "Step 6: Sorting Results"
-    print_info "Searching for electronics"
-    
-    execute_cmd "FT.SEARCH product_catalog electronics" "Category search"
-    pause
-
-    print_header "Step 7: Limiting Results"
-    print_info "Searching for wireless products with limit"
-    
-    execute_cmd "FT.SEARCH product_catalog wireless LIMIT 0 3" "Limited results"
-    pause
-
-    print_header "Step 8: Complex Query"
-    print_info "Finding audio products with noise cancellation"
-    
-    execute_cmd "FT.SEARCH product_catalog 'noise cancellation'" "Complex query"
-    pause
-
-    print_header "Step 9: Geographic Search"
-    print_info "Searching for meditation products"
-    
-    execute_cmd "FT.SEARCH product_catalog meditation" "Text search"
-    pause
-
-    print_header "Step 10: Aggregation Example"
-    print_info "Getting index information and statistics"
-    
-    execute_cmd "FT.INFO product_catalog" "Index information"
-    pause
-
-    print_header "Step 11: Search Comparison"
-    print_info "Comparing Tantivy search vs simple key matching"
-    
-    echo -e "${YELLOW}Tantivy Full-Text Search:${NC}"
-    execute_cmd "FT.SEARCH product_catalog 'battery life'" "Full-text search for 'battery life'"
-    
-    echo
-    echo -e "${YELLOW}Simple Key Pattern Matching:${NC}"
-    execute_cmd "KEYS *battery*" "Simple pattern matching for 'battery'"
-    
-    print_info "Notice how full-text search finds relevant results even when exact words don't match keys"
-    pause
-
-    print_header "Step 12: Fuzzy Search"
-    print_info "Searching for headphones"
-    
-    execute_cmd "FT.SEARCH product_catalog headphones" "Text search"
-    pause
-
-    print_header "Step 13: Phrase Search"
-    print_info "Searching for coffee products"
-    
-    execute_cmd "FT.SEARCH product_catalog coffee" "Text search"
-    pause
-
-    print_header "Step 14: Boolean Queries"
-    print_info "Searching for gaming products"
-    
-    execute_cmd "FT.SEARCH product_catalog gaming" "Text search"
-    echo
-    execute_cmd "FT.SEARCH product_catalog tea" "Text search"
-    pause
-
-    print_header "Step 15: Cleanup"
-    print_info "Removing test data"
-    
-    # Delete the search index
-    execute_cmd "FT.DROP product_catalog" "Dropping search index"
-    
-    # Clean up documents from search index
-    for i in {1..8}; do
-        execute_cmd "FT.DEL product_catalog product:$i" "Deleting product:$i from index"
-    done
-    
-    print_success "Cleanup completed"
-    echo
-
-    print_header "Demo Summary"
-    echo "This demonstration showed:"
-    echo "• Creating search indexes with different field types"
-    echo "• Adding documents to the search index"
-    echo "• Basic and advanced text search queries"
-    echo "• Filtering by categories and numeric ranges"
-    echo "• Sorting and limiting results"
-    echo "• Geographic searches"
-    echo "• Fuzzy matching and phrase searches"
-    echo "• Boolean query operators"
-    echo "• Comparison with simple pattern matching"
-    echo
-    print_success "HeroDB Tantivy search demo completed successfully!"
-    echo
-    print_info "Key advantages of Tantivy full-text search:"
-    echo "  - Relevance scoring and ranking"
-    echo "  - Fuzzy matching and typo tolerance"
-    echo "  - Complex boolean queries"
-    echo "  - Field-specific searches and filters"
-    echo "  - Geographic and numeric range queries"
-    echo "  - Much faster than pattern matching on large datasets"
-    echo
-    print_info "To run HeroDB server: cargo run -- --port 6381"
-    print_info "To connect with redis-cli: redis-cli -h localhost -p 6381"
-}
-
-# Run the demo
-main "$@"

examples/test_tantivy_integration.sh

@@ -1,101 +0,0 @@
-#!/bin/bash
-
-# Simple Tantivy Search Integration Test for HeroDB
-# This script tests the full-text search functionality we just integrated
-
-set -e
-
-echo "🔍 Testing Tantivy Search Integration..."
-
-# Build the project first
-echo "📦 Building HeroDB..."
-cargo build --release
-
-# Start the server in the background
-echo "🚀 Starting HeroDB server on port 6379..."
-cargo run --release -- --port 6379 --dir ./test_data &
-SERVER_PID=$!
-
-# Wait for server to start
-sleep 3
-
-# Function to cleanup on exit
-cleanup() {
-    echo "🧹 Cleaning up..."
-    kill $SERVER_PID 2>/dev/null || true
-    rm -rf ./test_data
-    exit
-}
-
-# Set trap for cleanup
-trap cleanup EXIT INT TERM
-
-# Function to execute Redis command
-execute_cmd() {
-    local cmd="$1"
-    local description="$2"
-    
-    echo "📝 $description"
-    echo "   Command: $cmd"
-    
-    if result=$(redis-cli -p 6379 $cmd 2>&1); then
-        echo "   ✅ Result: $result"
-        echo
-        return 0
-    else
-        echo "   ❌ Failed: $result"
-        echo
-        return 1
-    fi
-}
-
-echo "🧪 Running Tantivy Search Tests..."
-echo
-
-# Test 1: Create a search index
-execute_cmd "ft.create books SCHEMA title TEXT description TEXT author TEXT category TAG price NUMERIC" \
-           "Creating search index 'books'"
-
-# Test 2: Add documents to the index
-execute_cmd "ft.add books book1 1.0 title \"The Great Gatsby\" description \"A classic American novel about the Jazz Age\" author \"F. Scott Fitzgerald\" category \"fiction,classic\" price \"12.99\"" \
-           "Adding first book"
-
-execute_cmd "ft.add books book2 1.0 title \"To Kill a Mockingbird\" description \"A novel about racial injustice in the American South\" author \"Harper Lee\" category \"fiction,classic\" price \"14.99\"" \
-           "Adding second book"
-
-execute_cmd "ft.add books book3 1.0 title \"Programming Rust\" description \"A comprehensive guide to Rust programming language\" author \"Jim Blandy\" category \"programming,technical\" price \"49.99\"" \
-           "Adding third book"
-
-execute_cmd "ft.add books book4 1.0 title \"The Rust Programming Language\" description \"The official book on Rust programming\" author \"Steve Klabnik\" category \"programming,technical\" price \"39.99\"" \
-           "Adding fourth book"
-
-# Test 3: Basic search
-execute_cmd "ft.search books Rust" \
-           "Searching for 'Rust'"
-
-# Test 4: Search with filters
-execute_cmd "ft.search books programming FILTER category programming" \
-           "Searching for 'programming' with category filter"
-
-# Test 5: Search with limit
-execute_cmd "ft.search books \"*\" LIMIT 0 2" \
-           "Getting first 2 documents"
-
-# Test 6: Get index info
-execute_cmd "ft.info books" \
-           "Getting index information"
-
-# Test 7: Delete a document
-execute_cmd "ft.del books book1" \
-           "Deleting book1"
-
-# Test 8: Search again to verify deletion
-execute_cmd "ft.search books Gatsby" \
-           "Searching for deleted book"
-
-# Test 9: Drop the index
-execute_cmd "ft.drop books" \
-           "Dropping the index"
-
-echo "🎉 All tests completed successfully!"
-echo "✅ Tantivy search integration is working correctly"

src/cmd.rs

@@ -1,7 +1,8 @@
-use crate::{error::DBError, protocol::Protocol, server::Server, search_cmd};
+use crate::{error::DBError, protocol::Protocol, server::Server};
 use tokio::time::{timeout, Duration};
 use futures::future::select_all;
-use std::collections::HashMap;
+use std::sync::Arc;
+use base64::Engine;
 
 #[derive(Debug, Clone)]
 pub enum Cmd {
@@ -86,39 +87,47 @@ pub enum Cmd {
     AgeVerifyName(String, String, String), // name, message, signature_b64
     AgeList,
 
-    // Full-text search commands with schema support
-    FtCreate {
-        index_name: String,
-        schema: Vec<(String, String, Vec<String>)>, // (field_name, field_type, options)
+    // Lance vector database commands
+    LanceCreate {
+        dataset: String,
+        dim: usize,
+        schema: Vec<(String, String)>, // field_name, field_type pairs
     },
-    FtAdd {
-        index_name: String,
-        doc_id: String,
-        score: f64,
-        fields: std::collections::HashMap<String, String>,
+    LanceStore {
+        dataset: String,
+        text: Option<String>,
+        image_base64: Option<String>,
+        metadata: std::collections::HashMap<String, String>,
     },
-    FtSearch {
-        index_name: String,
-        query: String,
-        filters: Vec<(String, String)>,  // field, value pairs
-        limit: Option<usize>,
-        offset: Option<usize>,
-        return_fields: Option<Vec<String>>,
+    LanceSearch {
+        dataset: String,
+        vector: Vec<f32>,
+        k: usize,
+        nprobes: Option<usize>,
+        refine_factor: Option<usize>,
     },
-    FtDel(String, String),  // index_name, doc_id
-    FtInfo(String),  // index_name
-    FtDrop(String),  // index_name
-    FtAlter {
-        index_name: String,
-        field_name: String,
-        field_type: String,
-        options: Vec<String>,
+    LanceSearchText {
+        dataset: String,
+        query_text: String,
+        k: usize,
+        nprobes: Option<usize>,
+        refine_factor: Option<usize>,
     },
-    FtAggregate {
-        index_name: String,
-        query: String,
-        group_by: Vec<String>,
-        reducers: Vec<String>,
+    LanceEmbedText {
+        texts: Vec<String>,
+    },
+    LanceCreateIndex {
+        dataset: String,
+        index_type: String,
+        num_partitions: Option<usize>,
+        num_sub_vectors: Option<usize>,
+    },
+    LanceList,
+    LanceDrop {
+        dataset: String,
+    },
+    LanceInfo {
+        dataset: String,
     },
 }
 
