Adding upstream version 3.1.0.upstream/3.1.0

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

29 files changed, 288 insertions, 281 deletions

diff --git a/pre_commit/all_languages.py b/pre_commit/all_languages.py
new file mode 100644
index 0000000..2bed706
--- /dev/null
+++ b/pre_commit/all_languages.py
@@ -0,0 +1,48 @@
+from __future__ import annotations
+
+from pre_commit.lang_base import Language
+from pre_commit.languages import conda
+from pre_commit.languages import coursier
+from pre_commit.languages import dart
+from pre_commit.languages import docker
+from pre_commit.languages import docker_image
+from pre_commit.languages import dotnet
+from pre_commit.languages import fail
+from pre_commit.languages import golang
+from pre_commit.languages import lua
+from pre_commit.languages import node
+from pre_commit.languages import perl
+from pre_commit.languages import pygrep
+from pre_commit.languages import python
+from pre_commit.languages import r
+from pre_commit.languages import ruby
+from pre_commit.languages import rust
+from pre_commit.languages import script
+from pre_commit.languages import swift
+from pre_commit.languages import system
+
+
+languages: dict[str, Language] = {
+    'conda': conda,
+    'coursier': coursier,
+    'dart': dart,
+    'docker': docker,
+    'docker_image': docker_image,
+    'dotnet': dotnet,
+    'fail': fail,
+    'golang': golang,
+    'lua': lua,
+    'node': node,
+    'perl': perl,
+    'pygrep': pygrep,
+    'python': python,
+    'r': r,
+    'ruby': ruby,
+    'rust': rust,
+    'script': script,
+    'swift': swift,
+    'system': system,
+    # TODO: fully deprecate `python_venv`
+    'python_venv': python,
+}
+language_names = sorted(languages)
diff --git a/pre_commit/clientlib.py b/pre_commit/clientlib.py
index e191d3a..9ff38c6 100644
--- a/pre_commit/clientlib.py
+++ b/pre_commit/clientlib.py
@@ -12,8 +12,8 @@ import cfgv
 from identify.identify import ALL_TAGS
 
 import pre_commit.constants as C
+from pre_commit.all_languages import language_names
 from pre_commit.errors import FatalError
-from pre_commit.languages.all import all_languages
 from pre_commit.yaml import yaml_load
 
 logger = logging.getLogger('pre_commit')
@@ -49,7 +49,7 @@ MANIFEST_HOOK_DICT = cfgv.Map(
     cfgv.Required('id', cfgv.check_string),
     cfgv.Required('name', cfgv.check_string),
     cfgv.Required('entry', cfgv.check_string),
-    cfgv.Required('language', cfgv.check_one_of(all_languages)),
+    cfgv.Required('language', cfgv.check_one_of(language_names)),
     cfgv.Optional('alias', cfgv.check_string, ''),
 
     cfgv.Optional('files', check_string_regex, ''),
@@ -281,8 +281,8 @@ CONFIG_REPO_DICT = cfgv.Map(
 )
 DEFAULT_LANGUAGE_VERSION = cfgv.Map(
     'DefaultLanguageVersion', None,
-    cfgv.NoAdditionalKeys(all_languages),
-    *(cfgv.Optional(x, cfgv.check_string, C.DEFAULT) for x in all_languages),
+    cfgv.NoAdditionalKeys(language_names),
+    *(cfgv.Optional(x, cfgv.check_string, C.DEFAULT) for x in language_names),
 )
 CONFIG_SCHEMA = cfgv.Map(
     'Config', None,
diff --git a/pre_commit/commands/migrate_config.py b/pre_commit/commands/migrate_config.py
index 6f7af4e..842fb3a 100644
--- a/pre_commit/commands/migrate_config.py
+++ b/pre_commit/commands/migrate_config.py
@@ -42,6 +42,14 @@ def _migrate_sha_to_rev(contents: str) -> str:
     return re.sub(r'(\n\s+)sha:', r'\1rev:', contents)
 
 
+def _migrate_python_venv(contents: str) -> str:
+    return re.sub(
+        r'(\n\s+)language: python_venv\b',
+        r'\1language: python',
+        contents,
+    )
+
+
 def migrate_config(config_file: str, quiet: bool = False) -> int:
     with open(config_file) as f:
         orig_contents = contents = f.read()
@@ -55,6 +63,7 @@ def migrate_config(config_file: str, quiet: bool = False) -> int:
 
     contents = _migrate_map(contents)
     contents = _migrate_sha_to_rev(contents)
+    contents = _migrate_python_venv(contents)
 
     if contents != orig_contents:
         with open(config_file, 'w') as f:
diff --git a/pre_commit/commands/run.py b/pre_commit/commands/run.py
index a7eb4f4..c9bc55b 100644
--- a/pre_commit/commands/run.py
+++ b/pre_commit/commands/run.py
@@ -19,9 +19,9 @@ from identify.identify import tags_from_path
 from pre_commit import color
 from pre_commit import git
 from pre_commit import output
+from pre_commit.all_languages import languages
 from pre_commit.clientlib import load_config
 from pre_commit.hook import Hook
-from pre_commit.languages.all import languages
 from pre_commit.repository import all_hooks
 from pre_commit.repository import install_hook_envs
 from pre_commit.staged_files_only import staged_files_only
diff --git a/pre_commit/languages/helpers.py b/pre_commit/lang_base.py
index d1be409..9480c55 100644
--- a/pre_commit/languages/helpers.py
+++ b/pre_commit/lang_base.py
@@ -7,8 +7,10 @@ import random
 import re
 import shlex
 from typing import Any
+from typing import ContextManager
 from typing import Generator
 from typing import NoReturn
+from typing import Protocol
 from typing import Sequence
 
 import pre_commit.constants as C
@@ -22,6 +24,42 @@ FIXED_RANDOM_SEED = 1542676187
 SHIMS_RE = re.compile(r'[/\\]shims[/\\]')
 
