yaohao 1.0.0

package/src/source/beijing/crawl.js

@@ -0,0 +1,156 @@
+// 北京交通委公告站抓取层：负责 HTTP 请求、编码处理、本地缓存。
+// 零外部依赖，依赖 Node 18+ 自带的 fetch。
+
+import fs from 'node:fs/promises';
+import path from 'node:path';
+import crypto from 'node:crypto';
+import { homedir } from 'node:os';
+
+const DEFAULT_HEADERS = {
+  'User-Agent':
+    'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 ' +
+    '(KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36',
+  Accept:
+    'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8',
+  'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8',
+};
+
+const CACHE_DIR = path.join(homedir(), '.yaohao', 'cache');
+
+/**
+ * 抓取一个 URL，返回 { url, status, html, contentType, fromCache, elapsedMs }
+ *
+ * @param {string} url
+ * @param {object} [opts]
+ * @param {boolean} [opts.useCache=true]  本地缓存，避免反复打官网
+ * @param {number}  [opts.timeoutMs=15000]
+ * @param {number}  [opts.retries=2]
+ * @param {number}  [opts.maxCacheAgeMs] 缓存最大有效期（毫秒），默认无限
+ */
+export async function fetchHtml(url, opts = {}) {
+  const { useCache = true, timeoutMs = 15000, retries = 2, maxCacheAgeMs } = opts;
+
+  if (useCache) {
+    const cached = await readCache(url, maxCacheAgeMs);
+    if (cached) {
+      return { ...cached, fromCache: true };
+    }
+  }
+
+  let lastErr = null;
+  for (let attempt = 0; attempt <= retries; attempt++) {
+    const t0 = Date.now();
+    const ac = new AbortController();
+    const timer = setTimeout(() => ac.abort(), timeoutMs);
+    try {
+      const res = await fetch(url, {
+        headers: DEFAULT_HEADERS,
+        signal: ac.signal,
+        redirect: 'follow',
+      });
+      const contentType = res.headers.get('content-type') || '';
+      const buf = Buffer.from(await res.arrayBuffer());
+      const html = decode(buf, contentType);
+      const elapsedMs = Date.now() - t0;
+      const payload = {
+        url,
+        status: res.status,
+        contentType,
+        html,
+        elapsedMs,
+        fromCache: false,
+        savedAt: Date.now(),
+      };
+      if (useCache && res.status === 200) {
+        await writeCache(url, payload);
+      }
+      return payload;
+    } catch (err) {
+      lastErr = err;
+      if (attempt < retries) {
+        await sleep(500 * (attempt + 1));
+      }
+    } finally {
+      clearTimeout(timer);
+    }
+  }
+  throw new Error(`fetchHtml(${url}) failed after ${retries + 1} tries: ${lastErr?.message || lastErr}`);
+}
+
+/**
+ * 抓取一个二进制资源（PDF 等），返回 { url, status, buffer }。
+ * 走单独缓存目录 ~/.yaohao/cache/pdf/，按 url sha1 命名。
+ */
+export async function fetchBinary(url, opts = {}) {
+  const { useCache = true, timeoutMs = 30000 } = opts;
+  const pdfDir = path.join(CACHE_DIR, 'binary');
+  await fs.mkdir(pdfDir, { recursive: true });
+  const file = path.join(pdfDir, crypto.createHash('sha1').update(url).digest('hex') + path.extname(url));
+
+  if (useCache) {
+    try {
+      const buf = await fs.readFile(file);
+      return { url, status: 200, buffer: buf, fromCache: true };
+    } catch {
+      /* miss */
+    }
+  }
+
+  const ac = new AbortController();
+  const timer = setTimeout(() => ac.abort(), timeoutMs);
+  try {
+    const res = await fetch(url, { headers: DEFAULT_HEADERS, signal: ac.signal });
