mnote/scripts/task802-pdf-attachments-restore-smoke.js

#!/usr/bin/env node
"use strict";

const assert = require("node:assert/strict");
const fs = require("node:fs");
const path = require("node:path");
const { request } = require("playwright");

const BASE_URL = (process.env.MNOTE_WEB_SMOKE_BASE_URL || "http://127.0.0.1:3000").replace(/\/+$/, "");
const ROOT_PATH = process.env.MNOTE_WOLAI_PDF_SMOKE_ROOT
  || "/mnt/Data1T/Mnote_data/users/liaibo/workspaces/my-space";
const LIMIT = Number(process.env.MNOTE_WOLAI_PDF_SMOKE_LIMIT || 10);
const USERNAME = process.env.MNOTE_WOLAI_PDF_SMOKE_USER || "mnote.e2e@example.com";
const PASSWORD = process.env.MNOTE_WOLAI_PDF_SMOKE_PASSWORD || "MnoteE2E123!";
const OUT_DIR = path.join(process.cwd(), "tmp", "task802-pdf-attachments-restore-smoke");
const RESULT_PATH = path.join(OUT_DIR, "result.json");

function fileUrl(localPath) {
  return `file://${localPath}`;
}

function encodeLocalIdSegment(value) {
  const bytes = Buffer.from(String(value || ""), "utf8");
  let encoded = "";
  for (const byte of bytes) {
    const character = String.fromCharCode(byte);
    if (
      (byte >= 48 && byte <= 57)
      || (byte >= 65 && byte <= 90)
      || (byte >= 97 && byte <= 122)
      || character === "."
      || character === "_"
      || character === "-"
    ) {
      encoded += character;
    } else {
      encoded += `~${byte.toString(16).toUpperCase().padStart(2, "0")}`;
    }
  }
  return encoded;
}

function markdownDocumentId(relativePath) {
  return `local-md:${encodeLocalIdSegment(relativePath)}`;
}

function walkMarkdownFiles(dir, out = []) {
  for (const entry of fs.readdirSync(dir, { withFileTypes: true })) {
    if (entry.name === ".assets" || entry.name === ".git" || entry.name === "node_modules") {
      continue;
    }
    const fullPath = path.join(dir, entry.name);
    if (entry.isDirectory()) {
      walkMarkdownFiles(fullPath, out);
    } else if (entry.isFile() && entry.name.toLowerCase().endsWith(".md")) {
      out.push(fullPath);
    }
  }
  return out;
}

function rootRelativePath(filePath) {
  return path.relative(ROOT_PATH, filePath).split(path.sep).join("/");
}

function normalizeHref(href) {
  return String(href || "")
    .trim()
    .replace(/^<|>$/g, "")
    .replace(/\\([\\`*_{}\[\]()#+\-.!|>])/g, "$1");
}

function pdfHrefs(markdown) {
  const refs = [];
  const source = String(markdown || "");
  const linkPattern = /\[[^\]]*?\.pdf[^\]]*?\]\(\s*(?:<([^>]+?\.pdf)>|([^)\s]+?\.pdf))(?:\s+["'][^"']*["'])?\s*\)/gi;
  for (const match of source.matchAll(linkPattern)) {
    const href = normalizeHref(match[1] || match[2] || "");
    if (!href || /^https?:\/\//i.test(href) || href.startsWith("#") || href.startsWith("mailto:")) {
      continue;
    }
    refs.push(href);
  }
  return refs;
}

function collectExistingPdfAttachments() {
  const candidates = [];
  for (const markdownPath of walkMarkdownFiles(ROOT_PATH)) {
    const markdown = fs.readFileSync(markdownPath, "utf8");
    const markdownRelativePath = rootRelativePath(markdownPath);
    for (const href of pdfHrefs(markdown)) {
      if (!href.includes(".assets/")) {
        continue;
      }
      const assetRelativePath = path.posix.normalize(path.posix.join(
        path.posix.dirname(markdownRelativePath),
        href,
      ));
      const assetPath = path.join(ROOT_PATH, ...assetRelativePath.split("/"));
      if (!fs.existsSync(assetPath)) {
        continue;
      }
      const bytes = fs.statSync(assetPath).size;
      if (bytes <= 0) {
        continue;
      }
      candidates.push({
        markdownRelativePath,
        href,
        assetRelativePath,
        assetPath,
        bytes,
      });
      if (candidates.length >= LIMIT) {
        return candidates;
      }
    }
  }
  return candidates;
}

