Fix masked license validation test names (#5257)

AUTHORS.rst

@@ -108,3 +108,4 @@ The following organizations or individuals have contributed to ScanCode:
 - Yash Sharma @yasharmaster
 - Yunus Rahbar @yns88
 - Stefano Zacchiroli @zacchiro
+- Harish Wargad @harishwargad

CHANGELOG.rst

@@ -8,6 +8,11 @@ Next release
   ``licensedcode-data``.
   https://github.com/aboutcode-org/scancode-toolkit/pull/5056
 
+- Ensure each license rule has a unique generated validation 
+  test name, by renaming colliding rule files and adding a 
+  regression test to prevent rule name collisions.
+  https://github.com/aboutcode-org/scancode-toolkit/issues/5257
+
 - Improve copyright detection for statements with parens or trailing "authors"
 
 

tests/licensedcode/test_rule_file_names.py

@@ -0,0 +1,20 @@
+from collections import defaultdict
+from pathlib import Path
+from commoncode.text import python_safe_name
+
+RULES_DATA_DIR = Path(__file__).parents[2] / "src" / "licensedcode" / "data" / "rules"
+
+def test_rule_file_names_generate_unique_python_names():
+    rule_names_by_python_name = defaultdict(list)
+
+    for rule_file in RULES_DATA_DIR.glob("*.RULE"):
+        python_name = python_safe_name(rule_file.name)
+        rule_names_by_python_name[python_name].append(rule_file.name)
+
+    duplicate_names = {
+        python_name: sorted(rule_names)
+        for python_name, rule_names in rule_names_by_python_name.items()
+        if len(rule_names) > 1
+    }
+
+    assert not duplicate_names, duplicate_names