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const buf = Buffer.from(await res.arrayBuffer());
+    await fs.writeFile(file, buf);
+    return { url, status: res.status, buffer: buf, fromCache: false };
+  } finally {
+    clearTimeout(timer);
+  }
+}
+
+function decode(buf, contentType) {
+  let charset = 'utf-8';
+  const m = /charset=([\w-]+)/i.exec(contentType);
+  if (m) charset = m[1].toLowerCase();
+  const head = buf.slice(0, 1024).toString('ascii');
+  const metaMatch = /charset=["']?([\w-]+)/i.exec(head);
+  if (metaMatch) charset = metaMatch[1].toLowerCase();
+
+  if (charset === 'utf-8' || charset === 'utf8') {
+    return buf.toString('utf8');
+  }
+  try {
+    return new TextDecoder(charset).decode(buf);
+  } catch {
+    return buf.toString('utf8');
+  }
+}
+
+function sleep(ms) {
+  return new Promise((r) => setTimeout(r, ms));
+}
+
+function cacheKey(url) {
+  return crypto.createHash('sha1').update(url).digest('hex') + '.json';
+}
+
+async function readCache(url, maxAgeMs) {
+  try {
+    const file = path.join(CACHE_DIR, cacheKey(url));
+    const raw = await fs.readFile(file, 'utf8');
+    const payload = JSON.parse(raw);
+    if (maxAgeMs && payload.savedAt && Date.now() - payload.savedAt > maxAgeMs) {
+      return null;
+    }
+    return payload;
+  } catch {
+    return null;
+  }
+}
+
+async function writeCache(url, payload) {
+  await fs.mkdir(CACHE_DIR, { recursive: true });
+  const file = path.join(CACHE_DIR, cacheKey(url));
+  await fs.writeFile(file, JSON.stringify(payload), 'utf8');
+}

package/src/source/beijing/eligibility.js

@@ -0,0 +1,110 @@
+import { input, select, confirm } from '@inquirer/prompts';
+
+// 资格判定基于《北京市小客车数量调控暂行规定》（2025 年修订）
+// 4 种身份类型代码取自官方系统 choosePerson 页：BSHJ / JJ / ZZZ / GZJZZ
+
+const PERSON_TYPES = {
+  BSHJ: '本市户籍居民',
+  JJ: '驻京部队现役军人和现役武警',
+  ZZZ: '持有效居住证的非本市户籍人员',
+  GZJZZ: '持北京市工作居住证人员',
+};
+
+export async function checkEligibility() {
+  try {
+    const personType = await select({
+      message: '户籍/居住类型？',
+      choices: [
+        ...Object.entries(PERSON_TYPES).map(([k, v]) => ({ name: v, value: k })),
+        { name: '以上都不是', value: 'NONE' },
+      ],
+    });
+    if (personType === 'NONE') {
+      return result(false, '不符合户籍/居住要求', { reason: 'huji_not_match' });
+    }
+
+    const age = Number(await input({
+      message: '年龄？',
+      default: '30',
+      validate: (v) => /^\d+$/.test(v) && Number(v) > 0 || '请输入正整数',
+    }));
+    if (age < 18) return result(false, '需年满 18 周岁', { age });
+
+    const hasLicense = await confirm({ message: '是否持有有效机动车驾驶证？', default: true });
+    if (!hasLicense) return result(false, '需持有有效机动车驾驶证');
+
+    const hasBjPlate = await confirm({ message: '名下是否已有京牌小客车？', default: false });
+    if (hasBjPlate) return result(false, '名下已有京牌小客车，须先处理');
+
+    const hasIndicator = await confirm({ message: '名下是否已持有有效的小客车指标（含指标确认通知书）？', default: false });
+    if (hasIndicator) return result(false, '已持有有效指标，不能重复申请');
+
+    // 非京籍走 5 年社保 + 个税
+    if (personType === 'ZZZ') {
+      const sbYears = Number(await input({
+        message: '在京连续缴纳社保和个税年数（按整年算，断缴会清零）？',
+        default: '0',
+        validate: (v) => /^\d+(\.\d+)?$/.test(v) || '请输入数字',
+      }));
+      if (sbYears < 5) {
+        return result(false, `非京籍连续社保和个税不足 5 年，还差 ${(5 - sbYears).toFixed(1)} 年`, { sbYears });
