@retrivora-ai/rag-engine 1.9.6 → 1.9.8

package/src/llm/providers/QwenProvider.ts

@@ -0,0 +1,191 @@
+/**
+ * Qwen LLM Provider (via Alibaba DashScope / ModelStudio compatible mode)
+ */
+
+import OpenAI from 'openai';
+import { ILLMProvider, EmbedOptions } from '../ILLMProvider';
+import { ChatMessage, ChatOptions } from '../../types';
+import { LLMConfig, EmbeddingConfig } from '../../config/RagConfig';
+import { IProviderValidator, IProviderHealthChecker, HealthCheckResult } from '../../core/ProviderInterfaces';
+import { ValidationError } from '../../core/ConfigValidator';
+import { buildSystemContent } from '../utils';
+
+export class QwenProvider implements ILLMProvider {
+  private readonly client: OpenAI;
+  private readonly llmConfig: LLMConfig;
+  private readonly embeddingConfig?: EmbeddingConfig;
+
+  constructor(llmConfig: LLMConfig, embeddingConfig?: EmbeddingConfig) {
+    const apiKey = llmConfig.apiKey || process.env.DASHSCOPE_API_KEY || process.env.QWEN_API_KEY;
+    if (!apiKey) throw new Error('[QwenProvider] llmConfig.apiKey, DASHSCOPE_API_KEY, or QWEN_API_KEY environment variable is required');
+    
+    this.client = new OpenAI({
+      apiKey,
+      baseURL: llmConfig.baseUrl || 'https://dashscope.aliyuncs.com/compatible-mode/v1',
+    });
+    this.llmConfig = llmConfig;
+    this.embeddingConfig = embeddingConfig;
+  }
+
+  static getValidator(): IProviderValidator {
+    return {
+      validate(config: Record<string, unknown>): ValidationError[] {
+        const errors: ValidationError[] = [];
+        const isEmbedding = config.provider === 'qwen' && 'dimensions' in config;
+        const prefix = isEmbedding ? 'embedding' : 'llm';
+
+        if (!config.apiKey && !process.env.DASHSCOPE_API_KEY && !process.env.QWEN_API_KEY) {
+          errors.push({
+            field: `${prefix}.apiKey`,
+            message: 'Qwen API key is required',
+            suggestion: 'Set DASHSCOPE_API_KEY or QWEN_API_KEY environment variable',
+            severity: 'error',
+          });
+        }
+        if (!config.model) {
+          errors.push({
+            field: `${prefix}.model`,
+            message: 'Qwen model name is required',
+            suggestion: isEmbedding ? 'e.g., "text-embedding-v3"' : 'e.g., "qwen-plus" or "qwen-max"',
+            severity: 'error',
+          });
+        }
+        return errors;
+      }
+    };
+  }
+
+  static getHealthChecker(): IProviderHealthChecker {
+    return {
+      async check(config: Record<string, unknown>): Promise<HealthCheckResult> {
+        const timestamp = Date.now();
+        const apiKey = (config.apiKey as string) || process.env.DASHSCOPE_API_KEY || process.env.QWEN_API_KEY || '';
+        const modelName = config.model as string;
+        
+        try {
+          const OpenAI = await import('openai');
+          const client = new OpenAI.default({
+            apiKey,
+            baseURL: (config.baseUrl as string) || 'https://dashscope.aliyuncs.com/compatible-mode/v1',
+          });
+          const models = await client.models.list();
+          const hasModel = models.data.some((m) => m.id === modelName);
+          return {
+            healthy: true,
+            provider: 'qwen',
+            capabilities: { model: modelName, available: hasModel, totalModels: models.data.length },
+            timestamp,
+          };
+        } catch (error) {
+          return {
+            healthy: false,
+            provider: 'qwen',
+            error: `Connection failed: ${error instanceof Error ? error.message : String(error)}`,
+            timestamp,
+          };
+        }
+      }
+    };
+  }
+
+  async chat(messages: ChatMessage[], context: string, options?: ChatOptions): Promise<string> {
+    const basePrompt =
+      options?.systemPrompt ??
+      this.llmConfig.systemPrompt ??
+      'You are a helpful assistant. Answer questions based on the provided context.';
+
+    const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam = {
+      role: 'system',
+      content: buildSystemContent(basePrompt, context),
+    };
+
+    const formattedMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
+      systemMessage,
+      ...messages.map((m) => ({
+        role: m.role as 'user' | 'assistant',
+        content: m.content,
+      })),
+    ];
+
+    const completion = await this.client.chat.completions.create({
+      model: this.llmConfig.model,
+      messages: formattedMessages,
+      max_tokens: options?.maxTokens ?? this.llmConfig.maxTokens ?? 1024,
+      temperature: options?.temperature ?? this.llmConfig.temperature ?? 0.7,
+      stop: options?.stop,
+    });
+
+    return completion.choices[0]?.message?.content ?? '';
+  }
+  
+  async *chatStream(messages: ChatMessage[], context: string, options?: ChatOptions): AsyncIterable<string> {
+    const basePrompt =
+      options?.systemPrompt ??
+      this.llmConfig.systemPrompt ??
+      'You are a helpful assistant. Answer questions based on the provided context.';
+
+    const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam = {
+      role: 'system',
+      content: buildSystemContent(basePrompt, context),
+    };
+
+    const formattedMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
+      systemMessage,
+      ...messages.map((m) => ({
+        role: m.role as 'user' | 'assistant',
+        content: m.content,
+      })),
+    ];
+
+    const stream = await this.client.chat.completions.create({
+      model: this.llmConfig.model,
+      messages: formattedMessages,
+      max_tokens: options?.maxTokens ?? this.llmConfig.maxTokens ?? 1024,
+      temperature: options?.temperature ?? this.llmConfig.temperature ?? 0.7,
+      stop: options?.stop,
+      stream: true,
+    });
+
+    if (!stream) {
+      throw new Error('[QwenProvider] completions.create stream is undefined');
+    }
+
+    for await (const chunk of stream) {
+      const content = chunk.choices[0]?.delta?.content || '';
+      if (content) yield content;
+    }
+  }
+
+  async embed(text: string, options?: EmbedOptions): Promise<number[]> {
+    const results = await this.batchEmbed([text], options);
+    return results[0];
+  }
+
+  async batchEmbed(texts: string[], options?: EmbedOptions): Promise<number[][]> {
+    const model =
+      options?.model ??
+      this.embeddingConfig?.model ??
+      'text-embedding-v1';
+
+    const apiKey = this.embeddingConfig?.apiKey ?? this.llmConfig.apiKey;
+    const client = apiKey !== this.llmConfig.apiKey
+      ? new OpenAI({
+          apiKey,
+          baseURL: this.embeddingConfig?.baseUrl || 'https://dashscope.aliyuncs.com/compatible-mode/v1',
+        })
+      : this.client;
+
+    const response = await client.embeddings.create({ model, input: texts });
+    return response.data.map((d) => d.embedding);
+  }
+
+  async ping(): Promise<boolean> {
+    try {
+      await this.client.models.list();
+      return true;
+    } catch (err) {
+      console.error('[QwenProvider] Ping failed:', err);
+      return false;
+    }
+  }
+}

