rag-lite-ts 1.0.1 → 2.0.0

package/dist/core/abstract-embedder.js

@@ -0,0 +1,264 @@
+/**
+ * CORE MODULE — Abstract Base Embedder
+ *
+ * Provides model-agnostic base functionality for all embedder implementations.
+ * This is an abstract base class, not a concrete implementation.
+ *
+ * ARCHITECTURAL NOTE:
+ * While this contains implementation logic, it remains in the core layer because:
+ * 1. It's model-agnostic (no knowledge of specific models or transformers.js)
+ * 2. It's shared by multiple implementation layers (text, multimodal)
+ * 3. It provides common infrastructure (lifecycle, validation, batch processing)
+ * 4. Moving it would create awkward cross-layer dependencies
+ *
+ * This follows the "shared base class" pattern common in framework design,
+ * similar to React.Component, Django Model, or other framework base classes.
+ *
+ * RESPONSIBILITIES:
+ * - Model lifecycle management (loading, cleanup, disposal)
+ * - Batch processing coordination
+ * - Input validation and text truncation
+ * - Error handling with helpful messages
+ * - Embedding ID generation
+ * - Common utility methods
+ *
+ * IMPLEMENTATION LAYERS:
+ * - Text: SentenceTransformerEmbedder extends this class
+ * - Multimodal: CLIPEmbedder extends this class
+ */
+import { ModelRegistry } from './model-registry.js';
+import { validateContentType, createEnhancedEmbeddingResult } from './universal-embedder.js';
+// =============================================================================
+// BASE EMBEDDER ABSTRACT CLASS
+// =============================================================================
+/**
+ * Abstract base class for universal embedders
+ * Provides common functionality and lifecycle management
+ */
+export class BaseUniversalEmbedder {
+    modelName;
+    options;
+    _isLoaded = false;
+    _modelInfo;
+    constructor(modelName, options = {}) {
+        this.modelName = modelName;
+        this.options = options;
+        const modelInfo = ModelRegistry.getModelInfo(modelName);
+        if (!modelInfo) {
+            throw new Error(`Model '${modelName}' is not supported. ` +
+                `Supported models: ${ModelRegistry.getSupportedModels().join(', ')}`);
+        }
+        this._modelInfo = modelInfo;
+    }
+    // =============================================================================
+    // PUBLIC INTERFACE IMPLEMENTATION
+    // =============================================================================
+    get modelType() {
+        return this._modelInfo.type;
+    }
+    get dimensions() {
+        return this._modelInfo.dimensions;
+    }
+    get supportedContentTypes() {
+        return this._modelInfo.supportedContentTypes;
+    }
+    isLoaded() {
+        return this._isLoaded;
+    }
+    getModelInfo() {
+        return { ...this._modelInfo }; // Return a copy to prevent mutation
+    }
+    /**
+     * Dispose of all resources and prepare for garbage collection
+     * This method should be called when the embedder is no longer needed
+     */
+    async dispose() {
+        try {
+            // Call the specific cleanup implementation
+            await this.cleanup();
+            // Clear internal state
+            this._isLoaded = false;
+            // Force garbage collection if available
+            if (global.gc) {
+                global.gc();
+            }
+            this.logModelLoading('Resources disposed and garbage collection triggered');
+        }
+        catch (error) {
+            console.warn(`Error during resource disposal: ${error instanceof Error ? error.message : 'Unknown error'}`);
+        }
+    }
+    // =============================================================================
+    // DEFAULT IMPLEMENTATIONS
+    // =============================================================================
+    /**
+     * Batch embedding with default implementation
+     * Subclasses can override for more efficient batch processing
+     */
+    async embedBatch(items) {
+        if (!this._isLoaded) {
+            await this.loadModel();
+        }
+        const results = [];
+        const batchSize = this.options.maxBatchSize || this._modelInfo.capabilities.maxBatchSize || 8;
+        // Process in batches to avoid memory issues
+        for (let i = 0; i < items.length; i += batchSize) {
+            const batch = items.slice(i, i + batchSize);
+            const batchResults = await this.processBatch(batch);
+            results.push(...batchResults);
+        }
+        return results;
+    }
