the-mews/app/test/parser_test.dart

import 'package:flutter_test/flutter_test.dart';

import 'package:mews/mews/document.dart';
import 'package:mews/mews/parser.dart';

const conformingPage = '''
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//WAPFORUM//DTD XHTML Mobile 1.2//EN"
  "http://www.openmobilealliance.org/tech/DTD/xhtml-mobile12.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  <head>
    <title>Signal Bars</title>
    <meta name="mews-profile" content="0.1" />
    <meta name="viewport" content="width=device-width" />
    <meta name="description" content="Notes on old phones." />
    <link rel="alternate" type="application/atom+xml" href="feed.xml" title="Posts" />
  </head>
  <body class="mews-green">
    <h1>Signal Bars</h1>
    <p>Notes on <em>old phones</em> and <a href="about.html">the about page</a>.</p>
    <ul>
      <li><a href="2026-10-08-retro-phones.html">2026-10-08 Collecting retro phones</a></li>
      <li><a href="2026-09-21-first-post.html">2026-09-21 First post</a></li>
    </ul>
  </body>
</html>
''';

void main() {
  test('parses a conforming Mews page', () {
    final page = parseMewsPage(source: conformingPage, url: 'https://example.com/index.html');

    expect(page.isMews, isTrue);
    expect(page.specVersion, '0.1');
    expect(page.title, 'Signal Bars');
    expect(page.description, 'Notes on old phones.');
    expect(page.variant, MewsVariant.green);
    expect(page.lang, 'en');

    expect(page.entries, hasLength(2));
    expect(page.entries.first.url, 'https://example.com/2026-10-08-retro-phones.html');
    expect(page.entries.first.title, 'Collecting retro phones');
    expect(page.entries.first.date, DateTime(2026, 10, 8));
    // Entries are listed newest first.
    expect(page.entries.last.date, DateTime(2026, 9, 21));

    expect(page.body, hasLength(3));
    expect(page.body[0], isA<HeadingBlock>());
    final paragraph = page.body[1] as ParagraphBlock;
    expect(paragraph.inline, hasLength(5));
    final link = paragraph.inline[3] as MewsLink;
    expect(link.href, 'https://example.com/about.html');
  });

  test('rejects a page that is not well-formed', () {
    expect(
      () => parseMewsPage(
          source: '<html><head><title>x</title></head><body><p>oops</body></html>',
          url: 'https://example.com/'),
      throwsA(isA<MewsParseException>()),
    );
  });

  test('treats a page without the conformance marker as non-Mews', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head><title>Plain</title></head>
  <body><p>Just XHTML.</p></body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/plain.html');
    expect(page.isMews, isFalse);
    expect(page.specVersion, isNull);
    expect(page.variant, MewsVariant.plain);
  });

  test('resolves relative URLs against the page', () {
    final page = parseMewsPage(source: conformingPage, url: 'https://example.com/posts/index.html');
    final paragraph = page.body[1] as ParagraphBlock;
    final link = paragraph.inline[3] as MewsLink;
    expect(link.href, 'https://example.com/posts/about.html');
  });

  test('allows same-site images and refuses off-site and data URIs', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head><title>Pics</title></head>
  <body>
    <p><img src="/cat.png" alt="A cat" /> and <img src="https://img.example.com/dog.png" alt="A dog" /></p>
    <p><img src="https://other.example.net/bird.png" alt="A bird" /></p>
    <p><img src="data:image/png;base64,AAAA" alt="Inline" /></p>
  </body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/pics.html');
    final images = page.body.whereType<ImageBlock>().toList();
    expect(images, hasLength(4));
    expect(images[0].src, 'https://example.com/cat.png');
    expect(images[0].allowed, isTrue);
    expect(images[1].src, 'https://img.example.com/dog.png');
    expect(images[1].allowed, isTrue, reason: 'subdomains share the registered domain');
    expect(images[2].allowed, isFalse, reason: 'another site');
    expect(images[3].allowed, isFalse, reason: 'data URI');
  });

  test('drops scripts and styles entirely', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head><title>Spooky</title><script>alert(1)</script></head>
  <body><p>Keep this.</p><script>var x = 1;</script><style>p{color:red}</style></body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/spooky.html');
    expect(page.body, hasLength(1));
    final paragraph = page.body[0] as ParagraphBlock;
    expect((paragraph.inline[0] as MewsText).text, 'Keep this.');
  });

  test('honors dir=rtl', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml" dir="rtl" lang="he">
  <head><title>Shalom</title></head>
  <body><p>Text.</p></body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/he.html');
    expect(page.direction, TextDirection.rtl);
  });

  test('builds form fields with labels', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head><title>Guestbook</title></head>
  <body>
    <form action="/sign" method="post">
      <label for="name">Your name</label>
      <input type="text" id="name" name="name" />
      <input type="hidden" name="page" value="1" />
      <input type="submit" value="Sign guestbook" />
    </form>
  </body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/guest.html');
    final form = page.body.single as FormBlock;
    expect(form.action, 'https://example.com/sign');
    expect(form.method, 'post');
    final nameField = form.fields[0] as MewsTextField;
    expect(nameField.name, 'name');
    expect(nameField.label, 'Your name');
    final hidden = form.fields[1] as MewsHiddenField;
    expect(hidden.value, '1');
    final submit = form.fields[2] as MewsSubmitField;
    expect(submit.label, 'Sign guestbook');
  });

  test('mewsPlainText recovers text from broken markup', () {
    const broken = '<html><head><title>Half</title></head><body><p>Visible &amp; intact.</p>';
    final text = mewsPlainText(broken);
    expect(text, contains('Visible & intact.'));
  });

  test('keeps bare inline content in list items and quotes', () {
    final source = '''
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml">
  <head><title>Bare</title><meta name="mews-profile" content="0.1" /></head>
  <body>
    <ul>
      <li><a href="2026-10-08-post.html">2026-10-08 A dated link</a></li>
      <li>Plain item</li>
      <li>Text then a list:<ul><li>Nested</li></ul></li>
    </ul>
    <blockquote>Quoted words <em>verbatim</em>.</blockquote>
  </body>
</html>
''';
    final page = parseMewsPage(source: source, url: 'https://example.com/bare.html');
    final list = page.body[0] as ListBlock;
    final first = (list.items[0].blocks.single as ParagraphBlock).inline;
    final link = first.single as MewsLink;
    expect(link.href, 'https://example.com/2026-10-08-post.html');
    expect((link.children.single as MewsText).text, '2026-10-08 A dated link');

    expect((list.items[1].blocks.single as ParagraphBlock).inline.single,
        isA<MewsText>());
    // Text before a nested list stays with the item; the list becomes a block.
    expect(list.items[2].blocks, hasLength(2));
    expect(list.items[2].blocks[0], isA<ParagraphBlock>());
    expect(list.items[2].blocks[1], isA<ListBlock>());

    final quote = page.body[1] as QuoteBlock;
    final quoteText = (quote.children.single as ParagraphBlock).inline;
    expect(quoteText, hasLength(3));
    expect((quoteText[0] as MewsText).text, 'Quoted words ');
  });
}