From e74bd387e4ac807d1d20be2f28bf71e296f69970 Mon Sep 17 00:00:00 2001
From: Raphael Fakhri <153192858+RaphaelFakhri@users.noreply.github.com>
Date: Wed, 21 Oct 2026 07:49:09 +0000
Subject: [PATCH] fix: normalizer adding empty list items for whitespace around
 blocks

---
 cpp/parser/GumboNormalizer.c             | 6 +++++-
 cpp/tests/GumboParserTest.cpp            | 7 +++++++
 src/web/__tests__/htmlNormalizer.test.ts | 6 ++++++
 src/web/normalization/htmlNormalizer.ts  | 7 ++++++-
 4 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/cpp/parser/GumboNormalizer.c b/cpp/parser/GumboNormalizer.c
index 5a6210f15..562a745f1 100644
--- a/cpp/parser/GumboNormalizer.c
+++ b/cpp/parser/GumboNormalizer.c
@@ -638,8 +638,12 @@ static void flatten_li_node(GumboNode *node, buffer_t *ib, buffer_t *out,
                             li_ctx_t *ctx);
 
 static void flush_li_buffer(buffer_t *ib, buffer_t *out, li_ctx_t *ctx) {
-  if (ib->len == 0)
+  /* Whitespace around blocks inside an <li> (e.g. `<li>\n<p>a</p>\n</li>`)
+   * is layout noise, not an extra list item. */
+  if (is_whitespace_only(ib->data, ib->len)) {
+    buffer_clear(ib);
     return;
+  }
   buffer_append_str(out, "<li");
   emit_attributes(ctx->el, "li", out);
   buffer_append_str(out, ">");
diff --git a/cpp/tests/GumboParserTest.cpp b/cpp/tests/GumboParserTest.cpp
index ff06232c9..07c6c4af9 100644
--- a/cpp/tests/GumboParserTest.cpp
+++ b/cpp/tests/GumboParserTest.cpp
@@ -607,6 +607,13 @@ TEST(GumboParserTest, InterBlockWhitespace) {
   EXPECT_EQ(GumboParser::normalizeHtml("<p>Asdasd</p> <p>Asdasd</p>"),
             "<p>Asdasd</p><p>Asdasd</p>");
 
+  // Whitespace around blocks inside a list item must not add empty items.
+  EXPECT_EQ(GumboParser::normalizeHtml(
+                "<ul><li>\n<p>a</p>\n</li><li>\n<p>b</p>\n</li></ul>"),
+            "<ul><li>a</li><li>b</li></ul>");
+  EXPECT_EQ(GumboParser::normalizeHtml("<ol><li><p>a</p> <p>b</p></li></ol>"),
+            "<ol><li>a</li><li>b</li></ol>");
+
   // Significant inline content between blocks is still wrapped in <p>.
   EXPECT_EQ(GumboParser::normalizeHtml("<p>a</p> hello <p>b</p>"),
             "<p>a</p><p> hello </p><p>b</p>");
diff --git a/src/web/__tests__/htmlNormalizer.test.ts b/src/web/__tests__/htmlNormalizer.test.ts
index 04547a2de..8501ea353 100644
--- a/src/web/__tests__/htmlNormalizer.test.ts
+++ b/src/web/__tests__/htmlNormalizer.test.ts
@@ -616,6 +616,12 @@ describe('htmlNormalizer', () => {
         '<p>Asdasd</p><p>Asdasd</p><p>Asdasda</p>',
       ],
       ['<p>Asdasd</p> <p>Asdasd</p>', '<p>Asdasd</p><p>Asdasd</p>'],
+      // Whitespace around blocks inside a list item must not add empty items.
+      [
+        '<ul><li>\n<p>a</p>\n</li><li>\n<p>b</p>\n</li></ul>',
+        '<ul><li>a</li><li>b</li></ul>',
+      ],
+      ['<ol><li><p>a</p> <p>b</p></li></ol>', '<ol><li>a</li><li>b</li></ol>'],
       // Significant inline content between blocks is still wrapped in <p>.
       ['<p>a</p> hello <p>b</p>', '<p>a</p><p> hello </p><p>b</p>'],
       // Spaces inside text / between inlines must be preserved.
diff --git a/src/web/normalization/htmlNormalizer.ts b/src/web/normalization/htmlNormalizer.ts
index 35e6ef8a2..bb4f96090 100644
--- a/src/web/normalization/htmlNormalizer.ts
+++ b/src/web/normalization/htmlNormalizer.ts
@@ -432,7 +432,12 @@ function flushLiBuffer(
   out: { buf: string },
   ctx: LiCtx
 ): void {
-  if (ib.buf.length === 0) return;
+  // Whitespace around blocks inside an <li> (e.g. `<li>\n<p>a</p>\n</li>`)
+  // is layout noise, not an extra list item.
+  if (isWhitespaceOnly(ib.buf)) {
+    ib.buf = '';
+    return;
+  }
   out.buf += `<li${emitAttributes(ctx.el, 'li')}>`;
   out.buf += emitStylesOpen(ctx.styles);
   out.buf += ib.buf;