 
+class Language(Protocol):
+    # Use `None` for no installation / environment
+    @property
+    def ENVIRONMENT_DIR(self) -> str | None: ...
+    # return a value to replace `'default` for `language_version`
+    def get_default_version(self) -> str: ...
+    # return whether the environment is healthy (or should be rebuilt)
+    def health_check(self, prefix: Prefix, version: str) -> str | None: ...
+
+    # install a repository for the given language and language_version
+    def install_environment(
+            self,
+            prefix: Prefix,
+            version: str,
+            additional_dependencies: Sequence[str],
+    ) -> None:
+        ...
+
+    # modify the environment for hook execution
+    def in_env(self, prefix: Prefix, version: str) -> ContextManager[None]: ...
+
+    # execute a hook and return the exit code and output
+    def run_hook(
+            self,
+            prefix: Prefix,
+            entry: str,
+            args: Sequence[str],
+            file_args: Sequence[str],
+            *,
+            is_local: bool,
+            require_serial: bool,
+            color: bool,
+    ) -> tuple[int, bytes]:
+        ...
+
+
 def exe_exists(exe: str) -> bool:
     found = parse_shebang.find_executable(exe)
     if found is None:  # exe exists
@@ -45,7 +83,7 @@ def exe_exists(exe: str) -> bool:
     )
 
 
-def run_setup_cmd(prefix: Prefix, cmd: tuple[str, ...], **kwargs: Any) -> None:
+def setup_cmd(prefix: Prefix, cmd: tuple[str, ...], **kwargs: Any) -> None:
     cmd_output_b(*cmd, cwd=prefix.prefix_dir, **kwargs)
 
 
diff --git a/pre_commit/languages/all.py b/pre_commit/languages/all.py
deleted file mode 100644
index d952ae1..0000000
--- a/pre_commit/languages/all.py
+++ /dev/null
@@ -1,99 +0,0 @@
-from __future__ import annotations
-
-from typing import ContextManager
-from typing import Protocol
-from typing import Sequence
-
-from pre_commit.languages import conda
-from pre_commit.languages import coursier
-from pre_commit.languages import dart
-from pre_commit.languages import docker
-from pre_commit.languages import docker_image
-from pre_commit.languages import dotnet
-from pre_commit.languages import fail
-from pre_commit.languages import golang
-from pre_commit.languages import lua
-from pre_commit.languages import node
-from pre_commit.languages import perl
-from pre_commit.languages import pygrep
-from pre_commit.languages import python
-from pre_commit.languages import r
-from pre_commit.languages import ruby
-from pre_commit.languages import rust
-from pre_commit.languages import script
-from pre_commit.languages import swift
-from pre_commit.languages import system
-from pre_commit.prefix import Prefix
-
-
-class Language(Protocol):
-    # Use `None` for no installation / environment
-    @property
-    def ENVIRONMENT_DIR(self) -> str | None: ...
-    # return a value to replace `'default` for `language_version`
-    def get_default_version(self) -> str: ...
-
-    # return whether the environment is healthy (or should be rebuilt)
-    def health_check(
-            self,
-            prefix: Prefix,
-            language_version: str,
-    ) -> str | None:
-        ...
-
-    # install a repository for the given language and language_version
-    def install_environment(
-            self,
-            prefix: Prefix,
-            version: str,
-            additional_dependencies: Sequence[str],
-    ) -> None:
-        ...
-
-    # modify the environment for hook execution
-    def in_env(
-            self,
-            prefix: Prefix,
-            version: str,
-    ) -> ContextManager[None]:
-        ...
-
-    # execute a hook and return the exit code and output
-    def run_hook(
-            self,
-            prefix: Prefix,
-            entry: str,
-            args: Sequence[str],
-            file_args: Sequence[str],
-            *,
-            is_local: bool,
-            require_serial: bool,
-            color: bool,
-    ) -> tuple[int, bytes]:
-        ...
-
-
-languages: dict[str, Language] = {
-    'conda': conda,
-    'coursier': coursier,
-    'dart': dart,
-    'docker': docker,
-    'docker_image': docker_image,
-    'dotnet': dotnet,
-    'fail': fail,
-    'golang': golang,
-    'lua': lua,
-    'node': node,
-    'perl': perl,
-    'pygrep': pygrep,
-    'python': python,
-    'r': r,
-    'ruby': ruby,
-    'rust': rust,
-    'script': script,
-    'swift': swift,
-    'system': system,
-    # TODO: fully deprecate `python_venv`
-    'python_venv': python,
-}
-all_languages = sorted(languages)
diff --git a/pre_commit/languages/conda.py b/pre_commit/languages/conda.py
index e2fb019..05f1d29 100644
--- a/pre_commit/languages/conda.py
+++ b/pre_commit/languages/conda.py
@@ -5,19 +5,19 @@ import os
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import SubstitutionT
 from pre_commit.envcontext import UNSET
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output_b
 
 ENVIRONMENT_DIR = 'conda'
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def get_env_patch(env: str) -> PatchesT:
@@ -41,7 +41,7 @@ def get_env_patch(env: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -60,11 +60,11 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    helpers.assert_version_default('conda', version)
+    lang_base.assert_version_default('conda', version)
 
     conda_exe = _conda_exe()
 
