xuyue/OJ2
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(名字): 界面上不再显示 ks + 班号前缀，混着几个班的地方名字后面写班级
Deploy / deploy (push) Canceled after 0s
Details

Browse Source 
- 学生名字一律只显示名字（老号 ks232张三 显示成 张三），顶栏、首页问候、悬停提示、请假名单、比赛成绩表都一样
- 全服 / 年级排名、全部提交、几个班同时举手时，名字后面淡写班级；提交列表的名字列窄，只写班号（陈长 258）
- AI 班级分析点名「要多关心的同学」时不再带 ks 前缀
- 建号查重补漏：同班同名也比真名（加过数字的张三2、编号制老号）；名字自己带数字（王芳1）也认得出被占；
  导入时后端再拦一次同班同名，建号页要等查重结果对上当前名单才能点
- 登录名单换成新接口 /classes/:c/members（名字按界面规则剥，换过班的老号不再露出全名），
  旧接口原样留给上线时还开着旧登录页的机房电脑
- 后台手写老格式 ks232张三 新建用户、没选班时按前缀补班级；PUT /me/profile 只带真名时不再 500

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
xuyueandClaude Opus 5.5 committed 2026-11-02 23:41:49 +08:00
1 parent 221d264701
commit 2bda05a978
32 files changed
+229 -111
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

apps/api/src/routes/account.ts
+7 -4
View File
@@ -263,10 +263,13 @@ accountRoutes.put('/me/profile', requireAuth, async (c) => {
const parsed = await parseBody(c, updateProfileRequestSchema, 'Invalid profile payload')
if (!parsed.success) return parsed.response
const values = Object.fromEntries(Object.entries(parsed.data).map(([key, value]) => [key, value === '' ? null : value]))
await db
.update(schema.userProfile)
.set(values)
.where(eq(schema.userProfile.userId, c.get('user')!.id))
// 只带了不收的字段（比如老页面还发 realName）：没什么可改的，别让 drizzle 的空 set 报成 500
if (Object.keys(values).length) {
await db
.update(schema.userProfile)
.set(values)
.where(eq(schema.userProfile.userId, c.get('user')!.id))
}
const profile = await getUserProfileById(c.get('user')!.id, true)
if (!profile) return failure(c, 404, 'profile-not-found', 'User profile does not exist')
return success(c, profile)
apps/api/src/routes/admin/account.ts
+49 -19
View File
@@ -198,30 +198,47 @@ adminAccountRoutes.get('/users', requireSuperAdmin, async (c) => {
})
/**
* 批量建号之前查重（批量建号页边粘名单边查）：这个班里已经有谁了（多半是同一个人粘了两次），
* 以及每个名字会建成什么用户名（名字被别的班、注册的号占了就加数字）。比较归一成小写，和导入、登录同一个口径
* 这个班里已经有的、叫这几个名字的人（多半是同一个学生粘了两次）。三种写法都要认：老号 `ks232张三`
* （剥掉前缀比）、名字被占过加了数字的新号 `张三2`、24 级数媒那批编号制用户名 —— 后两种名字只在真名里，
* 所以真名也比。比较归一成小写，和登录、导入同一个口径
*/
async function classmatesNamed(className: string, names: string[]) {
const wanted = new Set(names.map((name) => name.toLowerCase()))
const rows = await db
.select({ username: schema.user.username, realName: schema.userProfile.realName, createTime: schema.user.createTime })
.from(schema.user)
.leftJoin(schema.userProfile, eq(schema.userProfile.userId, schema.user.id))
.where(eq(schema.user.className, className))
return rows.flatMap((row) => {
const name = [displayName(row.username), row.realName ?? ''].find((candidate) => candidate && wanted.has(candidate.toLowerCase()))
return name ? [{ name, username: row.username, createTime: row.createTime }] : []
})
}
/**
* 全站已经被占的用户名里，和这几个名字撞的：名字本身，以及「名字 + 数字」（给新号加数字时要绕开）。
* 名字自己就带数字（王芳1）也照样认 —— 不能先把库里的尾数剥掉再比，那样 `王芳1` 剥成 `王芳` 就漏了
*/
async function takenUsernames(names: string[]) {
const escape = (text: string) => text.replace(/[\\^$.|?*+()[\]{}]/g, '\\$&')
const pattern = `^(${[...new Set(names.map((name) => escape(name.toLowerCase())))].join('|')})[0-9]*$`
const rows = await db
.select({ username: sql<string>`lower(${schema.user.username})` })
.from(schema.user)
.where(sql`lower(${schema.user.username}) ~ ${pattern}`)