+      }
+    }
+
+    // 家庭摇号判断
+    const considerFamily = await confirm({
+      message: '是否考虑家庭摇号（积分制，比个人摇号中签率高）？',
+      default: true,
+    });
+    let familyEligible = false;
+    let familySize = 1;
+    if (considerFamily) {
+      familySize = Number(await input({
+        message: '家庭主申请人 + 配偶 + 子女 + 双方父母合计有效成员数？',
+        default: '2',
+        validate: (v) => /^\d+$/.test(v) || '请输入整数',
+      }));
+      familyEligible = familySize >= 2;
+    }
+
+    const lines = ['符合申请条件：'];
+    lines.push('  ✓ 个人普通指标（PTC）— 阶梯倍率制，参与年限越长倍率越高');
+    lines.push('  ✓ 个人新能源指标（XNY）— 排队轮候制');
+    if (familyEligible) {
+      lines.push('  ✓ 家庭普通指标（PTC）— 家庭积分制，比个人中签率高');
+      lines.push('  ✓ 家庭新能源指标（XNY）— 家庭积分轮候');
+    } else if (considerFamily) {
+      lines.push('  ✗ 家庭摇号：需至少 2 名家庭成员');
+    }
+    lines.push('');
+    lines.push('下一步：在申请窗口期（每年 1/1-3/8、8/1-10/8）登录 xkczb.jtw.beijing.gov.cn 提交申请。');
+    lines.push('提示：本判定为简化版，复杂情况以官方说明为准。');
+
+    return result(true, lines.join('\n'), {
+      personType,
+      personTypeLabel: PERSON_TYPES[personType],
+      age,
+      familySize,
+      eligibility: {
+        personal_ptc: true,
+        personal_xny: true,
+        family_ptc: familyEligible,
+        family_xny: familyEligible,
+      },
+    });
+  } catch (err) {
+    if (err && err.name === 'ExitPromptError') {
+      return { pass: null, cancelled: true, lines: ['已取消'] };
+    }
+    throw err;
+  }
+}
+
+function result(pass, message, extra = {}) {
+  return {
+    pass,
+    message,
+    ...extra,
+    lines: [pass ? message : `不符合：${message}`],
+  };
+}

package/src/source/beijing/index.js

@@ -0,0 +1,23 @@
+// 北京小客车摇号 source 入口
+// 统一接口规范：meta / getCalendar / checkEligibility / extractMarketMetrics / watchTargets
+
+import { SYSTEM_URL } from './constants.js';
+import { getCalendar } from './calendar.js';
+import { checkEligibility } from './eligibility.js';
+import { extractMarketMetrics, watchTargets } from './service.js';
+
+export const meta = {
+  name: 'beijing',
+  label: '北京',
+  systemUrl: SYSTEM_URL,
+  applyTypes: ['person', 'family'],
+  regTypes: ['PTC', 'XNY'],
+  supported: true,
+};
+
+export {
+  getCalendar,
+  checkEligibility,
+  extractMarketMetrics,
+  watchTargets,
+};

package/src/source/beijing/parse.js

@@ -0,0 +1,206 @@
+// 北京交通委公告站解析层：把 HTML 字符串转成结构化数据。
+// 零依赖，纯正则。
+
+/* ---------- 通用工具 ---------- */
+
+export function stripTags(html) {
+  if (!html) return '';
+  let s = html.replace(/<script[\s\S]*?<\/script>/gi, '')
+              .replace(/<style[\s\S]*?<\/style>/gi, '');
+  s = s.replace(/<br\s*\/?>/gi, '\n');
+  s = s.replace(/<[^>]+>/g, '');
+  s = s.replace(/&nbsp;/g, ' ')
+       .replace(/&amp;/g, '&')
+       .replace(/&lt;/g, '<')
+       .replace(/&gt;/g, '>')
+       .replace(/&quot;/g, '"')
+       .replace(/&ldquo;|&rdquo;/g, '"')
+       .replace(/&lsquo;|&rsquo;/g, "'")
+       .replace(/&#(\d+);/g, (_, n) => String.fromCodePoint(Number(n)));
+  s = s.replace(/[ \t]+/g, ' ')
+       .replace(/\n\s*\n+/g, '\n')
+       .trim();
+  return s;
+}
+
+export function normalizeNumberLike(s) {
+  return s
+    .replace(/[０-９]/g, (c) => String.fromCharCode(c.charCodeAt(0) - 0xfee0))
+    .replace(/[，]/g, ',')
+    .replace(/[．]/g, '.')