-    env_dir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    env_dir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     cmd_output_b(
         conda_exe, 'env', 'create', '-p', env_dir, '--file',
         'environment.yml', cwd=prefix.prefix_dir,
diff --git a/pre_commit/languages/coursier.py b/pre_commit/languages/coursier.py
index 6075758..9c5fbfe 100644
--- a/pre_commit/languages/coursier.py
+++ b/pre_commit/languages/coursier.py
@@ -5,19 +5,19 @@ import os.path
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
 from pre_commit.errors import FatalError
-from pre_commit.languages import helpers
 from pre_commit.parse_shebang import find_executable
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = 'coursier'
 
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def install_environment(
@@ -25,7 +25,7 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    helpers.assert_version_default('coursier', version)
+    lang_base.assert_version_default('coursier', version)
 
     # Support both possible executable names (either "cs" or "coursier")
     cs = find_executable('cs') or find_executable('coursier')
@@ -35,12 +35,12 @@ def install_environment(
             'executables in the application search path',
         )
 
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     def _install(*opts: str) -> None:
         assert cs is not None
-        helpers.run_setup_cmd(prefix, (cs, 'fetch', *opts))
-        helpers.run_setup_cmd(prefix, (cs, 'install', '--dir', envdir, *opts))
+        lang_base.setup_cmd(prefix, (cs, 'fetch', *opts))
+        lang_base.setup_cmd(prefix, (cs, 'install', '--dir', envdir, *opts))
 
     with in_env(prefix, version):
         channel = prefix.path('.pre-commit-channel')
@@ -71,6 +71,6 @@ def get_env_patch(target_dir: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
diff --git a/pre_commit/languages/dart.py b/pre_commit/languages/dart.py
index e3c1c58..e8539ca 100644
--- a/pre_commit/languages/dart.py
+++ b/pre_commit/languages/dart.py
@@ -7,19 +7,19 @@ import tempfile
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import win_exe
 from pre_commit.yaml import yaml_load
 
 ENVIRONMENT_DIR = 'dartenv'
 
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def get_env_patch(venv: str) -> PatchesT:
@@ -30,7 +30,7 @@ def get_env_patch(venv: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -40,9 +40,9 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    helpers.assert_version_default('dart', version)
+    lang_base.assert_version_default('dart', version)
 
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     bin_dir = os.path.join(envdir, 'bin')
 
     def _install_dir(prefix_p: Prefix, pub_cache: str) -> None:
@@ -51,10 +51,10 @@ def install_environment(
         with open(prefix_p.path('pubspec.yaml')) as f:
             pubspec_contents = yaml_load(f)
 
-        helpers.run_setup_cmd(prefix_p, ('dart', 'pub', 'get'), env=dart_env)
+        lang_base.setup_cmd(prefix_p, ('dart', 'pub', 'get'), env=dart_env)
 
         for executable in pubspec_contents['executables']:
-            helpers.run_setup_cmd(
+            lang_base.setup_cmd(
                 prefix_p,
                 (
                     'dart', 'compile', 'exe',
@@ -77,7 +77,7 @@ def install_environment(
             else:
                 dep_cmd = (dep,)
 
-            helpers.run_setup_cmd(
+            lang_base.setup_cmd(
                 prefix,
                 ('dart', 'pub', 'cache', 'add', *dep_cmd),
                 env={**os.environ, 'PUB_CACHE': dep_tmp},
diff --git a/pre_commit/languages/docker.py b/pre_commit/languages/docker.py
index e80c959..8e53ca9 100644
--- a/pre_commit/languages/docker.py
+++ b/pre_commit/languages/docker.py
@@ -5,16 +5,16 @@ import json
 import os
 from typing import Sequence
 
-from pre_commit.languages import helpers
+from pre_commit import lang_base
 from pre_commit.prefix import Prefix
 from pre_commit.util import CalledProcessError
 from pre_commit.util import cmd_output_b
 
 ENVIRONMENT_DIR = 'docker'
 PRE_COMMIT_LABEL = 'PRE_COMMIT'
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-in_env = helpers.no_env  # no special environment for docker
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+in_env = lang_base.no_env  # no special environment for docker
 
 
 def _is_in_docker() -> bool:
@@ -84,16 +84,16 @@ def build_docker_image(
         cmd += ('--pull',)
     # This must come last for old versions of docker.  See #477
     cmd += ('.',)
-    helpers.run_setup_cmd(prefix, cmd)
+    lang_base.setup_cmd(prefix, cmd)
 
 
 def install_environment(
         prefix: Prefix, version: str, additional_dependencies: Sequence[str],
 ) -> None:  # pragma: win32 no cover
-    helpers.assert_version_default('docker', version)
-    helpers.assert_no_additional_deps('docker', additional_dependencies)
+    lang_base.assert_version_default('docker', version)
+    lang_base.assert_no_additional_deps('docker', additional_dependencies)
 
-    directory = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    directory = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     # Docker doesn't really have relevant disk environment, but pre-commit
     # still needs to cleanup its state files on failure
@@ -135,12 +135,11 @@ def run_hook(
     # automated cleanup of docker images.
     build_docker_image(prefix, pull=False)
 
