Merge pull request #1227 from pre-commit/pre-commit-ci-update-config

[pre-commit.ci] pre-commit autoupdate

.github/workflows/main.yml

@@ -0,0 +1,19 @@
+name: main
+
+on:
+  push:
+    branches: [main, test-me-*]
+    tags: '*'
+  pull_request:
+
+jobs:
+  main-windows:
+    uses: asottile/workflows/.github/workflows/tox.yml@v1.8.1
+    with:
+      env: '["py310"]'
+      os: windows-latest
+  main-linux:
+    uses: asottile/workflows/.github/workflows/tox.yml@v1.8.1
+    with:
+      env: '["py310", "py311", "py312", "py313"]'
+      os: ubuntu-latest

.gitignore

@@ -3,9 +3,4 @@
 .*.sw[a-z]
 .coverage
 .tox
-.venv.touch
-/.mypy_cache
-/.pytest_cache
-/venv*
-coverage-html
 dist

.pre-commit-config.yaml

@@ -1,50 +1,41 @@
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.1.0
+    rev: v6.0.0
     hooks:
     -   id: trailing-whitespace
     -   id: end-of-file-fixer
-    -   id: check-docstring-first
-    -   id: check-json
-    -   id: check-added-large-files
     -   id: check-yaml
     -   id: debug-statements
-    -   id: name-tests-test
     -   id: double-quote-string-fixer
+    -   id: name-tests-test
     -   id: requirements-txt-fixer
--   repo: https://gitlab.com/pycqa/flake8
-    rev: 3.8.0
-    hooks:
-    -   id: flake8
-        additional_dependencies: [flake8-typing-imports==1.7.0]
--   repo: https://github.com/pre-commit/mirrors-autopep8
-    rev: v1.5.2
-    hooks:
-    -   id: autopep8
--   repo: https://github.com/pre-commit/pre-commit
-    rev: v2.4.0
-    hooks:
-    -   id: validate_manifest
--   repo: https://github.com/asottile/reorder_python_imports
-    rev: v2.3.0
-    hooks:
-    -   id: reorder-python-imports
-        args: [--py3-plus]
--   repo: https://github.com/asottile/pyupgrade
-    rev: v2.4.1
-    hooks:
-    -   id: pyupgrade
-        args: [--py36-plus]
--   repo: https://github.com/asottile/add-trailing-comma
-    rev: v2.0.1
-    hooks:
-    -   id: add-trailing-comma
-        args: [--py36-plus]
 -   repo: https://github.com/asottile/setup-cfg-fmt
-    rev: v1.9.0
+    rev: v3.2.0
     hooks:
     -   id: setup-cfg-fmt
+-   repo: https://github.com/asottile/reorder-python-imports
+    rev: v3.16.0
+    hooks:
+    -   id: reorder-python-imports
+        args: [--py310-plus, --add-import, 'from __future__ import annotations']
+-   repo: https://github.com/asottile/add-trailing-comma
+    rev: v4.0.0
+    hooks:
+    -   id: add-trailing-comma
+-   repo: https://github.com/asottile/pyupgrade
+    rev: v3.21.2
+    hooks:
+    -   id: pyupgrade
+        args: [--py310-plus]
+-   repo: https://github.com/hhatto/autopep8
+    rev: v2.3.2
+    hooks:
+    -   id: autopep8
+-   repo: https://github.com/PyCQA/flake8
+    rev: 7.3.0
+    hooks:
+    -   id: flake8
 -   repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v0.770
+    rev: v1.19.1
     hooks:
     -   id: mypy