@@ -652,148 +661,237 @@ impl Cmd {
                                 _ => return Err(DBError(format!("unsupported AGE subcommand {:?}", cmd))),
                             }
                         }
-                       "ft.create" => {
-                           if cmd.len() < 4 || cmd[2].to_uppercase() != "SCHEMA" {
-                               return Err(DBError("ERR FT.CREATE requires: indexname SCHEMA field1 type1 [options] ...".to_string()));
-                           }
-                           
-                           let index_name = cmd[1].clone();
-                           let mut schema = Vec::new();
-                           let mut i = 3;
-                           
-                           while i < cmd.len() {
-                               if i + 1 >= cmd.len() {
-                                   return Err(DBError("ERR incomplete field definition".to_string()));
-                               }
-                               
-                               let field_name = cmd[i].clone();
-                               let field_type = cmd[i + 1].to_uppercase();
-                               let mut options = Vec::new();
-                               i += 2;
-                               
-                               // Parse field options until we hit another field name or end
-                               while i < cmd.len() && !["TEXT", "NUMERIC", "TAG", "GEO"].contains(&cmd[i].to_uppercase().as_str()) {
-                                   options.push(cmd[i].to_uppercase());
-                                   i += 1;
-                                   
-                                   // If this option takes a value, consume it too
-                                   if i > 0 && ["SEPARATOR", "WEIGHT"].contains(&cmd[i-1].to_uppercase().as_str()) && i < cmd.len() {
-                                       options.push(cmd[i].clone());
-                                       i += 1;
-                                   }
-                               }
-                               
-                               schema.push((field_name, field_type, options));
-                           }
-                           
-                           Cmd::FtCreate {
-                               index_name,
-                               schema,
-                           }
-                       }
-                       "ft.add" => {
-                           if cmd.len() < 5 {
-                               return Err(DBError("ERR FT.ADD requires: index_name doc_id score field value ...".to_string()));
-                           }
-                           
-                           let index_name = cmd[1].clone();
-                           let doc_id = cmd[2].clone();
-                           let score = cmd[3].parse::<f64>()
-                               .map_err(|_| DBError("ERR score must be a number".to_string()))?;
-                           
-                           let mut fields = HashMap::new();
-                           let mut i = 4;
-                           
-                           while i + 1 < cmd.len() {
-                               fields.insert(cmd[i].clone(), cmd[i + 1].clone());
-                               i += 2;
-                           }
-                           
-                           Cmd::FtAdd {
-                               index_name,
-                               doc_id,
-                               score,
-                               fields,
-                           }
-                       }
-                       "ft.search" => {
-                           if cmd.len() < 3 {
-                               return Err(DBError("ERR FT.SEARCH requires: index_name query [options]".to_string()));
-                           }
-                           
-                           let index_name = cmd[1].clone();
-                           let query = cmd[2].clone();
-                           
-                           let mut filters = Vec::new();
-                           let mut limit = None;
-                           let mut offset = None;
-                           let mut return_fields = None;
-                           
-                           let mut i = 3;
-                           while i < cmd.len() {
-                               match cmd[i].to_uppercase().as_str() {
-                                   "FILTER" => {
-                                       if i + 3 >= cmd.len() {
-                                           return Err(DBError("ERR FILTER requires field and value".to_string()));
-                                       }
-                                       filters.push((cmd[i + 1].clone(), cmd[i + 2].clone()));
-                                       i += 3;
-                                   }
-                                   "LIMIT" => {
-                                       if i + 2 >= cmd.len() {
-                                           return Err(DBError("ERR LIMIT requires offset and num".to_string()));
-                                       }
-                                       offset = Some(cmd[i + 1].parse().unwrap_or(0));
-                                       limit = Some(cmd[i + 2].parse().unwrap_or(10));
-                                       i += 3;
-                                   }
-                                   "RETURN" => {
-                                       if i + 1 >= cmd.len() {
-                                           return Err(DBError("ERR RETURN requires field count".to_string()));
-                                       }
-                                       let count: usize = cmd[i + 1].parse().unwrap_or(0);
-                                       i += 2;
-                                       
-                                       let mut fields = Vec::new();
-                                       for _ in 0..count {
-                                           if i < cmd.len() {
-                                               fields.push(cmd[i].clone());
-                                               i += 1;
-                                           }
-                                       }
-                                       return_fields = Some(fields);
-                                   }
-                                   _ => i += 1,
-                               }
-                           }
-                           
-                           Cmd::FtSearch {
-                               index_name,
-                               query,
-                               filters,
-                               limit,
-                               offset,
-                               return_fields,
-                           }
-                       }
-                       "ft.del" => {
-                           if cmd.len() != 3 {
-                               return Err(DBError("ERR FT.DEL requires: index_name doc_id".to_string()));
-                           }
-                           Cmd::FtDel(cmd[1].clone(), cmd[2].clone())
-                       }
-                       "ft.info" => {
-                           if cmd.len() != 2 {
-                               return Err(DBError("ERR FT.INFO requires: index_name".to_string()));
-                           }
-                           Cmd::FtInfo(cmd[1].clone())
-                       }
-                       "ft.drop" => {
-                           if cmd.len() != 2 {
-                               return Err(DBError("ERR FT.DROP requires: index_name".to_string()));
-                           }
-                           Cmd::FtDrop(cmd[1].clone())
-                       }
+                        "lance" => {
+                            if cmd.len() < 2 {
+                                return Err(DBError("wrong number of arguments for LANCE".to_string()));
+                            }
+                            match cmd[1].to_lowercase().as_str() {
+                                "create" => {
+                                    if cmd.len() < 4 {
+                                        return Err(DBError("LANCE CREATE <dataset> DIM <dimension> [SCHEMA field:type ...]".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    
+                                    // Parse DIM parameter
+                                    if cmd[3].to_lowercase() != "dim" {
+                                        return Err(DBError("Expected DIM after dataset name".to_string()));
+                                    }
+                                    if cmd.len() < 5 {
+                                        return Err(DBError("Missing dimension value".to_string()));
+                                    }
+                                    let dim = cmd[4].parse::<usize>().map_err(|_| DBError("Invalid dimension value".to_string()))?;
+                                    
+                                    // Parse optional SCHEMA
+                                    let mut schema = Vec::new();
+                                    let mut i = 5;
+                                    if i < cmd.len() && cmd[i].to_lowercase() == "schema" {
+                                        i += 1;
+                                        while i < cmd.len() {
+                                            let field_spec = &cmd[i];
+                                            let parts: Vec<&str> = field_spec.split(':').collect();
+                                            if parts.len() != 2 {
+                                                return Err(DBError("Schema fields must be in format field:type".to_string()));
+                                            }
+                                            schema.push((parts[0].to_string(), parts[1].to_string()));
+                                            i += 1;
+                                        }
+                                    }
+                                    
+                                    Cmd::LanceCreate { dataset, dim, schema }
+                                }
+                                "store" => {
+                                    if cmd.len() < 3 {
+                                        return Err(DBError("LANCE STORE <dataset> [TEXT <text>] [IMAGE <base64>] [metadata...]".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    let mut text = None;
+                                    let mut image_base64 = None;
+                                    let mut metadata = std::collections::HashMap::new();
+                                    
+                                    let mut i = 3;
+                                    while i < cmd.len() {
+                                        match cmd[i].to_lowercase().as_str() {
+                                            "text" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("TEXT requires a value".to_string()));
+                                                }
+                                                text = Some(cmd[i + 1].clone());
+                                                i += 2;
+                                            }
+                                            "image" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("IMAGE requires a base64 value".to_string()));
+                                                }
+                                                image_base64 = Some(cmd[i + 1].clone());
+                                                i += 2;
+                                            }
+                                            _ => {
+                                                // Parse as metadata key:value
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("Metadata requires key value pairs".to_string()));
+                                                }
+                                                metadata.insert(cmd[i].clone(), cmd[i + 1].clone());
+                                                i += 2;
+                                            }
+                                        }
+                                    }
+                                    
+                                    Cmd::LanceStore { dataset, text, image_base64, metadata }
+                                }
+                                "search" => {
+                                    if cmd.len() < 5 {
+                                        return Err(DBError("LANCE SEARCH <dataset> VECTOR <vector> K <k> [NPROBES <n>] [REFINE <r>]".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    
+                                    if cmd[3].to_lowercase() != "vector" {
+                                        return Err(DBError("Expected VECTOR after dataset name".to_string()));
+                                    }
+                                    
+                                    // Parse vector - expect comma-separated floats in brackets or just comma-separated
+                                    let vector_str = &cmd[4];
+                                    let vector_str = vector_str.trim_start_matches('[').trim_end_matches(']');
+                                    let vector: Result<Vec<f32>, _> = vector_str
+                                        .split(',')
+                                        .map(|s| s.trim().parse::<f32>())
+                                        .collect();
+                                    let vector = vector.map_err(|_| DBError("Invalid vector format".to_string()))?;
+                                    
+                                    if cmd.len() < 7 || cmd[5].to_lowercase() != "k" {
+                                        return Err(DBError("Expected K after vector".to_string()));
+                                    }
+                                    let k = cmd[6].parse::<usize>().map_err(|_| DBError("Invalid K value".to_string()))?;
+                                    
+                                    let mut nprobes = None;
+                                    let mut refine_factor = None;
+                                    let mut i = 7;
+                                    while i < cmd.len() {
+                                        match cmd[i].to_lowercase().as_str() {
+                                            "nprobes" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("NPROBES requires a value".to_string()));
+                                                }
+                                                nprobes = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid NPROBES value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            "refine" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("REFINE requires a value".to_string()));
+                                                }
+                                                refine_factor = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid REFINE value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            _ => {
+                                                return Err(DBError(format!("Unknown parameter: {}", cmd[i])));
+                                            }
+                                        }
+                                    }
+                                    
+                                    Cmd::LanceSearch { dataset, vector, k, nprobes, refine_factor }
+                                }
+                                "search.text" => {
+                                    if cmd.len() < 6 {
+                                        return Err(DBError("LANCE SEARCH.TEXT <dataset> <query_text> K <k> [NPROBES <n>] [REFINE <r>]".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    let query_text = cmd[3].clone();
+                                    
+                                    if cmd[4].to_lowercase() != "k" {
+                                        return Err(DBError("Expected K after query text".to_string()));
+                                    }