-    entry_exe, *cmd_rest = helpers.hook_cmd(entry, args)
+    entry_exe, *cmd_rest = lang_base.hook_cmd(entry, args)
 
     entry_tag = ('--entrypoint', entry_exe, docker_tag(prefix))
-    cmd = (*docker_cmd(), *entry_tag, *cmd_rest)
-    return helpers.run_xargs(
-        cmd,
+    return lang_base.run_xargs(
+        (*docker_cmd(), *entry_tag, *cmd_rest),
         file_args,
         require_serial=require_serial,
         color=color,
diff --git a/pre_commit/languages/docker_image.py b/pre_commit/languages/docker_image.py
index 8e5f2c0..26f006e 100644
--- a/pre_commit/languages/docker_image.py
+++ b/pre_commit/languages/docker_image.py
@@ -2,15 +2,15 @@ from __future__ import annotations
 
 from typing import Sequence
 
-from pre_commit.languages import helpers
+from pre_commit import lang_base
 from pre_commit.languages.docker import docker_cmd
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = None
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-install_environment = helpers.no_install
-in_env = helpers.no_env
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+install_environment = lang_base.no_install
+in_env = lang_base.no_env
 
 
 def run_hook(
@@ -23,8 +23,8 @@ def run_hook(
         require_serial: bool,
         color: bool,
 ) -> tuple[int, bytes]:  # pragma: win32 no cover
-    cmd = docker_cmd() + helpers.hook_cmd(entry, args)
-    return helpers.run_xargs(
+    cmd = docker_cmd() + lang_base.hook_cmd(entry, args)
+    return lang_base.run_xargs(
         cmd,
         file_args,
         require_serial=require_serial,
diff --git a/pre_commit/languages/dotnet.py b/pre_commit/languages/dotnet.py
index 4c3955e..e9568f2 100644
--- a/pre_commit/languages/dotnet.py
+++ b/pre_commit/languages/dotnet.py
@@ -9,18 +9,18 @@ import zipfile
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = 'dotnetenv'
 BIN_DIR = 'bin'
 
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def get_env_patch(venv: str) -> PatchesT:
@@ -31,7 +31,7 @@ def get_env_patch(venv: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -57,19 +57,19 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    helpers.assert_version_default('dotnet', version)
-    helpers.assert_no_additional_deps('dotnet', additional_dependencies)
+    lang_base.assert_version_default('dotnet', version)
+    lang_base.assert_no_additional_deps('dotnet', additional_dependencies)
 
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
-    build_dir = 'pre-commit-build'
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    build_dir = prefix.path('pre-commit-build')
 
     # Build & pack nupkg file
-    helpers.run_setup_cmd(
+    lang_base.setup_cmd(
         prefix,
         (
             'dotnet', 'pack',
             '--configuration', 'Release',
-            '--output', build_dir,
+            '--property', f'PackageOutputPath={build_dir}',
         ),
     )
 
@@ -99,7 +99,7 @@ def install_environment(
 
         # Install to bin dir
         with _nuget_config_no_sources() as nuget_config:
-            helpers.run_setup_cmd(
+            lang_base.setup_cmd(
                 prefix,
                 (
                     'dotnet', 'tool', 'install',
@@ -109,7 +109,3 @@ def install_environment(
                     tool_id,
                 ),
             )
-
-    # Clean the git dir, ignoring the environment dir
-    clean_cmd = ('git', 'clean', '-ffxd', '-e', f'{ENVIRONMENT_DIR}-*')
-    helpers.run_setup_cmd(prefix, clean_cmd)
diff --git a/pre_commit/languages/fail.py b/pre_commit/languages/fail.py
index 33df067..a8ec6a5 100644
--- a/pre_commit/languages/fail.py
+++ b/pre_commit/languages/fail.py
@@ -2,14 +2,14 @@ from __future__ import annotations
 
 from typing import Sequence
 
-from pre_commit.languages import helpers
+from pre_commit import lang_base
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = None
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-install_environment = helpers.no_install
-in_env = helpers.no_env
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+install_environment = lang_base.no_install
+in_env = lang_base.no_env
 
 
 def run_hook(
diff --git a/pre_commit/languages/golang.py b/pre_commit/languages/golang.py
index 3c4b652..bea91e9 100644
--- a/pre_commit/languages/golang.py
+++ b/pre_commit/languages/golang.py
@@ -19,17 +19,17 @@ from typing import Protocol
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output
 from pre_commit.util import rmtree
 
 ENVIRONMENT_DIR = 'golangenv'
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 _ARCH_ALIASES = {
     'x86_64': 'amd64',
@@ -60,7 +60,7 @@ else:  # pragma: win32 no cover
 
 @functools.lru_cache(maxsize=1)
 def get_default_version() -> str:
-    if helpers.exe_exists('go'):
+    if lang_base.exe_exists('go'):
         return 'system'
     else:
         return C.DEFAULT
@@ -121,7 +121,7 @@ def _install_go(version: str, dest: str) -> None:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir, version)):
         yield
 
@@ -131,7 +131,7 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    env_dir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    env_dir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     if version != 'system':
         _install_go(version, env_dir)
@@ -149,9 +149,9 @@ def install_environment(
             os.path.join(env_dir, '.go', 'bin'), os.environ['PATH'],
         ))
 
-    helpers.run_setup_cmd(prefix, ('go', 'install', './...'), env=env)
+    lang_base.setup_cmd(prefix, ('go', 'install', './...'), env=env)
     for dependency in additional_dependencies:
-        helpers.run_setup_cmd(prefix, ('go', 'install', dependency), env=env)
+        lang_base.setup_cmd(prefix, ('go', 'install', dependency), env=env)
 
     # save some disk space -- we don't need this after installation
     pkgdir = os.path.join(env_dir, 'pkg')
diff --git a/pre_commit/languages/lua.py b/pre_commit/languages/lua.py
index ffc40b5..12d0661 100644
--- a/pre_commit/languages/lua.py
+++ b/pre_commit/languages/lua.py
@@ -6,17 +6,17 @@ import sys
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output
 
 ENVIRONMENT_DIR = 'lua_env'
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def _get_lua_version() -> str:  # pragma: win32 no cover
@@ -45,7 +45,7 @@ def get_env_patch(d: str) -> PatchesT:  # pragma: win32 no cover
 
