tests: switch some test modules to pytest

Change-Id: I524b5ff2d77f8232f94e21921b00ba4027d2ac4f
Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/563081
Tested-by: Mike Frysinger <vapier@google.com>
Reviewed-by: Gavin Mak <gavinmak@google.com>
Commit-Queue: Mike Frysinger <vapier@google.com>

12 files changed, 717 insertions, 700 deletions

diff --git a/tests/test_color.py b/tests/test_color.py
index 91a1bf250..923f7e355 100644
--- a/tests/test_color.py
+++ b/tests/test_color.py
@@ -15,60 +15,66 @@
 """Unittests for the color.py module."""
 
 import os
-import unittest
+
+import pytest
 
 import color
 import git_config
 
 
-def fixture(*paths):
+def fixture(*paths: str) -> str:
     """Return a path relative to test/fixtures."""
     return os.path.join(os.path.dirname(__file__), "fixtures", *paths)
 
 
-class ColoringTests(unittest.TestCase):
-    """tests of the Coloring class."""
-
-    def setUp(self):
-        """Create a GitConfig object using the test.gitconfig fixture."""
-        config_fixture = fixture("test.gitconfig")
-        self.config = git_config.GitConfig(config_fixture)
-        color.SetDefaultColoring("true")
-        self.color = color.Coloring(self.config, "status")
-
-    def test_Color_Parse_all_params_none(self):
-        """all params are None"""
-        val = self.color._parse(None, None, None, None)
-        self.assertEqual("", val)
-
-    def test_Color_Parse_first_parameter_none(self):
-        """check fg & bg & attr"""
-        val = self.color._parse(None, "black", "red", "ul")
-        self.assertEqual("\x1b[4;30;41m", val)
-
-    def test_Color_Parse_one_entry(self):
-        """check fg"""
-        val = self.color._parse("one", None, None, None)
-        self.assertEqual("\033[33m", val)
-
-    def test_Color_Parse_two_entry(self):
-        """check fg & bg"""
-        val = self.color._parse("two", None, None, None)
-        self.assertEqual("\033[35;46m", val)
-
-    def test_Color_Parse_three_entry(self):
-        """check fg & bg & attr"""
-        val = self.color._parse("three", None, None, None)
-        self.assertEqual("\033[4;30;41m", val)
-
-    def test_Color_Parse_reset_entry(self):
-        """check reset entry"""
-        val = self.color._parse("reset", None, None, None)
-        self.assertEqual("\033[m", val)
-
-    def test_Color_Parse_empty_entry(self):
-        """check empty entry"""
-        val = self.color._parse("none", "blue", "white", "dim")
-        self.assertEqual("\033[2;34;47m", val)
-        val = self.color._parse("empty", "green", "white", "bold")
-        self.assertEqual("\033[1;32;47m", val)
+@pytest.fixture
+def coloring() -> color.Coloring:
+    """Create a Coloring object for testing."""
+    config_fixture = fixture("test.gitconfig")
+    config = git_config.GitConfig(config_fixture)
+    color.SetDefaultColoring("true")
+    return color.Coloring(config, "status")
+
+
+def test_Color_Parse_all_params_none(coloring: color.Coloring) -> None:
+    """all params are None"""
+    val = coloring._parse(None, None, None, None)
+    assert val == ""
+
+
+def test_Color_Parse_first_parameter_none(coloring: color.Coloring) -> None:
+    """check fg & bg & attr"""
+    val = coloring._parse(None, "black", "red", "ul")
+    assert val == "\x1b[4;30;41m"
+
+
+def test_Color_Parse_one_entry(coloring: color.Coloring) -> None:
+    """check fg"""
+    val = coloring._parse("one", None, None, None)
+    assert val == "\033[33m"
+
+
+def test_Color_Parse_two_entry(coloring: color.Coloring) -> None:
+    """check fg & bg"""
+    val = coloring._parse("two", None, None, None)
+    assert val == "\033[35;46m"
+
+
+def test_Color_Parse_three_entry(coloring: color.Coloring) -> None:
+    """check fg & bg & attr"""
+    val = coloring._parse("three", None, None, None)
+    assert val == "\033[4;30;41m"
+
+
+def test_Color_Parse_reset_entry(coloring: color.Coloring) -> None:
+    """check reset entry"""
+    val = coloring._parse("reset", None, None, None)
+    assert val == "\033[m"
+
+
+def test_Color_Parse_empty_entry(coloring: color.Coloring) -> None:
+    """check empty entry"""
+    val = coloring._parse("none", "blue", "white", "dim")
+    assert val == "\033[2;34;47m"
+    val = coloring._parse("empty", "green", "white", "bold")
+    assert val == "\033[1;32;47m"
diff --git a/tests/test_editor.py b/tests/test_editor.py
index 8f5d160e2..d51765dd1 100644
--- a/tests/test_editor.py
+++ b/tests/test_editor.py
@@ -14,43 +14,32 @@
 
 """Unittests for the editor.py module."""
 
-import unittest
+import pytest
 
 from editor import Editor
 
 
-class EditorTestCase(unittest.TestCase):
+@pytest.fixture(autouse=True)
+def reset_editor() -> None:
     """Take care of resetting Editor state across tests."""
+    Editor._editor = None
+    yield
+    Editor._editor = None
 
-    def setUp(self):
-        self.setEditor(None)
 
-    def tearDown(self):
-        self.setEditor(None)
+def test_basic() -> None:
+    """Basic checking of _GetEditor."""
+    Editor._editor = ":"
+    assert Editor._GetEditor() == ":"
 
-    @staticmethod
-    def setEditor(editor):
-        Editor._editor = editor
 
+def test_no_editor() -> None:
+    """Check behavior when no editor is available."""
+    Editor._editor = ":"
+    assert Editor.EditString("foo") == "foo"
 
-class GetEditor(EditorTestCase):
-    """Check GetEditor behavior."""
 
-    def test_basic(self):
-        """Basic checking of _GetEditor."""
-        self.setEditor(":")
-        self.assertEqual(":", Editor._GetEditor())
-
-
-class EditString(EditorTestCase):
-    """Check EditString behavior."""
-
-    def test_no_editor(self):
-        """Check behavior when no editor is available."""
-        self.setEditor(":")
-        self.assertEqual("foo", Editor.EditString("foo"))
-
-    def test_cat_editor(self):
-        """Check behavior when editor is `cat`."""
-        self.setEditor("cat")
-        self.assertEqual("foo", Editor.EditString("foo"))
+def test_cat_editor() -> None:
+    """Check behavior when editor is `cat`."""
+    Editor._editor = "cat"
+    assert Editor.EditString("foo") == "foo"
diff --git a/tests/test_error.py b/tests/test_error.py
index a0ff32c31..3eb82835e 100644
--- a/tests/test_error.py
+++ b/tests/test_error.py
@@ -16,7 +16,9 @@
 
 import inspect
 import pickle
-import unittest
+from typing import Iterator, Type
+
+import pytest
 
 import command
 import error
@@ -26,7 +28,7 @@ import project
 from subcmds import all_modules
 
 
-imports = all_modules + [
+_IMPORTS = all_modules + [
     error,
     project,
     git_command,
@@ -35,36 +37,35 @@ imports = all_modules + [
 ]
 