package/src/server.ts

@@ -25,30 +25,20 @@ export type { ILLMProvider, EmbedOptions } from './llm/ILLMProvider';
 export type { Chunk, ChunkOptions } from './rag/DocumentChunker';
 export type { HealthCheckResult, IProviderValidator, IProviderHealthChecker } from './core/ProviderInterfaces';
 export type { ValidationError } from './core/ConfigValidator';
+export type { LicensePayload } from './core/LicenseVerifier';
 export type { BatchOptions, BatchResult } from './core/BatchProcessor';
 
 // ── Core Orchestration ─────────────────────────────────────────
-export { VectorPlugin } from './core/VectorPlugin';
 export { Retrivora } from './core/Retrivora';
+export { VectorPlugin } from './core/VectorPlugin';
 export { Pipeline } from './core/Pipeline';
 export { ConfigResolver } from './core/ConfigResolver';
 export { ProviderRegistry } from './core/ProviderRegistry';
 export { ConfigValidator } from './core/ConfigValidator';
+export { LicenseVerifier } from './core/LicenseVerifier';
 export { ProviderHealthCheck } from './core/ProviderHealthCheck';
 export { BatchProcessor } from './core/BatchProcessor';
 
-// ── Exceptions ────────────────────────────────────────────────
-export {
-  RetrivoraError,
-  ProviderNotFoundException,
-  EmbeddingFailedException,
-  RetrievalException,
-  RateLimitException,
-  ConfigurationException,
-  AuthenticationException,
-} from './exceptions';
-export type { RetrivoraErrorCode } from './exceptions';
-
 // ── Configuration Helpers ──────────────────────────────────────
 export { getRagConfig } from './config/serverConfig';
 export { ConfigBuilder, PRESETS, createFromPreset } from './config/ConfigBuilder';
@@ -77,6 +67,8 @@ export { LLMFactory } from './llm/LLMFactory';
 export { OpenAIProvider } from './llm/providers/OpenAIProvider';
 export { AnthropicProvider } from './llm/providers/AnthropicProvider';
 export { OllamaProvider } from './llm/providers/OllamaProvider';
+export { GroqProvider } from './llm/providers/GroqProvider';
+export { QwenProvider } from './llm/providers/QwenProvider';
 export { UniversalLLMAdapter } from './llm/providers/UniversalLLMAdapter';
 