return new Set(rows.map((row) => row.username))
}
/**
* 批量建号之前查重（批量建号页边粘名单边查）：这个班里已经有谁了，以及每个名字会建成什么用户名
* （名字被别的班、注册的号占了就加数字）。导入时后端还会再查一遍同班同名，这里只是让老师提前看到
*/
adminAccountRoutes.post('/users/check', requireSuperAdmin, async (c) => {
const parsed = await parseBody(c, checkUsernamesRequestSchema)
if (!parsed.success) return parsed.response
const { className, names } = parsed.data
const wanted = [...new Set(names.map((name) => name.toLowerCase()))]
// 同班同名：老号是 ks232张三，按剥掉前缀的名字比
const classmates = await db
.select({ username: schema.user.username, createTime: schema.user.createTime })
.from(schema.user)
.where(eq(schema.user.className, className))
const existing = classmates
.map((row) => ({ name: displayName(row.username), username: row.username, createTime: row.createTime }))
.filter((row) => wanted.includes(row.name.toLowerCase()))
// 用户名全站唯一：名字本身和「名字+数字」占了哪些
const taken = await db
.select({ username: sql<string>`lower(${schema.user.username})` })
.from(schema.user)
.where(inArray(sql`regexp_replace(lower(${schema.user.username}), '[0-9]+$', '')`, wanted))
return success(c, { existing, usernames: suggestUsernames(names, new Set(taken.map((row) => row.username))) } satisfies CheckUsernamesResponse)
const [existing, taken] = await Promise.all([classmatesNamed(className, names), takenUsernames(names)])
return success(c, { existing, usernames: suggestUsernames(names, taken) } satisfies CheckUsernamesResponse)
})
/**
@@ -379,6 +396,19 @@ adminAccountRoutes.post('/users', requireSuperAdmin, async (c) => {
return failure(c, 409, 'email-exists', `这批名单里邮箱重复：${batchMails.join('、')}`)
}
// 同班同名再查一遍：用户名不再带班号以后，「同一个学生粘了两次」撞不上用户名唯一约束了（新号会被加成张三2），
// 只靠批量建号页的查重挡 —— 那边有半秒防抖，粘完马上点就绕过去了
const byClass = new Map<string, string[]>()
for (const item of prepared) {
if (item.className) byClass.set(item.className, [...(byClass.get(item.className) ?? []), item.realName || displayName(item.username)])
}
for (const [className, names] of byClass) {
const clash = await classmatesNamed(className, names)
if (clash.length) {
return failure(c, 409, 'name-exists-in-class', `${className} 班已经有${clash.map((row) => `「${row.name}」`).join('、')}了`)
}
}
const existing = await db
.select({ username: schema.user.username, email: schema.user.email })
.from(schema.user)
apps/api/src/routes/ai.ts
+4 -1
View File
@@ -4,6 +4,7 @@ import {
aiHintRequestSchema,
classAnalysisRequestSchema,
classPkAnalysisRequestSchema,
displayName,
HINT_MIN_FAILURES,
LEARNING_PERIOD_LABELS,
learningPeriodSchema,
@@ -354,6 +355,8 @@ aiRoutes.post('/ai/class-analysis', requireAuth, async (c) => {
// 数字和抽屉里看到的是同一份（services/ranking.ts 的 classDetail），不再由前端整包传上来
const detail = await classDetail(parsed.data.className, true)
if (!detail.members) return failure(c, 404, 'class-not-found', '这个班没有学生')
// 「要多关心的同学」会被点名：老号的用户名带 ks252 前缀，喂给模型之前剥掉，报告里写的才是界面上那个名字
const forModel = { ...detail, care: detail.care?.map((row) => ({ ...row, user: { ...row.user, username: displayName(row.user.username) } })) ?? null }
return streamChat(
[
'你是编程课老师的助教。下面是一个班这学期在 OJ 上的做题数据（JSON，人名是用户名）。',
@@ -362,7 +365,7 @@ aiRoutes.post('/ai/class-analysis', requireAuth, async (c) => {
'**分化**（前后 10%、人最多的那一档同分）、**建议**（点名 care 里的同学，给下节课能直接做的事）。',
'用大白话，不要出现四分位数、标准差、方差这类统计词，不要编数据里没有的数字。',
].join(''),
JSON.stringify(detail),