 @contextlib.contextmanager  # pragma: win32 no cover
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -55,9 +55,9 @@ def install_environment(
     version: str,
     additional_dependencies: Sequence[str],
 ) -> None:  # pragma: win32 no cover
-    helpers.assert_version_default('lua', version)
+    lang_base.assert_version_default('lua', version)
 
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with in_env(prefix, version):
         # luarocks doesn't bootstrap a tree prior to installing
         # so ensure the directory exists.
@@ -66,10 +66,10 @@ def install_environment(
         # Older luarocks (e.g., 2.4.2) expect the rockspec as an arg
         for rockspec in prefix.star('.rockspec'):
             make_cmd = ('luarocks', '--tree', envdir, 'make', rockspec)
-            helpers.run_setup_cmd(prefix, make_cmd)
+            lang_base.setup_cmd(prefix, make_cmd)
 
         # luarocks can't install multiple packages at once
         # so install them individually.
         for dependency in additional_dependencies:
             cmd = ('luarocks', '--tree', envdir, 'install', dependency)
-            helpers.run_setup_cmd(prefix, cmd)
+            lang_base.setup_cmd(prefix, cmd)
diff --git a/pre_commit/languages/node.py b/pre_commit/languages/node.py
index 9688da3..66d6136 100644
--- a/pre_commit/languages/node.py
+++ b/pre_commit/languages/node.py
@@ -8,11 +8,11 @@ from typing import Generator
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import UNSET
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.languages.python import bin_dir
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output
@@ -20,7 +20,7 @@ from pre_commit.util import cmd_output_b
 from pre_commit.util import rmtree
 
 ENVIRONMENT_DIR = 'node_env'
-run_hook = helpers.basic_run_hook
+run_hook = lang_base.basic_run_hook
 
 
 @functools.lru_cache(maxsize=1)
@@ -30,7 +30,7 @@ def get_default_version() -> str:
         return C.DEFAULT
     # if node is already installed, we can save a bunch of setup time by
     # using the installed version
-    elif all(helpers.exe_exists(exe) for exe in ('node', 'npm')):
+    elif all(lang_base.exe_exists(exe) for exe in ('node', 'npm')):
         return 'system'
     else:
         return C.DEFAULT
@@ -60,13 +60,13 @@ def get_env_patch(venv: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
 
-def health_check(prefix: Prefix, language_version: str) -> str | None:
-    with in_env(prefix, language_version):
+def health_check(prefix: Prefix, version: str) -> str | None:
+    with in_env(prefix, version):
         retcode, _, _ = cmd_output_b('node', '--version', check=False)
         if retcode != 0:  # pragma: win32 no cover
             return f'`node --version` returned {retcode}'
@@ -78,7 +78,7 @@ def install_environment(
         prefix: Prefix, version: str, additional_dependencies: Sequence[str],
 ) -> None:
     assert prefix.exists('package.json')
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     # https://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx?f=255&MSPPError=-2147217396#maxpath
     if sys.platform == 'win32':  # pragma: no cover
@@ -96,13 +96,13 @@ def install_environment(
             'npm', 'install', '--dev', '--prod',
             '--ignore-prepublish', '--no-progress', '--no-save',
         )
-        helpers.run_setup_cmd(prefix, local_install_cmd)
+        lang_base.setup_cmd(prefix, local_install_cmd)
 
         _, pkg, _ = cmd_output('npm', 'pack', cwd=prefix.prefix_dir)
         pkg = prefix.path(pkg.strip())
 
         install = ('npm', 'install', '-g', pkg, *additional_dependencies)
-        helpers.run_setup_cmd(prefix, install)
+        lang_base.setup_cmd(prefix, install)
 
         # clean these up after installation
         if prefix.exists('node_modules'):  # pragma: win32 no cover
diff --git a/pre_commit/languages/perl.py b/pre_commit/languages/perl.py
index 2530c0e..2a7f162 100644
--- a/pre_commit/languages/perl.py
+++ b/pre_commit/languages/perl.py
@@ -6,16 +6,16 @@ import shlex
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = 'perl_env'
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def get_env_patch(venv: str) -> PatchesT:
@@ -34,7 +34,7 @@ def get_env_patch(venv: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -42,9 +42,9 @@ def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
 def install_environment(
         prefix: Prefix, version: str, additional_dependencies: Sequence[str],
 ) -> None:
-    helpers.assert_version_default('perl', version)
+    lang_base.assert_version_default('perl', version)
 
     with in_env(prefix, version):
-        helpers.run_setup_cmd(
+        lang_base.setup_cmd(
             prefix, ('cpan', '-T', '.', *additional_dependencies),
         )
diff --git a/pre_commit/languages/pygrep.py b/pre_commit/languages/pygrep.py
index f0eb9a9..ec55560 100644
--- a/pre_commit/languages/pygrep.py
+++ b/pre_commit/languages/pygrep.py
@@ -7,16 +7,16 @@ from typing import NamedTuple
 from typing import Pattern
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit import output
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.xargs import xargs
 
 ENVIRONMENT_DIR = None
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-install_environment = helpers.no_install
-in_env = helpers.no_env
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+install_environment = lang_base.no_install
+in_env = lang_base.no_env
 
 
 def _process_filename_by_line(pattern: Pattern[bytes], filename: str) -> int:
diff --git a/pre_commit/languages/python.py b/pre_commit/languages/python.py
index c373646..976674e 100644
--- a/pre_commit/languages/python.py
+++ b/pre_commit/languages/python.py
@@ -8,11 +8,11 @@ from typing import Generator
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import UNSET
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.parse_shebang import find_executable
 from pre_commit.prefix import Prefix
 from pre_commit.util import CalledProcessError
@@ -21,7 +21,7 @@ from pre_commit.util import cmd_output_b
 from pre_commit.util import win_exe
 
