linke-sdufe 0.5.0 → 0.7.0

package/README.md

@@ -17,7 +17,7 @@
 
 | 能力 | 签名 | 说明 |
 |---|---|---|
-| fetch | `fetch(url, {method, headers, body, redirect, timeoutMs, expect}) → Promise<Response-like>` | `expect='buffer'` 表示将调 `arrayBuffer()`（uni.request 单 responseType 现实约束）；Response-like 需 `{status, ok, headers.getSetCookie(), text(), arrayBuffer()}`；网络层失败 reject |
+| fetch | `fetch(url, {method, headers, body, redirect, timeoutMs, expect}) → Promise<Response-like>` | `expect='buffer'` 表示将调 `arrayBuffer()`（uni.request 单 responseType 现实约束）；Response-like 需 `{status, ok, headers.getSetCookie(), text(), arrayBuffer()}`，如宿主能取得重定向终点应同时提供 `url`；网络层失败 reject |
 | toBase64 | `(bytes:Uint8Array) → string` | 验证码图片上行 |
 | bytesToText | `(bytes:Uint8Array) → string` | 前 200 字节嗅探（宽松实现即可，ASCII 特征判定） |
 | progress | `(msg:string) → void`（可缺省） | 进度输出 |
@@ -38,6 +38,12 @@ const { cookie, userInfo } = await adapter.login(
 )
 const schedule = await adapter.fetchSchedule(cookie, { term: '', week: '' })
 
+// 公开通知：旧方法继续返回数组；需要可靠续页时消费原页真实下页链接
+const first = await adapter.fetchPortalNoticePage('jwc')
+const second = first.pagination.nextPageUrl
+  ? await adapter.fetchPortalNoticePage('jwc', { pageUrl: first.pagination.nextPageUrl })
+  : null
+
 // uni-app（垫片注入）
 import { initSdufe } from 'linke-sdufe'
 import { uniEnv } from '@/services/sdufePilot/uniEnv.js'

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "linke-sdufe",
-  "version": "0.5.0",
+  "version": "0.7.0",
   "description": "山东财经大学强智教务（jw.sdufe.edu.cn）适配器共享包：登录、教务查询与页面解析。linke-cli 与 Linke App 同源引用；教务请求一律从用户设备直发。",
   "type": "module",
   "main": "./src/index.js",

package/src/adapter.js

@@ -21,10 +21,13 @@ import {
   parseCurrentTerm,
   parseScheduleHtml,
   parseScoresHtml,
+  parseScoresHtmlWithMeta,
   parseCreditsHtml,
   parseCoursesHtml,
   parseGpaHtml,
+  parseGpaHtmlWithMeta,
   parseXjHtml,
+  parseXjHtmlWithMeta,
   parsePlanHtml,
   parsePyfaHtml,
   parseExamsHtml,
@@ -41,15 +44,173 @@ import {
 const USER_AGENT = 'Apifox/1.0.0 (https://apifox.com)'
 const REQUEST_TIMEOUT_MS = 20000
 