+    // =============================================================================
+    // PROTECTED HELPER METHODS
+    // =============================================================================
+    /**
+     * Process a single batch of items
+     * Can be overridden by subclasses for more efficient batch processing
+     */
+    async processBatch(batch) {
+        const results = [];
+        for (const item of batch) {
+            try {
+                validateContentType(item.contentType, this.supportedContentTypes);
+                let result;
+                if (item.contentType === 'text') {
+                    result = await this.embedText(item.content);
+                }
+                else if (item.contentType === 'image' && this.embedImage) {
+                    result = await this.embedImage(item.content);
+                }
+                else {
+                    throw new Error(`Content type '${item.contentType}' not supported by model '${this.modelName}'`);
+                }
+                // Enhance the result with content type and metadata
+                const enhancedResult = createEnhancedEmbeddingResult(result.embedding_id, result.vector, item.contentType, item.metadata);
+                results.push(enhancedResult);
+            }
+            catch (error) {
+                // Log error but continue processing other items
+                console.warn(`Failed to embed item: ${error instanceof Error ? error.message : 'Unknown error'}`);
+                // Create a placeholder result with zero vector for failed items
+                const zeroVector = new Float32Array(this.dimensions).fill(0);
+                const failedResult = createEnhancedEmbeddingResult(`failed_${Date.now()}_${Math.random()}`, zeroVector, item.contentType, { ...item.metadata, error: error instanceof Error ? error.message : 'Unknown error' });
+                results.push(failedResult);
+            }
+        }
+        return results;
+    }
+    /**
+     * Validate that the model is loaded before operations
+     */
+    ensureLoaded() {
+        if (!this._isLoaded) {
+            throw new Error(`Model '${this.modelName}' is not loaded. Call loadModel() first.`);
+        }
+    }
+    /**
+     * Generate a unique embedding ID
+     */
+    generateEmbeddingId(content, contentType = 'text') {
+        const timestamp = Date.now();
+        const random = Math.random().toString(36).substring(2);
+        const contentHash = this.simpleHash(content);
+        return `${contentType}_${contentHash}_${timestamp}_${random}`;
+    }
+    /**
+     * Simple hash function for content identification
+     */
+    simpleHash(str) {
+        let hash = 0;
+        for (let i = 0; i < str.length; i++) {
+            const char = str.charCodeAt(i);
+            hash = ((hash << 5) - hash) + char;
+            hash = hash & hash; // Convert to 32-bit integer
+        }
+        return Math.abs(hash).toString(36);
+    }
+    /**
+     * Validate text length against model constraints
+     */
+    validateTextLength(text) {
+        const maxLength = this._modelInfo.capabilities.maxTextLength;
+        if (maxLength && text.length > maxLength) {
+            console.warn(`Text length (${text.length}) exceeds model maximum (${maxLength}). ` +
+                `Text will be truncated.`);
+        }
+    }
+    /**
+     * Truncate text to model's maximum length
+     */
+    truncateText(text) {
+        const maxLength = this._modelInfo.capabilities.maxTextLength;
+        if (maxLength && text.length > maxLength) {
+            return text.substring(0, maxLength);
+        }
+        return text;
+    }
+    /**
+     * Log model loading progress
+     */
+    logModelLoading(stage, details) {
+        const message = `[${this.modelName}] ${stage}`;
+        if (details) {
+            console.log(`${message}: ${details}`);
+        }
+        else {
+            console.log(message);
+        }
+    }
+    /**
+     * Handle model loading errors with helpful messages
+     */
+    handleLoadingError(error) {
+        const baseMessage = `Failed to load model '${this.modelName}': ${error.message}`;
+        // Provide specific guidance based on error type
+        if (error.message.includes('network') || error.message.includes('fetch')) {
+            return new Error(`${baseMessage}\n` +
+                `This appears to be a network error. Please check your internet connection ` +
+                `and ensure the model repository is accessible.`);
+        }
