fix: Fix salary Excel column detection for index headers (#64)

改动点：
修复 [tools/convert_salary_excel.py (line 47)](/Users/alwin/ai-job-search/tools/convert_salary_excel.py:47) 里列类型识别的问题：之前 n 被当作任意子串匹配，导致 Index / Engineering Index 这类列会被误判成 count。
同步修复类别名生成，避免 Engineering Count 里的 n 被删坏。
把 openpyxl 缺失报错延迟到实际运行转换命令时，这样纯函数可以被单元测试导入。
新增 [tests/test_convert_salary_excel.py (line 25)](/Users/alwin/ai-job-search/tests/test_convert_salary_excel.py:25)，覆盖 index/count 识别和 worksheet 解析。

Co-authored-by: Alwin.Zhang <alwin.zhang420@gmail.com>

tests/__init__.py

@@ -0,0 +1 @@
+

tests/test_convert_salary_excel.py

@@ -0,0 +1,52 @@
+import unittest
+from types import SimpleNamespace
+
+from tools.convert_salary_excel import detect_column_type, parse_sheet
+
+
+class FakeWorksheet:
+    title = "Sheet1"
+
+    def __init__(self, rows):
+        self.rows = rows
+
+    def iter_rows(self, min_row=1, max_row=None, values_only=False):
+        rows = self.rows[min_row - 1:max_row]
+        for row in rows:
+            if values_only:
+                yield row
+            else:
+                yield [SimpleNamespace(value=value) for value in row]
+
+    def __getitem__(self, row_number):
+        return [SimpleNamespace(value=value) for value in self.rows[row_number - 1]]
+
+
+class DetectColumnTypeTests(unittest.TestCase):
+    def test_index_headers_are_not_misclassified_as_count(self):
+        for header in ("Index", "Salary Index", "Engineering Index", "Median salary"):
+            with self.subTest(header=header):
+                self.assertEqual(detect_column_type(header), "index")
+
+    def test_single_letter_n_only_matches_as_a_token(self):
+        self.assertEqual(detect_column_type("Employee n"), "count")
+        self.assertEqual(detect_column_type("Engineering"), None)
+
+    def test_count_headers_still_match_common_labels(self):
+        for header in ("Count", "Engineering Count", "Antal medarbejdere"):
+            with self.subTest(header=header):
+                self.assertEqual(detect_column_type(header), "count")
+
+    def test_parse_sheet_preserves_category_name_with_letter_n(self):
+        ws = FakeWorksheet([
+            ("Company", "Engineering Count", "Engineering Index"),
+            ("Example Corp", 12, 105.5),
+        ])
+
+        companies = parse_sheet(ws)
+
+        self.assertEqual(companies[0]["categories"]["engineering"], {"count": 12, "index": 105.5})
+
+
+if __name__ == "__main__":
+    unittest.main()

tools/convert_salary_excel.py

@@ -28,13 +28,13 @@ with paired count/index columns per category, it groups them automatically.
 import json
 import sys
 import argparse
+import re
 from pathlib import Path
 
 try:
     import openpyxl
 except ImportError:
-    print("Error: openpyxl is required. Install it with: pip install openpyxl", file=sys.stderr)
+    openpyxl = None
-    sys.exit(1)
 
 
 # Column name patterns for auto-detection
@@ -44,15 +44,37 @@ COUNT_PATTERNS = {"antal", "count", "number", "n", "employees", "medarbejdere"}
 INDEX_PATTERNS = {"indeks", "index", "idx", "salary", "løn", "median", "average", "gennemsnit"}
 
 
+def header_matches(header, patterns):
+    """Return True when a header contains a meaningful pattern match."""
+    h = header.lower().strip()
+    tokens = set(re.findall(r"[a-zæøåöäü0-9]+", h))
+
+    for p in patterns:
+        if len(p) == 1:
+            if p in tokens:
+                return True
+        elif p in h:
+            return True
+    return False
+
+
+def strip_type_patterns(header, patterns):
+    """Remove count/index words from a header to derive a category name."""
+    name = header.lower()
+    for p in patterns:
+        if len(p) == 1:
+            name = re.sub(rf"\b{re.escape(p)}\b", "", name)
+        else:
+            name = name.replace(p, "")
+    return name.strip(" _-")
+
+
 def detect_column_type(header):
     """Detect whether a column header refers to count or index data."""
-    h = header.lower().strip()
+    if header_matches(header, COUNT_PATTERNS):
-    for p in COUNT_PATTERNS:
+        return "count"
-        if p in h:
+    if header_matches(header, INDEX_PATTERNS):
-            return "count"
+        return "index"
-    for p in INDEX_PATTERNS:
-        if p in h:
-            return "index"
     return None
 
 
@@ -113,9 +135,7 @@ def parse_sheet(ws, sheet_label=None):
             # If we have a count/index pair, group them
             if col_type == "count" and next_col_type == "index":
                 # Use the header minus the count/index suffix as category name
-                cat_name = col_header
+                cat_name = strip_type_patterns(col_header, COUNT_PATTERNS)
-                for p in COUNT_PATTERNS:
-                    cat_name = cat_name.lower().replace(p, "").strip(" _-")
                 if not cat_name:
                     cat_name = f"category_{len(categories)+1}"
                 categories.append({
@@ -126,9 +146,7 @@ def parse_sheet(ws, sheet_label=None):
                 i += 2
                 continue
             elif col_type == "index" and next_col_type == "count":
-                cat_name = col_header
+                cat_name = strip_type_patterns(col_header, INDEX_PATTERNS)
-                for p in INDEX_PATTERNS:
-                    cat_name = cat_name.lower().replace(p, "").strip(" _-")
                 if not cat_name:
                     cat_name = f"category_{len(categories)+1}"
                 categories.append({
@@ -219,6 +237,10 @@ def main():
         print(f"Error: File not found: {excel_path}", file=sys.stderr)
         sys.exit(1)
 
+    if openpyxl is None:
+        print("Error: openpyxl is required. Install it with: pip install openpyxl", file=sys.stderr)
+        sys.exit(1)
+
     output_path = Path(args.output) if args.output else Path(__file__).parent.parent / "salary_data.json"
 
     print(f"Reading: {excel_path}")