 
-class PickleTests(unittest.TestCase):
-    """Make sure all our custom exceptions can be pickled."""
+def get_exceptions() -> Iterator[Type[Exception]]:
+    """Return all our custom exceptions."""
+    for entry in _IMPORTS:
+        for name in dir(entry):
+            cls = getattr(entry, name)
+            if isinstance(cls, type) and issubclass(cls, Exception):
+                yield cls
+
 
-    def getExceptions(self):
-        """Return all our custom exceptions."""
-        for entry in imports:
-            for name in dir(entry):
-                cls = getattr(entry, name)
-                if isinstance(cls, type) and issubclass(cls, Exception):
-                    yield cls
+def test_exception_lookup() -> None:
+    """Make sure our introspection logic works."""
+    classes = list(get_exceptions())
+    assert error.HookError in classes
+    # Don't assert the exact number to avoid being a change-detector test.
+    assert len(classes) > 10
 
-    def testExceptionLookup(self):
-        """Make sure our introspection logic works."""
-        classes = list(self.getExceptions())
-        self.assertIn(error.HookError, classes)
-        # Don't assert the exact number to avoid being a change-detector test.
-        self.assertGreater(len(classes), 10)
 
-    def testPickle(self):
-        """Try to pickle all the exceptions."""
-        for cls in self.getExceptions():
-            args = inspect.getfullargspec(cls.__init__).args[1:]
-            obj = cls(*args)
-            p = pickle.dumps(obj)
-            try:
-                newobj = pickle.loads(p)
-            except Exception as e:  # pylint: disable=broad-except
-                self.fail(
-                    "Class %s is unable to be pickled: %s\n"
-                    "Incomplete super().__init__(...) call?" % (cls, e)
-                )
-            self.assertIsInstance(newobj, cls)
-            self.assertEqual(str(obj), str(newobj))
+@pytest.mark.parametrize("cls", get_exceptions())
+def test_pickle(cls: Type[Exception]) -> None:
+    """Try to pickle all the exceptions."""
+    args = inspect.getfullargspec(cls.__init__).args[1:]
+    obj = cls(*args)
+    p = pickle.dumps(obj)
+    try:
+        newobj = pickle.loads(p)
+    except Exception as e:
+        pytest.fail(
+            f"Class {cls} is unable to be pickled: {e}\n"
+            "Incomplete super().__init__(...) call?"
+        )
+    assert isinstance(newobj, cls)
+    assert str(obj) == str(newobj)
diff --git a/tests/test_git_config.py b/tests/test_git_config.py
index e1604bd76..496d97141 100644
--- a/tests/test_git_config.py
+++ b/tests/test_git_config.py
@@ -15,200 +15,219 @@
 """Unittests for the git_config.py module."""
 
 import os
-import tempfile
-import unittest
+from pathlib import Path
+from typing import Any
+
+import pytest
 
 import git_config
 
 
-def fixture(*paths):
+def fixture_path(*paths: str) -> str:
     """Return a path relative to test/fixtures."""
     return os.path.join(os.path.dirname(__file__), "fixtures", *paths)
 