+                                    let k = cmd[5].parse::<usize>().map_err(|_| DBError("Invalid K value".to_string()))?;
+                                    
+                                    let mut nprobes = None;
+                                    let mut refine_factor = None;
+                                    let mut i = 6;
+                                    while i < cmd.len() {
+                                        match cmd[i].to_lowercase().as_str() {
+                                            "nprobes" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("NPROBES requires a value".to_string()));
+                                                }
+                                                nprobes = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid NPROBES value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            "refine" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("REFINE requires a value".to_string()));
+                                                }
+                                                refine_factor = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid REFINE value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            _ => {
+                                                return Err(DBError(format!("Unknown parameter: {}", cmd[i])));
+                                            }
+                                        }
+                                    }
+                                    
+                                    Cmd::LanceSearchText { dataset, query_text, k, nprobes, refine_factor }
+                                }
+                                "embed.text" => {
+                                    if cmd.len() < 3 {
+                                        return Err(DBError("LANCE EMBED.TEXT <text1> [text2] ...".to_string()));
+                                    }
+                                    let texts = cmd[2..].to_vec();
+                                    Cmd::LanceEmbedText { texts }
+                                }
+                                "create.index" => {
+                                    if cmd.len() < 5 {
+                                        return Err(DBError("LANCE CREATE.INDEX <dataset> <index_type> [PARTITIONS <n>] [SUBVECTORS <n>]".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    let index_type = cmd[3].clone();
+                                    
+                                    let mut num_partitions = None;
+                                    let mut num_sub_vectors = None;
+                                    let mut i = 4;
+                                    while i < cmd.len() {
+                                        match cmd[i].to_lowercase().as_str() {
+                                            "partitions" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("PARTITIONS requires a value".to_string()));
+                                                }
+                                                num_partitions = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid PARTITIONS value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            "subvectors" => {
+                                                if i + 1 >= cmd.len() {
+                                                    return Err(DBError("SUBVECTORS requires a value".to_string()));
+                                                }
+                                                num_sub_vectors = Some(cmd[i + 1].parse::<usize>().map_err(|_| DBError("Invalid SUBVECTORS value".to_string()))?);
+                                                i += 2;
+                                            }
+                                            _ => {
+                                                return Err(DBError(format!("Unknown parameter: {}", cmd[i])));
+                                            }
+                                        }
+                                    }
+                                    
+                                    Cmd::LanceCreateIndex { dataset, index_type, num_partitions, num_sub_vectors }
+                                }
+                                "list" => {
+                                    if cmd.len() != 2 {
+                                        return Err(DBError("LANCE LIST takes no arguments".to_string()));
+                                    }
+                                    Cmd::LanceList
+                                }
+                                "drop" => {
+                                    if cmd.len() != 3 {
+                                        return Err(DBError("LANCE DROP <dataset>".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    Cmd::LanceDrop { dataset }
+                                }
+                                "info" => {
+                                    if cmd.len() != 3 {
+                                        return Err(DBError("LANCE INFO <dataset>".to_string()));
+                                    }
+                                    let dataset = cmd[2].clone();
+                                    Cmd::LanceInfo { dataset }
+                                }
+                                _ => return Err(DBError(format!("unsupported LANCE subcommand {:?}", cmd))),
+                            }
+                        }
                         _ => Cmd::Unknow(cmd[0].clone()),
                     },
                     protocol,
@@ -908,34 +1006,18 @@ impl Cmd {
             Cmd::AgeSignName(name, message) => Ok(crate::age::cmd_age_sign_name(server, &name, &message).await),
             Cmd::AgeVerifyName(name, message, sig_b64) => Ok(crate::age::cmd_age_verify_name(server, &name, &message, &sig_b64).await),
             Cmd::AgeList => Ok(crate::age::cmd_age_list(server).await),
-
-            // Full-text search commands
-            Cmd::FtCreate { index_name, schema } => {
-                search_cmd::ft_create_cmd(server, index_name, schema).await
-            }
-            Cmd::FtAdd { index_name, doc_id, score, fields } => {
-                search_cmd::ft_add_cmd(server, index_name, doc_id, score, fields).await
-            }
-            Cmd::FtSearch { index_name, query, filters, limit, offset, return_fields } => {
-                search_cmd::ft_search_cmd(server, index_name, query, filters, limit, offset, return_fields).await
-            }
-            Cmd::FtDel(index_name, doc_id) => {
-                search_cmd::ft_del_cmd(server, index_name, doc_id).await
-            }
-            Cmd::FtInfo(index_name) => {
-                search_cmd::ft_info_cmd(server, index_name).await
-            }
-            Cmd::FtDrop(index_name) => {
-                search_cmd::ft_drop_cmd(server, index_name).await
-            }
-            Cmd::FtAlter { .. } => {
-                // Not implemented yet
-                Ok(Protocol::err("FT.ALTER not implemented yet"))
-            }
-            Cmd::FtAggregate { .. } => {
-                // Not implemented yet
-                Ok(Protocol::err("FT.AGGREGATE not implemented yet"))
-            }
+            
+            // Lance vector database commands
+            Cmd::LanceCreate { dataset, dim, schema } => lance_create_cmd(server, &dataset, dim, &schema).await,
+            Cmd::LanceStore { dataset, text, image_base64, metadata } => lance_store_cmd(server, &dataset, text.as_deref(), image_base64.as_deref(), &metadata).await,
+            Cmd::LanceSearch { dataset, vector, k, nprobes, refine_factor } => lance_search_cmd(server, &dataset, &vector, k, nprobes, refine_factor).await,
+            Cmd::LanceSearchText { dataset, query_text, k, nprobes, refine_factor } => lance_search_text_cmd(server, &dataset, &query_text, k, nprobes, refine_factor).await,
+            Cmd::LanceEmbedText { texts } => lance_embed_text_cmd(server, &texts).await,
+            Cmd::LanceCreateIndex { dataset, index_type, num_partitions, num_sub_vectors } => lance_create_index_cmd(server, &dataset, &index_type, num_partitions, num_sub_vectors).await,
+            Cmd::LanceList => lance_list_cmd(server).await,
+            Cmd::LanceDrop { dataset } => lance_drop_cmd(server, &dataset).await,
+            Cmd::LanceInfo { dataset } => lance_info_cmd(server, &dataset).await,
+            
             Cmd::Unknow(s) => Ok(Protocol::err(&format!("ERR unknown command `{}`", s))),
         }
     }
@@ -1719,3 +1801,243 @@ fn command_cmd(args: &[String]) -> Result<Protocol, DBError> {
         _ => Ok(Protocol::Array(vec![])),
     }
 }
+
+// Helper function to create Arrow schema from field specifications
+fn create_schema_from_fields(dim: usize, fields: &[(String, String)]) -> arrow::datatypes::Schema {
+    let mut schema_fields = Vec::new();
+    
+    // Always add the vector field first
+    let vector_field = arrow::datatypes::Field::new(
+        "vector",
+        arrow::datatypes::DataType::FixedSizeList(
+            Arc::new(arrow::datatypes::Field::new("item", arrow::datatypes::DataType::Float32, true)),
+            dim as i32
+        ),
+        false
+    );
+    schema_fields.push(vector_field);
+    
+    // Add custom fields
+    for (name, field_type) in fields {
+        let data_type = match field_type.to_lowercase().as_str() {
+            "string" | "text" => arrow::datatypes::DataType::Utf8,
+            "int" | "integer" => arrow::datatypes::DataType::Int64,
+            "float" => arrow::datatypes::DataType::Float64,
+            "bool" | "boolean" => arrow::datatypes::DataType::Boolean,
+            _ => arrow::datatypes::DataType::Utf8, // Default to string
+        };
+        schema_fields.push(arrow::datatypes::Field::new(name, data_type, true));
+    }
+    
+    arrow::datatypes::Schema::new(schema_fields)
+}
+
+// Lance vector database command implementations
+async fn lance_create_cmd(
+    server: &Server,
+    dataset: &str,
+    dim: usize,
+    schema: &[(String, String)],
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.create_dataset(dataset, create_schema_from_fields(dim, schema)).await {
+                Ok(_) => Ok(Protocol::SimpleString("OK".to_string())),
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_store_cmd(
+    server: &Server,
+    dataset: &str,
+    text: Option<&str>,
+    image_base64: Option<&str>,
+    metadata: &std::collections::HashMap<String, String>,
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.store_multimodal(server, dataset, text.map(|s| s.to_string()),
+                image_base64.and_then(|s| base64::engine::general_purpose::STANDARD.decode(s).ok()),
+                metadata.clone()).await {
+                Ok(id) => Ok(Protocol::BulkString(id)),
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_search_cmd(
+    server: &Server,
+    dataset: &str,
+    vector: &[f32],
+    k: usize,
+    nprobes: Option<usize>,
+    refine_factor: Option<usize>,
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.search_vectors(dataset, vector.to_vec(), k, nprobes, refine_factor).await {
+                Ok(results) => {
+                    let mut response = Vec::new();
+                    for (distance, metadata) in results {
+                        let mut item = Vec::new();
+                        item.push(Protocol::BulkString("distance".to_string()));
+                        item.push(Protocol::BulkString(distance.to_string()));
+                        for (key, value) in metadata {
+                            item.push(Protocol::BulkString(key));
+                            item.push(Protocol::BulkString(value));
+                        }
+                        response.push(Protocol::Array(item));
+                    }
+                    Ok(Protocol::Array(response))
+                }
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_search_text_cmd(
+    server: &Server,
+    dataset: &str,
+    query_text: &str,
+    k: usize,
+    nprobes: Option<usize>,
+    refine_factor: Option<usize>,
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.search_with_text(server, dataset, query_text.to_string(), k, nprobes, refine_factor).await {
+                Ok(results) => {
+                    let mut response = Vec::new();
+                    for (distance, metadata) in results {
+                        let mut item = Vec::new();
+                        item.push(Protocol::BulkString("distance".to_string()));
+                        item.push(Protocol::BulkString(distance.to_string()));
+                        for (key, value) in metadata {
+                            item.push(Protocol::BulkString(key));
+                            item.push(Protocol::BulkString(value));
+                        }
+                        response.push(Protocol::Array(item));
+                    }
+                    Ok(Protocol::Array(response))
+                }
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+// Helper function to sanitize error messages for Redis protocol
+fn sanitize_error_message(msg: &str) -> String {
+    // Remove newlines, carriage returns, and limit length
+    let sanitized = msg
+        .replace('\n', " ")
+        .replace('\r', " ")
+        .replace('\t', " ");
+    
+    // Limit to 200 characters to avoid overly long error messages
+    if sanitized.len() > 200 {
+        format!("{}...", &sanitized[..197])
+    } else {
+        sanitized
+    }
+}
+
+async fn lance_embed_text_cmd(
+    server: &Server,
+    texts: &[String],
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.embed_text(server, texts.to_vec()).await {
+                Ok(embeddings) => {
+                    let mut response = Vec::new();
+                    for embedding in embeddings {
+                        let vector_strings: Vec<Protocol> = embedding
+                            .iter()
+                            .map(|f| Protocol::BulkString(f.to_string()))
+                            .collect();
+                        response.push(Protocol::Array(vector_strings));
+                    }
+                    Ok(Protocol::Array(response))
+                }
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_create_index_cmd(
+    server: &Server,
+    dataset: &str,
+    index_type: &str,
+    num_partitions: Option<usize>,
+    num_sub_vectors: Option<usize>,
+) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.create_index(dataset, index_type, num_partitions, num_sub_vectors).await {
+                Ok(_) => Ok(Protocol::SimpleString("OK".to_string())),
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_list_cmd(server: &Server) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.list_datasets().await {
+                Ok(datasets) => {
+                    let response: Vec<Protocol> = datasets
+                        .into_iter()
+                        .map(Protocol::BulkString)
+                        .collect();
+                    Ok(Protocol::Array(response))
+                }
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_drop_cmd(server: &Server, dataset: &str) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.drop_dataset(dataset).await {
+                Ok(_) => Ok(Protocol::SimpleString("OK".to_string())),
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}
+
+async fn lance_info_cmd(server: &Server, dataset: &str) -> Result<Protocol, DBError> {
+    match server.lance_store() {
+        Ok(lance_store) => {
+            match lance_store.get_dataset_info(dataset).await {
+                Ok(info) => {
+                    let mut response = Vec::new();
+                    for (key, value) in info {
+                        response.push(Protocol::BulkString(key));
+                        response.push(Protocol::BulkString(value));
+                    }
+                    Ok(Protocol::Array(response))
+                }
+                Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR {}", e)))),
+            }
+        }
+        Err(e) => Ok(Protocol::err(&sanitize_error_message(&format!("ERR Lance store not available: {}", e)))),
+    }
+}

