前言

經過前幾篇的介紹以及講解後，接下來要開始在 Spring 上架設 Semantic Kernel，將這個功能建立在後端，讓前端及後端可以分工合作

Spring 架構

首先照前面建立 Spring boot + Spring Data 新增一個專案
接著新增以下的類別

pom.xml

我們要將 Semantic Kernel 的依賴加進專案，輸入以下的程式碼

    </dependencies>

        <dependency>
            <groupId>com.microsoft.semantic-kernel</groupId>
            <artifactId>semantickernel-api</artifactId>
            <version>1.2.2</version>
        </dependency>
        <dependency>
            <groupId>com.microsoft.semantic-kernel</groupId>
            <artifactId>semantickernel-aiservices-openai</artifactId>
            <version>1.2.2</version>
        </dependency>

    </dependencies>

    <dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>com.microsoft.semantic-kernel</groupId>
                <artifactId>semantickernel-bom</artifactId>
                <version>1.2.2</version>
                <type>pom</type>
                <scope>import</scope>
            </dependency>
        </dependencies>
    </dependencyManagement>

記得點選 Load Maven

application.properties

接著要設定一些基本的項目，這部分就跟前面介紹的是一樣的

spring.application.name=SementicKernelDemo
# PostgreSQL
spring.datasource.url=jdbc:postgresql://localhost:5432/semantic_kernel
spring.datasource.username=postgres
spring.datasource.password=postgres
spring.datasource.driver-class-name=org.postgresql.Driver


# JPA
spring.jpa.database-platform=org.hibernate.dialect.PostgreSQLDialect
spring.jpa.show-sql=true
spring.jpa.hibernate.ddl-auto=update
spring.jpa.properties.hibernate.format_sql=true
spring.jpa.database=postgresql

資料庫設定

資料庫的部分跟之前也一樣，到 DB 新增資料庫
首先要確認連線正常

接著建立一個新的 Database，名成要跟上方 application.propertise 設定的一樣

Controller

這次先進行一個簡單的測試

@Slf4j
@RestController
@RequiredArgsConstructor
@RequestMapping("/semantic-kernel/demo")
public class SemanticKernelController {
    private final SemanticKernelService service;

    @GetMapping
    public ResponseEntity<Object> TestChat() {
        return ResponseEntity.ok(service.TestChat());
    }

}

Service

一些基本的設定

@Slf4j
@Service
@RequiredArgsConstructor
public class SemanticKernelService {

    private final String modelId = "gpt-4o-mini";
    private final String OPEN_AI_KEY = "API KEY";

    AzureKeyCredential openAICredential = new AzureKeyCredential(OPEN_AI_KEY);

    // Create the client
    OpenAIAsyncClient client = new OpenAIClientBuilder()
            .credential(openAICredential)
            .buildAsyncClient();

    // Create the chat completion service
    ChatCompletionService openAIChatCompletion = OpenAIChatCompletion.builder()
            .withOpenAIAsyncClient(client)
            .withModelId(modelId)
            .build();

    // Initialize the kernel
    Kernel kernel = Kernel.builder()
            .withAIService(ChatCompletionService.class, openAIChatCompletion)
            .build();

    ChatCompletionService chatCompletionService;

    {
        try {
            chatCompletionService = kernel.getService(ChatCompletionService.class);
        } catch (ServiceNotFoundException e) {
            log.error("Service : " + e.getMessage());
        }
    }

TestChat

    public Object TestChat(){
        ChatHistory history = new ChatHistory();
        history.addUserMessage("Hello, how are you?");

        List<ChatMessageContent<?>> results = chatCompletionService.getChatMessageContentsAsync(
                history,
                kernel,
                null
        ).block();

        System.out.println(results.get(0).getContent());

        return results.get(0).getContent();
    }

接著只要使用 Postman 對這個方法的路徑測試就可以看到效果囉