 ENVIRONMENT_DIR = 'py_env'
-run_hook = helpers.basic_run_hook
+run_hook = lang_base.basic_run_hook
 
 
 @functools.lru_cache(maxsize=None)
@@ -153,13 +153,13 @@ def norm_version(version: str) -> str | None:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
 
-def health_check(prefix: Prefix, language_version: str) -> str | None:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, language_version)
+def health_check(prefix: Prefix, version: str) -> str | None:
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     pyvenv_cfg = os.path.join(envdir, 'pyvenv.cfg')
 
     # created with "old" virtualenv
@@ -202,7 +202,7 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     venv_cmd = [sys.executable, '-mvirtualenv', envdir]
     python = norm_version(version)
     if python is not None:
@@ -211,4 +211,4 @@ def install_environment(
 
     cmd_output_b(*venv_cmd, cwd='/')
     with in_env(prefix, version):
-        helpers.run_setup_cmd(prefix, install_cmd)
+        lang_base.setup_cmd(prefix, install_cmd)
diff --git a/pre_commit/languages/r.py b/pre_commit/languages/r.py
index e238365..138a26e 100644
--- a/pre_commit/languages/r.py
+++ b/pre_commit/languages/r.py
@@ -7,18 +7,18 @@ import shutil
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import UNSET
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output_b
 from pre_commit.util import win_exe
 
 ENVIRONMENT_DIR = 'renv'
 RSCRIPT_OPTS = ('--no-save', '--no-restore', '--no-site-file', '--no-environ')
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
 
 
 def get_env_patch(venv: str) -> PatchesT:
@@ -30,7 +30,7 @@ def get_env_patch(venv: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -93,7 +93,7 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    env_dir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    env_dir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     os.makedirs(env_dir, exist_ok=True)
     shutil.copy(prefix.path('renv.lock'), env_dir)
     shutil.copytree(prefix.path('renv'), os.path.join(env_dir, 'renv'))
@@ -166,7 +166,7 @@ def run_hook(
         color: bool,
 ) -> tuple[int, bytes]:
     cmd = _cmd_from_hook(prefix, entry, args, is_local=is_local)
-    return helpers.run_xargs(
+    return lang_base.run_xargs(
         cmd,
         file_args,
         require_serial=require_serial,
diff --git a/pre_commit/languages/ruby.py b/pre_commit/languages/ruby.py
index 4416f72..76631f2 100644
--- a/pre_commit/languages/ruby.py
+++ b/pre_commit/languages/ruby.py
@@ -2,30 +2,35 @@ from __future__ import annotations
 
 import contextlib
 import functools
+import importlib.resources
 import os.path
 import shutil
 import tarfile
 from typing import Generator
+from typing import IO
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import UNSET
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import CalledProcessError
-from pre_commit.util import resource_bytesio
 
 ENVIRONMENT_DIR = 'rbenv'
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
+
+
+def _resource_bytesio(filename: str) -> IO[bytes]:
+    return importlib.resources.open_binary('pre_commit.resources', filename)
 
 
 @functools.lru_cache(maxsize=1)
 def get_default_version() -> str:
-    if all(helpers.exe_exists(exe) for exe in ('ruby', 'gem')):
+    if all(lang_base.exe_exists(exe) for exe in ('ruby', 'gem')):
         return 'system'
     else:
         return C.DEFAULT
@@ -68,13 +73,13 @@ def get_env_patch(
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir, version)):
         yield
 
 
 def _extract_resource(filename: str, dest: str) -> None:
-    with resource_bytesio(filename) as bio:
+    with _resource_bytesio(filename) as bio:
         with tarfile.open(fileobj=bio) as tf:
             tf.extractall(dest)
 
@@ -83,7 +88,7 @@ def _install_rbenv(
         prefix: Prefix,
         version: str,
 ) -> None:  # pragma: win32 no cover
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     _extract_resource('rbenv.tar.gz', prefix.path('.'))
     shutil.move(prefix.path('rbenv'), envdir)
@@ -100,10 +105,10 @@ def _install_ruby(
         version: str,
 ) -> None:  # pragma: win32 no cover
     try:
-        helpers.run_setup_cmd(prefix, ('rbenv', 'download', version))
+        lang_base.setup_cmd(prefix, ('rbenv', 'download', version))
     except CalledProcessError:  # pragma: no cover (usually find with download)
         # Failed to download from mirror for some reason, build it instead
-        helpers.run_setup_cmd(prefix, ('rbenv', 'install', version))
+        lang_base.setup_cmd(prefix, ('rbenv', 'install', version))
 
 
 def install_environment(
@@ -114,17 +119,17 @@ def install_environment(
         with in_env(prefix, version):
             # Need to call this before installing so rbenv's directories
             # are set up
-            helpers.run_setup_cmd(prefix, ('rbenv', 'init', '-'))
+            lang_base.setup_cmd(prefix, ('rbenv', 'init', '-'))
             if version != C.DEFAULT:
                 _install_ruby(prefix, version)
             # Need to call this after installing to set up the shims
-            helpers.run_setup_cmd(prefix, ('rbenv', 'rehash'))
+            lang_base.setup_cmd(prefix, ('rbenv', 'rehash'))
 
     with in_env(prefix, version):
-        helpers.run_setup_cmd(
+        lang_base.setup_cmd(
             prefix, ('gem', 'build', *prefix.star('.gemspec')),
         )
-        helpers.run_setup_cmd(
+        lang_base.setup_cmd(
             prefix,
             (
                 'gem', 'install',
diff --git a/pre_commit/languages/rust.py b/pre_commit/languages/rust.py
index 391fd86..e98e0d0 100644
--- a/pre_commit/languages/rust.py
+++ b/pre_commit/languages/rust.py
@@ -11,19 +11,19 @@ from typing import Generator
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit import lang_base
 from pre_commit import parse_shebang
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output_b
 from pre_commit.util import make_executable
 from pre_commit.util import win_exe
 