src/error.rs

@@ -9,6 +9,12 @@ use bincode;
 #[derive(Debug)]
 pub struct DBError(pub String);
 
+impl std::fmt::Display for DBError {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        write!(f, "{}", self.0)
+    }
+}
+
 impl From<std::io::Error> for DBError {
     fn from(item: std::io::Error) -> Self {
         DBError(item.to_string().clone())
@@ -92,3 +98,40 @@ impl From<chacha20poly1305::Error> for DBError {
         DBError(item.to_string())
     }
 }
+
+// Lance and related dependencies error handling
+impl From<lance::Error> for DBError {
+    fn from(item: lance::Error) -> Self {
+        DBError(item.to_string())
+    }
+}
+
+impl From<arrow::error::ArrowError> for DBError {
+    fn from(item: arrow::error::ArrowError) -> Self {
+        DBError(item.to_string())
+    }
+}
+
+impl From<reqwest::Error> for DBError {
+    fn from(item: reqwest::Error) -> Self {
+        DBError(item.to_string())
+    }
+}
+
+impl From<image::ImageError> for DBError {
+    fn from(item: image::ImageError) -> Self {
+        DBError(item.to_string())
+    }
+}
+
+impl From<uuid::Error> for DBError {
+    fn from(item: uuid::Error) -> Self {
+        DBError(item.to_string())
+    }
+}
+
+impl From<base64::DecodeError> for DBError {
+    fn from(item: base64::DecodeError) -> Self {
+        DBError(item.to_string())
+    }
+}