JSON.stringify(forModel),
)
})
apps/api/src/routes/flowchart.ts
+8 -16
View File
@@ -1,6 +1,7 @@
import { randomBytes } from 'node:crypto'
import {
displayName,
createFlowchartRequestSchema,
isFlowchartPass,
type CreateFlowchartResponse,
@@ -27,18 +28,7 @@ import { consumeToken } from '../services/throttling'
import { buildWordFrequencies } from '../services/word-frequency'
import { dayStart } from '../time'
import { unknownDisplayIds } from '../services/submission-filters'
import {
classCondition,
isAdminRole,
isTeacherOrAbove,
matchedUsers,
asRecord,
parseDisplayIds,
queryInteger,
scopedUsers,
rounded,
stripClassPrefix,
} from './helpers'
import { classCondition, isAdminRole, isTeacherOrAbove, matchedUsers, asRecord, parseDisplayIds, queryInteger, scopedUsers, rounded } from './helpers'
export const flowchartRoutes = new Hono<AppEnv>()
@@ -293,6 +283,7 @@ const flowchartListColumns = {
evaluationTime: schema.flowchartSubmission.evaluationTime,
},
username: schema.user.username,
className: schema.user.className,
problem: {
displayId: schema.problem.displayId,
title: schema.problem.title,
@@ -349,10 +340,11 @@ flowchartRoutes.get('/flowcharts', requireAuth, async (c) => {
)
return success(c, {
results: rows.map(
({ flowchart, username, problem }) =>
({ flowchart, username, className, problem }) =>
({
id: flowchart.id,
username,
className,
problemDisplayId: problem.displayId,
problemTitle: problem.title,
status: flowchart.status,
@@ -540,7 +532,7 @@ flowchartRoutes.get('/flowcharts/statistics', requireTeacher, async (c) => {
const peopleTruncated = peopleRows.length > PEOPLE_LIMIT
const people = peopleRows.slice(0, PEOPLE_LIMIT).map((row) => ({
username: row.username,
realName: stripClassPrefix(row.username, row.className),
realName: displayName(row.username),
bestScore: row.bestScore === null ? null : Number(row.bestScore),
bestGrade: row.bestGrade,
count: row.count,
@@ -561,7 +553,7 @@ flowchartRoutes.get('/flowcharts/statistics', requireTeacher, async (c) => {
// 于是一节课刚开始、最该点名的时候，教师面板反而一个名字都不给
dataUnaccepted: roster.map((row) => ({
username: row.username,
realName: stripClassPrefix(row.username, row.className),
realName: displayName(row.username),
})),
people: [],
peopleTruncated: false,
@@ -639,7 +631,7 @@ flowchartRoutes.get('/flowcharts/statistics', requireTeacher, async (c) => {
.filter((row) => !submitted.has(row.id))
.map((row) => ({
username: row.username,
realName: stripClassPrefix(row.username, row.className),
realName: displayName(row.username),
})),
people,
peopleTruncated,
apps/api/src/routes/helpers.ts
+3 -3
View File
@@ -29,9 +29,9 @@ export function sampleUser(
}
/**
* 去掉老号用户名里的 `ks<班级号>` 前缀，得到学生本人那一段：`ks251张三` + `251` → `张三`。
* 2026-11 起新建的号不带前缀（用户名就是名字），原样返回。和契约 `displayName` 是同一件事，
* 这里按这个人自己的班级剥，比正则更严：班号对不上的不剥。
* 去掉老号用户名里的 `ks<班级号>` 前缀：`ks251张三` + `251` → `张三`。只剩旧版登录名单（site.ts 的
* /classes/:className/usernames）在用 —— 旧登录页拿名字拼回「ks + 班号 + 名字」去登录，所以必须严格按这个班的班号剥。
* 界面上显示名字一律用契约的 `displayName`。
*
* 用 startsWith + slice 而不是 replace：replace 会删掉字符串中间的匹配，
* 前缀对不上时从中间截出乱码。前缀不匹配就原样返回。
apps/api/src/routes/site.ts
+26 -3
View File
@@ -1,4 +1,4 @@
import { CLASS_NAME_RE, type ClassMember, type Quote, type WebsiteConfig } from '@oj2/contract'
