阶段性修改

yelaymaths_api/tools/legacy/build.ts

@@ -8,6 +8,7 @@
 import { asString, type LegacyRows } from './load';
 import {
   HOME_COURSE_TITLE_MAP,
+  OPENID_RE,
   transformArticle,
   transformBanner,
   transformCourse,
@@ -29,7 +30,16 @@ export function buildDataset(db: LegacyRows, staticBase: string) {
   const courseIdByCode = new Map(courses.map((c, i) => [c.code, i + 1]));
   const courseIdByTitle = new Map(courses.map((c, i) => [c.title, i + 1]));
 
-  const users = db.用户信息.map(transformUser);
+  // 过滤旧库里混入的攻击载荷：只有符合 openid 形态的才当作真实用户
+  const skippedUsers: Array<{ openid: string; reason: string }> = [];
+  const users = db.用户信息.flatMap((row) => {
+    const openid = asString(row.OpenID);
+    if (!OPENID_RE.test(openid)) {
+      skippedUsers.push({ openid: openid.slice(0, 60), reason: '非法 openid 形态（疑似攻击载荷）' });
+      return [];
+    }
+    return [transformUser(row)];
+  });
   const userIdByOpenid = new Map(users.map((u, i) => [u.openid, i + 1]));
 
   const skippedLessonOrphans: string[] = [];
@@ -45,10 +55,14 @@ export function buildDataset(db: LegacyRows, staticBase: string) {
   // 权益：按 (user, course) 去重，重复的只保留第一条
   const seen = new Set<string>();
   const duplicateEntitlements: string[] = [];
+  const droppedEntitlements: string[] = [];
   const entitlements = db.用户已购课程.flatMap((row) => {
     const userId = userIdByOpenid.get(asString(row.OpenID));
     const courseId = courseIdByCode.get(asString(row.课程id));
-    if (!userId || !courseId) return [];
+    if (!userId || !courseId) {
+      droppedEntitlements.push(`${asString(row.OpenID).slice(0, 30)}::${asString(row.课程id).slice(0, 12)}`);
+      return [];
+    }
     const key = `${userId}::${courseId}`;
     if (seen.has(key)) {
       duplicateEntitlements.push(key);
@@ -81,7 +95,12 @@ export function buildDataset(db: LegacyRows, staticBase: string) {
     settings: db.cloudconfig
       .map(transformSetting)
       .filter((v): v is NonNullable<typeof v> => v !== null),
-    diagnostics: { skippedLessonOrphans, duplicateEntitlements, unmappedHomeCourses },
+    diagnostics: {
+      skippedLessonOrphans,
+      duplicateEntitlements,
+      unmappedHomeCourses,
+      skippedUsers,
+      droppedEntitlements,
+    },
   };
 }
-