+function missingFields(rows, fields) {
+  const missing = {}
+  for (const field of fields) {
+    const count = rows.filter((row) => row == null || String(row[field] == null ? '' : row[field]).trim() === '').length
+    if (count) missing[field] = count
+  }
+  return missing
+}
+
+function personalReadMeta({ endpoint, method, status, scope, rows, requiredFields, parse, rangeReason, rangeIssue = null }) {
+  const requestComplete = Number(status) >= 200 && Number(status) < 300
+  const missing = missingFields(rows, requiredFields)
+  const summary = parse && typeof parse === 'object' ? parse : {}
+  const pagination = summary.pagination && typeof summary.pagination === 'object'
+    ? summary.pagination
+    : { state: 'unknown', nextPageUrl: null, reason: '解析器未返回来源分页证明' }
+  const candidates = Number.isInteger(summary.candidateRowCount) ? summary.candidateRowCount : null
+  const parsed = Number.isInteger(summary.parsedRowCount) ? summary.parsedRowCount : rows.length
+  const skipped = Number.isInteger(summary.skippedRowCount) ? summary.skippedRowCount : null
+  const unresolvedSkipped = Number.isInteger(summary.unresolvedSkippedRowCount)
+    ? summary.unresolvedSkippedRowCount
+    : skipped
+  let rangeState = 'unknown'
+  let reason = requestComplete ? '请求完成，但解析器没有给出可核验的完整范围证明' : `教务读取返回 HTTP ${status}`
+  if (requestComplete && rangeIssue) {
+    rangeState = rangeIssue.state === 'unknown' ? 'unknown' : 'partial'
+    reason = rangeIssue.reason || '实际返回范围与请求范围不一致'
+  } else if (requestComplete && pagination.state === 'partial') {
+    rangeState = 'partial'
+    reason = pagination.reason || '来源仍有未读分页'
+  } else if (requestComplete && pagination.state === 'unknown') {
+    rangeState = 'unknown'
+    reason = pagination.reason || '来源分页范围无法确定'
+  } else if (requestComplete && unresolvedSkipped != null && unresolvedSkipped > 0) {
+    rangeState = 'partial'
+    reason = `识别到 ${candidates == null ? '?' : candidates} 条候选记录，解析 ${parsed} 条，仍有 ${unresolvedSkipped} 条无法可靠解析`
+  } else if (requestComplete && summary.tableRecognized !== false && summary.cardRecognized !== false && candidates != null && candidates === parsed + (skipped || 0) && unresolvedSkipped === 0 && ['complete', 'not_applicable'].includes(pagination.state)) {
+    rangeState = 'complete'
+    reason = rangeReason
+  }
+  return {
+    source: { system: '山东财经大学强智教务系统', endpoint, method },
+    fetchedAt: new Date().toISOString(),
+    request: { state: requestComplete ? 'complete' : 'failed', httpStatus: Number(status) || null },
+    scope,
+    range: { state: requestComplete ? rangeState : 'failed', reason },
+    fields: { state: Object.keys(missing).length ? 'unknown' : 'complete', required: requiredFields, missing },
+    rowCount: rows.length,
+    parse: {
+      tableRecognized: summary.tableRecognized == null ? null : !!summary.tableRecognized,
+      cardRecognized: summary.cardRecognized == null ? null : !!summary.cardRecognized,
+      candidateRowCount: candidates,
+      parsedRowCount: parsed,
+      skippedRowCount: skipped,
+      unresolvedSkippedRowCount: unresolvedSkipped,
+      skippedReasons: summary.skippedReasons || {}
+    },
+    pagination: { ...pagination, nextCursor: null }
+  }
+}
+
+function stableTextFingerprint(text) {
+  let hash = 0x811c9dc5
+  const input = String(text || '')
+  for (let i = 0; i < input.length; i++) {
+    hash ^= input.charCodeAt(i)
+    hash = Math.imul(hash, 0x01000193)
+  }
+  return `fnv1a32:${(hash >>> 0).toString(16).padStart(8, '0')}:${input.length}`
+}
+
+function portalNoticeRecordId(url) {
+  const match = String(url || '').match(/\/info\/(\d+)\/([^/?#]+)\.htm(?:[?#]|$)/i)
+  return match ? `${match[1]}:${match[2]}` : ''
+}
+
+function portalNoticeContinuationFingerprint({ site, finalUrl, items, pagination }) {
+  // 续读只绑定列表事实和实际分页位置；整页 HTML hash 另作诊断。
+  // 因此 nonce、页脚时钟或统计脚本变化不会让合法游标失效，
+  // 但记录增删/重排/标题/日期/URL 及下一页目标变化仍会改变指纹。
+  const projection = {
+    v: 1,
+    site: String(site || ''),
+    position: {
+      finalUrl: String(finalUrl || ''),
+      page: pagination && pagination.page == null ? null : pagination.page,
+      totalPages: pagination && pagination.totalPages == null ? null : pagination.totalPages,
+      totalItems: pagination && pagination.totalItems == null ? null : pagination.totalItems,
+      nextPageUrl: String(pagination && pagination.nextPageUrl || ''),
+    },
+    items: (items || []).map((item) => ({
+      id: portalNoticeRecordId(item && item.url),
+      url: String(item && item.url || ''),
+      title: String(item && item.title || '').trim(),
+      date: String(item && item.date || '').trim(),
+    })),
+  }
+  return `notice-list-v1:${stableTextFingerprint(JSON.stringify(projection))}`
+}
+
+function portalPageUrl(src, page = 1) {
+  if (page > 1 && !src.pagePath) return null
+  const path = page <= 1
+    ? src.listPath
+    : `${src.pagePath}/${src.pageNumbered ? page : page - 1}.htm`
+  return src.origin + path
+}
+
+function validatePortalListUrl(src, rawUrl, baseUrl) {
+  let parsed
+  try {
+    parsed = new URL(rawUrl, baseUrl || src.origin + src.listPath)
+  } catch (_) {
+    return null
+  }
+  if (parsed.origin !== src.origin || parsed.username || parsed.password || parsed.port) return null
+  const allowed = parsed.pathname === src.listPath
+    || (src.pagePath && parsed.pathname.startsWith(`${src.pagePath}/`) && /^\d+\.htm$/.test(parsed.pathname.slice(src.pagePath.length + 1)))
+  if (!allowed) return null
+  parsed.hash = ''
+  return parsed.toString()
+}
+
+function parsePortalPagination(html, src, pageUrl) {
+  const source = String(html || '')
+  const summary = source.match(/共\s*(\d+)\s*条(?:&nbsp;|\s)*?(\d+)\s*\/\s*(\d+)/i)
+  const page = summary ? Number(summary[2]) : null
+  const totalPages = summary ? Number(summary[3]) : null
+  const totalItems = summary ? Number(summary[1]) : null
+  let nextPageUrl = null
+  let rejectedNextUrl = null
+  const anchorRe = /<a\b[^>]*href=["']([^"']+)["'][^>]*>([\s\S]*?)<\/a>/gi
+  let match
+  while ((match = anchorRe.exec(source))) {
+    const label = match[2].replace(/<[^>]+>/g, '').replace(/&nbsp;/gi, ' ').trim()
+    if (!/^(下一页|下页)$/.test(label)) continue
+    const validated = validatePortalListUrl(src, match[1], pageUrl)
+    if (validated) nextPageUrl = validated
+    else rejectedNextUrl = match[1]
+    break
+  }
+  const terminal = !!(page && totalPages && page >= totalPages && !nextPageUrl)
+  let state = 'unknown'
+  let reason = '页面未提供可验证的页码汇总或下一页链接'
+  if (rejectedNextUrl) {
+    state = 'failed'
+    reason = '页面的下一页链接越出已注册站点或栏目边界'
+  } else if (nextPageUrl) {
+    state = 'partial'
+    reason = '页面提供了已校验的下一页链接'
+  } else if (terminal) {
+    state = 'complete'
+    reason = '页码汇总证明当前页为末页且没有下一页链接'
+  }
+  return { page, totalPages, totalItems, nextPageUrl, terminal, state, reason, rejectedNextUrl }
+}
+
 /**
- * 校园官网公开通知三大主源（T46；博达 CMS 同构，勘察实锤 2026-09-04）。
+ * 校园官网公开通知主源（T46/T55；博达 CMS 同构，勘察实锤 2026-09-04/05）。
  * 翻页：listPath=栏目首页，page N → pagePath/{N-1}.htm。
  * 学工部通知栏目实测=info/1093（数据源地图旧记 1090 实测 404）。
  */
 /**
  * 校园官网公开通知源注册表（T46 起 36 源；数据源地图 v4 全站探测 +
  * 2026-09-04 执行者逐站实勘：列表/栏目/翻页/详情容器逐站 curl 实证）。
- * 翻页：listPath 去掉 .htm 即 pagerPath（page N → {pagerPath}/{N-1}.htm）；
+ * 翻页：默认 listPath 去掉 .htm 即 pagerPath（page N → {pagerPath}/{N-1}.htm）；
+ * pageNumbered=true 的站点使用站内页码（page N → {pagerPath}/{N}.htm）。
  * pagePath 为空=该源翻页形态特殊未实证（如 jinrong），page>1 明确报错。
  * infoPaths：列表条目栏目过滤（防导航混入）；详情白名单按 origin（不限栏目）。
  * 不接：zgjjyjy（无通知栏目）、xcb（内容发新闻网）、tuanwei/grads 等
@@ -95,6 +256,8 @@ export const PORTAL_NOTICE_SOURCES = {
   sjc: { name: '审计处', origin: 'https://sjc.sdufe.edu.cn', listPath: '/gzdt.htm', pagePath: '/gzdt', infoPaths: ['1006'] },
   cwc: { name: '财务处', origin: 'https://cwc.sdufe.edu.cn', listPath: '/tzgg.htm', pagePath: '/tzgg', infoPaths: ['1024'] },
   sclx: { name: '出国留学培训基地', origin: 'https://sclx.sdufe.edu.cn', listPath: '/index/tzgg.htm', pagePath: '/index/tzgg', infoPaths: ['1078'] },
+  alumni: { name: '校友会新闻动态', origin: 'https://alumni.sdufe.edu.cn', listPath: '/index/xwdt.htm', pagePath: '/index/xwdt', pageNumbered: true, infoPaths: ['1034'] },
+  international: { name: '国际交流与合作处通知公告', origin: 'https://international.sdufe.edu.cn', listPath: '/index/tzgg.htm', pagePath: '/index/tzgg', pageNumbered: true, infoPaths: ['1024'] },
 }
 
