fix(salary): parse Unicode whitespace grouping (#592)

* fix(salary): parse Unicode whitespace grouping

* docs(changelog): record fix for #592

CHANGELOG.md

@@ -200,6 +200,8 @@ per-file diff commands.
   command that does not exist and prompted instead; the string now names the real path, matching
   the six shipped `SKILL.md` files.
 
+- **Salary workbook conversion accepts Unicode whitespace grouping** (#592) - numeric cells written as text with non-breaking or narrow non-breaking grouping spaces now parse without losing their values.
+
 - **`robots_check` unquotes incoming query strings and restricts `$` anchors to pattern end**
   (#543, `tools/robots_check.py`, `tests/test_robots_check.py`) - two path-matching edge cases in
   the robots.txt gate failed open. First, `gate(url)` called `unquote(parts.path)` but concatenated

tests/test_salary_grouping_cli.py

@@ -0,0 +1,64 @@
+"""Preserve space-grouped numbers through the real workbook converter CLI."""
+import json
+from pathlib import Path
+import subprocess
+import sys
+from tempfile import TemporaryDirectory
+import unittest
+
+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 GroupingCLI(unittest.TestCase):
+    def convert(self, rows):
+        with TemporaryDirectory() as directory:
+            root = Path(directory)
+            workbook, output = root / "salary.xlsx", root / "salary_data.json"
+            book = openpyxl.Workbook()
+            try:
+                book.active.append(["Company", "Count", "Salary"])
+                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)],
+                cwd=root, capture_output=True, text=True,
+            )
+            self.assertEqual(result.returncode, 0, result.stdout + result.stderr)
+            return json.loads(output.read_text(encoding="utf-8"))["companies"]
+
+    def test_unicode_space_grouping_preserves_count_and_salary(self):
+        rows = [
+            ["Narrow Space Corp", "1\u202f234", "42\u202f000,50"],
+            ["Thin Space Corp", "2\u2009234", "52\u2009000.50"],
+        ]
+        companies = self.convert(rows)
+        self.assertEqual([entry["categories"] for entry in companies], [
+            {"all_employees": {"count": 1234, "index": 42000.5}},
+            {"all_employees": {"count": 2234, "index": 52000.5}},
+        ])
+
+    def test_existing_space_groups_and_ambiguous_separators_are_unchanged(self):
+        rows = [
+            ["ASCII Corp", "1 234", "42 000,50"],
+            ["NBSP Corp", "1\u00a0234", "42\u00a0000.50"],
+            ["Numeric Corp", 1234, 42000.5],
+            ["Ambiguous Corp", "1,234", "42.000"],
+        ]
+        companies = self.convert(rows)
+        expected = {"all_employees": {"count": 1234, "index": 42000.5}}
+        self.assertEqual([entry["categories"] for entry in companies[:3]], [expected] * 3)
+        self.assertEqual(companies[3]["categories"], {})
+
+
+if __name__ == "__main__":
+    unittest.main()

tools/convert_salary_excel.py

@@ -65,7 +65,8 @@ def parse_numeric_cell(value):
     if not isinstance(value, str):
         raise ValueError("not numeric")
 
-    text = value.strip().replace("\u00a0", " ").replace(" ", "")
+    # Localized exports also use narrow no-break and thin spaces for grouping.
+    text = "".join(value.split())
     if not text:
         raise ValueError("not numeric")
     if "," in text and "." in text: