[wahidrizka]

Marked version: 18.1.1 (cdbdc01)

Markdown flavor: CommonMark|GitHub Flavored Markdown

Description

When a list in a blockquote takes lazy continuation lines, blockquote() cuts the old list raw off the end of its own raw and appends the new list raw. The list raw has no > markers, so that only works for a one-line list. Otherwise the blockquote raw stops matching the source:

marked.lexer('> - a\n> - b\nc')[0].raw
// '> - - a\n- b\nc\n'

A blockquote around this one uses that raw to find out which continuation lines the inner blockquote took, so lazy lines end up outside both blockquotes, and sometimes text is lost:

Expectation

> > - a
> b
c

> > - a
> > - b
>   c
d

<blockquote>
<blockquote>
<ul>
<li>a
b
c</li>
</ul>
</blockquote>
</blockquote>
<blockquote>
<blockquote>
<ul>
<li>a</li>
<li>b
c
d</li>
</ul>
</blockquote>
</blockquote>

as commonmark.js renders it.

Result

<blockquote>
<blockquote>
<ul>
<li>a
b</li>
</ul>
</blockquote>
</blockquote>
<p>c</p>
<blockquote>
<blockquote>
<ul>
<li><ul>
<li>a</li>
</ul>
</li>
</ul>
</blockquote>
</blockquote>
<p>d</p>

The second list loses b and c.

What was attempted

The list branch now counts the continuation lines the new list took and adds those lines from the source to raw and text, as the nested blockquote branch above it already does. The lines it did not take are passed on as they are, without the empty first line that splitting the rest of newText used to add.

Found by checking that the raw values of the top-level tokens join back to the source. On 60,000 distinct generated documents with blockquotes, lists and trailing whitespace (gfm: false), raw mismatches drop from 5,345 to 1,206, and 73 render differently: 31 now match commonmark.js and did not before, 42 match it neither before nor after, and none stopped matching. In 27 of the 73, master's output is missing text that commonmark.js has, and the fix has all of it. One goes the other way: in > > - x\n> bar\n # foo the indented lazy line is now read as part of the list item and becomes a heading, where master moved it out of the blockquotes as a code block (commonmark.js keeps it as paragraph text). That comes from lazy lines being passed to list() as item lines, which this does not change. One document gets a raw mismatch that master does not have: the fix reads its last line - as an empty list item, as commonmark.js does, and the list's raw ends in -\n, which master also gives for - on its own. All 652 CommonMark and 28 GFM spec examples render exactly as before.

(Edited: my first numbers counted some generated documents twice and missed the raw mismatch above. These are from 60,000 distinct documents.)

The new lazy line in a nested list lexer test and test/specs/new/blockquote_nested_list_lazy_lines both fail on master. Specs, unit, umd, cjs, types and lint pass locally, except the five bin exec tests, which fail on master too because my checkout path has a space in it.

Contributor

• Test(s) exist to ensure functionality and minimize regression (if no tests added, list tests covering this PR); or,
• no tests required for this PR.
• If submitting new feature, it has been documented in the appropriate places.

Committer

In most cases, this should be a different person than the contributor.

• CI is green (no forced merge required).
• Squash and Merge PR following conventional commit guidelines.

[vercel]

@wahidrizka is attempting to deploy a commit to the MarkedJS Team on Vercel.

A member of the Team first needs to authorize it.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
marked-website	Ready	Preview	Oct 13, 2026 3:49am UTC

[UziTech]

Nice catch! 💯