 /**
@@ -109,6 +272,10 @@ export const PORTAL_PAGES = {
   xqbc: { name: '校车班线', url: 'https://www.sdufe.edu.cn/xyfw/xqbc.htm' },
   xydh: { name: '校园电话', url: 'https://www.sdufe.edu.cn/xyfw/xydh.htm' },
   zxxl: { name: '校历', url: 'https://www.sdufe.edu.cn/xyfw/zxxl.htm' },
+  // T56 主站学校概况具名页：逐页实勘 2026-09-05，正文走 parsePortalPage。
+  lrld: { name: '历任领导', url: 'https://www.sdufe.edu.cn/xxgk_/lrld.htm' },
+  cdbs: { name: '财大标识', url: 'https://www.sdufe.edu.cn/xxgk_/cdbs.htm' },
+  cdzc: { name: '财大章程', url: 'https://www.sdufe.edu.cn/xxgk_/cdzc.htm' },
   // T47 学院简介/领导页（16 站 24 key，照单注册清单 docs/plan/crawl/t47-page-key-registry.md
   // ——地图 v9+批次 8 全量遍历双向验证；命名 {学院缩写}-xyjj=学院简介 / -ldr=学院领导）。
   // 领导页为表格版式（姓名/职务/分工三列）无标准博达容器，走 parsePortalPage body 兜底。
@@ -331,14 +498,28 @@ export function createSdufeAdapter(env) {
       return parseScheduleHtml(res.text || '')
     },
 
-    /** 抓成绩：term 空 = 全部学期 */
-    async fetchScores(cookie, { term = '' } = {}) {
+    /** 抓成绩及范围证明：term 空 = 全部学期；旧 fetchScores 仍只返回 rows[]。 */
+    async fetchScoresWithMeta(cookie, { term = '' } = {}) {
       const form = `kksj=${encodeURIComponent(term)}&xsfs=all`
-      const res = await this.request(`${this.baseUrl}/jsxsd/kscj/cjcx_list`, 'POST', {
+      const endpoint = '/jsxsd/kscj/cjcx_list'
+      const res = await this.request(`${this.baseUrl}${endpoint}`, 'POST', {
         body: form,
         cookie,
       })
-      return parseScoresHtml(res.text || '')
+      const parsed = parseScoresHtmlWithMeta(res.text || '')
+      const rows = parsed.rows
+      const actualTerms = [...new Set(rows.map((row) => String(row && row.term || '').trim()).filter(Boolean))].sort()
+      const requestedTerm = String(term || '').trim()
+      const mismatchedTerms = requestedTerm ? actualTerms.filter((actual) => actual !== requestedTerm) : []
+      return { data: rows, meta: personalReadMeta({ endpoint, method: 'POST', status: res.status,
+        scope: { term: requestedTerm || 'all', actualTerms, termMatch: requestedTerm ? mismatchedTerms.length === 0 : true, scoreDisplay: 'all' }, rows, parse: parsed.meta,
+        requiredFields: ['term', 'courseCode', 'courseName', 'scoreText', 'credit', 'nature'],
+        rangeIssue: mismatchedTerms.length ? { state: 'partial', reason: `请求学期 ${requestedTerm}，但实际返回包含 ${mismatchedTerms.join('、')}，不能证明指定学期范围已满足` } : null,
+        rangeReason: term ? `已用 kksj=${term} 且 xsfs=all 读取指定学期成绩页；解析零行会失败` : '已用空 kksj 且 xsfs=all 读取全部学期成绩页；解析零行会失败' }) }
+    },
+
+    async fetchScores(cookie, { term = '' } = {}) {
+      return (await this.fetchScoresWithMeta(cookie, { term })).data
     },
 
