feat(round-16): AI意图情绪分析/展会名片双向互换/雷达爬虫清洗/离线多媒体缓存 (103套件1101测试全绿)

backend/src/modules/crm/service/bot-filter.ts

@@ -0,0 +1,102 @@
+import { Provide } from '@midwayjs/core';
+import { BaseService } from '@cool-midway/core';
+
+export type BotClassification = 'SEARCH_CRAWLER' | 'SOCIAL_PREVIEW' | 'SCRIPTER';
+
+export interface BotDetectResult {
+  isBot: boolean;
+  botType?: BotClassification;
+  botName?: string;
+}
+
+export interface TrafficFilterResult {
+  isBot: boolean;
+  category: 'HUMAN' | 'CRAWLER' | 'SUSPICIOUS_SPAM';
+  shouldRecordLead: boolean;
+  reason?: string;
+}
+
+const SEARCH_BOTS = ['googlebot', 'bingbot', 'baiduspider', 'yandexbot', 'duckduckbot', 'sogou'];
+const SOCIAL_BOTS = ['whatsapp', 'twitterbot', 'facebookexternalhit', 'linkedinbot', 'telegrambot', 'slackbot'];
+const SCRIPT_BOTS = ['curl', 'wget', 'python-requests', 'postman', 'httpclient', 'scrapy', 'axios'];
+
+/**
+ * 基于 User-Agent 识别机器人与爬虫
+ */
+export function detectBotUserAgent(userAgent?: string): BotDetectResult {
+  if (!userAgent || typeof userAgent !== 'string') {
+    return { isBot: false };
+  }
+
+  const ua = userAgent.toLowerCase();
+
+  for (const bot of SEARCH_BOTS) {
+    if (ua.includes(bot)) {
+      return { isBot: true, botType: 'SEARCH_CRAWLER', botName: bot };
+    }
+  }
+
+  for (const bot of SOCIAL_BOTS) {
+    if (ua.includes(bot)) {
+      return { isBot: true, botType: 'SOCIAL_PREVIEW', botName: bot };
+    }
+  }
+
+  for (const bot of SCRIPT_BOTS) {
+    if (ua.includes(bot)) {
+      return { isBot: true, botType: 'SCRIPTER', botName: bot };
+    }
+  }
+
+  return { isBot: false };
+}
+
+/**
+ * 检测 1 秒内的频次是否异常超高
+ */
+export function isRateLimitSuspicious(timestamps: number[], thresholdPerSec = 10): boolean {
+  if (!timestamps || timestamps.length < thresholdPerSec) {
+    return false;
+  }
+  const sorted = [...timestamps].sort((a, b) => a - b);
+  const now = sorted[sorted.length - 1];
+  const countInLastSecond = sorted.filter(t => now - t <= 1000).length;
+  return countInLastSecond >= thresholdPerSec;
+}
+
+/**
+ * 访客雷达异常流量与爬虫清洗过滤服务
+ */
+@Provide()
+export class RadarBotFilterService extends BaseService {
+  async filterIncomingTraffic(input: {
+    ip: string;
+    userAgent: string;
+    recentTimestamps?: number[];
+  }): Promise<TrafficFilterResult> {
+    const botCheck = detectBotUserAgent(input.userAgent);
+    if (botCheck.isBot) {
+      return {
+        isBot: true,
+        category: 'CRAWLER',
+        shouldRecordLead: false,
+        reason: `检测到 ${botCheck.botType}: ${botCheck.botName}`,
+      };
+    }
+
+    if (input.recentTimestamps && isRateLimitSuspicious(input.recentTimestamps)) {
+      return {
+        isBot: true,
+        category: 'SUSPICIOUS_SPAM',
+        shouldRecordLead: false,
+        reason: '1秒内突发异常高频请求，疑似自动化刷量攻击',
+      };
+    }
+
+    return {
+      isBot: false,
+      category: 'HUMAN',
+      shouldRecordLead: true,
+    };
+  }
+}