 ENVIRONMENT_DIR = 'rustenv'
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 @functools.lru_cache(maxsize=1)
@@ -63,7 +63,7 @@ def get_env_patch(target_dir: str, version: str) -> PatchesT:
 
 @contextlib.contextmanager
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir, version)):
         yield
 
@@ -78,7 +78,7 @@ def _add_dependencies(
         crate = f'{name}@{spec or "*"}'
         crates.append(crate)
 
-    helpers.run_setup_cmd(prefix, ('cargo', 'add', *crates))
+    lang_base.setup_cmd(prefix, ('cargo', 'add', *crates))
 
 
 def install_rust_with_toolchain(toolchain: str) -> None:
@@ -116,7 +116,7 @@ def install_environment(
         version: str,
         additional_dependencies: Sequence[str],
 ) -> None:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     # There are two cases where we might want to specify more dependencies:
     # as dependencies for the library being built, and as binary packages
diff --git a/pre_commit/languages/script.py b/pre_commit/languages/script.py
index 08325f4..89a3ab2 100644
--- a/pre_commit/languages/script.py
+++ b/pre_commit/languages/script.py
@@ -2,14 +2,14 @@ from __future__ import annotations
 
 from typing import Sequence
 
-from pre_commit.languages import helpers
+from pre_commit import lang_base
 from pre_commit.prefix import Prefix
 
 ENVIRONMENT_DIR = None
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-install_environment = helpers.no_install
-in_env = helpers.no_env
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+install_environment = lang_base.no_install
+in_env = lang_base.no_env
 
 
 def run_hook(
@@ -22,9 +22,9 @@ def run_hook(
         require_serial: bool,
         color: bool,
 ) -> tuple[int, bytes]:
-    cmd = helpers.hook_cmd(entry, args)
+    cmd = lang_base.hook_cmd(entry, args)
     cmd = (prefix.path(cmd[0]), *cmd[1:])
-    return helpers.run_xargs(
+    return lang_base.run_xargs(
         cmd,
         file_args,
         require_serial=require_serial,
diff --git a/pre_commit/languages/swift.py b/pre_commit/languages/swift.py
index c66ad5f..8250ab7 100644
--- a/pre_commit/languages/swift.py
+++ b/pre_commit/languages/swift.py
@@ -5,10 +5,10 @@ import os
 from typing import Generator
 from typing import Sequence
 
+from pre_commit import lang_base
 from pre_commit.envcontext import envcontext
 from pre_commit.envcontext import PatchesT
 from pre_commit.envcontext import Var
-from pre_commit.languages import helpers
 from pre_commit.prefix import Prefix
 from pre_commit.util import cmd_output_b
 
@@ -16,9 +16,9 @@ BUILD_DIR = '.build'
 BUILD_CONFIG = 'release'
 
 ENVIRONMENT_DIR = 'swift_env'
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+run_hook = lang_base.basic_run_hook
 
 
 def get_env_patch(venv: str) -> PatchesT:  # pragma: win32 no cover
@@ -28,7 +28,7 @@ def get_env_patch(venv: str) -> PatchesT:  # pragma: win32 no cover
 
 @contextlib.contextmanager  # pragma: win32 no cover
 def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
     with envcontext(get_env_patch(envdir)):
         yield
 
@@ -36,9 +36,9 @@ def in_env(prefix: Prefix, version: str) -> Generator[None, None, None]:
 def install_environment(
         prefix: Prefix, version: str, additional_dependencies: Sequence[str],
 ) -> None:  # pragma: win32 no cover
-    helpers.assert_version_default('swift', version)
-    helpers.assert_no_additional_deps('swift', additional_dependencies)
-    envdir = helpers.environment_dir(prefix, ENVIRONMENT_DIR, version)
+    lang_base.assert_version_default('swift', version)
+    lang_base.assert_no_additional_deps('swift', additional_dependencies)
+    envdir = lang_base.environment_dir(prefix, ENVIRONMENT_DIR, version)
 
     # Build the swift package
     os.mkdir(envdir)
diff --git a/pre_commit/languages/system.py b/pre_commit/languages/system.py
index 204cad7..f6ad688 100644
--- a/pre_commit/languages/system.py
+++ b/pre_commit/languages/system.py
@@ -1,10 +1,10 @@
 from __future__ import annotations
 
-from pre_commit.languages import helpers
+from pre_commit import lang_base
 
 ENVIRONMENT_DIR = None
-get_default_version = helpers.basic_get_default_version
-health_check = helpers.basic_health_check
-install_environment = helpers.no_install
-in_env = helpers.no_env
-run_hook = helpers.basic_run_hook
+get_default_version = lang_base.basic_get_default_version
+health_check = lang_base.basic_health_check
+install_environment = lang_base.no_install
+in_env = lang_base.no_env
+run_hook = lang_base.basic_run_hook
diff --git a/pre_commit/repository.py b/pre_commit/repository.py
index 616faf5..040f238 100644
--- a/pre_commit/repository.py
+++ b/pre_commit/repository.py
@@ -3,17 +3,18 @@ from __future__ import annotations
 import json
 import logging
 import os
+import shlex
 from typing import Any
 from typing import Sequence
 
 import pre_commit.constants as C
+from pre_commit.all_languages import languages
 from pre_commit.clientlib import load_manifest
 from pre_commit.clientlib import LOCAL
 from pre_commit.clientlib import META
 from pre_commit.clientlib import parse_version
 from pre_commit.hook import Hook
-from pre_commit.languages.all import languages
-from pre_commit.languages.helpers import environment_dir
+from pre_commit.lang_base import environment_dir
 from pre_commit.prefix import Prefix
 from pre_commit.store import Store
 from pre_commit.util import clean_path_on_failure
@@ -32,7 +33,7 @@ def _state_filename_v2(venv: str) -> str:
 
 
 def _state(additional_deps: Sequence[str]) -> object:
-    return {'additional_dependencies': sorted(additional_deps)}
+    return {'additional_dependencies': additional_deps}
 
 
 def _read_state(venv: str) -> object | None:
@@ -68,6 +69,14 @@ def _hook_install(hook: Hook) -> None:
     logger.info('Once installed this environment will be reused.')
     logger.info('This may take a few minutes...')
 