 
-class GitConfigReadOnlyTests(unittest.TestCase):
-    """Read-only tests of the GitConfig class."""
-
-    def setUp(self):
-        """Create a GitConfig object using the test.gitconfig fixture."""
-        config_fixture = fixture("test.gitconfig")
-        self.config = git_config.GitConfig(config_fixture)
-
-    def test_GetString_with_empty_config_values(self):
-        """
-        Test config entries with no value.
-
-        [section]
-            empty
-
-        """
-        val = self.config.GetString("section.empty")
-        self.assertEqual(val, None)
-
-    def test_GetString_with_true_value(self):
-        """
-        Test config entries with a string value.
-
-        [section]
-            nonempty = true
-
-        """
-        val = self.config.GetString("section.nonempty")
-        self.assertEqual(val, "true")
-
-    def test_GetString_from_missing_file(self):
-        """
-        Test missing config file
-        """
-        config_fixture = fixture("not.present.gitconfig")
-        config = git_config.GitConfig(config_fixture)
-        val = config.GetString("empty")
-        self.assertEqual(val, None)
-
-    def test_GetBoolean_undefined(self):
-        """Test GetBoolean on key that doesn't exist."""
-        self.assertIsNone(self.config.GetBoolean("section.missing"))
-
-    def test_GetBoolean_invalid(self):
-        """Test GetBoolean on invalid boolean value."""
-        self.assertIsNone(self.config.GetBoolean("section.boolinvalid"))
-
-    def test_GetBoolean_true(self):
-        """Test GetBoolean on valid true boolean."""
-        self.assertTrue(self.config.GetBoolean("section.booltrue"))
-
-    def test_GetBoolean_false(self):
-        """Test GetBoolean on valid false boolean."""
-        self.assertFalse(self.config.GetBoolean("section.boolfalse"))
-
-    def test_GetInt_undefined(self):
-        """Test GetInt on key that doesn't exist."""
-        self.assertIsNone(self.config.GetInt("section.missing"))
-
-    def test_GetInt_invalid(self):
-        """Test GetInt on invalid integer value."""
-        self.assertIsNone(self.config.GetBoolean("section.intinvalid"))
-
-    def test_GetInt_valid(self):
-        """Test GetInt on valid integers."""
-        TESTS = (
-            ("inthex", 16),
-            ("inthexk", 16384),
-            ("int", 10),
-            ("intk", 10240),
-            ("intm", 10485760),
-            ("intg", 10737418240),
-        )
-        for key, value in TESTS:
-            self.assertEqual(value, self.config.GetInt(f"section.{key}"))
-
-
-class GitConfigReadWriteTests(unittest.TestCase):
-    """Read/write tests of the GitConfig class."""
-
-    def setUp(self):
-        self.tmpfile = tempfile.NamedTemporaryFile()
-        self.config = self.get_config()
-
-    def get_config(self):
-        """Get a new GitConfig instance."""
-        return git_config.GitConfig(self.tmpfile.name)
-
-    def test_SetString(self):
-        """Test SetString behavior."""
-        # Set a value.
-        self.assertIsNone(self.config.GetString("foo.bar"))
-        self.config.SetString("foo.bar", "val")
-        self.assertEqual("val", self.config.GetString("foo.bar"))
-
-        # Make sure the value was actually written out.
-        config = self.get_config()
-        self.assertEqual("val", config.GetString("foo.bar"))
-
-        # Update the value.
-        self.config.SetString("foo.bar", "valll")
-        self.assertEqual("valll", self.config.GetString("foo.bar"))
-        config = self.get_config()
-        self.assertEqual("valll", config.GetString("foo.bar"))
-
-        # Delete the value.
-        self.config.SetString("foo.bar", None)
-        self.assertIsNone(self.config.GetString("foo.bar"))
-        config = self.get_config()
-        self.assertIsNone(config.GetString("foo.bar"))
-
-    def test_SetBoolean(self):
-        """Test SetBoolean behavior."""
-        # Set a true value.
-        self.assertIsNone(self.config.GetBoolean("foo.bar"))
-        for val in (True, 1):
-            self.config.SetBoolean("foo.bar", val)
-            self.assertTrue(self.config.GetBoolean("foo.bar"))
-
-        # Make sure the value was actually written out.
-        config = self.get_config()
-        self.assertTrue(config.GetBoolean("foo.bar"))
-        self.assertEqual("true", config.GetString("foo.bar"))
-
-        # Set a false value.
-        for val in (False, 0):
-            self.config.SetBoolean("foo.bar", val)
-            self.assertFalse(self.config.GetBoolean("foo.bar"))
-
-        # Make sure the value was actually written out.
-        config = self.get_config()
-        self.assertFalse(config.GetBoolean("foo.bar"))
-        self.assertEqual("false", config.GetString("foo.bar"))
-
-        # Delete the value.
-        self.config.SetBoolean("foo.bar", None)
-        self.assertIsNone(self.config.GetBoolean("foo.bar"))
-        config = self.get_config()
-        self.assertIsNone(config.GetBoolean("foo.bar"))
-
-    def test_SetInt(self):
-        """Test SetInt behavior."""
-        # Set a value.
-        self.assertIsNone(self.config.GetInt("foo.bar"))
-        self.config.SetInt("foo.bar", 10)
-        self.assertEqual(10, self.config.GetInt("foo.bar"))
-
-        # Make sure the value was actually written out.
-        config = self.get_config()
-        self.assertEqual(10, config.GetInt("foo.bar"))
-        self.assertEqual("10", config.GetString("foo.bar"))
-
-        # Update the value.
-        self.config.SetInt("foo.bar", 20)
-        self.assertEqual(20, self.config.GetInt("foo.bar"))
-        config = self.get_config()
-        self.assertEqual(20, config.GetInt("foo.bar"))
-
-        # Delete the value.
-        self.config.SetInt("foo.bar", None)
-        self.assertIsNone(self.config.GetInt("foo.bar"))
-        config = self.get_config()
-        self.assertIsNone(config.GetInt("foo.bar"))
-
-    def test_GetSyncAnalysisStateData(self):
-        """Test config entries with a sync state analysis data."""
-        superproject_logging_data = {}
-        superproject_logging_data["test"] = False
-        options = type("options", (object,), {})()
-        options.verbose = "true"
-        options.mp_update = "false"
-        TESTS = (
-            ("superproject.test", "false"),
-            ("options.verbose", "true"),
-            ("options.mpupdate", "false"),
-            ("main.version", "1"),
-        )
-        self.config.UpdateSyncAnalysisState(options, superproject_logging_data)
-        sync_data = self.config.GetSyncAnalysisStateData()
-        for key, value in TESTS:
-            self.assertEqual(
-                sync_data[f"{git_config.SYNC_STATE_PREFIX}{key}"], value
-            )
-        self.assertTrue(
-            sync_data[f"{git_config.SYNC_STATE_PREFIX}main.synctime"]
-        )
+@pytest.fixture
+def readonly_config() -> git_config.GitConfig:
+    """Create a GitConfig object using the test.gitconfig fixture."""
+    config_fixture = fixture_path("test.gitconfig")
+    return git_config.GitConfig(config_fixture)
+
+
+def test_get_string_with_empty_config_values(
+    readonly_config: git_config.GitConfig,
+) -> None:
+    """Test config entries with no value.
+
+    [section]
+        empty
+
+    """
+    val = readonly_config.GetString("section.empty")
+    assert val is None
+
+
+def test_get_string_with_true_value(
+    readonly_config: git_config.GitConfig,
+) -> None:
+    """Test config entries with a string value.
+
+    [section]
+        nonempty = true
+
+    """
+    val = readonly_config.GetString("section.nonempty")
+    assert val == "true"
+
+
+def test_get_string_from_missing_file() -> None:
+    """Test missing config file."""
+    config_fixture = fixture_path("not.present.gitconfig")
+    config = git_config.GitConfig(config_fixture)
+    val = config.GetString("empty")
+    assert val is None
+
+
+def test_get_boolean_undefined(readonly_config: git_config.GitConfig) -> None:
+    """Test GetBoolean on key that doesn't exist."""
+    assert readonly_config.GetBoolean("section.missing") is None
+
+
+def test_get_boolean_invalid(readonly_config: git_config.GitConfig) -> None:
+    """Test GetBoolean on invalid boolean value."""
+    assert readonly_config.GetBoolean("section.boolinvalid") is None
+
+
+def test_get_boolean_true(readonly_config: git_config.GitConfig) -> None:
+    """Test GetBoolean on valid true boolean."""
+    assert readonly_config.GetBoolean("section.booltrue") is True
+
+
+def test_get_boolean_false(readonly_config: git_config.GitConfig) -> None:
+    """Test GetBoolean on valid false boolean."""
+    assert readonly_config.GetBoolean("section.boolfalse") is False
+
+
+def test_get_int_undefined(readonly_config: git_config.GitConfig) -> None:
+    """Test GetInt on key that doesn't exist."""
+    assert readonly_config.GetInt("section.missing") is None
+
+
+def test_get_int_invalid(readonly_config: git_config.GitConfig) -> None:
+    """Test GetInt on invalid integer value."""
+    assert readonly_config.GetInt("section.intinvalid") is None
+
+
+@pytest.mark.parametrize(
+    "key, expected",
+    (
+        ("inthex", 16),
+        ("inthexk", 16384),
+        ("int", 10),
+        ("intk", 10240),
+        ("intm", 10485760),
+        ("intg", 10737418240),
+    ),
+)
+def test_get_int_valid(
+    readonly_config: git_config.GitConfig, key: str, expected: int
+) -> None:
+    """Test GetInt on valid integers."""
+    assert readonly_config.GetInt(f"section.{key}") == expected
+
+
+@pytest.fixture
+def rw_config_file(tmp_path: Path) -> Path:
+    """Return a path to a temporary config file."""
+    return tmp_path / "config"
+
+
+def test_set_string(rw_config_file: Path) -> None:
+    """Test SetString behavior."""
+    config = git_config.GitConfig(str(rw_config_file))
+
+    # Set a value.
+    assert config.GetString("foo.bar") is None
+    config.SetString("foo.bar", "val")
+    assert config.GetString("foo.bar") == "val"
+
+    # Make sure the value was actually written out.
+    config2 = git_config.GitConfig(str(rw_config_file))
+    assert config2.GetString("foo.bar") == "val"
+
+    # Update the value.
+    config.SetString("foo.bar", "valll")
+    assert config.GetString("foo.bar") == "valll"
+    config3 = git_config.GitConfig(str(rw_config_file))
+    assert config3.GetString("foo.bar") == "valll"
+
+    # Delete the value.
+    config.SetString("foo.bar", None)
+    assert config.GetString("foo.bar") is None
+    config4 = git_config.GitConfig(str(rw_config_file))
+    assert config4.GetString("foo.bar") is None
+
+
+def test_set_boolean(rw_config_file: Path) -> None:
+    """Test SetBoolean behavior."""
+    config = git_config.GitConfig(str(rw_config_file))
+
+    # Set a true value.
+    assert config.GetBoolean("foo.bar") is None
+    for val in (True, 1):
+        config.SetBoolean("foo.bar", val)
+        assert config.GetBoolean("foo.bar") is True
+
+    # Make sure the value was actually written out.
+    config2 = git_config.GitConfig(str(rw_config_file))
+    assert config2.GetBoolean("foo.bar") is True
+    assert config2.GetString("foo.bar") == "true"
+
+    # Set a false value.
+    for val in (False, 0):
+        config.SetBoolean("foo.bar", val)
+        assert config.GetBoolean("foo.bar") is False
+
+    # Make sure the value was actually written out.
+    config3 = git_config.GitConfig(str(rw_config_file))
+    assert config3.GetBoolean("foo.bar") is False
+    assert config3.GetString("foo.bar") == "false"
+
+    # Delete the value.
+    config.SetBoolean("foo.bar", None)
+    assert config.GetBoolean("foo.bar") is None
+    config4 = git_config.GitConfig(str(rw_config_file))
+    assert config4.GetBoolean("foo.bar") is None
+
+
+def test_set_int(rw_config_file: Path) -> None:
+    """Test SetInt behavior."""
+    config = git_config.GitConfig(str(rw_config_file))
+
+    # Set a value.
+    assert config.GetInt("foo.bar") is None
+    config.SetInt("foo.bar", 10)
+    assert config.GetInt("foo.bar") == 10
+
+    # Make sure the value was actually written out.
+    config2 = git_config.GitConfig(str(rw_config_file))
+    assert config2.GetInt("foo.bar") == 10
+    assert config2.GetString("foo.bar") == "10"
+
+    # Update the value.
+    config.SetInt("foo.bar", 20)
+    assert config.GetInt("foo.bar") == 20
+    config3 = git_config.GitConfig(str(rw_config_file))
+    assert config3.GetInt("foo.bar") == 20
+
+    # Delete the value.
+    config.SetInt("foo.bar", None)
+    assert config.GetInt("foo.bar") is None
+    config4 = git_config.GitConfig(str(rw_config_file))
+    assert config4.GetInt("foo.bar") is None
+
+
+def test_get_sync_analysis_state_data(rw_config_file: Path) -> None:
+    """Test config entries with a sync state analysis data."""
+    config = git_config.GitConfig(str(rw_config_file))
+    superproject_logging_data: dict[str, Any] = {"test": False}
+
+    class Options:
+        """Container for testing."""
+
+    options = Options()
+    options.verbose = "true"
+    options.mp_update = "false"
+
+    TESTS = (
+        ("superproject.test", "false"),
+        ("options.verbose", "true"),
+        ("options.mpupdate", "false"),
+        ("main.version", "1"),
+    )
+    config.UpdateSyncAnalysisState(options, superproject_logging_data)
+    sync_data = config.GetSyncAnalysisStateData()
+    for key, value in TESTS:
+        assert sync_data[f"{git_config.SYNC_STATE_PREFIX}{key}"] == value
+    assert sync_data[f"{git_config.SYNC_STATE_PREFIX}main.synctime"]
diff --git a/tests/test_hooks.py b/tests/test_hooks.py
index 76e928f7a..9d52d1849 100644
--- a/tests/test_hooks.py
+++ b/tests/test_hooks.py
@@ -14,42 +14,47 @@
 
