From d491a755c2b543426c78d9b95ff5a28b87e5e54c Mon Sep 17 00:00:00 2001
From: kevin9327 <5299031+kevin9327@users.noreply.github.com>
Date: Tue, 29 Sep 2026 06:38:15 +0900
Subject: [PATCH] Keep a one-letter or one-digit value instead of dropping it
 as a bullet

The last pass of mergeText drops every text block of one character, to
spend no tokens on list bullets and separators. A table cell is a block
of its own, so a value that is a single digit or letter went too: the
Wikipedia infobox for Barack Obama reads "Children" and then "Parents",
with the 2 in between gone, and a grade, a count, or a one-character CJK
word in a cell vanished the same way.

A one-character block is now dropped only when it is not a letter or a
digit. Bullets, pipes, and middle dots are punctuation or symbols, so
they still go.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---
 src/distill.js        |  6 ++++--
 tests/distill.test.js | 17 +++++++++++++++++
 2 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/src/distill.js b/src/distill.js
index d9274c1..b686a10 100644
--- a/src/distill.js
+++ b/src/distill.js
@@ -1445,9 +1445,11 @@ function mergeText(blocks) {
       out.push(b);
     }
   }
-  // Single stray characters (list bullets, separators) cost tokens and say nothing.
+  // Single stray characters (list bullets, separators) cost tokens and say
+  // nothing. A single letter or digit is a value, though: a table cell holds
+  // one as a block of its own, and an infobox read "Children" with no number.
   return out
     .filter((b) => b.type !== 'break')
-    .filter((b) => b.type !== 'text' || b.text.length > 1)
+    .filter((b) => b.type !== 'text' || b.text.length > 1 || /[\p{L}\p{N}]/u.test(b.text))
     .map(({ host, pre, post, ...block }) => block);
 }
diff --git a/tests/distill.test.js b/tests/distill.test.js
index d365fda..6ad47f7 100644
--- a/tests/distill.test.js
+++ b/tests/distill.test.js
@@ -347,6 +347,23 @@ test('an entity does not put spaces inside a word', () => {
   assert.ok(text.includes('interface & its restraint'), 'a real space was swallowed');
 });
 
+test('a one-character value is kept, a lone bullet or separator is not', () => {
+  // A Wikipedia infobox puts each value in a cell of its own, so a single
+  // digit is a whole block: "Children" was printed with no number after it.
+  const p = distill(`<html><body><table>
+    <tr><th>Spouse</th><td>Michelle Robinson</td></tr>
+    <tr><th>Children</th><td>2</td></tr>
+    <tr><th>Grade</th><td>A</td></tr>
+    <tr><th>Available</th><td>是</td></tr>
+  </table><ul><li>Home</li><li>|</li><li>•</li><li>About</li></ul></body></html>`, 'https://example.test/');
+  const texts = p.blocks.map((b) => b.text);
+  const after = (label) => texts[texts.indexOf(label) + 1];
+  assert.equal(after('Children'), '2', `the value went missing: ${JSON.stringify(texts)}`);
+  assert.equal(after('Grade'), 'A');
+  assert.equal(after('Available'), '是');
+  assert.ok(!texts.includes('|') && !texts.includes('•'), 'a separator was kept as content');
+});
+
 test('an icon button is named by its aria-label, a nameless one is dropped', () => {
   const blocks = social().blocks;
   assert.ok(blocks.some((b) => b.type === 'button' && b.text === 'Follow'), 'a labelled button went missing');