src/lance_store.rs

@@ -0,0 +1,609 @@
+use std::collections::HashMap;
+use std::path::PathBuf;
+use std::sync::Arc;
+use tokio::sync::RwLock;
+
+use arrow::array::{Float32Array, StringArray, ArrayRef, FixedSizeListArray, Array};
+use arrow::datatypes::{DataType, Field, Schema, SchemaRef};
+use arrow::record_batch::{RecordBatch, RecordBatchReader};
+use arrow::error::ArrowError;
+use lance::dataset::{Dataset, WriteParams, WriteMode};
+use lance::index::vector::VectorIndexParams;
+use lance_index::vector::pq::PQBuildParams;
+use lance_index::vector::ivf::IvfBuildParams;
+use lance_index::DatasetIndexExt;
+use lance_linalg::distance::MetricType;
+use futures::TryStreamExt;
+use base64::Engine;
+
+use serde::{Deserialize, Serialize};
+use crate::error::DBError;
+
+// Simple RecordBatchReader implementation for Vec<RecordBatch>
+struct VecRecordBatchReader {
+    batches: std::vec::IntoIter<Result<RecordBatch, ArrowError>>,
+}
+
+impl VecRecordBatchReader {
+    fn new(batches: Vec<RecordBatch>) -> Self {
+        let result_batches = batches.into_iter().map(Ok).collect::<Vec<_>>();
+        Self {
+            batches: result_batches.into_iter(),
+        }
+    }
+}
+
+impl Iterator for VecRecordBatchReader {
+    type Item = Result<RecordBatch, ArrowError>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        self.batches.next()
+    }
+}
+
+impl RecordBatchReader for VecRecordBatchReader {
+    fn schema(&self) -> SchemaRef {
+        // This is a simplified implementation - in practice you'd want to store the schema
+        Arc::new(Schema::empty())
+    }
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+struct EmbeddingRequest {
+    texts: Option<Vec<String>>,
+    images: Option<Vec<String>>, // base64 encoded
+    model: Option<String>,
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+struct EmbeddingResponse {
+    embeddings: Vec<Vec<f32>>,
+    model: String,
+    usage: Option<HashMap<String, u32>>,
+}
+
+// Ollama-specific request/response structures
+#[derive(Debug, Serialize, Deserialize)]
+struct OllamaEmbeddingRequest {
+    model: String,
+    prompt: String,
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+struct OllamaEmbeddingResponse {
+    embedding: Vec<f32>,
+}
+
+pub struct LanceStore {
+    datasets: Arc<RwLock<HashMap<String, Arc<Dataset>>>>,
+    data_dir: PathBuf,
+    http_client: reqwest::Client,
+}
+
+impl LanceStore {
+    pub async fn new(data_dir: PathBuf) -> Result<Self, DBError> {
+        // Create data directory if it doesn't exist
+        std::fs::create_dir_all(&data_dir)
+            .map_err(|e| DBError(format!("Failed to create Lance data directory: {}", e)))?;
+        
+        let http_client = reqwest::Client::builder()
+            .timeout(std::time::Duration::from_secs(30))
+            .build()
+            .map_err(|e| DBError(format!("Failed to create HTTP client: {}", e)))?;
+        
+        Ok(Self {
+            datasets: Arc::new(RwLock::new(HashMap::new())),
+            data_dir,
+            http_client,
+        })
+    }
+    
+    /// Get embedding service URL from Redis config, default to local Ollama
+    async fn get_embedding_url(&self, server: &crate::server::Server) -> Result<String, DBError> {
+        // Get the embedding URL from Redis config directly from storage
+        let storage = server.current_storage()?;
+        match storage.hget("config:core:aiembed", "url")? {
+            Some(url) => Ok(url),
+            None => Ok("http://localhost:11434".to_string()), // Default to local Ollama
+        }
+    }
+    
+    /// Check if we're using Ollama (default) or custom embedding service
+    async fn is_ollama_service(&self, server: &crate::server::Server) -> Result<bool, DBError> {
+        let url = self.get_embedding_url(server).await?;
+        Ok(url.contains("localhost:11434") || url.contains("127.0.0.1:11434"))
+    }
+    
+    /// Call external embedding service (Ollama or custom)
+    async fn call_embedding_service(
+        &self,
+        server: &crate::server::Server,
+        texts: Option<Vec<String>>,
+        images: Option<Vec<String>>,
+    ) -> Result<Vec<Vec<f32>>, DBError> {
+        let base_url = self.get_embedding_url(server).await?;
+        let is_ollama = self.is_ollama_service(server).await?;
+        
+        if is_ollama {
+            // Use Ollama API format
+            if let Some(texts) = texts {
+                let mut embeddings = Vec::new();
+                for text in texts {
+                    let url = format!("{}/api/embeddings", base_url);
+                    let request = OllamaEmbeddingRequest {
+                        model: "nomic-embed-text".to_string(),
+                        prompt: text,
+                    };
+                    
+                    let response = self.http_client
+                        .post(&url)
+                        .json(&request)
+                        .send()
+                        .await
+                        .map_err(|e| DBError(format!("Failed to call Ollama embedding service: {}", e)))?;
+                    
+                    if !response.status().is_success() {
+                        let status = response.status();
+                        let error_text = response.text().await.unwrap_or_default();
+                        return Err(DBError(format!(
+                            "Ollama embedding service returned error {}: {}",
+                            status, error_text
+                        )));
+                    }
+                    
+                    let ollama_response: OllamaEmbeddingResponse = response
+                        .json()
+                        .await
+                        .map_err(|e| DBError(format!("Failed to parse Ollama embedding response: {}", e)))?;
+                    
+                    embeddings.push(ollama_response.embedding);
+                }
+                Ok(embeddings)
+            } else if let Some(_images) = images {
+                // Ollama doesn't support image embeddings with this API yet
+                Err(DBError("Image embeddings not supported with Ollama. Please configure a custom embedding service.".to_string()))
+            } else {
+                Err(DBError("No text or images provided for embedding".to_string()))
+            }
+        } else {
+            // Use custom embedding service API format
+            let request = EmbeddingRequest {
+                texts,
+                images,
+                model: None, // Let the service use its default
+            };
+            
+            let response = self.http_client
+                .post(&base_url)
+                .json(&request)
+                .send()
+                .await
+                .map_err(|e| DBError(format!("Failed to call embedding service: {}", e)))?;
+            
+            if !response.status().is_success() {
+                let status = response.status();
+                let error_text = response.text().await.unwrap_or_default();
+                return Err(DBError(format!(
+                    "Embedding service returned error {}: {}",
+                    status, error_text
+                )));
+            }
+            
+            let embedding_response: EmbeddingResponse = response
+                .json()
+                .await
+                .map_err(|e| DBError(format!("Failed to parse embedding response: {}", e)))?;
+            
+            Ok(embedding_response.embeddings)
+        }
+    }
+    
+    pub async fn embed_text(
+        &self, 
+        server: &crate::server::Server,
+        texts: Vec<String>
+    ) -> Result<Vec<Vec<f32>>, DBError> {
+        if texts.is_empty() {
+            return Ok(Vec::new());
+        }
+        
+        self.call_embedding_service(server, Some(texts), None).await
+    }
+    
+    pub async fn embed_image(
+        &self,
+        server: &crate::server::Server,
+        image_bytes: Vec<u8>
+    ) -> Result<Vec<f32>, DBError> {
+        // Convert image bytes to base64
+        let base64_image = base64::engine::general_purpose::STANDARD.encode(&image_bytes);
+        
+        let embeddings = self.call_embedding_service(
+            server, 
+            None, 
+            Some(vec![base64_image])
+        ).await?;
+        
+        embeddings.into_iter()
+            .next()
+            .ok_or_else(|| DBError("No embedding returned for image".to_string()))
+    }
+    
+    pub async fn create_dataset(
+        &self,
+        name: &str,
+        schema: Schema,
+    ) -> Result<(), DBError> {
+        let dataset_path = self.data_dir.join(format!("{}.lance", name));
+        
+        // Create empty dataset with schema
+        let write_params = WriteParams {
+            mode: WriteMode::Create,
+            ..Default::default()
+        };
+        
+        // Create an empty RecordBatch with the schema
+        let empty_batch = RecordBatch::new_empty(Arc::new(schema));
+        
+        // Use RecordBatchReader for Lance 0.33
+        let reader = VecRecordBatchReader::new(vec![empty_batch]);
+        let dataset = Dataset::write(
+            reader,
+            dataset_path.to_str().unwrap(),
+            Some(write_params)
+        ).await
+        .map_err(|e| DBError(format!("Failed to create dataset: {}", e)))?;
+        
+        let mut datasets = self.datasets.write().await;
+        datasets.insert(name.to_string(), Arc::new(dataset));
+        
+        Ok(())
+    }
+    
+    pub async fn write_vectors(
+        &self,
+        dataset_name: &str,
+        vectors: Vec<Vec<f32>>,
+        metadata: Option<HashMap<String, Vec<String>>>,
+    ) -> Result<usize, DBError> {
+        let dataset_path = self.data_dir.join(format!("{}.lance", dataset_name));
+        
+        // Open or get cached dataset
+        let _dataset = self.get_or_open_dataset(dataset_name).await?;
+        
+        // Build RecordBatch
+        let num_vectors = vectors.len();
+        if num_vectors == 0 {
+            return Ok(0);
+        }
+        
+        let dim = vectors.first()
+            .ok_or_else(|| DBError("Empty vectors".to_string()))?
+            .len();
+        
+        // Flatten vectors
+        let flat_vectors: Vec<f32> = vectors.into_iter().flatten().collect();
+        let values_array = Float32Array::from(flat_vectors);
+        let field = Arc::new(Field::new("item", DataType::Float32, true));
+        let vector_array = FixedSizeListArray::try_new(
+            field,
+            dim as i32,
+            Arc::new(values_array),
+            None
+        ).map_err(|e| DBError(format!("Failed to create vector array: {}", e)))?;
+        
+        let mut arrays: Vec<ArrayRef> = vec![Arc::new(vector_array)];
+        let mut fields = vec![Field::new(
+            "vector",
+            DataType::FixedSizeList(
+                Arc::new(Field::new("item", DataType::Float32, true)),
+                dim as i32
+            ),
+            false
+        )];
+        
+        // Add metadata columns if provided
+        if let Some(metadata) = metadata {
+            for (key, values) in metadata {
+                if values.len() != num_vectors {
+                    return Err(DBError(format!(
+                        "Metadata field '{}' has {} values but expected {}", 
+                        key, values.len(), num_vectors
+                    )));
+                }
+                let array = StringArray::from(values);
+                arrays.push(Arc::new(array));
+                fields.push(Field::new(&key, DataType::Utf8, true));
+            }
+        }
+        
+        let schema = Arc::new(Schema::new(fields));
+        let batch = RecordBatch::try_new(schema, arrays)
+            .map_err(|e| DBError(format!("Failed to create RecordBatch: {}", e)))?;
+        
+        // Append to dataset
+        let write_params = WriteParams {
+            mode: WriteMode::Append,
+            ..Default::default()
+        };
+        
+        let reader = VecRecordBatchReader::new(vec![batch]);
+        Dataset::write(
+            reader,
+            dataset_path.to_str().unwrap(),
+            Some(write_params)
+        ).await
+        .map_err(|e| DBError(format!("Failed to write to dataset: {}", e)))?;
+        
+        // Refresh cached dataset
+        let mut datasets = self.datasets.write().await;
+        datasets.remove(dataset_name);
+        
+        Ok(num_vectors)
+    }
+    
+    pub async fn search_vectors(
+        &self,
+        dataset_name: &str,
+        query_vector: Vec<f32>,
+        k: usize,
+        nprobes: Option<usize>,
+        _refine_factor: Option<usize>,
+    ) -> Result<Vec<(f32, HashMap<String, String>)>, DBError> {
+        let dataset = self.get_or_open_dataset(dataset_name).await?;
+        
+        // Build query
+        let query_array = Float32Array::from(query_vector.clone());
+        let mut query = dataset.scan();
+        query.nearest(
+            "vector",
+            &query_array,
+            k,
+        ).map_err(|e| DBError(format!("Failed to build search query: {}", e)))?;
+        
+        if let Some(nprobes) = nprobes {
+            query.nprobs(nprobes);
+        }
+        
+        // Note: refine_factor might not be available in this Lance version
+        // if let Some(refine) = refine_factor {
+        //     query.refine_factor(refine);
+        // }
+        
+        // Execute search
+        let results = query
+            .try_into_stream()
+            .await
+            .map_err(|e| DBError(format!("Failed to execute search: {}", e)))?
+            .try_collect::<Vec<_>>()
+            .await
+            .map_err(|e| DBError(format!("Failed to collect results: {}", e)))?;
+        
+        // Process results
+        let mut output = Vec::new();
+        for batch in results {
+            // Get distances
+            let distances = batch
+                .column_by_name("_distance")
+                .ok_or_else(|| DBError("No distance column".to_string()))?
+                .as_any()
+                .downcast_ref::<Float32Array>()
+                .ok_or_else(|| DBError("Invalid distance type".to_string()))?;
+            
+            // Get metadata
+            for i in 0..batch.num_rows() {
+                let distance = distances.value(i);
+                let mut metadata = HashMap::new();
+                
+                for field in batch.schema().fields() {
+                    if field.name() != "vector" && field.name() != "_distance" {
+                        if let Some(col) = batch.column_by_name(field.name()) {
+                            if let Some(str_array) = col.as_any().downcast_ref::<StringArray>() {
+                                if !str_array.is_null(i) {
+                                    metadata.insert(
+                                        field.name().to_string(),
+                                        str_array.value(i).to_string()
+                                    );
+                                }
+                            }
+                        }
+                    }
+                }
+                
+                output.push((distance, metadata));
+            }
+        }
+        
+        Ok(output)
+    }
+    
+    pub async fn store_multimodal(
+        &self,
+        server: &crate::server::Server,
+        dataset_name: &str,
+        text: Option<String>,
+        image_bytes: Option<Vec<u8>>,
+        metadata: HashMap<String, String>,
+    ) -> Result<String, DBError> {
+        // Generate ID
+        let id = uuid::Uuid::new_v4().to_string();
+        
+        // Generate embeddings using external service
+        let embedding = if let Some(text) = text.as_ref() {
+            self.embed_text(server, vec![text.clone()]).await?
+                .into_iter()
+                .next()
+                .ok_or_else(|| DBError("No embedding returned".to_string()))?
+        } else if let Some(img) = image_bytes.as_ref() {
+            self.embed_image(server, img.clone()).await?
+        } else {
+            return Err(DBError("No text or image provided".to_string()));
+        };
+        
+        // Prepare metadata
+        let mut full_metadata = metadata;
+        full_metadata.insert("id".to_string(), id.clone());
+        if let Some(text) = text {
+            full_metadata.insert("text".to_string(), text);
+        }
+        if let Some(img) = image_bytes {
+            full_metadata.insert("image_base64".to_string(), base64::engine::general_purpose::STANDARD.encode(img));
+        }
+        
+        // Convert metadata to column vectors
+        let mut metadata_cols = HashMap::new();
+        for (key, value) in full_metadata {
+            metadata_cols.insert(key, vec![value]);
+        }
+        
+        // Write to dataset
+        self.write_vectors(dataset_name, vec![embedding], Some(metadata_cols)).await?;
+        
+        Ok(id)
+    }
+    
+    pub async fn search_with_text(
+        &self,
+        server: &crate::server::Server,
+        dataset_name: &str,
+        query_text: String,
+        k: usize,
+        nprobes: Option<usize>,
+        refine_factor: Option<usize>,
+    ) -> Result<Vec<(f32, HashMap<String, String>)>, DBError> {
+        // Embed the query text using external service
+        let embeddings = self.embed_text(server, vec![query_text]).await?;
+        let query_vector = embeddings.into_iter()
+            .next()
+            .ok_or_else(|| DBError("No embedding returned for query".to_string()))?;
+        
+        // Search with the embedding
+        self.search_vectors(dataset_name, query_vector, k, nprobes, refine_factor).await
+    }
+    
+    pub async fn create_index(
+        &self,
+        dataset_name: &str,
+        index_type: &str,
+        num_partitions: Option<usize>,
+        num_sub_vectors: Option<usize>,
+    ) -> Result<(), DBError> {
+        let _dataset = self.get_or_open_dataset(dataset_name).await?;
+        
+        match index_type.to_uppercase().as_str() {
+            "IVF_PQ" => {
+                let ivf_params = IvfBuildParams {
+                    num_partitions: num_partitions.unwrap_or(256),
+                    ..Default::default()
+                };
+                let pq_params = PQBuildParams {
+                    num_sub_vectors: num_sub_vectors.unwrap_or(16),
+                    ..Default::default()
+                };
+                let params = VectorIndexParams::with_ivf_pq_params(
+                    MetricType::L2,
+                    ivf_params,
+                    pq_params,
+                );
+                
+                // Get a mutable reference to the dataset
+                let mut dataset_mut = Dataset::open(self.data_dir.join(format!("{}.lance", dataset_name)).to_str().unwrap())
+                    .await
+                    .map_err(|e| DBError(format!("Failed to open dataset for indexing: {}", e)))?;
+                
+                dataset_mut.create_index(
+                    &["vector"],
+                    lance_index::IndexType::Vector,
+                    None,
+                    &params,
+                    true
+                ).await
+                .map_err(|e| DBError(format!("Failed to create index: {}", e)))?;
+            }
+            _ => return Err(DBError(format!("Unsupported index type: {}", index_type))),
+        }
+        
+        Ok(())
+    }
+    
+    async fn get_or_open_dataset(&self, name: &str) -> Result<Arc<Dataset>, DBError> {
+        let mut datasets = self.datasets.write().await;
+        
+        if let Some(dataset) = datasets.get(name) {
+            return Ok(dataset.clone());
+        }
+        
+        let dataset_path = self.data_dir.join(format!("{}.lance", name));
+        if !dataset_path.exists() {
+            return Err(DBError(format!("Dataset '{}' does not exist", name)));
+        }
+        
+        let dataset = Dataset::open(dataset_path.to_str().unwrap())
+            .await
+            .map_err(|e| DBError(format!("Failed to open dataset: {}", e)))?;
+        
+        let dataset = Arc::new(dataset);
+        datasets.insert(name.to_string(), dataset.clone());
+        
+        Ok(dataset)
+    }
+    
+    pub async fn list_datasets(&self) -> Result<Vec<String>, DBError> {
+        let mut datasets = Vec::new();
+        
+        let entries = std::fs::read_dir(&self.data_dir)
+            .map_err(|e| DBError(format!("Failed to read data directory: {}", e)))?;
+        
+        for entry in entries {
+            let entry = entry.map_err(|e| DBError(format!("Failed to read entry: {}", e)))?;
+            let path = entry.path();
+            
+            if path.is_dir() {
+                if let Some(name) = path.file_name() {
+                    if let Some(name_str) = name.to_str() {
+                        if name_str.ends_with(".lance") {
+                            let dataset_name = name_str.trim_end_matches(".lance");
+                            datasets.push(dataset_name.to_string());
+                        }
+                    }
+                }
+            }
+        }
+        
+        Ok(datasets)
+    }
+    
+    pub async fn drop_dataset(&self, name: &str) -> Result<(), DBError> {
+        // Remove from cache
+        let mut datasets = self.datasets.write().await;
+        datasets.remove(name);
+        
+        // Delete from disk
+        let dataset_path = self.data_dir.join(format!("{}.lance", name));
+        if dataset_path.exists() {
+            std::fs::remove_dir_all(dataset_path)
+                .map_err(|e| DBError(format!("Failed to delete dataset: {}", e)))?;
+        }
+        
+        Ok(())
+    }
+    
+    pub async fn get_dataset_info(&self, name: &str) -> Result<HashMap<String, String>, DBError> {
+        let dataset = self.get_or_open_dataset(name).await?;
+        
+        let mut info = HashMap::new();
+        info.insert("name".to_string(), name.to_string());
+        info.insert("version".to_string(), dataset.version().version.to_string());
+        info.insert("num_rows".to_string(), dataset.count_rows(None).await?.to_string());
+        
+        // Get schema info
+        let schema = dataset.schema();
+        let fields: Vec<String> = schema.fields
+            .iter()
+            .map(|f| format!("{}:{}", f.name, f.data_type()))
+            .collect();
+        info.insert("schema".to_string(), fields.join(", "));
+        
+        Ok(info)
+    }
+}

