hololake-system-architecture/language-world/reconstruction/tools/validate-reconstruction.mjs

#!/usr/bin/env node

import { createHash } from "node:crypto";
import { readdir, readFile, stat } from "node:fs/promises";
import path from "node:path";
import { fileURLToPath } from "node:url";

const here = path.dirname(fileURLToPath(import.meta.url));
const root = path.resolve(here, "..");
const knowledgeRoot = path.resolve(root, "../..");
const worldRoot = path.join(root, "world");
const manifest = JSON.parse(await readFile(path.join(root, "selected-sources-manifest.json"), "utf8"));
const errors = [];
let linksChecked = 0;
let jsonFilesChecked = 0;
let sourceHashesChecked = 0;

async function filesUnder(directory) {
  const found = [];
  for (const entry of await readdir(directory, { withFileTypes: true })) {
    if (entry.name === ".git") continue;
    const full = path.join(directory, entry.name);
    if (entry.isDirectory()) found.push(...await filesUnder(full));
    else found.push(full);
  }
  return found;
}

for (const file of (await filesUnder(root)).filter((item) => item.endsWith(".json"))) {
  try {
    JSON.parse(await readFile(file, "utf8"));
    jsonFilesChecked += 1;
  } catch (error) {
    errors.push({ type: "INVALID_JSON", file, message: error.message });
  }
}

for (const source of manifest.sources) {
  const imported = path.join(root, source.imported_path);
  for (const [kind, file] of [["source", source.source_path], ["imported", imported]]) {
    try {
      const bytes = await readFile(file);
      const sha256 = createHash("sha256").update(bytes).digest("hex");
      if (sha256 !== source.sha256) {
        errors.push({ type: "HASH_MISMATCH", id: source.id, kind, file });
      }
    } catch (error) {
      errors.push({ type: "SOURCE_READ_FAILED", id: source.id, kind, file, message: error.message });
    }
  }
  sourceHashesChecked += 1;
}

const markdownFiles = (await filesUnder(knowledgeRoot)).filter((item) => item.endsWith(".md"));
for (const file of markdownFiles) {
  const text = await readFile(file, "utf8");
  const linkPattern = /\[[^\]]*]\(([^)]+)\)/g;
  for (const match of text.matchAll(linkPattern)) {
    let target = match[1].trim();
    if (!target || target.startsWith("#") || /^(https?:|mailto:)/.test(target)) continue;
    if (target.startsWith("<") && target.endsWith(">")) target = target.slice(1, -1);
    target = decodeURIComponent(target.split("#")[0]);
    const resolved = path.resolve(path.dirname(file), target);
    linksChecked += 1;
    try {
      await stat(resolved);
    } catch {
      errors.push({ type: "BROKEN_RELATIVE_LINK", file, target });
    }
  }
}

for (const source of manifest.sources) {
  if (/^domains\/(main|sub|zero|zero-sense)/.test(source.group) && source.privacy === "PRIVATE_FIFTH") {
    errors.push({ type: "PRIVATE_SOURCE_IN_PUBLIC_DOMAIN", id: source.id, group: source.group });
  }
}

for (const domain of ["main", "sub", "zero", "zero-sense"]) {
  const file = path.join(worldRoot, "domains", domain, "INDEX.md");
  const text = await readFile(file, "utf8");
  if (text.includes("sources/domains/fifth") || text.includes("sources/supplemental/fd-mig-001")) {
    errors.push({ type: "PUBLIC_DOMAIN_LINKS_PRIVATE_FIFTH", file });
  }
}

const routes = JSON.parse(await readFile(path.join(worldRoot, "routing/login-route-map.json"), "utf8"));
const routeById = new Map(routes.routes.map((route) => [route.route_id, route]));
const expected = {
  "GLW-LOGIN-BINGSHUO-FIFTH": ["光湖语言世界", "第五域", "永恒湖心系统", "心跳核心频道"],
  "GLW-LOGIN-ZHIZHI-FIFTH": ["光湖语言世界", "第五域", "永恒湖心系统", "爱之核心子系统", "明天见频道"]
};
for (const [routeId, steps] of Object.entries(expected)) {
  if (JSON.stringify(routeById.get(routeId)?.steps) !== JSON.stringify(steps)) {
    errors.push({ type: "CURRENT_ROUTE_MISMATCH", route_id: routeId });
  }
}

console.log(JSON.stringify({
  schema: "guanghu.reconstruction-validation/v0.1",
  status: errors.length === 0 ? "PASS" : "FAIL",
  markdown_pages_checked: markdownFiles.length,
  relative_links_checked: linksChecked,
  json_files_checked: jsonFilesChecked,
  source_hash_pairs_checked: sourceHashesChecked,
  privacy_boundary_checked: true,
  current_routes_checked: Object.keys(expected).length,
  errors
}, null, 2));

process.exitCode = errors.length === 0 ? 0 : 1;