+    .replace(/[％]/g, '%');
+}
+
+/* ---------- 列表页：jggb / xwzz / bszn 同一套模板 ---------- */
+
+export function parseListPage(html, baseUrl) {
+  const items = [];
+  const sectionMatch = /<div\s+class="subpage_list">([\s\S]*?)<\/div>\s*(?:<div\s+class="pageturn"|<\/div>)/i.exec(html);
+  const scope = sectionMatch ? sectionMatch[1] : html;
+
+  const ddRe = /<dd[^>]*>([\s\S]*?)<\/dd>/gi;
+  for (const m of scope.matchAll(ddRe)) {
+    const block = m[1];
+    const aMatch = /<a\s+class="text"\s+href="([^"]+)"[^>]*>([\s\S]*?)<\/a>/i.exec(block);
+    const dMatch = /<span\s+class="date">([\s\S]*?)<\/span>/i.exec(block);
+    if (!aMatch) continue;
+    const rawTitle = stripTags(aMatch[2]);
+    const url = absUrl(aMatch[1], baseUrl);
+    const date = dMatch ? stripTags(dMatch[1]) : null;
+    const period = extractPeriod(rawTitle);
+    const kind = classifyTitle(rawTitle);
+    items.push({ title: rawTitle, date, url, period, kind });
+  }
+  return items;
+}
+
+export function parsePagination(html, baseUrl) {
+  const block = /<div\s+class="pageturn">([\s\S]*?)<\/div>/i.exec(html);
+  if (!block) return { current: 1, totalPages: 1 };
+  const seg = block[1];
+  const cur = /当前第(\d+)页/.exec(seg);
+  const total = /共(\d+)页/.exec(seg);
+  const next = /<a\s+href="([^"]+)"[^>]*>下一页<\/a>/i.exec(seg);
+  return {
+    current: cur ? Number(cur[1]) : 1,
+    totalPages: total ? Number(total[1]) : 1,
+    nextUrl: next ? absUrl(next[1], baseUrl) : null,
+  };
+}
+
+function extractPeriod(title) {
+  const m = /(20\d{2})年(?:第)?\s*([一二三四五六七八九十0-9]+)\s*期/.exec(title);
+  if (!m) return null;
+  const cnNum = { 一: 1, 二: 2, 三: 3, 四: 4, 五: 5, 六: 6, 七: 7, 八: 8, 九: 9, 十: 10 };
+  const year = Number(m[1]);
+  const rawNo = m[2];
+  const no = /^\d+$/.test(rawNo) ? Number(rawNo) : cnNum[rawNo] ?? null;
+  return no == null ? null : { year, no, label: `${year}年第${no}期` };
+}
+
+function classifyTitle(t) {
+  if (/配置结果/.test(t)) return 'result';
+  if (/申请审核结果和配置工作/.test(t)) return 'config_notice';
+  if (/亲属关系和婚姻状况核查/.test(t)) return 'family_check';
+  if (/资格审核结果/.test(t)) return 'qualify_review';
+  if (/指标配额/.test(t)) return 'quota';
+  if (/买卖|出租|承租|出借|借用/.test(t)) return 'penalty';
+  if (/十问十答|温馨提示/.test(t)) return 'faq';
+  return 'other';
+}
+
+/* ---------- 详情页：subpage_article ---------- */
+
+export function parseDetailPage(html, url) {
+  const article =
+    /<div\s+class="subpage_article">([\s\S]*?)<div\s+class="clearboth">/i.exec(html);