async function signIn(api) {
  const response = await api.post("/api/auth", {
    headers: { "content-type": "application/json", accept: "application/json" },
    data: {
      action: "auth:signIn",
      args: {
        provider: "password",
        params: {
          account: USERNAME,
          password: PASSWORD,
          flow: "signIn",
        },
      },
    },
  });
  assert.equal(response.status(), 200, `登录失败: ${response.status()} ${await response.text()}`);
}

function flattenBlockDocumentBlocks(blockDocument) {
  const direct = Array.isArray(blockDocument?.blocks) ? blockDocument.blocks : [];
  return direct;
}

function pageHasMediaSource(pageAggregate, href, assetRelativePath) {
  const content = Array.isArray(pageAggregate?.body?.content) ? pageAggregate.body.content : [];
  if (content.some((block) => block?.type === "media" && block?.props?.sourcePath === href)) {
    return true;
  }
  const blocks = flattenBlockDocumentBlocks(pageAggregate?.body?.blockDocument);
  return blocks.some((block) => (
    block?.type === "media"
    && (block?.attrs?.sourcePath === href || block?.attrs?.sourcePath === assetRelativePath)
  ));
}

async function main() {
  fs.mkdirSync(OUT_DIR, { recursive: true });
  const candidates = collectExistingPdfAttachments();
  assert.ok(candidates.length >= Math.min(5, LIMIT), `可验证 PDF 附件不足: ${candidates.length}`);

  const api = await request.newContext({
    baseURL: BASE_URL,
    extraHTTPHeaders: { accept: "application/json" },
  });
  try {
    await signIn(api);
    const checked = [];
    for (const candidate of candidates) {
      const documentId = markdownDocumentId(candidate.markdownRelativePath);
      const aggregateUrl = new URL(`/api/page-aggregate/${encodeURIComponent(documentId)}`, BASE_URL);
      aggregateUrl.searchParams.set("sourceKind", "local_folder");
      aggregateUrl.searchParams.set("rootUri", fileUrl(ROOT_PATH));
      const response = await api.get(`${aggregateUrl.pathname}${aggregateUrl.search}`);
      assert.equal(
        response.status(),
        200,
        `page aggregate 失败 ${candidate.markdownRelativePath}: ${response.status()} ${await response.text()}`,
      );
      const payload = await response.json();
      const aggregate = payload.result;
      const attachmentRefs = Array.isArray(aggregate?.body?.attachmentRefs)
        ? aggregate.body.attachmentRefs
        : [];
      const ref = attachmentRefs.find((item) => (
        item?.relativePath === candidate.assetRelativePath
        || item?.normalizedHref === candidate.href
        || item?.rawHref === candidate.href
      ));
      assert.ok(ref, `aggregate 缺少 PDF attachmentRef: ${candidate.assetRelativePath}`);
      assert.equal(String(ref.ext || "").toLowerCase(), "pdf", JSON.stringify(ref));
      assert.equal(ref.exists, true, JSON.stringify(ref));
      assert.equal(ref.authorized, true, JSON.stringify(ref));
      assert.equal(ref.fileSize, candidate.bytes, JSON.stringify({ ref, candidate }));
      assert.ok(
        pageHasMediaSource(aggregate, candidate.href, candidate.assetRelativePath),
        `aggregate 未还原 PDF media block: ${candidate.assetRelativePath}`,
      );
      checked.push({
        markdownRelativePath: candidate.markdownRelativePath,
        assetRelativePath: candidate.assetRelativePath,
        bytes: candidate.bytes,
        label: ref.label,
      });
    }
    const result = { ok: true, checked };
    fs.writeFileSync(RESULT_PATH, `${JSON.stringify(result, null, 2)}\n`, "utf8");
    console.log(JSON.stringify(result, null, 2));
  } finally {
    await api.dispose();
  }
}

main().catch((error) => {
  console.error(error && error.stack ? error.stack : error);
  process.exit(1);
});