From 991bda5dd22d3c2dd501152c58a52dd46b4045d6 Mon Sep 17 00:00:00 2001
From: Rudy Mizrahi Celekli <47457359+rudycelekli@users.noreply.github.com>
Date: Sun, 18 Oct 2026 01:19:24 -0400
Subject: [PATCH] fix(salary): reject non-finite workbook numbers (#572)

* fix(salary): reject non-finite workbook numbers

* docs(changelog): record fix for #572
---
 CHANGELOG.md                       |  2 ++
 tests/test_salary_nonfinite_cli.py | 53 ++++++++++++++++++++++++++++++
 tools/convert_salary_excel.py      | 12 +++++--
 3 files changed, 65 insertions(+), 2 deletions(-)
 create mode 100644 tests/test_salary_nonfinite_cli.py

diff --git a/CHANGELOG.md b/CHANGELOG.md
index b7948527..1847804d 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -423,6 +423,8 @@ per-file diff commands.
   harmless, and valid UTF-8 rules decide as before. Nine new tests; the seven that pin the
   fail-opens fail on master.
 
+- **Salary workbook conversion rejects non-finite text cells** (#572) - hand-typed `Infinity` or `NaN` text cells now produce validation errors instead of numeric-conversion tracebacks.
+
 - **`verify_layout.py` finds an orphaned entry header by where the text starts**
   (`tools/verify_layout.py`, `tests/test_verify_layout.py`) - the orphan rule compared line
   left edges against the document margin, and a list marker moves a line's left edge without
diff --git a/tests/test_salary_nonfinite_cli.py b/tests/test_salary_nonfinite_cli.py
new file mode 100644
index 00000000..34c6a011
--- /dev/null
+++ b/tests/test_salary_nonfinite_cli.py
@@ -0,0 +1,53 @@
+"""Non-finite workbook cells cannot crash conversion or escape as JSON numbers."""
+import json
+import subprocess
+import sys
+import unittest
+from pathlib import Path
+from tempfile import TemporaryDirectory
+
+try:
+    import openpyxl
+except ImportError:
+    openpyxl = None
+
+TOOL = Path(__file__).resolve().parent.parent / "tools" / "convert_salary_excel.py"
+
+
+@unittest.skipUnless(openpyxl, "requires optional openpyxl (installed in CI)")
+class NonFiniteWorkbookCLI(unittest.TestCase):
+    def convert(self, headers, rows):
+        with TemporaryDirectory() as directory:
+            root = Path(directory)
+            workbook, output = root / "salary.xlsx", root / "salary.json"
+            book = openpyxl.Workbook()
+            try:
+                book.active.append(headers)
+                for row in rows:
+                    book.active.append(row)
+                book.save(workbook)
+            finally:
+                book.close()
+            result = subprocess.run(
+                [sys.executable, str(TOOL), str(workbook), "--output", str(output)],
+                capture_output=True, text=True, encoding="utf-8",
+            )
+            self.assertEqual(result.returncode, 0, result.stderr)
+            def reject_constant(value):
+                self.fail(f"converter emitted non-JSON numeric constant {value}")
+            return json.loads(output.read_text(encoding="utf-8"), parse_constant=reject_constant)["companies"]
+
+    def test_infinite_paired_count_does_not_abort_the_workbook(self):
+        companies = self.convert(["Company", "Count", "Index"], [
+            ["Invalid Count", "Infinity", 108.5], ["Valid Control", 12, 102.5]
+        ])
+        self.assertEqual(companies[0]["categories"], {"all_employees": {"count": None, "index": 108.5}})
+        self.assertEqual(companies[1]["categories"], {"all_employees": {"count": 12, "index": 102.5}})
+
+    def test_nonfinite_paired_index_is_skipped_instead_of_emitted_as_nan(self):
+        companies = self.convert(["Company", "Count", "Index"], [["Acme", 12, "NaN"]])
+        self.assertEqual(companies[0]["categories"], {"all_employees": {"count": 12, "index": None}})
+
+    def test_nonfinite_standalone_count_is_skipped(self):
+        companies = self.convert(["Company", "Count"], [["Acme", "NaN"]])
+        self.assertEqual(companies[0]["categories"], {})
diff --git a/tools/convert_salary_excel.py b/tools/convert_salary_excel.py
index fa01414b..3d4822cf 100644
--- a/tools/convert_salary_excel.py
+++ b/tools/convert_salary_excel.py
@@ -26,6 +26,7 @@ with paired count/index columns per category, it groups them automatically.
 """
 
 import json
+import math
 import sys
 import argparse
 import re
@@ -58,10 +59,17 @@ ID_PATTERNS = {"id", "personnummer"}
 DEFAULT_CATEGORY = "all_employees"
 
 
+def _finite_float(value):
+    number = float(value)
+    if not math.isfinite(number):
+        raise ValueError("salary values must be finite")
+    return number
+
+
 def parse_numeric_cell(value):
     """Parse numeric Excel values, including localized string cells."""
     if isinstance(value, (int, float)):
-        return float(value)
+        return _finite_float(value)
     if not isinstance(value, str):
         raise ValueError("not numeric")
 
@@ -84,7 +92,7 @@ def parse_numeric_cell(value):
     elif "." in text:
         if re.fullmatch(r"[+-]?\d+\.\d{3}", text):
             raise ValueError("ambiguous dot separator")
-    return float(text)
+    return _finite_float(text)
 
 
 def header_matches(header, patterns):