src/lib.rs

@@ -2,11 +2,10 @@ pub mod age;   // NEW
 pub mod cmd;
 pub mod crypto;
 pub mod error;
+pub mod lance_store;    // Add Lance store module
 pub mod options;
 pub mod protocol;
-pub mod search_cmd;  // Add this
 pub mod server;
 pub mod storage;
 pub mod storage_trait;  // Add this
 pub mod storage_sled;   // Add this
-pub mod tantivy_search;

src/search_cmd.rs

@@ -1,272 +0,0 @@
-use crate::{
-    error::DBError,
-    protocol::Protocol,
-    server::Server,
-    tantivy_search::{
-        TantivySearch, FieldDef, NumericType, IndexConfig,
-        SearchOptions, Filter, FilterType
-    },
-};
-use std::collections::HashMap;
-use std::sync::Arc;
-
-pub async fn ft_create_cmd(
-    server: &Server,
-    index_name: String,
-    schema: Vec<(String, String, Vec<String>)>,
-) -> Result<Protocol, DBError> {
-    // Parse schema into field definitions
-    let mut field_definitions = Vec::new();
-    
-    for (field_name, field_type, options) in schema {
-        let field_def = match field_type.to_uppercase().as_str() {
-            "TEXT" => {
-                let mut weight = 1.0;
-                let mut sortable = false;
-                let mut no_index = false;
-                
-                for opt in &options {
-                    match opt.to_uppercase().as_str() {
-                        "WEIGHT" => {
-                            // Next option should be the weight value
-                            if let Some(idx) = options.iter().position(|x| x == opt) {
-                                if idx + 1 < options.len() {
-                                    weight = options[idx + 1].parse().unwrap_or(1.0);
-                                }
-                            }
-                        }
-                        "SORTABLE" => sortable = true,
-                        "NOINDEX" => no_index = true,
-                        _ => {}
-                    }
-                }
-                
-                FieldDef::Text {
-                    stored: true,
-                    indexed: !no_index,
-                    tokenized: true,
-                    fast: sortable,
-                }
-            }
-            "NUMERIC" => {
-                let mut sortable = false;
-                
-                for opt in &options {
-                    if opt.to_uppercase() == "SORTABLE" {
-                        sortable = true;
-                    }
-                }
-                
-                FieldDef::Numeric {
-                    stored: true,
-                    indexed: true,
-                    fast: sortable,
-                    precision: NumericType::F64,
-                }
-            }
-            "TAG" => {
-                let mut separator = ",".to_string();
-                let mut case_sensitive = false;
-                
-                for i in 0..options.len() {
-                    match options[i].to_uppercase().as_str() {
-                        "SEPARATOR" => {
-                            if i + 1 < options.len() {
-                                separator = options[i + 1].clone();
-                            }
-                        }
-                        "CASESENSITIVE" => case_sensitive = true,
-                        _ => {}
-                    }
-                }
-                
-                FieldDef::Tag {
-                    stored: true,
-                    separator,
-                    case_sensitive,
-                }
-            }
-            "GEO" => {
-                FieldDef::Geo { stored: true }
-            }
-            _ => {
-                return Err(DBError(format!("Unknown field type: {}", field_type)));
-            }
-        };
-        
-        field_definitions.push((field_name, field_def));
-    }
-    
-    // Create the search index
-    let search_path = server.search_index_path();
-    let config = IndexConfig::default();
-    
-    println!("Creating search index '{}' at path: {:?}", index_name, search_path);
-    println!("Field definitions: {:?}", field_definitions);
-    
-    let search_index = TantivySearch::new_with_schema(
-        search_path,
-        index_name.clone(),
-        field_definitions,
-        Some(config),
-    )?;
-    
-    println!("Search index '{}' created successfully", index_name);
-    
-    // Store in registry
-    let mut indexes = server.search_indexes.write().unwrap();
-    indexes.insert(index_name, Arc::new(search_index));
-    
-    Ok(Protocol::SimpleString("OK".to_string()))
-}
-
-pub async fn ft_add_cmd(
-    server: &Server,
-    index_name: String,
-    doc_id: String,
-    _score: f64,
-    fields: HashMap<String, String>,
-) -> Result<Protocol, DBError> {
-    let indexes = server.search_indexes.read().unwrap();
-    
-    let search_index = indexes.get(&index_name)
-        .ok_or_else(|| DBError(format!("Index '{}' not found", index_name)))?;
-    
-    search_index.add_document_with_fields(&doc_id, fields)?;
-    
-    Ok(Protocol::SimpleString("OK".to_string()))
-}
-
-pub async fn ft_search_cmd(
-    server: &Server,
-    index_name: String,
-    query: String,
-    filters: Vec<(String, String)>,
-    limit: Option<usize>,
-    offset: Option<usize>,
-    return_fields: Option<Vec<String>>,
-) -> Result<Protocol, DBError> {
-    let indexes = server.search_indexes.read().unwrap();
-    
-    let search_index = indexes.get(&index_name)
-        .ok_or_else(|| DBError(format!("Index '{}' not found", index_name)))?;
-    
-    // Convert filters to search filters
-    let search_filters = filters.into_iter().map(|(field, value)| {
-        Filter {
-            field,
-            filter_type: FilterType::Equals(value),
-        }
-    }).collect();
-    
-    let options = SearchOptions {
-        limit: limit.unwrap_or(10),
-        offset: offset.unwrap_or(0),
-        filters: search_filters,
-        sort_by: None,
-        return_fields,
-        highlight: false,
-    };
-    
-    let results = search_index.search_with_options(&query, options)?;
-    
-    // Format results as Redis protocol
-    let mut response = Vec::new();
-    
-    // First element is the total count
-    response.push(Protocol::SimpleString(results.total.to_string()));
-    
-    // Then each document
-    for doc in results.documents {
-        let mut doc_array = Vec::new();
-        
-        // Add document ID if it exists
-        if let Some(id) = doc.fields.get("_id") {
-            doc_array.push(Protocol::BulkString(id.clone()));
-        }
-        
-        // Add score
-        doc_array.push(Protocol::BulkString(doc.score.to_string()));
-        
-        // Add fields as key-value pairs
-        for (field_name, field_value) in doc.fields {
-            if field_name != "_id" {
-                doc_array.push(Protocol::BulkString(field_name));
-                doc_array.push(Protocol::BulkString(field_value));
-            }
-        }
-        
-        response.push(Protocol::Array(doc_array));
-    }
-    
-    Ok(Protocol::Array(response))
-}
-
-pub async fn ft_del_cmd(
-    server: &Server,
-    index_name: String,
-    doc_id: String,
-) -> Result<Protocol, DBError> {
-    let indexes = server.search_indexes.read().unwrap();
-    
-    let _search_index = indexes.get(&index_name)
-        .ok_or_else(|| DBError(format!("Index '{}' not found", index_name)))?;
-    
-    // For now, return success
-    // In a full implementation, we'd need to add a delete method to TantivySearch
-    println!("Deleting document '{}' from index '{}'", doc_id, index_name);
-    
-    Ok(Protocol::SimpleString("1".to_string()))
-}
-
-pub async fn ft_info_cmd(
-    server: &Server,
-    index_name: String,
-) -> Result<Protocol, DBError> {
-    let indexes = server.search_indexes.read().unwrap();
-    
-    let search_index = indexes.get(&index_name)
-        .ok_or_else(|| DBError(format!("Index '{}' not found", index_name)))?;
-    
-    let info = search_index.get_info()?;
-    
-    // Format info as Redis protocol
-    let mut response = Vec::new();
-    
-    response.push(Protocol::BulkString("index_name".to_string()));
-    response.push(Protocol::BulkString(info.name));
-    
-    response.push(Protocol::BulkString("num_docs".to_string()));
-    response.push(Protocol::BulkString(info.num_docs.to_string()));
-    
-    response.push(Protocol::BulkString("num_fields".to_string()));
-    response.push(Protocol::BulkString(info.fields.len().to_string()));
-    
-    response.push(Protocol::BulkString("fields".to_string()));
-    let fields_str = info.fields.iter()
-        .map(|f| format!("{}:{}", f.name, f.field_type))
-        .collect::<Vec<_>>()
-        .join(", ");
-    response.push(Protocol::BulkString(fields_str));
-    
-    Ok(Protocol::Array(response))
-}
-
-pub async fn ft_drop_cmd(
-    server: &Server,
-    index_name: String,
-) -> Result<Protocol, DBError> {
-    let mut indexes = server.search_indexes.write().unwrap();
-    
-    if indexes.remove(&index_name).is_some() {
-        // Also remove the index files from disk
-        let index_path = server.search_index_path().join(&index_name);
-        if index_path.exists() {
-            std::fs::remove_dir_all(index_path)
-                .map_err(|e| DBError(format!("Failed to remove index files: {}", e)))?;
-        }
-        Ok(Protocol::SimpleString("OK".to_string()))
-    } else {
-        Err(DBError(format!("Index '{}' not found", index_name)))
-    }
-}

src/server.rs

@@ -4,23 +4,21 @@ use std::sync::Arc;
 use tokio::io::AsyncReadExt;
 use tokio::io::AsyncWriteExt;
 use tokio::sync::{Mutex, oneshot};
-use std::sync::RwLock;
 
 use std::sync::atomic::{AtomicU64, Ordering};
 
 use crate::cmd::Cmd;
 use crate::error::DBError;
+use crate::lance_store::LanceStore;
 use crate::options;
 use crate::protocol::Protocol;
 use crate::storage::Storage;
 use crate::storage_sled::SledStorage;
 use crate::storage_trait::StorageBackend;
-use crate::tantivy_search::TantivySearch;
 
