[refactor] Separate assertion from setup in functional test's test

1 files changed, 32 insertions, 21 deletions

diff --git a/tests/testutils/test_lint_module_output_update.py b/tests/testutils/test_lint_module_output_update.py
index 22da7b0f7..4874c9713 100644
--- a/tests/testutils/test_lint_module_output_update.py
+++ b/tests/testutils/test_lint_module_output_update.py
@@ -90,10 +90,10 @@ def test_lint_module_output_update_remove_useless_txt(
 )
 @pytest.mark.parametrize("directory_path", DIRECTORIES, ids=DIRECTORIES_NAMES)
 def test_update_of_functional_output(directory_path: Path, tmp_path: Path) -> None:
-    """Functional test for the functional tests helper."""
+    """Functional test for the functional tests' helper."""
 
-    def _check_expected_output(ftf: FunctionalTestFile) -> None:
-        new_output_path = ftf.expected_output
+    def _check_expected_output(_ftf: FunctionalTestFile) -> None:
+        new_output_path = _ftf.expected_output
         assert Path(
             new_output_path
         ).exists(), "The expected output file does not exists"
@@ -103,43 +103,54 @@ def test_update_of_functional_output(directory_path: Path, tmp_path: Path) -> No
             new_output == "exec-used:7:0:7:14::Use of exec:UNDEFINED\n"
         ), f"The content was wrongly updated in {new_output_path}"
 
-    new_path = tmp_path / directory_path.name
-    shutil.copytree(directory_path, new_path)
-    for filename in new_path.iterdir():
-        if filename.suffix != ".py":
-            continue
-        new_path_str = str(new_path)
-        ftf = FunctionalTestFile(directory=new_path_str, filename=filename.name)
-        # Standard functional test helper
-        lint_module = LintModuleTest(ftf)
-        # Functional test helper that automatically update the output
-        lint_module_output_update = LintModuleOutputUpdate(ftf)
+    def _assert_behavior_is_correct(
+        _ftf: FunctionalTestFile,
+        _lint_module: LintModuleTest,
+        _lint_module_output_update: LintModuleOutputUpdate,
+        _new_path: Path,
+    ) -> None:
+        new_path_str = str(_new_path)
         if "wrong_test" in new_path_str:
             expected = r'Wrong message\(s\) raised for "exec_used.py"'
             with pytest.raises(AssertionError, match=expected):
-                lint_module.runTest()
+                _lint_module.runTest()
             # When the tests are wrong we do not update the output at all
             # and the test should fail
             with pytest.raises(AssertionError, match=expected):
-                lint_module_output_update.runTest()
+                _lint_module_output_update.runTest()
         elif "ok_test" in new_path_str:
             if any(f"{x}_output" in new_path_str for x in ("wrong", "no", "broken")):
                 with pytest.raises(
                     AssertionError, match='Wrong output for "exec_used.txt"'
                 ):
-                    lint_module.runTest()
+                    _lint_module.runTest()
             elif "ok_output" in new_path_str:
-                lint_module.runTest()
-                _check_expected_output(ftf)
+                _lint_module.runTest()
+                _check_expected_output(_ftf)
             else:
                 raise AssertionError(f"Unhandled test case: {new_path_str}")
 
             # When the tests are ok we update the output whatever it's state
             # was originally
-            lint_module_output_update.runTest()
-            _check_expected_output(ftf)
+            _lint_module_output_update.runTest()
+            _check_expected_output(_ftf)
         else:
             raise AssertionError(
                 f"Do not pollute '{FIXTURE_DIRECTORY}' with unrelated "
                 f"or badly named test files."
             )
+
+    new_path = tmp_path / directory_path.name
+    shutil.copytree(directory_path, new_path)
+    for filename in new_path.iterdir():
+        if filename.suffix != ".py":
+            continue
+        ftf = FunctionalTestFile(directory=str(new_path), filename=filename.name)
+        # Standard functional test helper
+        lint_module = LintModuleTest(ftf)
+        # Functional test helper that automatically update the output
+        lint_module_output_update = LintModuleOutputUpdate(ftf)
+
+        _assert_behavior_is_correct(
+            ftf, lint_module, lint_module_output_update, new_path
+        )