example: add rag application example (#3743)

Signed-off-by: yuluo-yx <[email protected]>

spring-cloud-alibaba-examples/spring-cloud-ai-example/pom.xml → spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-example/pom.xml

@@ -24,7 +24,7 @@
 		<artifactId>spring-cloud-alibaba-examples</artifactId>
 		<groupId>com.alibaba.cloud</groupId>
 		<version>${revision}</version>
-		<relativePath>../pom.xml</relativePath>
+		<relativePath>../../pom.xml</relativePath>
 	</parent>
 	<modelVersion>4.0.0</modelVersion>
 

spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-example/src/main/java/com/alibaba/cloud/ai/example/tongyi/service/impl/textembedding/README.md

@@ -0,0 +1,22 @@
+# Spring Cloud Alibaba AI Text Embedding
+
+`TongYiController` 接受一个 HTTP GET 请求 `http://localhost:8080/ai/audio`。
+`controller` 将会调用 `TongYiService` 中的 `genAudio` 方法，完成服务请求得到响应。
+
+有一个可选的 `text` 参数，其默认值为“Spring Cloud Alibaba AI 框架！”。 请求响应来自 Alibaba TongYi Text Embedding 服务。
+
+## 构建和运行
+
+1. 修改配置文件 `application.yml` 中的 apikey 为有效的 apikey；
+2. 通过 IDE 或者 `./mvnw spring-boot:run` 运行应用程序。
+
+## 访问接口
+
+使用 curl 工具或者使用浏览器对接口发起请求：
+
+```shell
+$ curl http://localhost:8080/ai/textEmbedding
+
+# Response: 
+为一组向量集合
+```

spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-rag-example/README-en.md

@@ -0,0 +1,126 @@
+# Spring Cloud Alibaba AI RAG Example
+
+This sample describes how to implement a RAG (Retrieval Augmented Generation) application using SCA AI and Spring AI RedisVector Store.
+
+> RAG is a generative model based on retrieval, which combines retrieval and generation to produce more accurate and diverse texts.
+> SCA AI: Spring Cloud Alibaba AI, adapting TongYi LLM big model through Spring AI API.
+> Spring AI: The Spring AI project aims to simplify the development of applications that include artificial intelligence features and avoid unnecessary complexity.
+> Spring AI RedisVector Store: Redis extends the core functionality of Redis OSS to allow Redis to be used as a vector database. Spring AI provides the RedisVector Store adapter.
+> Project Code Address: [spring-cloud-ai-rag-example](https://github.com/alibaba/spring-cloud-alibaba/tree/2023.x/spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-rag-example)
+
+## 1. Environmental preparation
+
+Use Docker Compose to deploy a Redis service to store vector data.
+
+```yaml
+version: '3.8'
+
+services:
+  redis:
+    image: redis/redis-stack-server
+    container_name: redis
+    hostname: redis
+    ports:
+      - 6379:6379
+```
+
+Start with `docker compose up -d`, and then you can `docker ps | grep redis` check to see if the container is running properly.
+
+## 2. Project dependency
+
+> This project introduces `spring-cloud-alibaba-ai-starter` and `spring-ai-redis-spring-boot-starter` realizes RAG application.
+
+You need to introduce the following dependencies in the POM. XML:
+
+```xml
+<dependency>
+    <groupId>com.alibaba.cloud</groupId>
+    <artifactId>spring-cloud-starter-alibaba-ai</artifactId>
+</dependency>
+
+<dependency>
+    <groupId>org.springframework.ai</groupId>
+    <artifactId>spring-ai-redis-spring-boot-starter</artifactId>
+    <version>${spring.ai.version}</version>
+</dependency>
+```
+
+## 3. Configuration
+
+Configure the following information in the application. Yml:
+
+```yaml
+spring:
+  ai:
+    vectorstore:
+      redis:
+        index: peer
+        prefix: peer
+```
+
+## 4. Write the code
+
+The `loader` classes are as follows:
+
+```java 
+@Override
+public void run(ApplicationArguments args) throws Exception {
+
+    Map<String, Object> indexInfo = vectorStore.getJedis().ftInfo(properties.getIndex());
+    int numDocs = Integer.parseInt((String) indexInfo.getOrDefault("num_docs", "0"));
+    if (numDocs > 20000) {
+        logger.info("Embeddings already loaded. Skipping");
+        return;
+    }
+
+    Resource file = data;
+    if (Objects.requireNonNull(data.getFilename()).endsWith(".gz")) {
+        GZIPInputStream inputStream = new GZIPInputStream(data.getInputStream());
+        file = new InputStreamResource(inputStream, "beers.json.gz");
+    }
+
+    logger.info("Creating Embeddings...");
+    JsonReader loader = new JsonReader(file, KEYS);
+    vectorStore.add(loader.get());
+    logger.info("Embeddings created.");
+}
+```
+
+The `Service` classes are as follows:
+
+```java
+public Generation retrieve(String message) {
+
+    SearchRequest request = SearchRequest.query(message).withTopK(topK);
+    List<Document> docs = store.similaritySearch(request);
+
+    Message systemMessage = getSystemMessage(docs);
+    UserMessage userMessage = new UserMessage(message);
+
+    Prompt prompt = new Prompt(List.of(systemMessage, userMessage));
+    ChatResponse response = client.call(prompt);
+
+    return response.getResult();
+}
+
+private Message getSystemMessage(List<Document> similarDocuments) {
+
+    String documents = similarDocuments.stream()
+            .map(Document::getContent)
+            .collect(Collectors.joining("\n"));
+    SystemPromptTemplate systemPromptTemplate = new SystemPromptTemplate(systemBeerPrompt);
+
+    return systemPromptTemplate.createMessage(Map.of("documents", documents));
+}
+```
+
+## 5. Run and verify
+
+You can start the SprigBoot main class and then use a browser to access:
+
+```shell
+# request params is prompt，the default value：What ber pairs well with smoked meats?"
+http://localhost:8080/rag/chat
+```
+
+To experience the RAG application.

spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-rag-example/README.md

@@ -0,0 +1,126 @@
+# Spring Cloud Alibaba AI RAG Example
+
+本示例介绍如何使用 SCA AI 和 Spring AI RedisVectorStore 实现 RAG（Retrieval Augmented Generation）应用。
+
+> RAG 是一个基于检索的生成模型，它将检索和生成结合在一起，以生成更加准确和多样化的文本。
+> SCA AI: Spring Cloud Alibaba AI, 通过 Spring AI API 适配 TongYi LLM 大模型。
+> Spring AI: Spring AI项目旨在简化包含人工智能功能的应用程序的开发，避免不必要的复杂性。
+> Spring AI RedisVectorStore: Redis 扩展了 Redis OSS 的核心功能，允许将 Redis 用作矢量数据库，Spring AI 提供了 RedisVectorStore 适配器。
+> 项目代码地址：[spring-cloud-ai-rag-example](https://github.com/alibaba/spring-cloud-alibaba/tree/2023.x/spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-rag-example)
+
+## 1. 环境准备
+
+使用 Docker Compose 部署一个 Redis 服务，用于存储向量数据。
+
+```yaml
+version: '3.8'
+
+services:
+  redis:
+    image: redis/redis-stack-server
+    container_name: redis
+    hostname: redis
+    ports:
+      - 6379:6379
+```
+
+使用 `docker compose up -d` 启动，然后您可以通过 `docker ps | grep redis` 查看容器是否正常运行。
+
+## 2. 项目依赖
+
+> 本项目通过引入 `spring-cloud-alibaba-ai-starter` 和 `spring-ai-redis-spring-boot-starter` 实现 RAG 应用。
+
+您需要在 pom.xml 中引入如下依赖：
+
+```xml
+<dependency>
+    <groupId>com.alibaba.cloud</groupId>
+    <artifactId>spring-cloud-starter-alibaba-ai</artifactId>
+</dependency>
+
+<dependency>
+    <groupId>org.springframework.ai</groupId>
+    <artifactId>spring-ai-redis-spring-boot-starter</artifactId>
+    <version>${spring.ai.version}</version>
+</dependency>
+```
+
+## 3. 配置
+
+在 application.yml 中配置如下信息：
+
+```yaml
+spring:
+  ai:
+    vectorstore:
+      redis:
+        index: peer
+        prefix: peer
+```
+
+## 4. 编写代码
+
+`loader` 类如下所示：
+
+```java 
+@Override
+public void run(ApplicationArguments args) throws Exception {
+
+    Map<String, Object> indexInfo = vectorStore.getJedis().ftInfo(properties.getIndex());
+    int numDocs = Integer.parseInt((String) indexInfo.getOrDefault("num_docs", "0"));
+    if (numDocs > 20000) {
+        logger.info("Embeddings already loaded. Skipping");
+        return;
+    }
+
+    Resource file = data;
+    if (Objects.requireNonNull(data.getFilename()).endsWith(".gz")) {
+        GZIPInputStream inputStream = new GZIPInputStream(data.getInputStream());
+        file = new InputStreamResource(inputStream, "beers.json.gz");
+    }
+
+    logger.info("Creating Embeddings...");
+    JsonReader loader = new JsonReader(file, KEYS);
+    vectorStore.add(loader.get());
+    logger.info("Embeddings created.");
+}
+```
+
+`Service` 类如下所示：
+
+```java
+public Generation retrieve(String message) {
+
+    SearchRequest request = SearchRequest.query(message).withTopK(topK);
+    List<Document> docs = store.similaritySearch(request);
+
+    Message systemMessage = getSystemMessage(docs);
+    UserMessage userMessage = new UserMessage(message);
+
+    Prompt prompt = new Prompt(List.of(systemMessage, userMessage));
+    ChatResponse response = client.call(prompt);
+
+    return response.getResult();
+}
+
+private Message getSystemMessage(List<Document> similarDocuments) {
+
+    String documents = similarDocuments.stream()
+            .map(Document::getContent)
+            .collect(Collectors.joining("\n"));
+    SystemPromptTemplate systemPromptTemplate = new SystemPromptTemplate(systemBeerPrompt);
+
+    return systemPromptTemplate.createMessage(Map.of("documents", documents));
+}
+```
+
+## 5. 运行并验证
+
+您可以通过启动 SprigBoot 主类，之后使用浏览器访问：
+
+```shell
+# 参数为 prompt，默认值为：What ber pairs well with smoked meats?"
+http://localhost:8080/rag/chat
+```
+
+来体验 RAG 应用。

spring-cloud-alibaba-examples/ai-example/spring-cloud-ai-rag-example/docker/docker-compose.yml

@@ -0,0 +1,9 @@
+version: '3.8'
+
+services:
+  redis:
+    image: redis/redis-stack-server
+    container_name: redis
+    hostname: redis
+    ports:
+      - 6379:6379