 #[derive(Clone)]
 pub struct Server {
-    pub db_cache: Arc<RwLock<HashMap<u64, Arc<dyn StorageBackend>>>>,
-    pub search_indexes: Arc<RwLock<HashMap<String, Arc<TantivySearch>>>>,
+    pub db_cache: std::sync::Arc<std::sync::RwLock<HashMap<u64, Arc<dyn StorageBackend>>>>,
     pub option: options::DBOption,
     pub client_name: Option<String>,
     pub selected_db: u64, // Changed from usize to u64
@@ -29,6 +27,9 @@ pub struct Server {
     // BLPOP waiter registry: per (db_index, key) FIFO of waiters
     pub list_waiters: Arc<Mutex<HashMap<u64, HashMap<String, Vec<Waiter>>>>>,
     pub waiter_seq: Arc<AtomicU64>,
+    
+    // Lance vector store
+    pub lance_store: Option<Arc<LanceStore>>,
 }
 
 pub struct Waiter {
@@ -45,9 +46,18 @@ pub enum PopSide {
 
 impl Server {
     pub async fn new(option: options::DBOption) -> Self {
+        // Initialize Lance store
+        let lance_data_dir = std::path::PathBuf::from(&option.dir).join("lance");
+        let lance_store = match LanceStore::new(lance_data_dir).await {
+            Ok(store) => Some(Arc::new(store)),
+            Err(e) => {
+                eprintln!("Warning: Failed to initialize Lance store: {}", e.0);
+                None
+            }
+        };
+
         Server {
-            db_cache: Arc::new(RwLock::new(HashMap::new())),
-            search_indexes: Arc::new(RwLock::new(HashMap::new())),
+            db_cache: Arc::new(std::sync::RwLock::new(HashMap::new())),
             option,
             client_name: None,
             selected_db: 0,
@@ -55,9 +65,17 @@ impl Server {
 
             list_waiters: Arc::new(Mutex::new(HashMap::new())),
             waiter_seq: Arc::new(AtomicU64::new(1)),
+            lance_store,
         }
     }
 
+    pub fn lance_store(&self) -> Result<Arc<LanceStore>, DBError> {
+        self.lance_store
+            .as_ref()
+            .cloned()
+            .ok_or_else(|| DBError("Lance store not initialized".to_string()))
+    }
+
     pub fn current_storage(&self) -> Result<Arc<dyn StorageBackend>, DBError> {
         let mut cache = self.db_cache.write().unwrap();
         
@@ -104,11 +122,6 @@ impl Server {
         // DB 0-9 are non-encrypted, DB 10+ are encrypted
         self.option.encrypt && db_index >= 10
     }
-    
-    // Add method to get search index path
-    pub fn search_index_path(&self) -> std::path::PathBuf {
-        std::path::PathBuf::from(&self.option.dir).join("search_indexes")
-    }
 