     /** 抓学分修读（通选课统计，GET 直出） */
@@ -370,16 +551,36 @@ export function createSdufeAdapter(env) {
       return parseCoursesHtml(res.text || '')
     },
 
-    /** 平均学分绩点（含辅修行，GET 直出） */
+    /** 平均学分绩点（含主修/辅修统计行）；旧 fetchGpa 返回形态不变。 */
+    async fetchGpaWithMeta(cookie) {
+      const endpoint = '/jsxsd/kscj/cjcx_avg'
+      const res = await this.request(`${this.baseUrl}${endpoint}`, 'GET', { cookie })
+      const parsed = parseGpaHtmlWithMeta(res.text || '')
+      const data = parsed.data
+      return { data, meta: personalReadMeta({ endpoint, method: 'GET', status: res.status,
+        scope: { rows: 'major_and_minor_statistics' }, rows: data.rows, parse: parsed.meta,
+        requiredFields: ['studentId', 'name', 'major', 'className', 'level', 'totalCredits', 'courseCount', 'averageScore', 'averageGrade', 'gpa', 'majorType'],
+        rangeReason: '已读取平均学分绩点整表；主修与页面返回的全部辅修统计行均保留，未找到表或零行会失败' }) }
+    },
+
     async fetchGpa(cookie) {
-      const res = await this.request(`${this.baseUrl}/jsxsd/kscj/cjcx_avg`, 'GET', { cookie })
-      return parseGpaHtml(res.text || '')
+      return (await this.fetchGpaWithMeta(cookie)).data
+    },
+
+    /** 学籍卡片及核心字段证明；旧 fetchXj 返回形态不变。 */
+    async fetchXjWithMeta(cookie, { full = false } = {}) {
+      const endpoint = '/jsxsd/grxx/xsxx'
+      const res = await this.request(`${this.baseUrl}${endpoint}`, 'GET', { cookie })
+      const parsed = parseXjHtmlWithMeta(res.text || '', { full })
+      const data = parsed.data
+      return { data, meta: personalReadMeta({ endpoint, method: 'GET', status: res.status,
+        scope: { card: 'logged_in_student', fields: full ? 'core_plus_non_sensitive_whitelist' : 'core_only' }, rows: [data], parse: parsed.meta,
+        requiredFields: ['studentId', 'department', 'major', 'duration', 'className', 'level', 'grade'],
+        rangeReason: '已读取本机登录学生学籍卡片；学号用于确认卡片有效但宿主可继续裁剪，敏感字段不进入返回' }) }
     },
 
-    /** 学籍卡片（默认裁剪敏感字段，full=true 输出白名单内扩展字段） */
     async fetchXj(cookie, { full = false } = {}) {
-      const res = await this.request(`${this.baseUrl}/jsxsd/grxx/xsxx`, 'GET', { cookie })
-      return parseXjHtml(res.text || '', { full })
+      return (await this.fetchXjWithMeta(cookie, { full })).data
     },
 
     /** 培养执行计划（GET 直出，逐学期课程列表） */
@@ -494,30 +695,47 @@ export function createSdufeAdapter(env) {
     },
 