.pre-commit-hooks.yaml

@@ -1,195 +1,212 @@
--   id: autopep8-wrapper
-    name: autopep8 wrapper (removed)
-    description: (removed) use pre-commit/mirrors-autopep8 instead.
-    entry: pre-commit-hooks-removed autopep8-wrapper autopep8 https://github.com/pre-commit/mirrors-autopep8
-    language: python
-    always_run: true
-    pass_filenames: false
 -   id: check-added-large-files
-    name: Check for added large files
-    description: Prevent giant files from being committed
+    name: check for added large files
+    description: prevents giant files from being committed.
     entry: check-added-large-files
     language: python
+    stages: [pre-commit, pre-push, manual]
+    minimum_pre_commit_version: 3.2.0
 -   id: check-ast
-    name: Check python ast
-    description: Simply check whether the files parse as valid python.
+    name: check python ast
+    description: simply checks whether the files parse as valid python.
     entry: check-ast
     language: python
     types: [python]
 -   id: check-byte-order-marker
-    name: Check for byte-order marker
-    description: Forbid files which have a UTF-8 byte-order marker
-    entry: check-byte-order-marker
+    name: check-byte-order-marker (removed)
+    description: (removed) use fix-byte-order-marker instead.
+    entry: pre-commit-hooks-removed check-byte-order-marker fix-byte-order-marker https://github.com/pre-commit/pre-commit-hooks
     language: python
     types: [text]
 -   id: check-builtin-literals
-    name: Check builtin type constructor use
-    description: Require literal syntax when initializing empty or zero Python builtin types.
+    name: check builtin type constructor use
+    description: requires literal syntax when initializing empty or zero python builtin types.
     entry: check-builtin-literals
     language: python
     types: [python]
 -   id: check-case-conflict
-    name: Check for case conflicts
-    description: Check for files that would conflict in case-insensitive filesystems
+    name: check for case conflicts
+    description: checks for files that would conflict in case-insensitive filesystems.
     entry: check-case-conflict
     language: python
 -   id: check-docstring-first
-    name: Check docstring is first
-    description: Checks a common error of defining a docstring after code.
+    name: check docstring is first (deprecated)
+    description: checks a common error of defining a docstring after code.
     entry: check-docstring-first
     language: python
     types: [python]
 -   id: check-executables-have-shebangs
-    name: Check that executables have shebangs
-    description: Ensures that (non-binary) executables have a shebang.
+    name: check that executables have shebangs
+    description: ensures that (non-binary) executables have a shebang.
     entry: check-executables-have-shebangs
     language: python
     types: [text, executable]
-    stages: [commit, push, manual]
+    stages: [pre-commit, pre-push, manual]
+    minimum_pre_commit_version: 3.2.0
+-   id: check-illegal-windows-names
+    name: check illegal windows names
+    entry: Illegal Windows filenames detected
+    language: fail
+    files: '(?i)((^|/)(CON|PRN|AUX|NUL|COM[\d¹²³]|LPT[\d¹²³])(\.|/|$)|[<>:\"\\|?*\x00-\x1F]|/[^/]*[\.\s]/|[^/]*[\.\s]$)'
 -   id: check-json
-    name: Check JSON
-    description: This hook checks json files for parseable syntax.
+    name: check json
+    description: checks json files for parseable syntax.
     entry: check-json
     language: python
     types: [json]
+-   id: check-shebang-scripts-are-executable
+    name: check that scripts with shebangs are executable
+    description: ensures that (non-binary) files with a shebang are executable.
+    entry: check-shebang-scripts-are-executable
+    language: python
+    types: [text]
+    stages: [pre-commit, pre-push, manual]
+    minimum_pre_commit_version: 3.2.0
 -   id: pretty-format-json
-    name: Pretty format JSON
-    description: This hook sets a standard for formatting JSON files.
