From 52f84e0f823780362964f270200463370ef3b3e7 Mon Sep 17 00:00:00 2001
From: Ayobami Adegoke <adegokeayobamiseun@gmail.com>
Date: Sat, 3 Oct 2026 18:33:48 +0100
Subject: [PATCH] fix(verify_pdf): --ascii-dates needs digit boundaries around
 the year (#512)

_YEAR matched (19|20)\d{2} anywhere, so `2000` inside `120000` or `12000`
made a budget or throughput range typed with `--` read as an en-dashed
date range: "Grew budget DKK 120000-200000" and "12000-15000 events/s"
both failed /apply Step 5d, and the agent was sent to fix a date argument
that does not exist. The guide promises a numeric range with no year is
left alone.

Bound the year with (?<!\d) and (?!\d). "2016-2024" next to a longer
number is still caught. Two new cases; the numeric one fails on the
unbounded pattern.
---
 CHANGELOG.md             |  8 ++++++++
 tests/test_verify_pdf.py | 10 ++++++++++
 tools/verify_pdf.py      |  4 +++-
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 36148725..10f14cbe 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -15,6 +15,14 @@ per-file diff commands.
 
 ### Fixed
 
+- **`verify_pdf.py --ascii-dates` no longer reads a year-like run inside a longer number as a
+  date** - the year pattern had no digit boundaries, so `2000` inside `120000` or `12000` made
+  `Grew budget DKK 120000–200000` and `12000–15000 events/s` fail `/apply` Step 5d as
+  en-dashed date ranges, and the agent was sent to fix a "date argument" that does not exist.
+  The guide promises a numeric range with no year is left alone; now it is. `2016–2024` next to
+  a longer number is still caught. Two new `test_verify_pdf.py` cases; the numeric one fails
+  on the unbounded pattern.
+
 - **`check_framework_version.py`'s git diff read no longer crashes on non-ASCII framework-file
   content** (`tools/check_framework_version.py`) - `run_git()` called
   `subprocess.run(text=True)` without an explicit `encoding`, so output decoded via the host
diff --git a/tests/test_verify_pdf.py b/tests/test_verify_pdf.py
index ee2680d6..8dc69740 100644
--- a/tests/test_verify_pdf.py
+++ b/tests/test_verify_pdf.py
@@ -99,6 +99,16 @@ class FindNonAsciiDateRangesTests(unittest.TestCase):
         # 05-cv-templates.md keeps `--` in prose ranges like EUR 600k--1M.
         self.assertEqual(find_non_ascii_date_ranges("EUR 600k\u20131M, 12\u201315 people"), [])
 
+    def test_year_like_digits_inside_a_longer_number_are_not_a_year(self):
+        # `2000` sits inside `120000`, `200000`, `12000`, `15000`: a budget or
+        # throughput range typed with `--` is not a date and must not fail 5d.
+        for text in ("Grew budget DKK 120000\u2013200000", "Scaled ingestion from 12000\u201315000 events/s"):
+            with self.subTest(text=text):
+                self.assertEqual(find_non_ascii_date_ranges(text), [])
+
+    def test_a_standalone_year_next_to_a_longer_number_is_still_caught(self):
+        self.assertEqual(find_non_ascii_date_ranges("2016\u20132024 (120000 users)")[0][1], "\u2013")
+
     def test_year_ending_a_line_is_not_joined_to_the_next_lines_dash(self):
         text = "Heading 1988-1994\n\u2013 note\n"
         self.assertEqual(find_non_ascii_date_ranges(text), [])
diff --git a/tools/verify_pdf.py b/tools/verify_pdf.py
index f75f03cf..b4abfef8 100644
--- a/tools/verify_pdf.py
+++ b/tools/verify_pdf.py
@@ -94,7 +94,9 @@ def normalize_text(text):
 # rest are the other Unicode dashes and the minus sign, which a parser that splits
 # a range only on U+002D treats the same way.
 NON_ASCII_DASHES = "\u2010\u2011\u2012\u2013\u2014\u2015\u2212"
-_YEAR = r"(?:19|20)\d{2}"
+# A year is four digits standing alone: without the digit boundaries, `2000`
+# inside `120000` or `12000` made a budget or throughput range read as a date.
+_YEAR = r"(?<!\d)(?:19|20)\d{2}(?!\d)"
 # Horizontal whitespace only between the year and the dash: `\s*` also matched
 # a newline, so a year ending one line joined a dash opening the next (a bullet,
 # a wrapped prose line) and an ASCII date was reported as U+2013.