     // ----- BLPOP waiter helpers -----
 

src/tantivy_search.rs

@@ -1,567 +0,0 @@
-use tantivy::{
-    collector::TopDocs,
-    directory::MmapDirectory,
-    query::{QueryParser, BooleanQuery, Query, TermQuery, Occur},
-    schema::{Schema, Field, TextOptions, TextFieldIndexing,
-             STORED, STRING, Value},
-    Index, IndexWriter, IndexReader, ReloadPolicy,
-    Term, DateTime, TantivyDocument,
-    tokenizer::{TokenizerManager},
-};
-use std::path::PathBuf;
-use std::sync::{Arc, RwLock};
-use std::collections::HashMap;
-use crate::error::DBError;
-use serde::{Serialize, Deserialize};
-
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub enum FieldDef {
-    Text {
-        stored: bool,
-        indexed: bool,
-        tokenized: bool,
-        fast: bool,
-    },
-    Numeric {
-        stored: bool,
-        indexed: bool,
-        fast: bool,
-        precision: NumericType,
-    },
-    Tag {
-        stored: bool,
-        separator: String,
-        case_sensitive: bool,
-    },
-    Geo {
-        stored: bool,
-    },
-}
-
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub enum NumericType {
-    I64,
-    U64,
-    F64,
-    Date,
-}
-
-pub struct IndexSchema {
-    schema: Schema,
-    fields: HashMap<String, (Field, FieldDef)>,
-    default_search_fields: Vec<Field>,
-}
-
-pub struct TantivySearch {
-    index: Index,
-    writer: Arc<RwLock<IndexWriter>>,
-    reader: IndexReader,
-    index_schema: IndexSchema,
-    name: String,
-    config: IndexConfig,
-}
-
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct IndexConfig {
-    pub language: String,
-    pub stopwords: Vec<String>,
-    pub stemming: bool,
-    pub max_doc_count: Option<usize>,
-    pub default_score: f64,
-}
-
-impl Default for IndexConfig {
-    fn default() -> Self {
-        IndexConfig {
-            language: "english".to_string(),
-            stopwords: vec![],
-            stemming: true,
-            max_doc_count: None,
-            default_score: 1.0,
-        }
-    }
-}
-
-impl TantivySearch {
-    pub fn new_with_schema(
-        base_path: PathBuf,
-        name: String,
-        field_definitions: Vec<(String, FieldDef)>,
-        config: Option<IndexConfig>,
-    ) -> Result<Self, DBError> {
-        let index_path = base_path.join(&name);
-        std::fs::create_dir_all(&index_path)
-            .map_err(|e| DBError(format!("Failed to create index dir: {}", e)))?;
-
-        // Build schema from field definitions
-        let mut schema_builder = Schema::builder();
-        let mut fields = HashMap::new();
-        let mut default_search_fields = Vec::new();
-
-        // Always add a document ID field
-        let id_field = schema_builder.add_text_field("_id", STRING | STORED);
-        fields.insert("_id".to_string(), (id_field, FieldDef::Text {
-            stored: true,
-            indexed: true,
-            tokenized: false,
-            fast: false,
-        }));
-
-        // Add user-defined fields
-        for (field_name, field_def) in field_definitions {
-            let field = match &field_def {
-                FieldDef::Text { stored, indexed, tokenized, fast: _fast } => {
-                    let mut text_options = TextOptions::default();
-                    
-                    if *stored {
-                        text_options = text_options.set_stored();
-                    }
-                    
-                    if *indexed {
-                        let indexing_options = if *tokenized {
-                            TextFieldIndexing::default()
-                                .set_tokenizer("default")
-                                .set_index_option(tantivy::schema::IndexRecordOption::WithFreqsAndPositions)
-                        } else {
-                            TextFieldIndexing::default()
-                                .set_tokenizer("raw")
-                                .set_index_option(tantivy::schema::IndexRecordOption::Basic)
-                        };
-                        text_options = text_options.set_indexing_options(indexing_options);
-                        
-                        let f = schema_builder.add_text_field(&field_name, text_options);
-                        if *tokenized {
-                            default_search_fields.push(f);
-                        }
-                        f
-                    } else {
-                        schema_builder.add_text_field(&field_name, text_options)
-                    }
-                }
-                FieldDef::Numeric { stored, indexed, fast, precision } => {
-                    match precision {
-                        NumericType::I64 => {
-                            let mut opts = tantivy::schema::NumericOptions::default();
-                            if *stored { opts = opts.set_stored(); }
-                            if *indexed { opts = opts.set_indexed(); }
-                            if *fast { opts = opts.set_fast(); }
-                            schema_builder.add_i64_field(&field_name, opts)
-                        }
-                        NumericType::U64 => {
-                            let mut opts = tantivy::schema::NumericOptions::default();
-                            if *stored { opts = opts.set_stored(); }
-                            if *indexed { opts = opts.set_indexed(); }
-                            if *fast { opts = opts.set_fast(); }
-                            schema_builder.add_u64_field(&field_name, opts)
-                        }
-                        NumericType::F64 => {
-                            let mut opts = tantivy::schema::NumericOptions::default();
-                            if *stored { opts = opts.set_stored(); }
-                            if *indexed { opts = opts.set_indexed(); }
-                            if *fast { opts = opts.set_fast(); }
-                            schema_builder.add_f64_field(&field_name, opts)
-                        }
-                        NumericType::Date => {
-                            let mut opts = tantivy::schema::DateOptions::default();
-                            if *stored { opts = opts.set_stored(); }
-                            if *indexed { opts = opts.set_indexed(); }
-                            if *fast { opts = opts.set_fast(); }
-                            schema_builder.add_date_field(&field_name, opts)
-                        }
-                    }
-                }
-                FieldDef::Tag { stored, separator: _, case_sensitive: _ } => {
-                    let mut text_options = TextOptions::default();
-                    if *stored {
-                        text_options = text_options.set_stored();
-                    }
-                    text_options = text_options.set_indexing_options(
-                        TextFieldIndexing::default()
-                            .set_tokenizer("raw")
-                            .set_index_option(tantivy::schema::IndexRecordOption::Basic)
-                    );
-                    schema_builder.add_text_field(&field_name, text_options)
-                }
-                FieldDef::Geo { stored } => {
-                    // For now, store as two f64 fields for lat/lon
-                    let mut opts = tantivy::schema::NumericOptions::default();
-                    if *stored { opts = opts.set_stored(); }
-                    opts = opts.set_indexed().set_fast();
-                    
-                    let lat_field = schema_builder.add_f64_field(&format!("{}_lat", field_name), opts.clone());
-                    let lon_field = schema_builder.add_f64_field(&format!("{}_lon", field_name), opts);
-                    
-                    fields.insert(format!("{}_lat", field_name), (lat_field, FieldDef::Numeric {
-                        stored: *stored,
-                        indexed: true,
-                        fast: true,
-                        precision: NumericType::F64,
-                    }));
-                    fields.insert(format!("{}_lon", field_name), (lon_field, FieldDef::Numeric {
-                        stored: *stored,
-                        indexed: true,
-                        fast: true,
-                        precision: NumericType::F64,
-                    }));
-                    continue; // Skip adding the geo field itself
-                }
-            };
-            
-            fields.insert(field_name.clone(), (field, field_def));
-        }
-
-        let schema = schema_builder.build();
-        let index_schema = IndexSchema {
-            schema: schema.clone(),
-            fields,
-            default_search_fields,
-        };
-
-        // Create or open index
-        let dir = MmapDirectory::open(&index_path)
-            .map_err(|e| DBError(format!("Failed to open index directory: {}", e)))?;
-        
-        let mut index = Index::open_or_create(dir, schema)
-            .map_err(|e| DBError(format!("Failed to create index: {}", e)))?;
-
-        // Configure tokenizers
-        let tokenizer_manager = TokenizerManager::default();
-        index.set_tokenizers(tokenizer_manager);
-
-        let writer = index.writer(1_000_000)
-            .map_err(|e| DBError(format!("Failed to create index writer: {}", e)))?;
-
-        let reader = index
-            .reader_builder()
-            .reload_policy(ReloadPolicy::OnCommitWithDelay)
-            .try_into()
-            .map_err(|e| DBError(format!("Failed to create reader: {}", e)))?;
-
-        let config = config.unwrap_or_default();
-
-        Ok(TantivySearch {
-            index,
-            writer: Arc::new(RwLock::new(writer)),
-            reader,
-            index_schema,
-            name,
-            config,
-        })
-    }
-
-    pub fn add_document_with_fields(
-        &self,
-        doc_id: &str,
-        fields: HashMap<String, String>,
-    ) -> Result<(), DBError> {
-        let mut writer = self.writer.write()
-            .map_err(|e| DBError(format!("Failed to acquire writer lock: {}", e)))?;
-
-        // Delete existing document with same ID
-        if let Some((id_field, _)) = self.index_schema.fields.get("_id") {
-            writer.delete_term(Term::from_field_text(*id_field, doc_id));
-        }
-
-        // Create new document
-        let mut doc = tantivy::doc!();
-        
-        // Add document ID
-        if let Some((id_field, _)) = self.index_schema.fields.get("_id") {
-            doc.add_text(*id_field, doc_id);
-        }
-
-        // Add other fields based on schema
-        for (field_name, field_value) in fields {
-            if let Some((field, field_def)) = self.index_schema.fields.get(&field_name) {
-                match field_def {
-                    FieldDef::Text { .. } => {
-                        doc.add_text(*field, &field_value);
-                    }
-                    FieldDef::Numeric { precision, .. } => {
-                        match precision {
-                            NumericType::I64 => {
-                                if let Ok(v) = field_value.parse::<i64>() {
-                                    doc.add_i64(*field, v);
-                                }
-                            }
-                            NumericType::U64 => {
-                                if let Ok(v) = field_value.parse::<u64>() {
-                                    doc.add_u64(*field, v);
-                                }
-                            }
-                            NumericType::F64 => {
-                                if let Ok(v) = field_value.parse::<f64>() {
-                                    doc.add_f64(*field, v);
-                                }
-                            }
-                            NumericType::Date => {
-                                if let Ok(v) = field_value.parse::<i64>() {
-                                    doc.add_date(*field, DateTime::from_timestamp_millis(v));
-                                }
-                            }
-                        }
-                    }
-                    FieldDef::Tag { separator, case_sensitive, .. } => {
-                        let tags = if !case_sensitive {
-                            field_value.to_lowercase()
-                        } else {
-                            field_value.clone()
-                        };
-                        
-                        // Store tags as separate terms for efficient filtering
-                        for tag in tags.split(separator.as_str()) {
-                            doc.add_text(*field, tag.trim());
-                        }
-                    }
-                    FieldDef::Geo { .. } => {
-                        // Parse "lat,lon" format
-                        let parts: Vec<&str> = field_value.split(',').collect();
-                        if parts.len() == 2 {
-                            if let (Ok(lat), Ok(lon)) = (parts[0].parse::<f64>(), parts[1].parse::<f64>()) {
-                                if let Some((lat_field, _)) = self.index_schema.fields.get(&format!("{}_lat", field_name)) {
-                                    doc.add_f64(*lat_field, lat);
-                                }
-                                if let Some((lon_field, _)) = self.index_schema.fields.get(&format!("{}_lon", field_name)) {
-                                    doc.add_f64(*lon_field, lon);
-                                }
-                            }
-                        }
-                    }
-                }
-            }
-        }
-
-        writer.add_document(doc).map_err(|e| DBError(format!("Failed to add document: {}", e)))?;
-
-        writer.commit()
-            .map_err(|e| DBError(format!("Failed to commit: {}", e)))?;
-
-        Ok(())
-    }
-
-    pub fn search_with_options(
-        &self,
-        query_str: &str,
-        options: SearchOptions,
-    ) -> Result<SearchResults, DBError> {
-        let searcher = self.reader.searcher();
-        
-        // Parse query based on search fields
-        let query: Box<dyn Query> = if self.index_schema.default_search_fields.is_empty() {
-            return Err(DBError("No searchable fields defined in schema".to_string()));
-        } else {
-            let query_parser = QueryParser::for_index(
-                &self.index,
-                self.index_schema.default_search_fields.clone(),
-            );
-            
-            Box::new(query_parser.parse_query(query_str)
-                .map_err(|e| DBError(format!("Failed to parse query: {}", e)))?)
-        };
-
-        // Apply filters if any
-        let final_query = if !options.filters.is_empty() {
-            let mut clauses: Vec<(Occur, Box<dyn Query>)> = vec![(Occur::Must, query)];
-
-            // Add filters
-            for filter in options.filters {
-                if let Some((field, _)) = self.index_schema.fields.get(&filter.field) {
-                    match filter.filter_type {
-                        FilterType::Equals(value) => {
-                            let term_query = TermQuery::new(
-                                Term::from_field_text(*field, &value),
-                                tantivy::schema::IndexRecordOption::Basic,
-                            );
-                            clauses.push((Occur::Must, Box::new(term_query)));
-                        }
-                        FilterType::Range { min: _, max: _ } => {
-                            // Would need numeric field handling here
-                            // Simplified for now
-                        }
-                        FilterType::InSet(values) => {
-                            let mut sub_clauses: Vec<(Occur, Box<dyn Query>)> = vec![];
-                            for value in values {
-                                let term_query = TermQuery::new(
-                                    Term::from_field_text(*field, &value),
-                                    tantivy::schema::IndexRecordOption::Basic,
-                                );
-                                sub_clauses.push((Occur::Should, Box::new(term_query)));
-                            }
-                            clauses.push((Occur::Must, Box::new(BooleanQuery::new(sub_clauses))));
-                        }
-                    }
-                }
-            }
-            
-            Box::new(BooleanQuery::new(clauses))
-        } else {
-            query
-        };
-
-        // Execute search
-        let top_docs = searcher.search(
-            &*final_query,
-            &TopDocs::with_limit(options.limit + options.offset)
-        ).map_err(|e| DBError(format!("Search failed: {}", e)))?;
-
-        let total_hits = top_docs.len();
-        let mut documents = Vec::new();
-
-        for (score, doc_address) in top_docs.iter().skip(options.offset).take(options.limit) {
-            let retrieved_doc: TantivyDocument = searcher.doc(*doc_address)
-                .map_err(|e| DBError(format!("Failed to retrieve doc: {}", e)))?;
-
-            let mut doc_fields = HashMap::new();
-            
-            // Extract all stored fields
-            for (field_name, (field, field_def)) in &self.index_schema.fields {
-                match field_def {
-                    FieldDef::Text { stored, .. } |
-                    FieldDef::Tag { stored, .. } => {
-                        if *stored {
-                            if let Some(value) = retrieved_doc.get_first(*field) {
-                                if let Some(text) = value.as_str() {
-                                    doc_fields.insert(field_name.clone(), text.to_string());
-                                }
-                            }
-                        }
-                    }
-                    FieldDef::Numeric { stored, precision, .. } => {
-                        if *stored {
-                            let value_str = match precision {
-                                NumericType::I64 => {
-                                    retrieved_doc.get_first(*field)
-                                        .and_then(|v| v.as_i64())
-                                        .map(|v| v.to_string())
-                                }
-                                NumericType::U64 => {
-                                    retrieved_doc.get_first(*field)
-                                        .and_then(|v| v.as_u64())
-                                        .map(|v| v.to_string())
-                                }
-                                NumericType::F64 => {
-                                    retrieved_doc.get_first(*field)
-                                        .and_then(|v| v.as_f64())
-                                        .map(|v| v.to_string())
-                                }
-                                NumericType::Date => {
-                                    retrieved_doc.get_first(*field)
-                                        .and_then(|v| v.as_datetime())
-                                        .map(|v| v.into_timestamp_millis().to_string())
-                                }
-                            };
-                            
-                            if let Some(v) = value_str {
-                                doc_fields.insert(field_name.clone(), v);
-                            }
-                        }
-                    }
-                    FieldDef::Geo { stored } => {
-                        if *stored {
-                            let lat_field = self.index_schema.fields.get(&format!("{}_lat", field_name)).unwrap().0;
-                            let lon_field = self.index_schema.fields.get(&format!("{}_lon", field_name)).unwrap().0;
-                            
-                            let lat = retrieved_doc.get_first(lat_field).and_then(|v| v.as_f64());
-                            let lon = retrieved_doc.get_first(lon_field).and_then(|v| v.as_f64());
-                            
-                            if let (Some(lat), Some(lon)) = (lat, lon) {
-                                doc_fields.insert(field_name.clone(), format!("{},{}", lat, lon));
-                            }
-                        }
-                    }
-                }
-            }
-
-            documents.push(SearchDocument {
-                fields: doc_fields,
-                score: *score,
-            });
-        }
-
-        Ok(SearchResults {
-            total: total_hits,
-            documents,
-        })
-    }
-
-    pub fn get_info(&self) -> Result<IndexInfo, DBError> {
-        let searcher = self.reader.searcher();
-        let num_docs = searcher.num_docs();
-        
-        let fields_info: Vec<FieldInfo> = self.index_schema.fields.iter().map(|(name, (_, def))| {
-            FieldInfo {
-                name: name.clone(),
-                field_type: format!("{:?}", def),
-            }
-        }).collect();
-
-        Ok(IndexInfo {
-            name: self.name.clone(),
-            num_docs,
-            fields: fields_info,
-            config: self.config.clone(),
-        })
-    }
-}
-
-#[derive(Debug, Clone)]
-pub struct SearchOptions {
-    pub limit: usize,
-    pub offset: usize,
-    pub filters: Vec<Filter>,
-    pub sort_by: Option<String>,
-    pub return_fields: Option<Vec<String>>,
-    pub highlight: bool,
-}
-
-impl Default for SearchOptions {
-    fn default() -> Self {
-        SearchOptions {
-            limit: 10,
-            offset: 0,
-            filters: vec![],
-            sort_by: None,
-            return_fields: None,
-            highlight: false,
-        }
-    }
-}
-
-#[derive(Debug, Clone)]
-pub struct Filter {
-    pub field: String,
-    pub filter_type: FilterType,
-}
-
-#[derive(Debug, Clone)]
-pub enum FilterType {
-    Equals(String),
-    Range { min: String, max: String },
-    InSet(Vec<String>),
-}
-
-#[derive(Debug)]
-pub struct SearchResults {
-    pub total: usize,
-    pub documents: Vec<SearchDocument>,
-}
-
-#[derive(Debug)]
-pub struct SearchDocument {
-    pub fields: HashMap<String, String>,
-    pub score: f32,
-}
-
-#[derive(Debug, Serialize, Deserialize)]
-pub struct IndexInfo {
-    pub name: String,
-    pub num_docs: u64,
-    pub fields: Vec<FieldInfo>,
-    pub config: IndexConfig,
-}
-
-#[derive(Debug, Serialize, Deserialize)]
-pub struct FieldInfo {
-    pub name: String,
-    pub field_type: String,
-}