     /**
-     * 校园官网公开通知源列表（T46 三源起步，T46 扩展 36 源；博达 CMS 同构
+     * 校园官网公开通知源列表（T46/T55，博达 CMS 同构
      * 一套解析，逐站实勘配置见 PORTAL_NOTICE_SOURCES）。
      * 翻页口径：page 1=listPath，page N={pagePath}/{N-1}.htm；
      * pagePath 为空=该源翻页未实证（page>1 明确报错，不静默回首页）。
      * 礼貌纪律：只抓请求的页，不做归档批量预取。
      */
     async fetchPortalNotices(site, page = 1) {
+      const result = await this.fetchPortalNoticePage(site, { page })
+      return result.items
+    },
+
+    /**
+     * 校园官网公开通知的单页事实合同。与 fetchPortalNotices 的数组返回并存：
+     * 新调用方可沿页面真实“下页”链接续读，并获得页码、末页与内容指纹证据。
+     */
+    async fetchPortalNoticePage(site, { page = 1, pageUrl = '' } = {}) {
       const src = PORTAL_NOTICE_SOURCES[site]
       if (!src) {
         throw new LinkeError('BAD_SOURCE', `未知通知源：${site}`, {
           exitCode: EXIT.GENERAL,
-          hint: 'source 取源注册表 key（jwc=教务处 / main=主站 / xgb=学工部 / lib=图书馆 / yjsy=研究生院 / news=新闻网 / 各学院与职能部门拼音缩写）',
+          hint: 'source 取源注册表 key（jwc=教务处 / alumni=校友会新闻动态 / international=国际交流与合作处通知公告 / main=主站 / xgb=学工部 / lib=图书馆 / yjsy=研究生院 / news=新闻网 / 各学院与职能部门拼音缩写）',
         })
       }
-      if (page > 1 && !src.pagePath) {
+      if (!pageUrl && page > 1 && !src.pagePath) {
         throw new LinkeError('NO_PAGER', `${src.name}暂不支持翻页（该站翻页形态特殊未实证）`, {
           exitCode: EXIT.GENERAL,
           hint: '本源仅可查第一页（page=1）',
         })
       }
-      const path = page <= 1 ? src.listPath : `${src.pagePath}/${page - 1}.htm`
+      const requestedUrl = pageUrl
+        ? validatePortalListUrl(src, pageUrl)
+        : portalPageUrl(src, page)
+      if (!requestedUrl) {
+        throw new LinkeError('BAD_PAGER_URL', `${src.name}分页地址不在已注册栏目边界`, {
+          exitCode: EXIT.GENERAL,
+          hint: '只能使用 fetchPortalNoticePage 上一次返回的 nextPageUrl',
+        })
+      }
       let response
       try {
-        response = await env.fetch(src.origin + path, {
+        response = await env.fetch(requestedUrl, {
           headers: { 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) linke-cli' },
           redirect: 'follow',
           timeoutMs: 15000,
@@ -528,10 +746,33 @@ export function createSdufeAdapter(env) {
       if (!response.ok) {
         throw networkError(`抓取${src.name}公告页（HTTP ${response.status}）`, null)
       }
+      const finalUrl = response.url ? validatePortalListUrl(src, response.url) : requestedUrl
+      if (!finalUrl) {
+        throw new LinkeError('BAD_REDIRECT', `${src.name}公告页重定向越出已注册栏目边界`, {
+          exitCode: EXIT.GENERAL,
+        })
+      }
+      const html = await response.text()
       // 栏目过滤：列表页含导航/侧栏的 info 链接（如主站「学校概况」），
       // 只保留本源通知栏目条目
-      return parsePortalNotices(await response.text(), src.origin)
+      const items = parsePortalNotices(html, src.origin)
         .filter((x) => !src.infoPaths || !src.infoPaths.length || src.infoPaths.some((c) => x.url.includes(`/info/${c}/`)))
+        .map((x) => ({ ...x, dateSource: x.date ? 'list_page' : 'unavailable' }))
+      const pagination = parsePortalPagination(html, src, finalUrl)
+      const continuationFingerprint = portalNoticeContinuationFingerprint({ site, finalUrl, items, pagination })
+      return {
+        site,
+        sourceName: src.name,
+        requestedUrl,
+        finalUrl,
+        redirectVerified: !!response.url,
+        fetchedAt: new Date().toISOString(),
+        contentFingerprint: stableTextFingerprint(html),
+        continuationFingerprintVersion: 1,
+        continuationFingerprint,
+        items,
+        pagination,
+      }
     },
 
     /**
@@ -544,7 +785,7 @@ export function createSdufeAdapter(env) {
       if (!page) {
         throw new LinkeError('BAD_PAGE', `未知具名页面：${pageKey}（可选：${Object.keys(PORTAL_PAGES).join('/')}）`, {
           exitCode: EXIT.GENERAL,
-          hint: '学校级：xxjj=学校简介 / xrld=现任领导 / zuzjg=组织机构 / xqbc=校车班线 / xydh=校园电话 / zxxl=校历；学院级（T47）：{学院缩写}-xyjj=学院简介 / {学院缩写}-ldr=学院领导（如 gkgc-ldr=管科学院领导）',
+          hint: '学校级：xxjj=学校简介 / xrld=现任领导 / lrld=历任领导 / zuzjg=组织机构 / cdbs=财大标识 / cdzc=财大章程 / xqbc=校车班线 / xydh=校园电话 / zxxl=校历；学院级（T47）：{学院缩写}-xyjj=学院简介 / {学院缩写}-ldr=学院领导（如 gkgc-ldr=管科学院领导）',
         })
       }
       let response
@@ -587,7 +828,7 @@ export function createSdufeAdapter(env) {
       if (!src || !/^\/info\/\d+\/\d+\.htm$/.test(parsed.pathname)) {
         throw new LinkeError('BAD_URL', `URL 不在校内通知源范围：${url}`, {
           exitCode: EXIT.GENERAL,
-          hint: `仅支持通知源注册表内各站（jwc/main/xgb/lib/yjsy/news/各学院等）的 info/{栏目}/{id}.htm 详情页`,
+          hint: `仅支持通知源注册表内各站（jwc/alumni/international/main/xgb/lib/yjsy/news/各学院等）的 info/{栏目}/{id}.htm 详情页`,
         })
       }
       let response

