package com.teacher.service.Impl;
import com.google.common.util.concurrent.RateLimiter;
import com.teacher.exception.AiServiceException;
import com.teacher.model.dto.AiRequest;
import com.teacher.model.dto.AiResponse;
import com.teacher.model.dto.TeachingPlanDto;
import com.teacher.service.AiService;
import com.teacher.util.QwenClient;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;
import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.TimeUnit;
@Service
public class AiServiceImpl implements AiService {
private static final double AI_RATE_LIMIT = 5.0; // 5 requests per second
private final QwenClient qwenClient;
private final RateLimiter rateLimiter = RateLimiter.create(AI_RATE_LIMIT);
@Autowired
public AiServiceImpl(QwenClient qwenClient) {
this.qwenClient = qwenClient;
}
@Override
public AiResponse chatWithAI(AiRequest request) {
try {
if (!rateLimiter.tryAcquire(1, TimeUnit.SECONDS)) {
throw new AiServiceException("AI服务请求过载，请稍后再试");
}
// 直接使用AiRequest调用
String response = qwenClient.callQwenApi(request);
return new AiResponse(response, System.currentTimeMillis());
} catch (Exception e) {
throw new AiServiceException("AI服务调用失败", e);
}
}
@Override
public String generateSummary(String content, int maxLength) {
String prompt = String.format(
"请为以下内容生成一个简洁的摘要，不超过%d个字符：\n\n%s",
maxLength, content
);
return callQwenApiWithRetry(prompt, 2);
}
@Override
public List<String> extractKnowledgePoints(String content) {
String prompt = String.format(
"请从以下教学内容中提取关键知识点，每个知识点用短句表示，按重要性排序：\n\n%s",
content
);
String response = callQwenApiWithRetry(prompt, 2);
return parseKnowledgePoints(response);
}
@Override
public String correctText(String text) {
String prompt = String.format(
"请校对以下文本，修正语法、拼写和标点错误，保持原意不变：\n\n%s",
text
);
return callQwenApiWithRetry(prompt, 1);
}
@Override
public String generateMultimediaContent(String prompt, String mediaType) {
String fullPrompt = String.format(
"根据以下描述生成%s内容：\n%s\n\n请使用适合%s的格式输出",
mediaType, prompt, mediaType
);
return callQwenApiWithRetry(fullPrompt, 2);
}
private String callQwenApiWithRetry(String prompt, int maxRetries) {
int attempt = 0;
while (attempt < maxRetries) {
try {
return qwenClient.callQwenApi(prompt);
} catch (Exception e) {
attempt++;
if (attempt >= maxRetries) {
throw new AiServiceException("AI服务调用失败，重试次数已用完", e);
}
// 指数退避重试
try {
Thread.sleep((long) (Math.pow(2, attempt) * 1000));
} catch (InterruptedException ie) {
Thread.currentThread().interrupt();
}
}
}
return "AI服务暂时不可用";
}
private List<String> parseKnowledgePoints(String response) {
List<String> points = new ArrayList<>();
String[] lines = response.split("\n");
for (String line : lines) {
if (line.matches("^\\d+\\.\\s+.+")) {
points.add(line.substring(line.indexOf('.') + 1).trim());
}
}
return points;
}
// 实现生成课程设计方法
@Override
public String generateCourseDesign(String subject, List<String> knowledgePoints) {
String knowledgeList = String.join("、", knowledgePoints);
String prompt = String.format(
"请为%s学科设计一个完整的课程，包含以下知识点：%s。要求包括教学目标、教学活动和评估方法。",
subject, knowledgeList
);
return callQwenApiWithRetry(prompt, 2);
}
// 实现生成教学计划方法
@Override
public TeachingPlanDto generateTeachingPlan(String subject, List<String> keywords) {
String keywordStr = String.join("、", keywords);
String prompt = String.format(
"请为%s学科生成一个详细的教学计划，关键词包括：%s。要求包括年级、计划标题、描述、时长和多个教学单元（每个单元包含主题、目标、知识点、资源和评估方法）。",
subject, keywordStr
);
String response = callQwenApiWithRetry(prompt, 2);
return parseTeachingPlan(response);
}
// 实现分析学生答案方法
@Override
public String analyzeAnswer(String question, String studentAnswer, String correctAnswer) {
String prompt = String.format(
"问题：%s\n学生的答案：%s\n正确答案：%s\n请分析学生答案中的错误，并给出改进建议。",
question, studentAnswer, correctAnswer
);
return callQwenApiWithRetry(prompt, 2);
}
// 解析教学计划响应的辅助方法
private TeachingPlanDto parseTeachingPlan(String response) {
TeachingPlanDto plan = new TeachingPlanDto();
// 简化的解析逻辑 - 实际项目中应使用更健壮的解析方法
if (response.contains("标题：")) {
plan.setTitle(response.substring(response.indexOf("标题：") + 3, response.indexOf("\n")));
}
// 设置其他字段...
return plan;
}
}
这是原有的AiServiceImpl，请给出修改后的AiServiceImpl
最新发布
本文详细解析了HBaseUI中RequestsPerSecond指标的含义，该指标显示当前regionserver每秒接收的RPC请求数量，是监控HBase系统性能的关键参数。通过对这一指标的分析，可以深入了解HBase集群的工作负载和潜在瓶颈。
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
3644