+        if (error.message.includes('memory') || error.message.includes('OOM')) {
+            return new Error(`${baseMessage}\n` +
+                `This appears to be a memory error. Try using a smaller model or ` +
+                `increase available memory. Required: ${this._modelInfo.requirements.minimumMemory}MB`);
+        }
+        if (error.message.includes('unsupported') || error.message.includes('not found')) {
+            const suggestions = ModelRegistry.getSupportedModels(this.modelType);
+            return new Error(`${baseMessage}\n` +
+                `Model may not be available. Supported ${this.modelType} models: ${suggestions.join(', ')}`);
+        }
+        return new Error(baseMessage);
+    }
+}
+// =============================================================================
+// UTILITY FUNCTIONS
+// =============================================================================
+/**
+ * Create embedder options with defaults
+ */
+export function createEmbedderOptions(options = {}) {
+    return {
+        maxBatchSize: 8,
+        timeout: 30000, // 30 seconds
+        enableGPU: false,
+        logLevel: 'info',
+        ...options
+    };
+}
+/**
+ * Validate embedder options
+ */
+export function validateEmbedderOptions(options) {
+    if (options.maxBatchSize && (options.maxBatchSize < 1 || options.maxBatchSize > 128)) {
+        throw new Error('maxBatchSize must be between 1 and 128');
+    }
+    if (options.timeout && options.timeout < 1000) {
+        throw new Error('timeout must be at least 1000ms');
+    }
+    const validLogLevels = ['debug', 'info', 'warn', 'error', 'silent'];
+    if (options.logLevel && !validLogLevels.includes(options.logLevel)) {
+        throw new Error(`logLevel must be one of: ${validLogLevels.join(', ')}`);
+    }
+}
+//# sourceMappingURL=abstract-embedder.js.map

package/dist/core/actionable-error-messages.d.ts

@@ -0,0 +1,60 @@
+/**
+ * CORE MODULE — Actionable Error Messages
+ * Provides user-friendly, actionable error messages with specific guidance
+ * Replaces technical error messages with helpful troubleshooting steps
+ */
+/**
+ * Configuration for actionable error messages
+ */
+export interface ActionableErrorConfig {
+    /** Include examples in error messages */
+    includeExamples?: boolean;
+    /** Include troubleshooting steps */
+    includeTroubleshooting?: boolean;
+    /** Context about the current operation */
+    operationContext?: string;
+}
+/**
+ * Create actionable error message for missing files
+ */
+export declare function createMissingFileError(filePath: string, fileType: 'index' | 'database' | 'config' | 'content', config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for invalid paths
+ */
+export declare function createInvalidPathError(paths: {
+    name: string;
+    value: string | undefined;
+}[], config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for model loading failures
+ */
+export declare function createModelLoadingError(modelName: string, originalError: string, config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for dimension mismatches
+ */
+export declare function createDimensionMismatchError(expected: number, actual: number, context: string, config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for mode mismatches
+ */
+export declare function createModeMismatchError(expectedMode: string, actualMode: string, config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for empty or invalid content
+ */
+export declare function createInvalidContentError(contentType: string, issue: 'empty' | 'invalid_format' | 'too_large' | 'unsupported', config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for missing dependencies
+ */
+export declare function createMissingDependencyError(dependencyName: string, dependencyType: 'function' | 'object' | 'service', config?: ActionableErrorConfig): Error;
+/**
+ * Create actionable error message for factory creation failures
+ */
+export declare function createFactoryCreationError(factoryName: string, originalError: string, config?: ActionableErrorConfig): Error;
+/**
+ * Enhance an existing error with actionable information
+ */
+export declare function enhanceError(originalError: Error, context: string, suggestions?: string[], config?: ActionableErrorConfig): Error;
+/**
+ * Create a user-friendly error message with context
+ */
+export declare function createContextualError(message: string, context: string, suggestions?: string[], examples?: string[], config?: ActionableErrorConfig): Error;
+//# sourceMappingURL=actionable-error-messages.d.ts.map