 """Unittests for the hooks.py module."""
 
-import unittest
+import pytest
 
 import hooks
 
 
-class RepoHookShebang(unittest.TestCase):
-    """Check shebang parsing in RepoHook."""
-
-    def test_no_shebang(self):
-        """Lines w/out shebangs should be rejected."""
-        DATA = ("", "#\n# foo\n", "# Bad shebang in script\n#!/foo\n")
-        for data in DATA:
-            self.assertIsNone(hooks.RepoHook._ExtractInterpFromShebang(data))
-
-    def test_direct_interp(self):
-        """Lines whose shebang points directly to the interpreter."""
-        DATA = (
-            ("#!/foo", "/foo"),
-            ("#! /foo", "/foo"),
-            ("#!/bin/foo ", "/bin/foo"),
-            ("#! /usr/foo ", "/usr/foo"),
-            ("#! /usr/foo -args", "/usr/foo"),
-        )
-        for shebang, interp in DATA:
-            self.assertEqual(
-                hooks.RepoHook._ExtractInterpFromShebang(shebang), interp
-            )
-
-    def test_env_interp(self):
-        """Lines whose shebang launches through `env`."""
-        DATA = (
-            ("#!/usr/bin/env foo", "foo"),
-            ("#!/bin/env foo", "foo"),
-            ("#! /bin/env /bin/foo ", "/bin/foo"),
-        )
-        for shebang, interp in DATA:
-            self.assertEqual(
-                hooks.RepoHook._ExtractInterpFromShebang(shebang), interp
-            )
+@pytest.mark.parametrize(
+    "data",
+    (
+        "",
+        "#\n# foo\n",
+        "# Bad shebang in script\n#!/foo\n",
+    ),
+)
+def test_no_shebang(data: str) -> None:
+    """Lines w/out shebangs should be rejected."""
+    assert hooks.RepoHook._ExtractInterpFromShebang(data) is None
+
+
+@pytest.mark.parametrize(
+    "shebang, interp",
+    (
+        ("#!/foo", "/foo"),
+        ("#! /foo", "/foo"),
+        ("#!/bin/foo ", "/bin/foo"),
+        ("#! /usr/foo ", "/usr/foo"),
+        ("#! /usr/foo -args", "/usr/foo"),
+    ),
+)
+def test_direct_interp(shebang: str, interp: str) -> None:
+    """Lines whose shebang points directly to the interpreter."""
+    assert hooks.RepoHook._ExtractInterpFromShebang(shebang) == interp
+
+
+@pytest.mark.parametrize(
+    "shebang, interp",
+    (
+        ("#!/usr/bin/env foo", "foo"),
+        ("#!/bin/env foo", "foo"),
+        ("#! /bin/env /bin/foo ", "/bin/foo"),
+    ),
+)
+def test_env_interp(shebang: str, interp: str) -> None:
+    """Lines whose shebang launches through `env`."""
+    assert hooks.RepoHook._ExtractInterpFromShebang(shebang) == interp
diff --git a/tests/test_platform_utils.py b/tests/test_platform_utils.py
index aab8a52ab..e89965391 100644
--- a/tests/test_platform_utils.py
+++ b/tests/test_platform_utils.py
@@ -14,39 +14,35 @@
 
 """Unittests for the platform_utils.py module."""
 
-import os
-import tempfile
-import unittest
+from pathlib import Path
 
-import platform_utils
+import pytest
 
+import platform_utils
 
-class RemoveTests(unittest.TestCase):
-    """Check remove() helper."""
 
-    def testMissingOk(self):
-        """Check missing_ok handling."""
-        with tempfile.TemporaryDirectory() as tmpdir:
-            path = os.path.join(tmpdir, "test")
+def test_remove_missing_ok(tmp_path: Path) -> None:
+    """Check missing_ok handling."""
+    path = tmp_path / "test"
 
-            # Should not fail.
-            platform_utils.remove(path, missing_ok=True)
+    # Should not fail.
+    platform_utils.remove(path, missing_ok=True)
 
-            # Should fail.
-            self.assertRaises(OSError, platform_utils.remove, path)
-            self.assertRaises(
-                OSError, platform_utils.remove, path, missing_ok=False
-            )
+    # Should fail.
+    with pytest.raises(OSError):
+        platform_utils.remove(path)
+    with pytest.raises(OSError):
+        platform_utils.remove(path, missing_ok=False)
 
-            # Should not fail if it exists.
-            open(path, "w").close()
-            platform_utils.remove(path, missing_ok=True)
-            self.assertFalse(os.path.exists(path))
+    # Should not fail if it exists.
+    path.touch()
+    platform_utils.remove(path, missing_ok=True)
+    assert not path.exists()
 
-            open(path, "w").close()
-            platform_utils.remove(path)
-            self.assertFalse(os.path.exists(path))
+    path.touch()
+    platform_utils.remove(path)
+    assert not path.exists()
 
