wiki-print/test/html-builder.test.js

import assert from 'node:assert/strict'
import test from 'node:test'
import * as cheerio from 'cheerio'
import { buildExportHtml } from '../src/html-builder.js'
import { normalizeBaseUrl } from '../src/utils.js'

const baseUrl = normalizeBaseUrl('https://wiki.example.com')

const pages = [
  {
    path: 'docs',
    locale: 'en',
    title: 'Docs',
    depth: 0,
    updatedAt: '2026-01-01T00:00:00.000Z',
    render: `
      <h1 id="welcome">Docs</h1>
      <p><a href="#local">Local</a></p>
      <p><a href="/en/docs/child#sub-heading">Child heading</a></p>
      <p><a href="/en/docs/child">Child</a></p>
      <p><a href="/en/docs-other">Outside subtree</a></p>
      <p><a href="/de/docs">German docs</a></p>
      <p><a href="https://example.com/page">External</a></p>
      <span id="local">Anchor target</span>
    `,
  },
  {
    path: 'docs/child',
    locale: 'en',
    title: 'Child',
    depth: 1,
    updatedAt: '2026-01-01T00:00:00.000Z',
    render: `
      <h2 id="sub-heading">Sub heading</h2>
      <h3>Auto heading</h3>
      <p aria-labelledby="sub-heading other-id"><a name="old-anchor">Back</a>
      <a href="/en/docs">Docs</a>
      <a href="/en/docs/child/grand#child-heading">Grand</a></p>
      <img src="/assets/pic.png" srcset="/assets/pic-2x.png 2x" loading="lazy" alt="Picture">
      <script>window.bad = true</script>
      <div onclick="alert(1)">Clickable</div>
      <iframe src="/embedded"></iframe>
      <form><input value="x"><p>Form content</p></form>
    `,
  },
  {
    path: 'docs/child/grand',
    locale: 'en',
    title: 'Grand',
    depth: 2,
    updatedAt: '2026-01-01T00:00:00.000Z',
    render: `<h2 id="child-heading">Grand child heading</h2>`,
  },
]

async function build(overrides = {}) {
  return buildExportHtml({
    pages,
    baseUrl,
    locale: 'en',
    startPath: 'docs',
    startPage: { title: 'Docs' },
    localization: { namespacing: true, namespaces: ['en'] },
    includeToc: true,
    inlineImages: false,
    apiKey: 'test-key',
    ...overrides,
  })
}

test('rewrites internal links to PDF anchors and keeps external links online', async () => {
  const html = await build()
  const $ = cheerio.load(html)

  assert.equal($('.wiki-content a[href="#wp1-sub-heading"]').length, 1)
  assert.equal($('.wiki-content a[href="#wiki-page-1"]').length, 1)
  assert.equal($('.wiki-content a[href="#wp2-child-heading"]').length, 1)
  assert.equal($('.wiki-content a[href="#wp0-local"]').length, 1)
  assert.ok(
    $('.wiki-content a[data-wiki-path="docs"]')
      .toArray()
      .map((element) => $(element).attr('href'))
      .includes('#wiki-page-0'),
  )

  assert.equal($('a[href="https://wiki.example.com/en/docs-other"]').length, 1)
  assert.equal($('a[href="https://wiki.example.com/de/docs"]').length, 1)
  assert.equal($('a[href="https://example.com/page"]').length, 1)

  assert.equal($('#wp1-sub-heading').length, 1)
  assert.equal($('#wp1-auto-heading').length, 1)
  assert.equal($('a[name="wp1-old-anchor"]').length, 1)
  assert.equal($('[aria-labelledby="wp1-sub-heading other-id"]').length, 1)

  const childHrefs = $('.wiki-content a[data-wiki-path="docs/child"]')
    .toArray()
    .map((element) => $(element).attr('href'))
  assert.ok(childHrefs.includes('#wp1-sub-heading'))
  assert.ok(childHrefs.includes('#wiki-page-1'))
  assert.equal(
    $('.wiki-content a[data-wiki-path="docs/child/grand"]').attr('href'),
    '#wp2-child-heading',
  )
  assert.ok(
    $('.wiki-content a[data-wiki-path="docs"]')
      .toArray()
      .map((element) => $(element).attr('href'))
      .includes('#wiki-page-0'),
  )
})

test('sanitizes scripts, event handlers, forms and iframes', async () => {
  const html = await build()
  const $ = cheerio.load(html)

  assert.equal($('script').length, 0)
  assert.equal($('[onclick]').length, 0)
  assert.equal($('form').length, 0)
  assert.equal($('input').length, 0)
  assert.equal($('iframe').length, 0)
  assert.equal($('.pdf-embed a[href="https://wiki.example.com/embedded"]').length, 1)
  assert.match($('.wiki-content').text(), /Form content/)
})

test('resolves images even when embedding is disabled', async () => {
  const html = await build()
  const $ = cheerio.load(html)

  const image = $('img[alt="Picture"]')
  assert.equal(image.attr('src'), 'https://wiki.example.com/assets/pic.png')
  assert.equal(image.attr('srcset'), undefined)
  assert.equal(image.attr('loading'), undefined)
})

test('supports wikis without URL locale namespacing', async () => {
  const withoutNamespacing = pages.map((page) => ({
    ...page,
    render: page.render.replaceAll('/en/', '/'),
  }))

  const html = await build({
    pages: withoutNamespacing,
    localization: { namespacing: false, namespaces: [] },
  })
  const $ = cheerio.load(html)

  const childHrefs = $('.wiki-content a[data-wiki-path="docs/child"]')
    .toArray()
    .map((element) => $(element).attr('href'))
  assert.ok(childHrefs.includes('#wp1-sub-heading'))
  assert.ok(childHrefs.includes('#wiki-page-1'))
  assert.equal(
    $('.wiki-content a[data-wiki-path="docs/child/grand"]').attr('href'),
    '#wp2-child-heading',
  )
  assert.ok(
    $('.wiki-content a[data-wiki-path="docs"]')
      .toArray()
      .map((element) => $(element).attr('href'))
      .includes('#wiki-page-0'),
  )
})

test('includes only the exported pages in the table of contents', async () => {
  const html = await build()
  const $ = cheerio.load(html)

  const tocHrefs = $('.toc-entry a').toArray().map((element) => $(element).attr('href'))
  assert.deepEqual(tocHrefs, ['#wiki-page-0', '#wiki-page-1', '#wiki-page-2'])
  assert.equal($('.wiki-page').length, 3)
})