package/src/parsers.js

@@ -146,7 +146,7 @@ export function findTableByHeaders(html, mustInclude) {
  * T10：补捕获学分列（真实页列序 序号/学期/编号/名称/成绩/学分/绩点/
  * 考试性质/课程性质/课程属性/辅修——学分紧跟成绩双闭合单元格后）。
  */
-export function parseScoresHtml(html) {
+function parseScoresRows(html) {
   if (isJwLoginExpired(html)) {
     const err = new Error('jw login expired')
     err.isJwLoginExpired = true
@@ -180,7 +180,7 @@ export function parseScoresHtml(html) {
   // 变体一：行首带序号列，第二列学期（现役主口径，PHP matchesWithLeading）
   const withLeading = Array.from(
     cleaned.matchAll(
-      /<tr><td>.*?<\/td><td>(.*?)<\/td><tdalign=.*?>(.*?)<\/td><tdalign=.*?>(.*?)<\/td><!--控制成绩显示--><tdstyle=.*?><ahref=.*?>(.*?)<\/a><\/td><\/td><td>(.*?)<\/td><!--控制绩点显示--><td>.*?<\/td><td>.*?<\/td><td>(.*?)<\/td><td>.*?<\/td><td>.*?<\/td><\/tr>/g
+      /<tr[^>]*><td>.*?<\/td><td>(.*?)<\/td><tdalign=.*?>(.*?)<\/td><tdalign=.*?>(.*?)<\/td><!--控制成绩显示--><tdstyle=.*?><ahref=.*?>(.*?)<\/a><\/td><\/td><td>(.*?)<\/td><!--控制绩点显示--><td>.*?<\/td><td>.*?<\/td><td>(.*?)<\/td><td>.*?<\/td><td>.*?<\/td><\/tr>/g
     )
   )
   for (const m of withLeading) {
@@ -193,6 +193,28 @@ export function parseScoresHtml(html) {
     }
   }
 
+  // 兼容同一成绩表中缺少“控制成绩显示”注释、但单元格结构仍完整的记录。
+  // 注释只是前端展示实现，不能同时充当候选记录清单和解析成功条件。
+  for (const candidate of scoreCandidateRows(html)) {
+    if (/控制成绩显示/.test(candidate.html)) continue
+    const cells = candidate.cells
+    const termIndex = cells.findIndex((cell) => TERM_RE.test(cell.text))
+    if (termIndex < 0 || cells.length < termIndex + 5) continue
+    const scoreCell = cells[termIndex + 3]
+    // 无表头兼容页仍要求成绩格保留链接结构，避免把普通布局表误认成成绩表。
+    if (!candidate.headerRecognized && !/<a\b/i.test(scoreCell.raw)) continue
+    const before = rows.length
+    addRow(
+      cells[termIndex].text,
+      cells[termIndex + 1].text,
+      cells[termIndex + 2].text,
+      cells[termIndex + 4].text,
+      scoreCell.text,
+      cells[termIndex + 7] ? cells[termIndex + 7].text : ''
+    )
+    if (rows.length === before) continue
+  }
+
   // 变体二：legacy 无前导学期列（PHP matchesLegacy，仅在变体一整页零命中时启用）
   if (rows.length === 0) {
     const legacy = Array.from(
@@ -211,6 +233,142 @@ export function parseScoresHtml(html) {
   return rows
 }
 