-            open(path, "w").close()
-            platform_utils.remove(path, missing_ok=False)
-            self.assertFalse(os.path.exists(path))
+    path.touch()
+    platform_utils.remove(path, missing_ok=False)
+    assert not path.exists()
diff --git a/tests/test_repo_logging.py b/tests/test_repo_logging.py
index e072039ed..c5bba6b8e 100644
--- a/tests/test_repo_logging.py
+++ b/tests/test_repo_logging.py
@@ -12,90 +12,96 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Unit test for repo_logging module."""
+"""Unittests for the repo_logging.py module."""
 
 import contextlib
 import io
 import logging
-import unittest
+import re
 from unittest import mock
 
+import pytest
+
 from color import SetDefaultColoring
 from error import RepoExitError
 from repo_logging import RepoLogger
 
 
-class TestRepoLogger(unittest.TestCase):
-    @mock.patch.object(RepoLogger, "error")
-    def test_log_aggregated_errors_logs_aggregated_errors(self, mock_error):
-        """Test if log_aggregated_errors logs a list of aggregated errors."""
-        logger = RepoLogger(__name__)
-        logger.log_aggregated_errors(
-            RepoExitError(
-                aggregate_errors=[
-                    Exception("foo"),
-                    Exception("bar"),
-                    Exception("baz"),
-                    Exception("hello"),
-                    Exception("world"),
-                    Exception("test"),
-                ]
-            )
-        )
-
-        mock_error.assert_has_calls(
-            [
-                mock.call("=" * 80),
-                mock.call(
-                    "Repo command failed due to the following `%s` errors:",
-                    "RepoExitError",
-                ),
-                mock.call("foo\nbar\nbaz\nhello\nworld"),
-                mock.call("+%d additional errors...", 1),
+@mock.patch.object(RepoLogger, "error")
+def test_log_aggregated_errors_logs_aggregated_errors(mock_error) -> None:
+    """Test if log_aggregated_errors logs a list of aggregated errors."""
+    logger = RepoLogger(__name__)
+    logger.log_aggregated_errors(
+        RepoExitError(
+            aggregate_errors=[
+                Exception("foo"),
+                Exception("bar"),
+                Exception("baz"),
+                Exception("hello"),
+                Exception("world"),
+                Exception("test"),
             ]
         )
+    )
 
-    @mock.patch.object(RepoLogger, "error")
-    def test_log_aggregated_errors_logs_single_error(self, mock_error):
-        """Test if log_aggregated_errors logs empty aggregated_errors."""
-        logger = RepoLogger(__name__)
-        logger.log_aggregated_errors(RepoExitError())
+    mock_error.assert_has_calls(
+        [
+            mock.call("=" * 80),
+            mock.call(
+                "Repo command failed due to the following `%s` errors:",
+                "RepoExitError",
+            ),
+            mock.call("foo\nbar\nbaz\nhello\nworld"),
+            mock.call("+%d additional errors...", 1),
+        ]
+    )
 
-        mock_error.assert_has_calls(
-            [
-                mock.call("=" * 80),
-                mock.call("Repo command failed: %s", "RepoExitError"),
-            ]
-        )
 
-    def test_log_with_format_string(self):
-        """Test different log levels with format strings."""
-
-        # Set color output to "always" for consistent test results.
-        # This ensures the logger's behavior is uniform across different
-        # environments and git configurations.
-        SetDefaultColoring("always")
-
-        # Regex pattern to match optional ANSI color codes.
-        # \033    - Escape character
-        # \[      - Opening square bracket
-        # [0-9;]* - Zero or more digits or semicolons
-        # m       - Ending 'm' character
-        # ?       - Makes the entire group optional
-        opt_color = r"(\033\[[0-9;]*m)?"
-
-        for level in (logging.INFO, logging.WARN, logging.ERROR):
-            name = logging.getLevelName(level)
-
-            with self.subTest(level=level, name=name):
-                output = io.StringIO()
-
-                with contextlib.redirect_stderr(output):
-                    logger = RepoLogger(__name__)
-                    logger.log(level, "%s", "100% pass")
-
-                self.assertRegex(
-                    output.getvalue().strip(),
-                    f"^{opt_color}100% pass{opt_color}$",
-                    f"failed for level {name}",
-                )
+@mock.patch.object(RepoLogger, "error")
+def test_log_aggregated_errors_logs_single_error(mock_error) -> None:
+    """Test if log_aggregated_errors logs empty aggregated_errors."""
+    logger = RepoLogger(__name__)
+    logger.log_aggregated_errors(RepoExitError())
+
+    mock_error.assert_has_calls(
+        [
+            mock.call("=" * 80),
+            mock.call("Repo command failed: %s", "RepoExitError"),
+        ]
+    )
+
+
+@pytest.mark.parametrize(
+    "level",
+    (
+        logging.INFO,
+        logging.WARN,
+        logging.ERROR,
+    ),
+)
+def test_log_with_format_string(level: int) -> None:
+    """Test different log levels with format strings."""
+    name = logging.getLevelName(level)
+
+    # Set color output to "always" for consistent test results.
+    # This ensures the logger's behavior is uniform across different
+    # environments and git configurations.
+    SetDefaultColoring("always")
+
+    # Regex pattern to match optional ANSI color codes.
+    # \033    - Escape character
+    # \[      - Opening square bracket
+    # [0-9;]* - Zero or more digits or semicolons
+    # m       - Ending 'm' character
+    # ?       - Makes the entire group optional
+    opt_color = r"(\033\[[0-9;]*m)?"
+
+    output = io.StringIO()
+
+    with contextlib.redirect_stderr(output):
+        logger = RepoLogger(__name__)
+        logger.log(level, "%s", "100% pass")
+
+    assert re.search(
+        f"^{opt_color}100% pass{opt_color}$", output.getvalue().strip()
+    ), f"failed for level {name}"
diff --git a/tests/test_repo_trace.py b/tests/test_repo_trace.py
index 2ea341025..3ec540b25 100644
--- a/tests/test_repo_trace.py
+++ b/tests/test_repo_trace.py
@@ -15,46 +15,37 @@
 """Unittests for the repo_trace.py module."""
 