+    name: pretty format json
+    description: sets a standard for formatting json files.
     entry: pretty-format-json
     language: python
     types: [json]
 -   id: check-merge-conflict
-    name: Check for merge conflicts
-    description: Check for files that contain merge conflict strings.
+    name: check for merge conflicts
+    description: checks for files that contain merge conflict strings.
     entry: check-merge-conflict
     language: python
     types: [text]
 -   id: check-symlinks
-    name: Check for broken symlinks
-    description: Checks for symlinks which do not point to anything.
+    name: check for broken symlinks
+    description: checks for symlinks which do not point to anything.
     entry: check-symlinks
     language: python
     types: [symlink]
 -   id: check-toml
-    name: Check Toml
-    description: This hook checks toml files for parseable syntax.
+    name: check toml
+    description: checks toml files for parseable syntax.
     entry: check-toml
     language: python
     types: [toml]
 -   id: check-vcs-permalinks
-    name: Check vcs permalinks
-    description: Ensures that links to vcs websites are permalinks.
+    name: check vcs permalinks
+    description: ensures that links to vcs websites are permalinks.
     entry: check-vcs-permalinks
     language: python
     types: [text]
 -   id: check-xml
-    name: Check Xml
-    description: This hook checks xml files for parseable syntax.
+    name: check xml
+    description: checks xml files for parseable syntax.
     entry: check-xml
     language: python
     types: [xml]
 -   id: check-yaml
-    name: Check Yaml
-    description: This hook checks yaml files for parseable syntax.
+    name: check yaml
+    description: checks yaml files for parseable syntax.
     entry: check-yaml
     language: python
     types: [yaml]
 -   id: debug-statements
-    name: Debug Statements (Python)
-    description: Check for debugger imports and py37+ `breakpoint()` calls in python source.
+    name: debug statements (python)
+    description: checks for debugger imports and py37+ `breakpoint()` calls in python source.
     entry: debug-statement-hook
     language: python
     types: [python]
+-   id: destroyed-symlinks
+    name: detect destroyed symlinks
+    description: detects symlinks which are changed to regular files with a content of a path which that symlink was pointing to.
+    entry: destroyed-symlinks
+    language: python
+    types: [file]
+    stages: [pre-commit, pre-push, manual]
 -   id: detect-aws-credentials
-    name: Detect AWS Credentials
-    description: Detects *your* aws credentials from the aws cli credentials file
+    name: detect aws credentials
+    description: detects *your* aws credentials from the aws cli credentials file.
     entry: detect-aws-credentials
     language: python
     types: [text]
 -   id: detect-private-key
-    name: Detect Private Key
-    description: Detects the presence of private keys
+    name: detect private key
+    description: detects the presence of private keys.
     entry: detect-private-key
     language: python
     types: [text]
 -   id: double-quote-string-fixer
-    name: Fix double quoted strings
-    description: This hook replaces double quoted strings with single quoted strings
+    name: fix double quoted strings
+    description: replaces double quoted strings with single quoted strings.
     entry: double-quote-string-fixer
     language: python
     types: [python]
 -   id: end-of-file-fixer
-    name: Fix End of Files
-    description: Ensures that a file is either empty, or ends with one newline.
+    name: fix end of files
+    description: ensures that a file is either empty, or ends with one newline.
     entry: end-of-file-fixer
     language: python
     types: [text]
-    stages: [commit, push, manual]
+    stages: [pre-commit, pre-push, manual]
+    minimum_pre_commit_version: 3.2.0
 -   id: file-contents-sorter
-    name: File Contents Sorter