 // ── Next.js Route Handler Factories ───────────────────────────
@@ -86,8 +78,26 @@ export {
   createIngestHandler,
   createHealthHandler,
   createUploadHandler,
+  createHistoryHandler,
+  createFeedbackHandler,
+  createSessionsHandler,
+  createRagHandler,
   sseFrame,
   sseTextFrame,
   sseMetaFrame,
   sseErrorFrame,
 } from './handlers';
+
+// ── Exceptions ────────────────────────────────────────────────
+export type { RetrivoraErrorCode } from './exceptions';
+export {
+  RetrivoraError,
+  ProviderNotFoundException,
+  EmbeddingFailedException,
+  RetrievalException,
+  RateLimitException,
+  ConfigurationException,
+  AuthenticationException,
+  wrapError,
+} from './exceptions';
+

package/src/types/chat.ts

@@ -41,6 +41,12 @@ export interface ChatOptions {
 export interface UseRagChatOptions {
   /** Override the chat API endpoint (default: /api/chat) */
   apiUrl?: string;
+  /**
+   * Base URL for the Retrivora SDK catch-all route used for history, feedback,
+   * and sessions endpoints. Defaults to '/api/retrivora'.
+   * e.g. if your catch-all is at /api/retrivora/[[...retrivora]], set this to '/api/retrivora'.
+   */
+  retrivoraApiBase?: string;
   /** Override project namespace */
   namespace?: string;
   /** Persist chat history to localStorage (default: true) */
@@ -49,6 +55,10 @@ export interface UseRagChatOptions {
   onReply?: (message: RagMessage) => void;
   /** Called on error */
   onError?: (error: string) => void;
+  /** Optional custom headers to send with the chat request */
+  headers?: Record<string, string>;
+  /** Session ID for history and feedback tracking (default: 'default') */
+  sessionId?: string;
 }
 
 export interface UseRagChatReturn {
@@ -68,4 +78,10 @@ export interface UseRagChatReturn {
   setMessages: React.Dispatch<React.SetStateAction<RagMessage[]>>;
   /** Abort the current active stream request */
   stop: () => void;
+  /** Load message history from database storage */
+  loadHistory: (sessionId: string) => Promise<void>;
+  /** Clear conversation history from database and locally */
+  clearHistory: (sessionId: string) => Promise<void>;
+  /** Submit rating and optional comment for a specific assistant message */
+  submitFeedback: (messageId: string, rating: 'thumbs_up' | 'thumbs_down', comment?: string) => Promise<void>;
 }

package/src/types/props.ts

@@ -1,8 +1,37 @@
-import { ReactNode, CSSProperties, MouseEvent } from 'react';
+import { ReactNode, CSSProperties, MouseEvent, ElementType } from 'react';
 import { Product, VectorMatch } from './index';
 import { RagMessage } from './chat';
 import { UIConfig } from '../config/RagConfig';
 
+export interface ArchitectureCardProps {
+  icon: ReactNode;
+  title: string;
+  description: string;
+  badge: string;
+  badgeColor: string;
+}
+
+export interface Snippet {
+  id: string;
+  title: string;
+  description: string;
+  code: string;
+  language: string;
+}
+
+export interface PipelineStep {
+  step: string;
+  Icon: ElementType;
+  title: string;
+  desc: string;
+  colors: { from: string; to: string };
+}
+
+export interface ProviderPill {
+  Icon: ElementType;
+  label: string;
+}
+
 export type ChatViewportSize = 'compact' | 'medium' | 'large';
 
 export interface ChatWindowProps {
@@ -26,6 +55,15 @@ export interface ChatWindowProps {
   isMaximized?: boolean;
   /** Called when the user clicks 'Add to Cart' on a product */
   onAddToCart?: (product: Product) => void;
+  /** Optional custom API URL to send chat messages to */
+  apiUrl?: string;
+  /**
+   * Base URL for Retrivora SDK history/feedback/sessions endpoints.
+   * Defaults to '/api/retrivora'. Set this to match your catch-all route location.
+   */
+  retrivoraApiBase?: string;
+  /** Optional custom headers to send with the chat request */
+  headers?: Record<string, string>;
 }
 
 export interface ChatWidgetProps {
@@ -33,6 +71,15 @@ export interface ChatWidgetProps {
   position?: 'bottom-right' | 'bottom-left';
   /** Called when the user clicks 'Add to Cart' on a product */
   onAddToCart?: (product: Product) => void;
+  /** Optional custom API URL to send chat messages to */
+  apiUrl?: string;
+  /**
+   * Base URL for Retrivora SDK history/feedback/sessions endpoints.
+   * Defaults to '/api/retrivora'.
+   */
+  retrivoraApiBase?: string;
+  /** Optional custom headers to send with the chat request */
+  headers?: Record<string, string>;
 }
 