 import os
-import unittest
-from unittest import mock
+
+import pytest
 
 import repo_trace
 
 
-class TraceTests(unittest.TestCase):
+def test_trace_max_size_enforced(monkeypatch: pytest.MonkeyPatch) -> None:
     """Check Trace behavior."""
-
-    def testTrace_MaxSizeEnforced(self):
-        content = "git chicken"
-
-        with repo_trace.Trace(content, first_trace=True):
-            pass
-        first_trace_size = os.path.getsize(repo_trace._TRACE_FILE)
-
-        with repo_trace.Trace(content):
-            pass
-        self.assertGreater(
-            os.path.getsize(repo_trace._TRACE_FILE), first_trace_size
-        )
-
-        # Check we clear everything is the last chunk is larger than _MAX_SIZE.
-        with mock.patch("repo_trace._MAX_SIZE", 0):
-            with repo_trace.Trace(content, first_trace=True):
-                pass
-            self.assertEqual(
-                first_trace_size, os.path.getsize(repo_trace._TRACE_FILE)
-            )
-
-            # Check we only clear the chunks we need to.
-            repo_trace._MAX_SIZE = (first_trace_size + 1) / (1024 * 1024)
-            with repo_trace.Trace(content, first_trace=True):
-                pass
-            self.assertEqual(
-                first_trace_size * 2, os.path.getsize(repo_trace._TRACE_FILE)
-            )
-
-            with repo_trace.Trace(content, first_trace=True):
-                pass
-            self.assertEqual(
-                first_trace_size * 2, os.path.getsize(repo_trace._TRACE_FILE)
-            )
+    content = "git chicken"
+
+    with repo_trace.Trace(content, first_trace=True):
+        pass
+    first_trace_size = os.path.getsize(repo_trace._TRACE_FILE)
+
+    with repo_trace.Trace(content):
+        pass
+    assert os.path.getsize(repo_trace._TRACE_FILE) > first_trace_size
+
+    # Check we clear everything if the last chunk is larger than _MAX_SIZE.
+    monkeypatch.setattr(repo_trace, "_MAX_SIZE", 0)
+    with repo_trace.Trace(content, first_trace=True):
+        pass
+    assert os.path.getsize(repo_trace._TRACE_FILE) == first_trace_size
+
+    # Check we only clear the chunks we need to.
+    new_max = (first_trace_size + 1) / (1024 * 1024)
+    monkeypatch.setattr(repo_trace, "_MAX_SIZE", new_max)
+    with repo_trace.Trace(content, first_trace=True):
+        pass
+    assert os.path.getsize(repo_trace._TRACE_FILE) == first_trace_size * 2
+
+    with repo_trace.Trace(content, first_trace=True):
+        pass
+    assert os.path.getsize(repo_trace._TRACE_FILE) == first_trace_size * 2
diff --git a/tests/test_ssh.py b/tests/test_ssh.py
index ce65fac45..6bb2c2f4e 100644
--- a/tests/test_ssh.py
+++ b/tests/test_ssh.py
@@ -16,72 +16,84 @@
 
 import multiprocessing
 import subprocess
-import unittest
+from typing import Tuple
 from unittest import mock
 
+import pytest
+
 import ssh
 
 
-class SshTests(unittest.TestCase):
-    """Tests the ssh functions."""
+@pytest.fixture(autouse=True)
+def clear_ssh_version_cache() -> None:
+    """Clear the ssh version cache before each test."""
+    ssh.version.cache_clear()
 
-    def setUp(self) -> None:
-        super().setUp()
-        ssh.version.cache_clear()
 
-    def test_parse_ssh_version(self):
-        """Check _parse_ssh_version() handling."""
-        ver = ssh._parse_ssh_version("Unknown\n")
-        self.assertEqual(ver, ())
-        ver = ssh._parse_ssh_version("OpenSSH_1.0\n")
-        self.assertEqual(ver, (1, 0))
-        ver = ssh._parse_ssh_version(
-            "OpenSSH_6.6.1p1 Ubuntu-2ubuntu2.13, OpenSSL 1.0.1f 6 Jan 2014\n"
-        )
-        self.assertEqual(ver, (6, 6, 1))
-        ver = ssh._parse_ssh_version(
-            "OpenSSH_7.6p1 Ubuntu-4ubuntu0.3, OpenSSL 1.0.2n  7 Dec 2017\n"
-        )
-        self.assertEqual(ver, (7, 6))
-        ver = ssh._parse_ssh_version("OpenSSH_9.0p1, LibreSSL 3.3.6\n")
-        self.assertEqual(ver, (9, 0))
-
-    def test_version(self):
-        """Check version() handling."""
-        with mock.patch("ssh._run_ssh_version", return_value="OpenSSH_1.2\n"):
-            self.assertEqual(ssh.version(), (1, 2))
-
-    def test_context_manager_empty(self):
-        """Verify context manager with no clients works correctly."""
-        with multiprocessing.Manager() as manager:
-            with ssh.ProxyManager(manager):
-                pass
-
-    def test_context_manager_child_cleanup(self):
-        """Verify orphaned clients & masters get cleaned up."""
-        with multiprocessing.Manager() as manager:
-            with mock.patch("ssh.version", return_value=(1, 2)):
-                with ssh.ProxyManager(manager) as ssh_proxy:
-                    client = subprocess.Popen(["sleep", "964853320"])
-                    ssh_proxy.add_client(client)
-                    master = subprocess.Popen(["sleep", "964853321"])
-                    ssh_proxy.add_master(master)
-        # If the process still exists, these will throw timeout errors.
-        client.wait(0)
-        master.wait(0)
-
-    def test_ssh_sock(self):
-        """Check sock() function."""
-        manager = multiprocessing.Manager()
+@pytest.mark.parametrize(
+    "input_str, expected",
+    (
+        ("Unknown\n", ()),
+        ("OpenSSH_1.0\n", (1, 0)),
+        (
+            "OpenSSH_6.6.1p1 Ubuntu-2ubuntu2.13, OpenSSL 1.0.1f 6 Jan 2014\n",
+            (6, 6, 1),
+        ),
+        (
+            "OpenSSH_7.6p1 Ubuntu-4ubuntu0.3, OpenSSL 1.0.2n  7 Dec 2017\n",
+            (7, 6),
+        ),
+        ("OpenSSH_9.0p1, LibreSSL 3.3.6\n", (9, 0)),
+    ),
+)
+def test_parse_ssh_version(input_str: str, expected: Tuple[int, ...]) -> None:
+    """Check _parse_ssh_version() handling."""
+    assert ssh._parse_ssh_version(input_str) == expected
+
+
+def test_version() -> None:
+    """Check version() handling."""
+    with mock.patch("ssh._run_ssh_version", return_value="OpenSSH_1.2\n"):
+        assert ssh.version() == (1, 2)
+
+
+def test_context_manager_empty() -> None:
+    """Verify context manager with no clients works correctly."""
+    with multiprocessing.Manager() as manager:
+        with ssh.ProxyManager(manager):
+            pass
+
+
+def test_context_manager_child_cleanup() -> None:
+    """Verify orphaned clients & masters get cleaned up."""
+    with multiprocessing.Manager() as manager:
+        with mock.patch("ssh.version", return_value=(1, 2)):
+            with ssh.ProxyManager(manager) as ssh_proxy:
+                client = subprocess.Popen(["sleep", "964853320"])
+                ssh_proxy.add_client(client)
+                master = subprocess.Popen(["sleep", "964853321"])
+                ssh_proxy.add_master(master)
+    # If the process still exists, these will throw timeout errors.
+    client.wait(0)
+    master.wait(0)
+
+
+def test_ssh_sock(monkeypatch: pytest.MonkeyPatch) -> None:
+    """Check sock() function."""
+    with multiprocessing.Manager() as manager:
         proxy = ssh.ProxyManager(manager)
-        with mock.patch("tempfile.mkdtemp", return_value="/tmp/foo"):
-            # Old ssh version uses port.
-            with mock.patch("ssh.version", return_value=(6, 6)):
-                with proxy as ssh_proxy:
-                    self.assertTrue(ssh_proxy.sock().endswith("%p"))
-
-            proxy._sock_path = None
-            # New ssh version uses hash.
-            with mock.patch("ssh.version", return_value=(6, 7)):
-                with proxy as ssh_proxy:
-                    self.assertTrue(ssh_proxy.sock().endswith("%C"))
+        monkeypatch.setattr(
+            "tempfile.mkdtemp", lambda *args, **kwargs: "/tmp/foo"
+        )
+
+        # Old ssh version uses port.
+        with mock.patch("ssh.version", return_value=(6, 6)):
+            with proxy as ssh_proxy:
+                assert ssh_proxy.sock().endswith("%p")
+
+        proxy._sock_path = None
+        # New ssh version uses hash.
+        with mock.patch("ssh.version", return_value=(6, 7)):
+            with proxy as ssh_proxy:
+                assert ssh_proxy.sock().endswith("%C")
+        proxy._sock_path = None
diff --git a/tests/test_subcmds.py b/tests/test_subcmds.py
index 3382d822d..12bbb2700 100644
--- a/tests/test_subcmds.py
+++ b/tests/test_subcmds.py
@@ -15,170 +15,164 @@
 """Unittests for the subcmds module (mostly __init__.py than subcommands)."""
 