+  const scope = article ? article[1] : html;
+
+  const titleM = /<h2[^>]*>([\s\S]*?)<\/h2>/i.exec(scope);
+  const title = titleM ? stripTags(titleM[1]) : null;
+
+  const h4M = /<h4[^>]*>([\s\S]*?)<\/h4>/i.exec(scope);
+  const h4Text = h4M ? stripTags(h4M[1]) : '';
+  const dateText = /发布日期[：:]\s*([0-9]{4}年[0-9]{1,2}月[0-9]{1,2}日)/.exec(h4Text)?.[1] || null;
+  const dateIso = dateText ? cnDateToIso(dateText) : null;
+
+  const attachments = [];
+  const aRe = /<a[^>]+href="([^"]+\.(?:pdf|docx?|xlsx?|zip))"[^>]*>([\s\S]*?)<\/a>/gi;
+  for (const m of scope.matchAll(aRe)) {
+    attachments.push({
+      url: absUrl(m[1], url),
+      name: stripTags(m[2]).replace(/^下载/, '').trim() || basename(m[1]),
+    });
+  }
+
+  const bodyText = stripTags(scope);
+  const metrics = extractMetricsFromText(bodyText);
+
+  return {
+    url,
+    title,
+    dateText,
+    dateIso,
+    bodyText,
+    attachments,
+    metrics,
+  };
+}
+
+/* ---------- 字段抽取器：从中文段落里捞数字 ---------- */
+
+export function extractMetricsFromText(rawText) {
+  const text = normalizeNumberLike(rawText.replace(/\s+/g, ' '));
+  const r = {};
+
+  const fam = /家庭普通小客车指标申请共?计?\s*([\d,]+)\s*个?有效编码/.exec(text);
+  if (fam) r.familyApplyCount = num(fam[1]);
+  const per = /个人普通小客车指标申请共?计?\s*([\d,]+)\s*个?有效编码/.exec(text);
+  if (per) r.personalApplyCount = num(per[1]);
+  const uni = /单位普通小客车指标申请共?计?\s*([\d,]+)\s*家/.exec(text);
+  if (uni) r.unitApplyCount = num(uni[1]);
+
+  const famPerAlloc = /家庭(?:和|与)个人普通小客车指标共?计?\s*([\d,]+)\s*个/.exec(text);
+  if (famPerAlloc) r.familyPersonAlloc = num(famPerAlloc[1]);
+  const unitAlloc = /配置单位普通小客车指标\s*([\d,]+)\s*个/.exec(text);
+  if (unitAlloc) r.unitAlloc = num(unitAlloc[1]);
+
+  const nevFam = /家庭新能源小客车指标申请共?计?\s*([\d,]+)/.exec(text);
+  if (nevFam) r.familyNevApplyCount = num(nevFam[1]);
+  const nevPer = /个人新能源小客车指标申请共?计?\s*([\d,]+)/.exec(text);
+  if (nevPer) r.personalNevApplyCount = num(nevPer[1]);
+  const nevWait =
+    /新能源(?:小客车)?指标(?:轮候|排队)(?:.*?)(?:约|预计)?\s*([\d.]+)\s*年/.exec(text);
+  if (nevWait) r.nevWaitYears = Number(nevWait[1]);
+
+  const minScore =
+    /家庭(?:摇号)?(?:最低)?中签(?:家庭)?(?:积分|分值)(?:为|是|:|：)\s*([\d.]+)/.exec(text) ||
+    /家庭(?:摇号)?(?:积分|分值)(?:最低|不低于).*?([\d.]+)\s*分/.exec(text);
+  if (minScore) r.familyMinScore = Number(minScore[1]);
+
+  const pRate = /个人(?:普通(?:小客车)?指标)?中签率(?:约)?(?:为|是|:|：)?\s*([\d.]+%|1[/／]\d+)/.exec(text);
+  if (pRate) r.personalRate = pRate[1];
