Skip to content

fix: preserve table rows with omitted end tags #322

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
DahanItamar wants to merge 2 commits into
base: main
Choose a base branch
from
Open

fix: preserve table rows with omitted end tags #322

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
23fe6ab
fix: close optional table tags at section boundaries
DahanItamar Oct 6, 2026
461a468
fix: keep omitted table rows and sections as siblings
DahanItamar Oct 6, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 25 additions & 12 deletions src/nodes/html.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -997,10 +997,17 @@ const kElementsClosedByOpening = {
P: { p: true, div: true, P: true, DIV: true },
b: { div: true, DIV: true },
B: { div: true, DIV: true },
td: { td: true, th: true, TD: true, TH: true },
TD: { td: true, th: true, TD: true, TH: true },
th: { td: true, th: true, TD: true, TH: true },
TH: { td: true, th: true, TD: true, TH: true },
// A new row or section also ends any open cell in the preceding row.
td: { td: true, th: true, tr: true, tbody: true, tfoot: true, TD: true, TH: true, TR: true, TBODY: true, TFOOT: true },
TD: { td: true, th: true, tr: true, tbody: true, tfoot: true, TD: true, TH: true, TR: true, TBODY: true, TFOOT: true },
th: { td: true, th: true, tr: true, tbody: true, tfoot: true, TD: true, TH: true, TR: true, TBODY: true, TFOOT: true },
TH: { td: true, th: true, tr: true, tbody: true, tfoot: true, TD: true, TH: true, TR: true, TBODY: true, TFOOT: true },
tr: { tr: true, tbody: true, tfoot: true, TR: true, TBODY: true, TFOOT: true },
TR: { tr: true, tbody: true, tfoot: true, TR: true, TBODY: true, TFOOT: true },
thead: { tbody: true, tfoot: true, TBODY: true, TFOOT: true },
THEAD: { tbody: true, tfoot: true, TBODY: true, TFOOT: true },
tbody: { tbody: true, tfoot: true, TBODY: true, TFOOT: true },
TBODY: { tbody: true, tfoot: true, TBODY: true, TFOOT: true },
h1: { h1: true, H1: true },
H1: { h1: true, H1: true },
h2: { h2: true, H2: true },
Expand Down Expand Up @@ -1029,10 +1036,17 @@ const kElementsClosedByClosing = {
I: { div: true, DIV: true },
p: { div: true, DIV: true },
P: { div: true, DIV: true },
td: { tr: true, table: true, TR: true, TABLE: true },
TD: { tr: true, table: true, TR: true, TABLE: true },
th: { tr: true, table: true, TR: true, TABLE: true },
TH: { tr: true, table: true, TR: true, TABLE: true },
// Cells and rows may omit their end tags at the end of a table section.
td: { tr: true, table: true, thead: true, tbody: true, tfoot: true, TR: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
TD: { tr: true, table: true, thead: true, tbody: true, tfoot: true, TR: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
th: { tr: true, table: true, thead: true, tbody: true, tfoot: true, TR: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
TH: { tr: true, table: true, thead: true, tbody: true, tfoot: true, TR: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
tr: { table: true, thead: true, tbody: true, tfoot: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
TR: { table: true, thead: true, tbody: true, tfoot: true, TABLE: true, THEAD: true, TBODY: true, TFOOT: true },
tbody: { table: true, TABLE: true },
TBODY: { table: true, TABLE: true },
tfoot: { table: true, TABLE: true },
TFOOT: { table: true, TABLE: true },
dt: { dl: true, body: true, html: true, DL: true, BODY: true, HTML: true },
DT: { dl: true, body: true, html: true, DL: true, BODY: true, HTML: true },
dd: { dl: true, body: true, html: true, DL: true, BODY: true, HTML: true },
Expand Down Expand Up @@ -1169,10 +1183,9 @@ export function base_parse(data: string, options = {} as Partial<Options>) {
attrs[key.toLowerCase()] = isQuoted ? val.slice(1, val.length - 1) : val;
}

const parentTagName = currentParent.rawTagName as IRawTagName;

if (!closingSlash && !options.preserveTagNesting && kElementsClosedByOpening[parentTagName]) {
if (kElementsClosedByOpening[parentTagName][tagName]) {
if (!closingSlash && !options.preserveTagNesting) {
// One opening tag can end a cell, its row, and its table section.
while (kElementsClosedByOpening[currentParent.rawTagName as IRawTagName]?.[tagName]) {
stack.pop();
currentParent = arr_back(stack);
}
Expand Down
119 changes: 119 additions & 0 deletions test/tests/issues/302.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,119 @@
const { parse, valid } = require('@test/test-target');