+function isDisabledPaginationControl(attrsValue) {
+  const attrs = String(attrsValue || '')
+  const nativeDisabled = /(?:^|\s)disabled(?:\s*=\s*(?:"[^"]*"|'[^']*'|[^\s]+))?(?=\s|$)/i.test(attrs)
+  const ariaDisabled = /(?:^|\s)aria-disabled\s*=\s*(?:"true"|'true'|true)(?=\s|$)/i.test(attrs)
+  const classValue = (attrs.match(/(?:^|\s)class\s*=\s*(?:"([^"]*)"|'([^']*)'|([^\s]+))/i) || []).slice(1).find((value) => value != null) || ''
+  const classDisabled = classValue.toLowerCase().split(/\s+/).includes('disabled')
+  return nativeDisabled || ariaDisabled || classDisabled
+}
+
+function paginationEvidence(html, candidateRowCount, label = '成绩') {
+  const source = String(html || '')
+  const visible = cleanCell(source)
+  const totalMatch = visible.match(/共\s*(\d+)\s*页/)
+  const totalPages = totalMatch ? Number(totalMatch[1]) : null
+  const currentMatch = visible.match(/第\s*(\d+)\s*页/)
+  const currentPage = currentMatch ? Number(currentMatch[1]) : null
+  const totalRowsMatch = visible.match(/共\s*(\d+)\s*(?:条|条记录|条数据)/)
+  const totalRows = totalRowsMatch ? Number(totalRowsMatch[1]) : null
+  const next = Array.from(source.matchAll(/<a\b([^>]*)>([\s\S]*?)<\/a>/gi))
+    .map((m) => {
+      const attrs = String(m[1] || '')
+      const href = (attrs.match(/\bhref=["']([^"']+)["']/i) || [])[1] || ''
+      const disabled = isDisabledPaginationControl(attrs)
+      return { href, text: cleanCell(m[2]), disabled }
+    })
+    .find((row) => /下一页|下页|next/i.test(row.text) && !row.disabled && row.href && row.href !== '#' && !/^javascript:/i.test(row.href))
+  const nextControls = Array.from(source.matchAll(/<(a|button|input)\b([^>]*)>([\s\S]*?)<\/\1>|<input\b([^>]*)\/?>/gi))
+    .map((m) => {
+      const attrs = String(m[2] || m[4] || '')
+      const text = cleanCell(m[3] || (attrs.match(/\bvalue=["']([^"']*)["']/i) || [])[1] || '')
+      const disabled = isDisabledPaginationControl(attrs)
+      return { text, disabled }
+    })
+    .filter((control) => /下一页|下页|next/i.test(control.text))
+  const enabledNextControl = nextControls.some((control) => !control.disabled)
+  const hasMoreByPageCount = Number.isInteger(totalPages) && totalPages > 1
+  const hasMoreByRowCount = Number.isInteger(totalRows) && totalRows > candidateRowCount
+  const contradictoryRowCount = Number.isInteger(totalRows) && totalRows < candidateRowCount
+  const hasMore = !!next || enabledNextControl || hasMoreByPageCount || hasMoreByRowCount
+  const ambiguousNextText = /下一页|下页/i.test(visible) && !next && nextControls.length === 0
+  const state = hasMore ? 'partial' : contradictoryRowCount || ambiguousNextText ? 'unknown' : 'complete'
+  return {
+    detected: totalPages != null || totalRows != null || /下一页|下页/i.test(source),
+    state,
+    currentPage,
+    totalPages,
+    totalRows,
+    nextPageUrl: next ? next.href : null,
+    reason: hasMore
+      ? `${label}响应明示多页、可用下一页控件或未交付记录；单次响应不能证明此前页与后续页均已读取`
+      : contradictoryRowCount
+        ? `${label}响应宣称共 ${totalRows} 条，但识别到 ${candidateRowCount} 条候选记录，来源计数矛盾`
+        : ambiguousNextText
+          ? `${label}响应出现分页提示，但无法判断下一页控件是否可用`
+          : `已识别${label}结果行，响应未显示未读分页或计数缺口`
+  }
+}
+
+function scoreCandidateRows(html) {
+  const candidates = []
+  for (const tableMatch of String(html || '').matchAll(/<table\b[\s\S]*?<\/table>/gi)) {
+    const tableHtml = tableMatch[0]
+    const headers = Array.from(tableHtml.matchAll(/<th\b[^>]*>([\s\S]*?)<\/th>/gi)).map((m) => cleanCell(m[1]))
+    const headerRecognized = headers.some((h) => h.includes('学期'))
+      && headers.some((h) => h.includes('课程'))
+      && headers.some((h) => h.includes('成绩'))
+    for (const rowMatch of tableHtml.matchAll(/<tr\b[^>]*>([\s\S]*?)<\/tr>/gi)) {
+      const rowHtml = rowMatch[0]
+      if (/<th\b/i.test(rowHtml)) continue
+      const cells = Array.from(rowMatch[1].matchAll(/<td\b[^>]*>([\s\S]*?)<\/td>/gi))
+        .map((cell) => ({ raw: cell[1], text: cleanCell(cell[1]) }))
+      const termIndex = cells.findIndex((cell) => TERM_RE.test(cell.text))
+      const legacyShape = termIndex >= 0 && termIndex <= 1 && cells.length >= termIndex + 8
+      if (headerRecognized || legacyShape || /控制成绩显示/.test(rowHtml)) {
+        candidates.push({ html: rowHtml, cells, headerRecognized })
+      }
+    }
+  }
+  return candidates
+}
+
+function scorePaginationEvidence(html, candidateRowCount) {
+  return paginationEvidence(html, candidateRowCount, '成绩')
+}
+
+function classifySkippedScoreRow(rowHtml) {
+  const cleaned = stripSpaces(rowHtml)
+  const scoreMatch = cleaned.match(/<!--控制成绩显示--><tdstyle=.*?><ahref=.*?>(.*?)<\/a>/)
+  const scoreText = scoreMatch ? cleanCell(scoreMatch[1]) : ''
+  if (INVALID_SCORE_TEXTS.has(scoreText)) return 'score_not_published_placeholder'
+  if (!scoreMatch) return 'score_cell_unrecognized'
+  if (scoreText === '') return 'score_value_empty'
+  if (/^\d+(?:\.\d+)?$/.test(scoreText) && (Number(scoreText) < 0 || Number(scoreText) > 100)) {
+    return 'score_value_out_of_range'
+  }
+  const visible = cleanCell(rowHtml)
+  if (!TERM_RE.test((visible.match(/\d{4}-\d{4}-\d/) || [])[0] || '')) return 'term_unrecognized'
+  return 'record_structure_or_value_unparsed'
+}
+
+/** 兼容 metadata 入口；旧 parseScoresHtml 仍只返回 rows[]。 */
+export function parseScoresHtmlWithMeta(html) {
+  const source = String(html || '')
+  const candidateRows = scoreCandidateRows(source).map((candidate) => candidate.html)
+  const rows = parseScoresRows(html)
+  const skippedReasons = {}
+  let parsedCandidates = 0
+  for (const candidate of candidateRows) {
+    try {
+      parsedCandidates += parseScoresRows(`<table>${candidate}</table>`).length
+    } catch {
+      const reason = classifySkippedScoreRow(candidate)
+      skippedReasons[reason] = (skippedReasons[reason] || 0) + 1
+    }
+  }
+  const skippedRowCount = Math.max(0, candidateRows.length - parsedCandidates)
+  const ignoredPlaceholderCount = Number(skippedReasons.score_not_published_placeholder || 0)
+  const unresolvedSkippedRowCount = Math.max(0, skippedRowCount - ignoredPlaceholderCount)
+  return {
+    rows,
+    meta: {
+      tableRecognized: candidateRows.length > 0,
+      candidateRowCount: candidateRows.length,
+      parsedRowCount: rows.length,
+      skippedRowCount,
+      unresolvedSkippedRowCount,
+      skippedReasons,
+      pagination: scorePaginationEvidence(html, candidateRows.length)
+    }
+  }
+}
+
+export function parseScoresHtml(html) {
+  return parseScoresHtmlWithMeta(html).rows
+}
+
 /**
  * 解析学分修读页（/jsxsd/xxwcqk/xstxkxdqk.do，GET 直出）。
  * 真实结构双表：
@@ -450,6 +608,50 @@ export function parseXjHtml(html, { full = false } = {}) {
   return result
 }
 