+  const uRate = /单位(?:普通(?:小客车)?指标)?中签率(?:约)?(?:为|是|:|：)?\s*([\d.]+%|1[/／]\d+)/.exec(text);
+  if (uRate) r.unitRate = uRate[1];
+
+  const bl = /共有\s*([\d,]+)\s*个失信被执行人/.exec(text);
+  if (bl) r.blackListCount = num(bl[1]);
+
+  // 配置结果 PDF 里的关键字段
+  const validEncode = /有效编码总数[：:]\s*([\d,]+)/.exec(text);
+  if (validEncode) r.validEncodeCount = num(validEncode[1]);
+  const baseSeed = /基数序号总数[：:]\s*([\d,]+)/.exec(text);
+  if (baseSeed) r.baseSeedTotal = num(baseSeed[1]);
+  const allocTotal = /指标配置总数[：:]\s*([\d,]+)/.exec(text);
+  if (allocTotal) r.allocTotal = num(allocTotal[1]);
+  const randSeed = /指标配置种子数[：:]\s*([\d,]+)/.exec(text);
+  if (randSeed) r.randomSeed = num(randSeed[1]);
+
+  return r;
+}
+
+/* ---------- 小工具 ---------- */
+
+function num(s) {
+  return Number(String(s).replace(/,/g, ''));
+}
+
+function cnDateToIso(s) {
+  const m = /(\d{4})年(\d{1,2})月(\d{1,2})日/.exec(s);
+  if (!m) return null;
+  return `${m[1]}-${m[2].padStart(2, '0')}-${m[3].padStart(2, '0')}`;
+}
+
+function absUrl(href, base) {
+  if (!href) return href;
+  if (/^https?:\/\//i.test(href)) return href;
+  try {
+    return new URL(href, base).toString();
+  } catch {
+    return href;
+  }
+}
+
+function basename(p) {
+  return String(p).split(/[\\/]/).pop() || p;
+}

package/src/source/beijing/pdfExtract.js

@@ -0,0 +1,41 @@
+// PDF 解析模块：把 attachment 里的 PDF 转纯文本，复用 parse.js 的 extractMetricsFromText。
+// 用 pdfjs-dist（纯 JS，无原生依赖）。
+
+import { fetchBinary } from './crawl.js';
+import { extractMetricsFromText } from './parse.js';
+
+let _pdfjs = null;
+
+async function getPdfjs() {
+  if (!_pdfjs) {
+    _pdfjs = await import('pdfjs-dist/legacy/build/pdf.mjs');
+  }
+  return _pdfjs;
+}
+
+/**
+ * 抓取 PDF 并抽取所有页面的纯文本。
+ * @returns {Promise<string>}
+ */
+export async function extractPdfText(pdfUrl) {
+  const pdfjs = await getPdfjs();
+  const { buffer } = await fetchBinary(pdfUrl);
+  const doc = await pdfjs.getDocument({ data: new Uint8Array(buffer) }).promise;
+  const pages = [];
+  for (let i = 1; i <= doc.numPages; i++) {
+    const page = await doc.getPage(i);
+    const content = await page.getTextContent();
+    pages.push(content.items.map((it) => it.str).join(' '));
+  }
+  await doc.destroy();
+  return pages.join('\n');
+}
+
+/**
+ * 抓取 PDF 并抽取关键指标（中签率、家庭最低分等）。
+ * @returns {Promise<object>}
+ */
+export async function extractPdfMetrics(pdfUrl) {
+  const text = await extractPdfText(pdfUrl);
+  return extractMetricsFromText(text);
+}