-    description: Sort the lines in specified files (defaults to alphabetical). You must provide list of target files as input in your .pre-commit-config.yaml file.
+    name: file contents sorter
+    description: sorts the lines in specified files (defaults to alphabetical). you must provide list of target files as input in your .pre-commit-config.yaml file.
     entry: file-contents-sorter
     language: python
     files: '^$'
+-   id: fix-byte-order-marker
+    name: fix utf-8 byte order marker
+    description: removes utf-8 byte order marker.
+    entry: fix-byte-order-marker
+    language: python
+    types: [text]
 -   id: fix-encoding-pragma
-    name: Fix python encoding pragma
+    name: fix python encoding pragma (removed)
+    description: (removed) use pyupgrade instead.
+    entry: pre-commit-hooks-removed fix-encoding-pragma pyupgrade https://github.com/asottile/pyupgrade
     language: python
-    entry: fix-encoding-pragma
-    description: 'Add # -*- coding: utf-8 -*- to the top of python files'
     types: [python]
--   id: flake8
-    name: Flake8 (removed)
-    description: (removed) use gitlab.com/pycqa/flake8 instead.
-    entry: pre-commit-hooks-removed flake8 flake8 https://gitlab.com/pycqa/flake8
-    language: python
-    always_run: true
-    pass_filenames: false
 -   id: forbid-new-submodules
-    name: Forbid new submodules
+    name: forbid new submodules
+    description: prevents addition of new git submodules.
     language: python
     entry: forbid-new-submodules
-    description: Prevent addition of new git submodules
+    types: [directory]
+-   id: forbid-submodules
+    name: forbid submodules
+    description: forbids any submodules in the repository
+    language: fail
+    entry: 'submodules are not allowed in this repository:'
+    types: [directory]
 -   id: mixed-line-ending
-    name: Mixed line ending
-    description: Replaces or checks mixed line ending
+    name: mixed line ending
+    description: replaces or checks mixed line ending.
     entry: mixed-line-ending
     language: python
     types: [text]
 -   id: name-tests-test
-    name: Tests should end in _test.py
-    description: This verifies that test files are named correctly
+    name: python tests naming
+    description: verifies that test files are named correctly.
     entry: name-tests-test
     language: python
     files: (^|/)tests/.+\.py$
 -   id: no-commit-to-branch
-    name: "Don't commit to branch"
+    name: "don't commit to branch"
     entry: no-commit-to-branch
     language: python
     pass_filenames: false
     always_run: true
--   id: pyflakes
-    name: Pyflakes (removed)
-    description: (removed) use gitlab.com/pycqa/flake8 instead.
-    entry: pre-commit-hooks-removed pyflakes flake8 https://gitlab.com/pycqa/flake8
-    language: python
-    always_run: true
-    pass_filenames: false
 -   id: requirements-txt-fixer
-    name: Fix requirements.txt
-    description: Sorts entries in requirements.txt
+    name: fix requirements.txt
+    description: sorts entries in requirements.txt.
     entry: requirements-txt-fixer
     language: python
-    files: requirements.*\.txt$
+    files: (requirements|constraints).*\.txt$
 -   id: sort-simple-yaml
-    name: Sort simple YAML files
+    name: sort simple yaml files
+    description: sorts simple yaml files which consist only of top-level keys, preserving comments and blocks.
     language: python
     entry: sort-simple-yaml
-    description: Sorts simple YAML files which consist only of top-level keys, preserving comments and blocks.
     files: '^$'
 -   id: trailing-whitespace
-    name: Trim Trailing Whitespace
-    description: This hook trims trailing whitespace.
+    name: trim trailing whitespace
+    description: trims trailing whitespace.
     entry: trailing-whitespace-fixer
     language: python
     types: [text]
-    stages: [commit, push, manual]
+    stages: [pre-commit, pre-push, manual]
+    minimum_pre_commit_version: 3.2.0