 import optparse
-import unittest
-
-import subcmds
+from typing import Type
 
+import pytest
 
-class AllCommands(unittest.TestCase):
-    """Check registered all_commands."""
+from command import Command
+import subcmds
 
-    def test_required_basic(self):
-        """Basic checking of registered commands."""
-        # NB: We don't test all subcommands as we want to avoid "change
-        # detection" tests, so we just look for the most common/important ones
-        # here that are unlikely to ever change.
-        for cmd in {"cherry-pick", "help", "init", "start", "sync", "upload"}:
-            self.assertIn(cmd, subcmds.all_commands)
 
-    def test_naming(self):
-        """Verify we don't add things that we shouldn't."""
-        for cmd in subcmds.all_commands:
-            # Reject filename suffixes like "help.py".
-            self.assertNotIn(".", cmd)
+# NB: We don't test all subcommands as we want to avoid "change detection"
+# tests, so we just look for the most common/important ones here that are
+# unlikely to ever change.
+@pytest.mark.parametrize(
+    "cmd", ("cherry-pick", "help", "init", "start", "sync", "upload")
+)
+def test_required_basic(cmd: str) -> None:
+    """Basic checking of registered commands."""
+    assert cmd in subcmds.all_commands
+
+
+@pytest.mark.parametrize("name", subcmds.all_commands.keys())
+def test_naming(name: str) -> None:
+    """Verify we don't add things that we shouldn't."""
+    # Reject filename suffixes like "help.py".
+    assert "." not in name
+
+    # Make sure all '_' were converted to '-'.
+    assert "_" not in name
+
+    # Reject internal python paths like "__init__".
+    assert not name.startswith("__")
+
+
+@pytest.mark.parametrize("name, cls", subcmds.all_commands.items())
+def test_help_desc_style(name: str, cls: Type[Command]) -> None:
+    """Force some consistency in option descriptions.
+
+    Python's optparse & argparse has a few default options like --help.
+    Their option description text uses lowercase sentence fragments, so
+    enforce our options follow the same style so UI is consistent.
+
+    We enforce:
+    * Text starts with lowercase.
+    * Text doesn't end with period.
+    """
+    cmd = cls()
+    parser = cmd.OptionParser
+    for option in parser.option_list:
+        if option.help == optparse.SUPPRESS_HELP or not option.help:
+            continue
+
+        c = option.help[0]
+        assert c.lower() == c, (
+            f"subcmds/{name}.py: {option.get_opt_string()}: "
+            f'help text should start with lowercase: "{option.help}"'
+        )
+
+        assert option.help[-1] != ".", (
+            f"subcmds/{name}.py: {option.get_opt_string()}: "
+            f'help text should not end in a period: "{option.help}"'
+        )
+
+
+@pytest.mark.parametrize("name, cls", subcmds.all_commands.items())
+def test_cli_option_style(name: str, cls: Type[Command]) -> None:
+    """Force some consistency in option flags."""
+    cmd = cls()
+    parser = cmd.OptionParser
+    for option in parser.option_list:
+        for opt in option._long_opts:
+            assert "_" not in opt, (
+                f"subcmds/{name}.py: {opt}: only use dashes in "
+                "options, not underscores"
+            )
 
-            # Make sure all '_' were converted to '-'.
-            self.assertNotIn("_", cmd)
 
-            # Reject internal python paths like "__init__".
-            self.assertFalse(cmd.startswith("__"))
+def test_cli_option_dest() -> None:
+    """Block redundant dest= arguments."""
+    bad_opts: list[tuple[str, str]] = []
 
-    def test_help_desc_style(self):
-        """Force some consistency in option descriptions.
+    def _check_dest(opt: optparse.Option) -> None:
+        """Check the dest= setting."""
+        # If the destination is not set, nothing to check.
+        # If long options are not set, then there's no implicit destination.
+        # If callback is used, then a destination might be needed because
+        # optparse cannot assume a value is always stored.
+        if opt.dest is None or not opt._long_opts or opt.callback:
+            return
 
-        Python's optparse & argparse has a few default options like --help.
-        Their option description text uses lowercase sentence fragments, so
-        enforce our options follow the same style so UI is consistent.
+        long = opt._long_opts[0]
+        assert long.startswith("--")
+        # This matches optparse's behavior.
+        implicit_dest = long[2:].replace("-", "_")
+        if implicit_dest == opt.dest:
+            bad_opts.append((str(opt), opt.dest))
 
-        We enforce:
-        * Text starts with lowercase.
-        * Text doesn't end with period.
-        """
-        for name, cls in subcmds.all_commands.items():
-            cmd = cls()
-            parser = cmd.OptionParser
-            for option in parser.option_list:
-                if option.help == optparse.SUPPRESS_HELP:
-                    continue
-
-                c = option.help[0]
-                self.assertEqual(
-                    c.lower(),
-                    c,
-                    msg=f"subcmds/{name}.py: {option.get_opt_string()}: "
-                    f'help text should start with lowercase: "{option.help}"',
-                )
-
-                self.assertNotEqual(
-                    option.help[-1],
-                    ".",
-                    msg=f"subcmds/{name}.py: {option.get_opt_string()}: "
-                    f'help text should not end in a period: "{option.help}"',
-                )
-
-    def test_cli_option_style(self):
-        """Force some consistency in option flags."""
-        for name, cls in subcmds.all_commands.items():
-            cmd = cls()
-            parser = cmd.OptionParser
-            for option in parser.option_list:
-                for opt in option._long_opts:
-                    self.assertNotIn(
-                        "_",
-                        opt,
-                        msg=f"subcmds/{name}.py: {opt}: only use dashes in "
-                        "options, not underscores",
-                    )
-
-    def test_cli_option_dest(self):
-        """Block redundant dest= arguments."""
-
-        def _check_dest(opt):
-            """Check the dest= setting."""
-            # If the destination is not set, nothing to check.
-            # If long options are not set, then there's no implicit destination.
-            # If callback is used, then a destination might be needed because
-            # optparse cannot assume a value is always stored.
-            if opt.dest is None or not opt._long_opts or opt.callback:
-                return
-
-            long = opt._long_opts[0]
-            assert long.startswith("--")
-            # This matches optparse's behavior.
-            implicit_dest = long[2:].replace("-", "_")
-            if implicit_dest == opt.dest:
-                bad_opts.append((str(opt), opt.dest))
-
-        # Hook the option check list.
-        optparse.Option.CHECK_METHODS.insert(0, _check_dest)
+    # Hook the option check list.
+    optparse.Option.CHECK_METHODS.insert(0, _check_dest)
 
