tihao-ai/claude-code-tihao-sourcing/scripts/export-proof-gap-software-form.js

#!/usr/bin/env node
const fs = require('fs');
const path = require('path');

const ROOT = path.resolve(__dirname, '..');
const DEFAULT_INPUT = path.join(ROOT, 'docs', 'ai-optimization-proof-gap-software-table.md');
const DEFAULT_OUTPUT = path.join(ROOT, 'outputs', 'tihao-proof-gap-software-form-latest');
const SOFTWARE_HEADER = [
  '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',
  '主页证据分',
  '视觉质感分',
  '调性一致分',
  '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'
];

function main() {
  const args = parseArgs(process.argv.slice(2));
  const input = path.resolve(args.input || DEFAULT_INPUT);
  const outputDir = path.resolve(args.output || DEFAULT_OUTPUT);
  const result = buildProofGapSoftwareForm({ input, outputDir });

  fs.mkdirSync(outputDir, { recursive: true });
  fs.writeFileSync(result.files.csv, withBom(renderCsv([SOFTWARE_HEADER, ...result.rows])), 'utf8');
  fs.writeFileSync(result.files.markdown, withBom(renderMarkdown(result)), 'utf8');
  fs.writeFileSync(result.files.summary, JSON.stringify(result.summary, null, 2), 'utf8');

  console.log(JSON.stringify({
    outputDir,
    csv: result.files.csv,
    markdown: result.files.markdown,
    summary: result.files.summary,
    rowCount: result.summary.rowCount,
    duplicateIdCount: result.summary.duplicateIdCount,
    headerMatches: result.summary.headerMatches
  }, null, 2));

  if (args.strict && !result.summary.passed) process.exitCode = 1;
}

function buildProofGapSoftwareForm({ input, outputDir }) {
  const text = fs.readFileSync(input, 'utf8').replace(/^\uFEFF/, '');
  const rows = text.split(/\r?\n/).map(splitMarkdownRow).filter(cells => cells.length);
  const header = rows.find(cells => cells[0] === 'brief编号') || [];
  const dataRows = rows.filter(cells => /^GAP-\d{3}$/.test(cells[0] || ''));
  const ids = dataRows.map(cells => cells[0]);
  const duplicateIds = ids.filter((id, index) => ids.indexOf(id) !== index);
  const headerMatches = JSON.stringify(header) === JSON.stringify(SOFTWARE_HEADER);
  const expectedTitles = [
    '真实历史 Brief 数据集',
    '真实视频资源表',
    '真实视频证据 A/B',
    '真实 live/provider 长跑证明',
    '商务复核和客户效果'
  ];
  const actualTitles = dataRows.map(cells => cells[4]);
  const missingTitles = expectedTitles.filter(title => !actualTitles.includes(title));
  const rowWidthOk = dataRows.every(cells => cells.length === SOFTWARE_HEADER.length);
  const summary = {
    generatedAt: new Date().toISOString(),
    input: path.relative(ROOT, input).replace(/\\/g, '/'),
    outputDir: path.relative(ROOT, outputDir).replace(/\\/g, '/'),
    headerMatches,
    rowWidthOk,
    rowCount: dataRows.length,
    duplicateIdCount: duplicateIds.length,
    duplicateIds: Array.from(new Set(duplicateIds)),
    expectedTitles,
    actualTitles,
    missingTitles,
    passed: headerMatches && rowWidthOk && dataRows.length === 5 && duplicateIds.length === 0 && missingTitles.length === 0,
    guardrails: [
      '本表单用于安排补证任务，不证明客户效果完成。',
      '不能把 sample、smoke、模板、provider fallback 或接口 200 当成业务证明。',
      '表单不得包含 sessionToken、Authorization、模型 token 或 npm token。'
    ]
  };
  return {
    rows: dataRows,
    summary,
    files: {
      csv: path.join(outputDir, 'tihao-proof-gap-software-form.csv'),
      markdown: path.join(outputDir, 'tihao-proof-gap-software-form.md'),
      summary: path.join(outputDir, 'tihao-proof-gap-software-form-summary.json')
    }
  };
}

function renderMarkdown(result) {
  const lines = [
    '# 最新提号补证表单（软件端格式）',
    '',
    `- 生成时间：${result.summary.generatedAt}`,
    `- 行数：${result.summary.rowCount}`,
    `- 表头匹配：${result.summary.headerMatches ? '是' : '否'}`,
    `- 是否通过：${result.summary.passed ? '是' : '否'}`,
    '',
    '| ' + SOFTWARE_HEADER.join(' | ') + ' |',
    '| ' + SOFTWARE_HEADER.map(() => '---').join(' | ') + ' |',
    ...result.rows.map(row => '| ' + row.map(escapeMarkdownCell).join(' | ') + ' |'),
    '',
    '## 边界',
    '',
    ...result.summary.guardrails.map(item => `- ${item}`)
  ];
  return lines.join('\n');
}

function renderCsv(rows) {
  return rows.map(row => row.map(csvCell).join(',')).join('\n');
}

function splitMarkdownRow(line) {
  if (!line || !line.trim().startsWith('|')) return [];
  return line.trim().replace(/^\|/, '').replace(/\|$/, '').split('|').map(cell => cell.trim());
}

function parseArgs(argv) {
  const args = {};
  for (let index = 0; index < argv.length; index += 1) {
    const raw = argv[index];
    if (!raw.startsWith('--')) continue;
    const key = raw.slice(2).replace(/-([a-z])/g, (_, char) => char.toUpperCase());
    const next = argv[index + 1];
    if (!next || next.startsWith('--')) args[key] = true;
    else {
      args[key] = next;
      index += 1;
    }
  }
  return args;
}

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

function escapeMarkdownCell(value) {
  return String(value ?? '').replace(/\|/g, '/').replace(/\r?\n/g, ' ');
}

function withBom(text) {
  return `\uFEFF${text}`;
}

if (require.main === module) main();

module.exports = {
  SOFTWARE_HEADER,
  buildProofGapSoftwareForm
};