CHANGELOG.md

@@ -1,3 +1,255 @@
+6.0.0 - 2025-08-09
+==================
+
+## Fixes
+- `check-shebang-scripts-are-executable`: improve error message.
+    - #1115 PR by @homebysix.
+
+## Migrating
+- now requires python >= 3.9.
+    - #1098 PR by @asottile.
+- `file-contents-sorter`: disallow `--unique` and `--ignore-case` at the same
+  time.
+    - #1095 PR by @nemacysts.
+    - #794 issue by @teksturi.
+- Removed `check-byte-order-marker` and `fix-encoding-pragma`.
+    - `check-byte-order-marker`: migrate to `fix-byte-order-marker`.
+    - `fix-encoding-pragma`: migrate to `pyupgrade`.
+    - #1034 PR by @mxr.
+    - #1032 issue by @mxr.
+    - #522 PR by @jgowdy.
+
+5.0.0 - 2024-10-05
+==================
+
+### Features
+- `requirements-txt-fixer`: also remove `pkg_resources==...`.
+    - #850 PR by @ericfrederich.
+    - #1030 issue by @ericfrederich.
+- `check-illegal-windows-names`: new hook!
+    - #1044 PR by @ericfrederich.
+    - #589 issue by @ericfrederich.
+    - #1049 PR by @Jeffrey-Lim.
+- `pretty-format-json`: continue processing even if a file has a json error.
+    - #1039 PR by @amarvin.
+    - #1038 issue by @amarvin.
+
+### Fixes
+- `destroyed-symlinks`: set `stages` to `[pre-commit, pre-push, manual]`
+    - PR #1085 by @AdrianDC.
+
+### Migrating
+- pre-commit-hooks now requires `pre-commit>=3.2.0`.
+- use non-deprecated names for `stages`.
+    - #1093 PR by @asottile.
+
+4.6.0 - 2024-04-06
+==================
+
+### Features
+- `requirements-txt-fixer`: remove duplicate packages.
+    - #1014 PR by @vhoulbreque-withings.
+    - #960 issue @csibe17.
+
+### Migrating
+- `fix-encoding-pragma`: deprecated -- will be removed in 5.0.0.  use
+  [pyupgrade](https://github.com/asottile/pyupgrade) or some other tool.
+    - #1033 PR by @mxr.
+    - #1032 issue by @mxr.
+
+4.5.0 - 2023-10-07
+==================
+
+### Features
+- `requirements-txt-fixer`: also sort `constraints.txt` by default.
+    - #857 PR by @lev-blit.
+    - #830 issue by @PLPeeters.
+- `debug-statements`: add `bpdb` debugger.
+    - #942 PR by @mwip.
+    - #941 issue by @mwip.
+
+### Fixes
+- `file-contents-sorter`: fix sorting an empty file.
+    - #944 PR by @RoelAdriaans.
+    - #935 issue by @paduszyk.
+- `double-quote-string-fixer`: don't rewrite inside f-strings in 3.12+.
+    - #973 PR by @asottile.
+    - #971 issue by @XuehaiPan.
+
+## Migrating
+- now requires python >= 3.8.
+    - #926 PR by @asottile.
+    - #927 PR by @asottile.
+
+4.4.0 - 2022-11-23
+==================
+
+### Features
+- `forbid-submodules`: new hook which outright bans submodules.
+    - #815 PR by @asottile.
+    - #707 issue by @ChiefGokhlayeh.
+
+4.3.0 - 2022-06-07
+==================
+
+### Features
+- `check-executables-have-shebangs`: use `git config core.fileMode` to
+  determine if it should query `git`.
+    - #730 PR by @Kurt-von-Laven.
+- `name-tests-test`: add `--pytest-test-first` test convention.
+    - #779 PR by @asottile.
+
+### Fixes
+- `check-shebang-scripts-are-executable`: update windows instructions.
+    - #774 PR by @mdeweerd.
+    - #770 issue by @mdeweerd.
+- `check-toml`: use stdlib `tomllib` when available.
+    - #771 PR by @DanielNoord.
+    - #755 issue by @sognetic.
+- `check-added-large-files`: don't run on non-file `stages`.
+    - #778 PR by @asottile.
+    - #777 issue by @skyj.
+
+4.2.0 - 2022-04-06
+==================
+
+### Features
+- `name-tests-test`: updated display text.
+    - #713 PR by @asottile.
+- `check-docstring-first`: make output more parsable.
+    - #748 PR by @asottile.
+- `check-merge-conflict`: make output more parsable.
+    - #748 PR by @asottile.
+- `debug-statements`: make output more parsable.
+    - #748 PR by @asottile.
+
+### Fixes
+- `check-merge-conflict`: fix detection of `======` conflict marker on windows.