+    if hook.language == 'python_venv':
+        logger.warning(
+            f'`repo: {hook.src}` uses deprecated `language: python_venv`.  '
+            f'This is an alias for `language: python`.  '
+            f'Often `pre-commit autoupdate --repo {shlex.quote(hook.src)}` '
+            f'will fix this.',
+        )
+
     lang = languages[hook.language]
     assert lang.ENVIRONMENT_DIR is not None
 
diff --git a/pre_commit/staged_files_only.py b/pre_commit/staged_files_only.py
index 172fb20..8812356 100644
--- a/pre_commit/staged_files_only.py
+++ b/pre_commit/staged_files_only.py
@@ -7,6 +7,7 @@ import time
 from typing import Generator
 
 from pre_commit import git
+from pre_commit.errors import FatalError
 from pre_commit.util import CalledProcessError
 from pre_commit.util import cmd_output
 from pre_commit.util import cmd_output_b
@@ -49,12 +50,16 @@ def _intent_to_add_cleared() -> Generator[None, None, None]:
 @contextlib.contextmanager
 def _unstaged_changes_cleared(patch_dir: str) -> Generator[None, None, None]:
     tree = cmd_output('git', 'write-tree')[1].strip()
-    retcode, diff_stdout_binary, _ = cmd_output_b(
+    diff_cmd = (
         'git', 'diff-index', '--ignore-submodules', '--binary',
         '--exit-code', '--no-color', '--no-ext-diff', tree, '--',
-        check=False,
     )
-    if retcode and diff_stdout_binary.strip():
+    retcode, diff_stdout, diff_stderr = cmd_output_b(*diff_cmd, check=False)
+    if retcode == 0:
+        # There weren't any staged files so we don't need to do anything
+        # special
+        yield
+    elif retcode == 1 and diff_stdout.strip():
         patch_filename = f'patch{int(time.time())}-{os.getpid()}'
         patch_filename = os.path.join(patch_dir, patch_filename)
         logger.warning('Unstaged files detected.')
@@ -62,7 +67,7 @@ def _unstaged_changes_cleared(patch_dir: str) -> Generator[None, None, None]:
         # Save the current unstaged changes as a patch
         os.makedirs(patch_dir, exist_ok=True)
         with open(patch_filename, 'wb') as patch_file:
-            patch_file.write(diff_stdout_binary)
+            patch_file.write(diff_stdout)
 
         # prevent recursive post-checkout hooks (#1418)
         no_checkout_env = dict(os.environ, _PRE_COMMIT_SKIP_POST_CHECKOUT='1')
@@ -86,10 +91,12 @@ def _unstaged_changes_cleared(patch_dir: str) -> Generator[None, None, None]:
                 _git_apply(patch_filename)
 
             logger.info(f'Restored changes from {patch_filename}.')
-    else:
-        # There weren't any staged files so we don't need to do anything
-        # special
-        yield
+    else:  # pragma: win32 no cover
+        # some error occurred while requesting the diff
+        e = CalledProcessError(retcode, diff_cmd, b'', diff_stderr)
+        raise FatalError(
+            f'pre-commit failed to diff -- perhaps due to permissions?\n\n{e}',
+        )
 
 
 @contextlib.contextmanager
diff --git a/pre_commit/store.py b/pre_commit/store.py
index 6ddc7c4..487e3e7 100644
--- a/pre_commit/store.py
+++ b/pre_commit/store.py
@@ -125,7 +125,7 @@ class Store:
     @classmethod
     def db_repo_name(cls, repo: str, deps: Sequence[str]) -> str:
         if deps:
-            return f'{repo}:{",".join(sorted(deps))}'
+            return f'{repo}:{",".join(deps)}'
         else:
             return repo
 
diff --git a/pre_commit/util.py b/pre_commit/util.py
index 8ea4844..ea0d4f5 100644
--- a/pre_commit/util.py
+++ b/pre_commit/util.py
@@ -12,7 +12,6 @@ from types import TracebackType
 from typing import Any
 from typing import Callable
 from typing import Generator
-from typing import IO
 
 from pre_commit import parse_shebang
 
@@ -36,10 +35,6 @@ def clean_path_on_failure(path: str) -> Generator[None, None, None]:
         raise
 
 
-def resource_bytesio(filename: str) -> IO[bytes]:
-    return importlib.resources.open_binary('pre_commit.resources', filename)
-
-
 def resource_text(filename: str) -> str:
     return importlib.resources.read_text('pre_commit.resources', filename)
 
@@ -67,7 +62,7 @@ class CalledProcessError(RuntimeError):
     def __bytes__(self) -> bytes:
         def _indent_or_none(part: bytes | None) -> bytes:
             if part:
-                return b'\n    ' + part.replace(b'\n', b'\n    ')
+                return b'\n    ' + part.replace(b'\n', b'\n    ').rstrip()
             else:
                 return b' (none)'