import { CLASS_NAME_RE, displayName, type ClassMember, type Quote, type WebsiteConfig } from '@oj2/contract'
import { asc, desc, eq } from 'drizzle-orm'
import { Hono } from 'hono'
import { resolve } from 'node:path'
@@ -92,6 +92,30 @@ siteRoutes.get('/quotes/random', async (c) => {
}
})
/**
* 登录框点名字的名单：名字给人点，用户名拿去登录（老号 ks232张三，新号就是名字）。名字按界面上的显示规则剥，
* 和 UserName 一致 —— 不能只剥「ks + 当前班号」：班级现在能在后台改，换过班的老号会露出整串
*/
siteRoutes.get('/classes/:className/members', async (c) => {
const className = c.req.param('className').trim()
if (!CLASS_NAME_RE.test(className)) {
return failure(c, 400, 'invalid-class', 'Class name must contain 3 or 4 digits')
}
const rows = await db
.select({ username: schema.user.username })
.from(schema.user)
.where(eq(schema.user.className, className))
.orderBy(desc(schema.user.createTime), asc(schema.user.id))
return success(
c,
rows.map(({ username }) => ({ name: displayName(username), username }) satisfies ClassMember),
)
})
/**
* 旧版登录框的名单：只回名字，前端拼「ks + 班号 + 名字」去登录。2026-11 的新前端改走上面的 /members；
* 这条原样留着，是给上线时还开着旧登录页、没刷新的机房电脑用的（老号照样登得上）。过一阵可以删
*/
siteRoutes.get('/classes/:className/usernames', async (c) => {
const className = c.req.param('className').trim()
if (!CLASS_NAME_RE.test(className)) {
@@ -102,9 +126,8 @@ siteRoutes.get('/classes/:className/usernames', async (c) => {
.from(schema.user)
.where(eq(schema.user.className, className))
.orderBy(desc(schema.user.createTime), asc(schema.user.id))
// 名字给人点，用户名拿去登录：老号是 ks232张三（点的是「张三」），新号两个一样
return success(
c,
rows.map(({ username }) => ({ name: stripClassPrefix(username, className), username }) satisfies ClassMember),
rows.map(({ username }) => stripClassPrefix(username, className)),
)
})
apps/api/src/routes/submission-statistics.ts
+5 -4
View File
@@ -7,6 +7,7 @@
*/
import {
displayName,
FLOWCHART_PASS_GRADES,
NO_CLASS,
STUDENT_ROLES,
@@ -27,7 +28,7 @@ import { type ContestEnv } from '../services/contest'
import { getBooleanOption } from '../services/options'
import { lessonDoneCounts } from '../services/lesson'
import { calendarDay, dayStart, localTime } from '../time'
import { isAdminRole, parseDisplayIds, rounded, scopedUsers, stripClassPrefix } from './helpers'
import { isAdminRole, parseDisplayIds, rounded, scopedUsers } from './helpers'
export const submissionStatisticsRoutes = new Hono<ContestEnv>()
@@ -345,7 +346,7 @@ submissionStatisticsRoutes.get('/submissions/statistics', requireTeacher, async
.filter((row) => !submittedUserIds.has(row.id))
.map((row) => ({
username: row.username,
realName: stripClassPrefix(row.username, row.className),
realName: displayName(row.username),
}))
/**
@@ -374,7 +375,7 @@ submissionStatisticsRoutes.get('/submissions/statistics', requireTeacher, async
* 不传用户名的全站视图里各班混在一起，剥完只剩一串重名的名字，反而认不出谁，
* 所以原样给完整用户名。班名取 perUser join 出来的那一列，和花名册同一份数据。
*/
realName: scoped ? stripClassPrefix(row.username, row.className) : row.username,
realName: scoped ? displayName(row.username) : row.username,
submissionCount: row.submissionCount,
solvedCount: row.solvedCount,
}))
@@ -481,7 +482,7 @@ submissionStatisticsRoutes.get('/submissions/statistics/grid', requireTeacher, a
if (!entry) {
entry = {
username: row.username,
realName: stripClassPrefix(row.username, row.className),
realName: displayName(row.username),
className: row.className,
submissions: [],
}