+    try:
         # Gather all the bad options up front so people can see all bad options
         # instead of failing at the first one.
-        all_bad_opts = {}
+        all_bad_opts: dict[str, list[tuple[str, str]]] = {}
         for name, cls in subcmds.all_commands.items():
-            bad_opts = all_bad_opts[name] = []
+            bad_opts = []
             cmd = cls()
             # Trigger construction of parser.
-            cmd.OptionParser
+            _ = cmd.OptionParser
+            all_bad_opts[name] = bad_opts
 
-        errmsg = None
-        for name, bad_opts in sorted(all_bad_opts.items()):
-            if bad_opts:
+        errmsg = ""
+        for name, bad_opts_list in sorted(all_bad_opts.items()):
+            if bad_opts_list:
                 if not errmsg:
                     errmsg = "Omit redundant dest= when defining options.\n"
                 errmsg += f"\nSubcommand {name} (subcmds/{name}.py):\n"
                 errmsg += "".join(
-                    f"    {opt}: dest='{dest}'\n" for opt, dest in bad_opts
+                    f"    {opt}: dest='{dest}'\n" for opt, dest in bad_opts_list
                 )
         if errmsg:
-            self.fail(errmsg)
-
+            pytest.fail(errmsg)
+    finally:
         # Make sure we aren't popping the wrong stuff.
         assert optparse.Option.CHECK_METHODS.pop(0) is _check_dest
 
-    def test_common_validate_options(self):
-        """Verify CommonValidateOptions sets up expected fields."""
-        for name, cls in subcmds.all_commands.items():
-            cmd = cls()
-            opts, args = cmd.OptionParser.parse_args([])
 
-            # Verify the fields don't exist yet.
-            self.assertFalse(
-                hasattr(opts, "verbose"),
-                msg=f"{name}: has verbose before validation",
-            )
-            self.assertFalse(
-                hasattr(opts, "quiet"),
-                msg=f"{name}: has quiet before validation",
-            )
+@pytest.mark.parametrize("name, cls", subcmds.all_commands.items())
+def test_common_validate_options(name: str, cls: Type[Command]) -> None:
+    """Verify CommonValidateOptions sets up expected fields."""
+    cmd = cls()
+    opts, args = cmd.OptionParser.parse_args([])
 
-            cmd.CommonValidateOptions(opts, args)
+    # Verify the fields don't exist yet.
+    assert not hasattr(
+        opts, "verbose"
+    ), f"{name}: has verbose before validation"
+    assert not hasattr(opts, "quiet"), f"{name}: has quiet before validation"
+
+    cmd.CommonValidateOptions(opts, args)
+
+    # Verify the fields exist now.
+    assert hasattr(opts, "verbose"), f"{name}: missing verbose after validation"
+    assert hasattr(opts, "quiet"), f"{name}: missing quiet after validation"
+    assert hasattr(
+        opts, "outer_manifest"
+    ), f"{name}: missing outer_manifest after validation"
 
-            # Verify the fields exist now.
-            self.assertTrue(
-                hasattr(opts, "verbose"),
-                msg=f"{name}: missing verbose after validation",
-            )
-            self.assertTrue(
-                hasattr(opts, "quiet"),
-                msg=f"{name}: missing quiet after validation",
-            )
-            self.assertTrue(
-                hasattr(opts, "outer_manifest"),
-                msg=f"{name}: missing outer_manifest after validation",
-            )
 
-    def test_attribute_error_repro(self):
-        """Confirm that accessing verbose before CommonValidateOptions fails."""
-        from subcmds.sync import Sync
+def test_attribute_error_repro() -> None:
+    """Confirm that accessing verbose before CommonValidateOptions fails."""
+    from subcmds.sync import Sync
 
-        cmd = Sync()
-        opts, args = cmd.OptionParser.parse_args([])
+    cmd = Sync()
+    opts, args = cmd.OptionParser.parse_args([])
 
-        # This confirms that without the fix in main.py, an AttributeError
-        # would be raised because CommonValidateOptions hasn't been called yet.
-        with self.assertRaises(AttributeError):
-            _ = opts.verbose
+    # This confirms that without the fix in main.py, an AttributeError
+    # would be raised because CommonValidateOptions hasn't been called yet.
+    with pytest.raises(AttributeError):
+        _ = opts.verbose
 
-        cmd.CommonValidateOptions(opts, args)
-        self.assertTrue(hasattr(opts, "verbose"))
+    cmd.CommonValidateOptions(opts, args)
+    assert hasattr(opts, "verbose")
diff --git a/tests/test_subcmds_init.py b/tests/test_subcmds_init.py
index 25e5be567..960fb93fb 100644
--- a/tests/test_subcmds_init.py
+++ b/tests/test_subcmds_init.py
@@ -14,33 +14,36 @@
 
 """Unittests for the subcmds/init.py module."""
 
-import unittest
+from typing import List
+
+import pytest
 
 from subcmds import init
 
 
-class InitCommand(unittest.TestCase):
-    """Check registered all_commands."""
-
-    def setUp(self):
-        self.cmd = init.Init()
-
-    def test_cli_parser_good(self):
-        """Check valid command line options."""
-        ARGV = ([],)
-        for argv in ARGV:
-            opts, args = self.cmd.OptionParser.parse_args(argv)
-            self.cmd.ValidateOptions(opts, args)
-
-    def test_cli_parser_bad(self):
-        """Check invalid command line options."""
-        ARGV = (
-            # Too many arguments.
-            ["url", "asdf"],
-            # Conflicting options.
-            ["--mirror", "--archive"],
-        )
-        for argv in ARGV:
-            opts, args = self.cmd.OptionParser.parse_args(argv)
-            with self.assertRaises(SystemExit):
-                self.cmd.ValidateOptions(opts, args)
+@pytest.mark.parametrize(
+    "argv",
+    ([],),
+)
+def test_cli_parser_good(argv: List[str]) -> None:
+    """Check valid command line options."""
+    cmd = init.Init()
+    opts, args = cmd.OptionParser.parse_args(argv)
+    cmd.ValidateOptions(opts, args)
+
+
+@pytest.mark.parametrize(
+    "argv",
+    (
+        # Too many arguments.
+        ["url", "asdf"],
+        # Conflicting options.
+        ["--mirror", "--archive"],
+    ),
+)
+def test_cli_parser_bad(argv: List[str]) -> None:
+    """Check invalid command line options."""
+    cmd = init.Init()
+    opts, args = cmd.OptionParser.parse_args(argv)
+    with pytest.raises(SystemExit):
+        cmd.ValidateOptions(opts, args)
diff --git a/tests/test_update_manpages.py b/tests/test_update_manpages.py
index 06e5d3f5f..f52f8575f 100644
--- a/tests/test_update_manpages.py
+++ b/tests/test_update_manpages.py
@@ -14,15 +14,10 @@
 
 """Unittests for the update_manpages module."""
 
-import unittest
-
 from release import update_manpages
 
 
-class UpdateManpagesTest(unittest.TestCase):
-    """Tests the update-manpages code."""
-
-    def test_replace_regex(self):
-        """Check that replace_regex works."""
-        data = "\n\033[1mSummary\033[m\n"
-        self.assertEqual(update_manpages.replace_regex(data), "\nSummary\n")
+def test_replace_regex() -> None:
+    """Check that replace_regex works."""
+    data = "\n\033[1mSummary\033[m\n"
+    assert update_manpages.replace_regex(data) == "\nSummary\n"