+    - #748 PR by @asottile.
+
+### Updating
+- Drop python<3.7.
+    - #719 PR by @asottile.
+- Changed default branch from `master` to `main`.
+    - #744 PR by @asottile.
+
+4.1.0 - 2021-12-22
+==================
+
+### Features
+- `debug-statements`: add `pdbr` debugger.
+    - #614 PR by @cansarigol.
+- `detect-private-key`: add detection for additional key types.
+    - #658 PR by @ljmf00.
+- `check-executables-have-shebangs`: improve messaging on windows.
+    - #689 PR by @pujitm.
+    - #686 issue by @jmerdich.
+- `check-added-large-files`: support `--enforce-all` with `git-lfs`.
+    - #674 PR by @amartani.
+    - #560 issue by @jeremy-coulon.
+
+### Fixes
+- `check-case-conflict`: improve performance.
+    - #626 PR by @guykisel.
+    - #625 issue by @guykisel.
+- `forbid-new-submodules`: fix false-negatives for `pre-push`.
+    - #619 PR by @m-khvoinitsky.
+    - #609 issue by @m-khvoinitsky.
+- `check-merge-conflict`: fix execution in git worktrees.
+    - #662 PR by @errsyn.
+    - #638 issue by @daschuer.
+
+### Misc.
+- Normalize case of hook names and descriptions.
+    - #671 PR by @dennisroche.
+    - #673 PR by @revolter.
+
+4.0.1 - 2021-05-16
+==================
+
+### Fixes
+- `check-shebang-scripts-are-executable` fix entry point.
+    - #602 issue by @Person-93.
+    - #603 PR by @scop.
+
+4.0.0 - 2021-05-14
+==================
+
+### Features
+- `check-json`: report duplicate keys.
+    - #558 PR by @AdityaKhursale.
+    - #554 issue by @adamchainz.
+- `no-commit-to-branch`: add `main` to default blocked branches.
+    - #565 PR by @ndevenish.
+- `check-case-conflict`: check conflicts in directory names as well.
+    - #575 PR by @slsyy.
+    - #70 issue by @andyjack.
+- `check-vcs-permalinks`: forbid other branch names.
+    - #582 PR by @jack1142.
+    - #581 issue by @jack1142.
+- `check-shebang-scripts-are-executable`: new hook which ensures shebang'd
+  scripts are executable.
+    - #545 PR by @scop.
+
+### Fixes
+- `check-executables-have-shebangs`: Short circuit shebang lookup on windows.
+    - #544 PR by @scop.
+- `requirements-txt-fixer`: Fix comments which have indentation
+    - #549 PR by @greshilov.
+    - #548 issue by @greshilov.
+- `pretty-format-json`: write to stdout using UTF-8 encoding.
+    - #571 PR by @jack1142.
+    - #570 issue by @jack1142.
+- Use more inclusive language.
+    - #599 PR by @asottile.
+
+### Breaking changes
+- Remove deprecated hooks: `flake8`, `pyflakes`, `autopep8-wrapper`.
+    - #597 PR by @asottile.
+
+
+3.4.0 - 2020-12-15
+==================
+
+### Features
+- `file-contents-sorter`: Add `--unique` argument
+    - #524 PR by @danielhoherd.
+- `check-vcs-permalinks`: Add `--additional-github-domain` option
+    - #530 PR by @youngminz.
+- New hook: `destroyed-symlinks` to detect unintentional symlink-breakages on
+  windows.
+    - #511 PR by @m-khvoinitsky.
+
+3.3.0 - 2020-10-20
+==================
+
+### Features
+- `file-contents-sorter`: add `--ignore-case` option for case-insensitive
+  sorting
+    - #514 PR by @Julian.
+- `check-added-large-files`: add `--enforce-all` option to check non-added
+  files as well
+    - #519 PR by @mshawcroft.
+    - #518 issue by @mshawcroft.
+- `fix-byte-order-marker`: new hook which fixes UTF-8 byte-order marker.
+    - #522 PR by @jgowdy.
+
+### Deprecations
+- `check-byte-order-marker` is now deprecated for `fix-byte-order-marker`
+
+3.2.0 - 2020-07-30
+==================
+
+### Features
+- `debug-statements`: add support for `pydevd_pycharm` debugger
+    - #502 PR by @jgeerds.
+
+### Fixes
+- `check-executables-have-shebangs`: fix git-quoted files on windows (spaces,
+  non-ascii, etc.)
+    - #509 PR by @pawamoy.
+    - #508 issue by @pawamoy.
+
 3.1.0 - 2020-05-20
 ==================
 