+/** GPA 生产者范围摘要：以实际命中的表头、候选数据行和解析行生成。 */
+export function parseGpaHtmlWithMeta(html) {
+  const table = findTableByHeaders(html, ['学号', '平均学分绩点'])
+  const data = parseGpaHtml(html)
+  const candidateRowCount = table ? table.rows.filter((row) => row.some((value) => String(value || '').trim())).length : 0
+  const skippedRowCount = Math.max(0, candidateRowCount - data.rows.length)
+  return {
+    data,
+    meta: {
+      tableRecognized: !!table,
+      candidateRowCount,
+      parsedRowCount: data.rows.length,
+      skippedRowCount,
+      unresolvedSkippedRowCount: skippedRowCount,
+      skippedReasons: skippedRowCount ? { empty_or_unrecognized_gpa_row: skippedRowCount } : {},
+      pagination: (() => {
+        const pagination = paginationEvidence(html, candidateRowCount, '平均学分绩点')
+        return pagination.detected
+          ? pagination
+          : { ...pagination, state: 'not_applicable', reason: '平均学分绩点响应未显示分页控件或页数信息' }
+      })()
+    }
+  }
+}
+
+/** 学籍生产者范围摘要：以实际识别的登录学生卡片和核心字段生成。 */
+export function parseXjHtmlWithMeta(html, { full = false } = {}) {
+  const data = parseXjHtml(html, { full })
+  const labels = Array.from(String(html || '').matchAll(/(?:^|>)(学号|院系|专业|学制|班级|学习层次)[：:]?/g)).map((m) => m[1])
+  return {
+    data,
+    meta: {
+      cardRecognized: !!data.studentId,
+      candidateRowCount: 1,
+      parsedRowCount: data.studentId ? 1 : 0,
+      skippedRowCount: 0,
+      unresolvedSkippedRowCount: 0,
+      skippedReasons: {},
+      recognizedLabels: [...new Set(labels)],
+      pagination: { detected: false, state: 'not_applicable', nextPageUrl: null, reason: '学籍端点返回当前登录学生单卡片' }
+    }
+  }
+}
+
 /**
  * 解析培养执行计划页（/jsxsd/pyfa/pyfa_query，GET 直出）。
  * 真实结构单表 11 列：序号/开课学期/课程编号/课程名称/开课单位/学分/
@@ -943,7 +1145,7 @@ function extractVNewsContent(html) {
 /**
  * 博达 CMS 通知详情页通用解析（T46 三源通吃）。
  * → { title, date, content, attachments: [{ name, url }] }
- * - 标题：h1 优先，fallback title（剥站名后缀）
+ * - 标题：h1/h2/h3 优先，fallback title（剥站名后缀）
  * - 日期：arttime（YY-MM-DD 补 20 前缀）/ 全页 YYYY-MM-DD / 发布时间元素
  * - 正文：.v_news_content 结构化文本（表格转 | 分列行）
  * - 附件：正文内 a[href 含 .doc/.pdf//system/_ 等] 列表化（不下载）
@@ -954,9 +1156,18 @@ export function parsePortalNoticeDetail(html, pageUrl = '') {
     throw parseError('通知详情正文（.v_news_content 容器缺失——页面可能已改版或非通知页）')
   }
   let title = ''
-  const h1 = html.match(/<h1[^>]*>([\s\S]*?)<\/h1>/)
-  if (h1) {
-    title = decodeEntities(h1[1].replace(/<[^>]+>/g, '').trim())
+  const headingRe = /<h[123][^>]*>([\s\S]*?)<\/h[123]>/gi
+  let heading
+  let headingMatch
+  while ((headingMatch = headingRe.exec(html))) {
+    const candidate = decodeEntities(headingMatch[1].replace(/<[^>]+>/g, '').trim())
+    if (candidate) {
+      heading = candidate
+      break
+    }
+  }
+  if (heading) {
+    title = heading
   } else {
     const t = html.match(/<title>([^<]*)<\/title>/)
     title = t ? decodeEntities(t[1].replace(/-[^-]*$/, '').trim()) : ''
@@ -968,6 +1179,10 @@ export function parsePortalNoticeDetail(html, pageUrl = '') {
     const pub = html.match(/(?:发布时间|发布日期)[：:\s]*(\d{4}-\d{2}-\d{2})/)
     if (pub) date = pub[1]
   }
+  if (!date) {
+    const pubCn = html.match(/(?:发布时间|发布日期)[：:\s]*(\d{4})年\s*(\d{1,2})月\s*(\d{1,2})日/)
+    if (pubCn) date = `${pubCn[1]}-${String(pubCn[2]).padStart(2, '0')}-${String(pubCn[3]).padStart(2, '0')}`
+  }
   if (!date) date = (html.match(/\d{4}-\d{2}-\d{2}/) || [])[0] || ''
 
   const inner = extractVNewsContent(html)

package/src/registry.js

@@ -27,7 +27,8 @@ export function initSdufe(env = nodeEnv()) {
  *   probeSession(cookie) → userInfo（过期抛 err.isJwLoginExpired）
  *   fetchCurrentTerm(cookie) → string | null
  *   fetchSchedule(cookie, { term, week }) → { weeks, remark? }
- *   fetchScores(cookie, { term }) → rows[]
+ *   fetchScores(cookie, { term }) → rows[]（兼容）；fetchScoresWithMeta → {data,meta}
+ *   fetchGpa/fetchXj 保持旧返回；对应 WithMeta 方法返回读取范围与字段证明
  */
 export function getAdapter(schoolId) {
   const adapter = adapters.get(schoolId)