 export interface MessageBubbleProps {
@@ -43,6 +90,8 @@ export interface MessageBubbleProps {
   accentColor: string;
   onAddToCart?: (product: Product) => void;
   viewportSize?: ChatViewportSize;
+  /** Called when the user clicks thumbs up or thumbs down on an assistant message */
+  onFeedback?: (messageId: string, rating: 'thumbs_up' | 'thumbs_down') => void;
 }
 
 export interface ProductCardProps {

package/.env.example

@@ -1,80 +0,0 @@
-# ─── Project ─────────────────────────────────────────────────
-RAG_PROJECT_ID=my-project
-
-# ─── Vector Database ──────────────────────────────────────────
-# Choose: pinecone | pgvector | rest | universal_rest
-VECTOR_DB_PROVIDER=pinecone
-VECTOR_DB_INDEX=rag-index
-
-# Pinecone (if provider=pinecone)
-PINECONE_API_KEY=your-pinecone-api-key
-PINECONE_ENVIRONMENT=us-east-1
-
-# pgVector (if provider=pgvector)
-PGVECTOR_CONNECTION_STRING=postgresql://user:password@localhost:5432/StagVectorDB
-
-# Generic REST (if provider=rest)
-VECTOR_DB_REST_URL=http://localhost:8080
-VECTOR_DB_REST_API_KEY=your-rest-api-key
-# VECTOR_DB_QUERY_PATH=/query
-# VECTOR_DB_UPSERT_PATH=/upsert
-
-# ─── LLM ──────────────────────────────────────────────────────
-# Choose: openai | anthropic | ollama | gemini | universal_rest
-LLM_PROVIDER=openai
-LLM_MODEL=gpt-4o
-LLM_MAX_TOKENS=1024
-LLM_TEMPERATURE=0.7
-LLM_SYSTEM_PROMPT=You are a helpful assistant. Use the provided context to answer questions accurately.
-
-# OpenAI (if provider=openai)
-OPENAI_API_KEY=sk-...
-
-# Anthropic (if provider=anthropic)
-ANTHROPIC_API_KEY=sk-ant-...
-
-# Gemini (if provider=gemini)
-GEMINI_API_KEY=AIza...
-
-# Ollama (if provider=ollama)
-LLM_BASE_URL=http://localhost:11434
-
-# ── Master Universal Provider (LiteLLM, Qdrant, etc.) ─────────
-# Use 'universal_rest' for ANY model or DB via pre-set profiles.
-LLM_PROVIDER=universal_rest
-LLM_MODEL=gpt-4o
-# Profile: litellm | ollama-standard | openai-compatible
-LLM_UNIVERSAL_PROFILE=litellm
-LLM_BASE_URL=https://proxy.example.com/api
-LLM_API_KEY=your-key
-
-VECTOR_DB_PROVIDER=universal_rest
-# Profile: qdrant | chromadb
-VECTOR_UNIVERSAL_PROFILE=qdrant
-VECTOR_BASE_URL=http://localhost:6333
-VECTOR_DB_INDEX=my-index
-
-# ─── Embedding ────────────────────────────────────────────────
-# Choose: openai | ollama | gemini | custom
-EMBEDDING_PROVIDER=openai
-EMBEDDING_MODEL=text-embedding-3-small
-EMBEDDING_DIMENSIONS=1536
-# EMBEDDING_BASE_URL=http://localhost:11434  # for ollama
-# EMBEDDING_API_KEY=your-embedding-key       # for custom embedding endpoints
-
-# ─── RAG Pipeline Tuning ──────────────────────────────────────
-RAG_TOP_K=5
-RAG_SCORE_THRESHOLD=0.0
-RAG_CHUNK_SIZE=1000
-RAG_CHUNK_OVERLAP=200
-
-# ─── UI Branding (NEXT_PUBLIC_ = safe to expose to browser) ───
-NEXT_PUBLIC_PROJECT_ID=my-project
-NEXT_PUBLIC_UI_TITLE=AI Assistant
-NEXT_PUBLIC_UI_SUBTITLE=Powered by RAG
-NEXT_PUBLIC_PRIMARY_COLOR=#6366f1
-NEXT_PUBLIC_ACCENT_COLOR=#8b5cf6
-NEXT_PUBLIC_PLACEHOLDER=Ask me anything…
-NEXT_PUBLIC_SHOW_SOURCES=true
-NEXT_PUBLIC_WELCOME_MESSAGE=Hello! I'm your AI assistant. Ask me anything about your documents.
-# NEXT_PUBLIC_LOGO_URL=https://your-domain.com/logo.png

package/LICENSE.txt

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) [2026] [Abhinav Alkuchi]
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.