tihao-ai/claude-code-tihao-sourcing/scripts/software-table-smoke.js

#!/usr/bin/env node
const fs = require('fs');
const os = require('os');
const path = require('path');
const { spawnSync } = require('child_process');
const { SOFTWARE_TABLE_HEADER } = require('../mcp/src/features/tihao-sourcing/report');

const root = path.resolve(__dirname, '..');
const fixtureDir = fs.mkdtempSync(path.join(os.tmpdir(), 'tihao-software-table-'));
const input = path.join(fixtureDir, 'manual-review-sample.csv');
const output = path.join(fixtureDir, 'software-client-list.dedup.csv');
const expectedHeader = SOFTWARE_TABLE_HEADER.join(',');

const rows = [
  [
    'brief编号',
    '策略',
    '排名',
    '平台',
    '博主名称',
    '综合分',
    'brief匹配分',
    '参考风格分',
    '证据加分',
    '证据风险扣分',
    '推荐理由',
    '风险提示',
    '主页链接',
    '人工复核标签'
  ],
  ['brief-a', 'result-first', '1', 'xiaohongshu', '账号A', '88', '90', '80', '1', '0', '标签覆盖 测评、测评、开箱，理由A', '风险A', 'http://m.xiaohongshu.com/user/profile/a?x=1', ''],
  ['brief-a', 'video-enhanced', '2', 'xhs', '账号A（备孕中）', '90', '91', '81', '2', '0', '标签覆盖 测评、开箱，理由A2', '风险A2', 'https://www.xiaohongshu.com/user/profile/a#note', ''],
  ['brief-a', 'result-first', '3', 'douyin', '账号B', '80', '82', '70', '0', '0', '理由B', '风险B', '', ''],
  ['brief-a', 'result-first-risk', '4', 'dy', '账号B', '79', '81', '69', '0', '0', '理由B2', '风险B2', '', ''],
  ['brief-b', 'result-first', '5', 'xiaohongshu', '账号C', '81', '83', '71', '0', '0', '理由C', '风险C', 'https://example.com/c', ''],
  ['brief-b', 'result-first', '6', 'xiaohongshu', '账号D', '79', '81', '69', '0', '0', '理由D', '风险D', 'https://example.com/d', ''],
  ['brief-b', 'result-first', '7', 'xiaohongshu', '账号A', '85', '86', '78', '1', '0', '跨 brief 重复账号', '风险A3', 'https://www.xiaohongshu.com/user/profile/a?brief=b', '']
];
fs.writeFileSync(input, '\uFEFF' + rows.map(row => row.map(csvCell).join(',')).join('\n'));

const result = spawnSync(process.execPath, [
  path.join(root, 'scripts', 'export-software-table.js'),
  '--input',
  input,
  '--output',
  output
], { cwd: root, encoding: 'utf8' });

if (result.status !== 0) {
  process.stderr.write(result.stderr || result.stdout || '');
  throw new Error('export-software-table 执行失败');
}

const text = fs.readFileSync(output, 'utf8').replace(/^\uFEFF/, '');
const lines = text.trim().split(/\r?\n/);
assert(lines[0] === expectedHeader, '软件端表头必须固定为中文表头');
assert(lines.length === 5, '重复博主必须被移除');
assert(lines[1].includes('video-enhanced'), '重复项应保留证据更完整且分数更高的记录');
assert(lines[1].includes(',小红书,'), '平台必须导出为中文展示名');
assert(lines[1].includes('标签覆盖 测评、开箱'), '推荐理由里的重复标签必须清理');
assert(lines.some(line => line.includes(',抖音,')), '抖音平台必须导出为中文展示名');

const body = parseCsv(text).slice(1);
assert(body.filter(row => row[0] === 'brief-a').map(row => row[2]).join(',') === '1,2', 'brief-a 去重后排名必须连续');
assert(body.filter(row => row[0] === 'brief-b').map(row => row[2]).join(',') === '1,2', 'brief-b 去重后排名必须连续');

const summary = JSON.parse(fs.readFileSync(output.replace(/\.csv$/i, '.summary.json'), 'utf8'));
assert(summary.duplicateCount === 3, '摘要应报告 3 条重复记录');
assert(summary.sourceDuplicateRemovedCount === 3, 'summary should expose source duplicate rows removed');
assert(summary.duplicateCountMeaning === 'source_duplicate_removed_count', 'summary should define legacy duplicateCount meaning');
assert(summary.finalDuplicateGroupCount === 0, 'summary should expose zero final duplicate groups');
assert(summary.rankContinuous === true, '摘要应报告排名连续');
assert(summary.duplicateAudit.sameBriefUrl.length === 0, '同 brief URL 重复组必须为 0');
assert(summary.duplicateAudit.sameBriefName.length === 0, '同 brief 名称重复组必须为 0');
assert(summary.duplicateAudit.globalUrl.length === 0, '全局 URL 重复组必须为 0');
assert(summary.duplicateAudit.globalNamePlatform.length === 0, '全局平台+名称重复组必须为 0');

const markdownOutput = output.replace(/\.csv$/i, '.md');
const markdown = spawnSync(process.execPath, [
  path.join(root, 'scripts', 'export-software-markdown.js'),
  '--input',
  output,
  '--output',
  markdownOutput
], { cwd: root, encoding: 'utf8' });
if (markdown.status !== 0) {
  process.stderr.write(markdown.stderr || markdown.stdout || '');
  throw new Error('export-software-markdown 执行失败');
}
const markdownText = fs.readFileSync(markdownOutput, 'utf8');
assert(markdownText.includes('# 软件端去重博主名单'), 'Markdown 标题必须为中文');
assert(markdownText.includes('源表去重丢弃行数：3'), 'Markdown should clarify removed rows are source duplicates');
assert(markdownText.includes('最终名单重复组数：0'), 'Markdown should expose final duplicate group count');
assert(markdownText.includes('同 brief URL 重复组：0'), 'Markdown 必须输出重复审计摘要');

console.log(JSON.stringify({
  ok: true,
  input,
  output,
  markdownOutput,
  duplicateCount: summary.duplicateCount,
  sourceDuplicateRemovedCount: summary.sourceDuplicateRemovedCount,
  finalDuplicateGroupCount: summary.finalDuplicateGroupCount,
  outputRows: summary.outputRows
}, null, 2));

function assert(condition, message) {
  if (!condition) throw new Error(message);
}

function csvCell(value) {
  const text = String(value ?? '');
  return /[",\n]/.test(text) ? `"${text.replace(/"/g, '""')}"` : text;
}

function parseCsv(text) {
  const rows = [];
  let row = [];
  let cell = '';
  let quoted = false;
  for (let index = 0; index < text.length; index += 1) {
    const char = text[index];
    if (char === '\r') continue;
    if (char === '"' && quoted && text[index + 1] === '"') {
      cell += '"';
      index += 1;
    } else if (char === '"') {
      quoted = !quoted;
    } else if (char === ',' && !quoted) {
      row.push(cell);
      cell = '';
    } else if (char === '\n' && !quoted) {
      row.push(cell);
      rows.push(row);
      row = [];
      cell = '';
    } else {
      cell += char;
    }
  }
  if (cell || row.length) {
    row.push(cell);
    rows.push(row);
  }
  return rows;
}