describe('issue 302 optional table end tags', function () {
for (const section of ['thead', 'tbody', 'tfoot']) {
for (const cell of ['td', 'th']) {
it(`closes an open ${cell} and tr when ${section} ends`, function () {
const html = `<table><${section}><tr><${cell}>value</${section}></table>`;
const root = parse(html);
valid(html).should.equal(true);
root.outerHTML.should.equal(`<table><${section}><tr><${cell}>value</${cell}></tr></${section}></table>`);
root.querySelector('tr').parentNode.should.equal(root.querySelector(section));
});
}
}

it('closes an open cell, row, and tbody when the table ends', function () {
const html = '<table><tbody><tr><td>value</table>';
const root = parse(html);
valid(html).should.equal(true);
root.outerHTML.should.equal('<table><tbody><tr><td>value</td></tr></tbody></table>');
});

it('keeps nested tables inside their outer cell', function () {
const html = '<table><tbody><tr><td>outer<table><tbody><tr><td>inner</table></table>';
const root = parse(html);
valid(html).should.equal(true);
root.querySelectorAll('table').length.should.equal(2);
const tables = root.querySelectorAll('table');
tables[1].parentNode.should.equal(tables[0].querySelector('td'));
root.outerHTML.should.equal(
'<table><tbody><tr><td>outer<table><tbody><tr><td>inner</td></tr></tbody></table></td></tr></tbody></table>'
);
});

for (const options of [{}, { parseNoneClosedTags: true }, { closeAllByClosing: true }]) {
it(`keeps successive rows as siblings with options ${JSON.stringify(options)}`, function () {
const html = '<table><tbody><tr><td>A<td>1<tr><td>B<td>2</tbody></table>';
const root = parse(html, options);
valid(html, options).should.equal(true);
const tbody = root.querySelector('tbody');
const rows = root.querySelectorAll('tr');
rows.map((row) => row.children.map((cell) => cell.text)).should.eql([
['A', '1'],
['B', '2'],
]);
rows.forEach((row) => row.parentNode.should.equal(tbody));
});
}

for (const [previous, next] of [
['thead', 'tbody'],
['thead', 'tfoot'],
['tbody', 'tbody'],
['tbody', 'tfoot'],
]) {
it(`closes an open cell, row, and ${previous} when ${next} starts`, function () {
const html = `<table><${previous}><tr><td>first<${next}><tr><td>second</${next}></table>`;
const root = parse(html);
valid(html).should.equal(true);
root.querySelector('table')
.children.map((section) => section.rawTagName)
.should.eql([previous, next]);
root.querySelectorAll('tr')
.map((row) => row.text)
.should.eql(['first', 'second']);
});
}

it('parses the table from issue 302 without losing its rows', function () {
const html = '<table><thead><tr><th>Name<th>Value<tbody><tr><td>A<td>1<tr><td>B<td>2</table>';
const root = parse(html);
valid(html).should.equal(true);
root.querySelectorAll('tr')
.map((row) => row.children.map((cell) => cell.text))
.should.eql([
['Name', 'Value'],
['A', '1'],
['B', '2'],
]);
root.querySelector('table')
.children.map((section) => section.rawTagName)
.should.eql(['thead', 'tbody']);
});

it('does not close outer rows when another row starts inside a nested table', function () {
const html = '<table><tbody><tr><td>outer<table><tbody><tr><td>inner<tr><td>next</table><tr><td>last</table>';
const root = parse(html);
valid(html).should.equal(true);
const tables = root.querySelectorAll('table');
const outerRows = tables[0].querySelector('tbody').children;
const innerRows = tables[1].querySelector('tbody').children;
outerRows.length.should.equal(2);
innerRows.map((row) => row.text).should.eql(['inner', 'next']);
tables[1].parentNode.should.equal(outerRows[0].firstElementChild);
outerRows[1].text.should.equal('last');
});

it('respects preserveTagNesting when a new row starts', function () {
const root = parse('<table><tr><td>first<tr><td>second</table>', { preserveTagNesting: true });
const rows = root.querySelectorAll('tr');
rows.length.should.equal(2);
rows[1].parentNode.should.equal(rows[0].firstElementChild);
});

for (const lowerCaseTagName of [false, true]) {
it(`handles uppercase table tags with lowerCaseTagName=${lowerCaseTagName}`, function () {
const html = '<TABLE><THEAD><TR><TH>Head<TBODY><TR><TD>first<TR><TD>second</TABLE>';
const options = { lowerCaseTagName };
const root = parse(html, options);
valid(html, options).should.equal(true);
root.querySelectorAll('tr')
.map((row) => row.text)
.should.eql(['Head', 'first', 'second']);
root.querySelector('table')
.children.map((section) => section.rawTagName)
.should.eql(lowerCaseTagName ? ['thead', 'tbody'] : ['THEAD', 'TBODY']);
});
}
});