README.md

@@ -1,5 +1,5 @@
-[![Build Status](https://asottile.visualstudio.com/asottile/_apis/build/status/pre-commit.pre-commit-hooks?branchName=master)](https://asottile.visualstudio.com/asottile/_build/latest?definitionId=17&branchName=master)
-[![Azure DevOps coverage](https://img.shields.io/azure-devops/coverage/asottile/asottile/17/master.svg)](https://dev.azure.com/asottile/asottile/_build/latest?definitionId=17&branchName=master)
+[![build status](https://github.com/pre-commit/pre-commit-hooks/actions/workflows/main.yml/badge.svg)](https://github.com/pre-commit/pre-commit-hooks/actions/workflows/main.yml)
+[![pre-commit.ci status](https://results.pre-commit.ci/badge/github/pre-commit/pre-commit-hooks/main.svg)](https://results.pre-commit.ci/latest/github/pre-commit/pre-commit-hooks/main)
 
 pre-commit-hooks
 ================
@@ -15,7 +15,7 @@ Add this to your `.pre-commit-config.yaml`
 
 ```yaml
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.1.0  # Use the ref you want to point at
+    rev: v6.0.0  # Use the ref you want to point at
     hooks:
     -   id: trailing-whitespace
     # -   id: ...
@@ -26,8 +26,11 @@ Add this to your `.pre-commit-config.yaml`
 #### `check-added-large-files`
 Prevent giant files from being committed.
   - Specify what is "too large" with `args: ['--maxkb=123']` (default=500kB).
+  - Limits checked files to those indicated as staged for addition by git.
   - If `git-lfs` is installed, lfs files will be skipped
     (requires `git-lfs>=2.2.1`)
+  - `--enforce-all` - Check all listed files not just those staged for
+    addition.
 
 #### `check-ast`
 Simply check whether files parse as valid python.
@@ -39,23 +42,24 @@ Require literal syntax when initializing empty or zero Python builtin types.
   - Ignore this requirement for specific builtin types with `--ignore=type1,type2,…`.
   - Forbid `dict` keyword syntax with `--no-allow-dict-kwargs`.
 
-#### `check-byte-order-marker`
-Forbid files which have a UTF-8 byte-order marker
-
 #### `check-case-conflict`
 Check for files with names that would conflict on a case-insensitive filesystem like MacOS HFS+ or Windows FAT.
 
-#### `check-docstring-first`
-Checks for a common error of placing code before the docstring.
-
 #### `check-executables-have-shebangs`
 Checks that non-binary executables have a proper shebang.
 
+#### `check-illegal-windows-names`
+Check for files that cannot be created on Windows.
+
 #### `check-json`
 Attempts to load all json files to verify syntax.
 
 #### `check-merge-conflict`
 Check for files that contain merge conflict strings.
+  - `--assume-in-merge` - Allows running the hook when there is no ongoing merge operation
+
+#### `check-shebang-scripts-are-executable`
+Checks that scripts with shebangs are executable.
 
 #### `check-symlinks`
 Checks for symlinks which do not point to anything.
@@ -65,6 +69,10 @@ Attempts to load all TOML files to verify syntax.
 
 #### `check-vcs-permalinks`
 Ensures that links to vcs websites are permalinks.
+  - `--additional-github-domain DOMAIN` - Add check for specified domain.
+    Can be repeated multiple times.  for example, if your company uses
+    GitHub Enterprise you may use something like
+    `--additional-github-domain github.example.com`
 
 #### `check-xml`
 Attempts to load all xml files to verify syntax.
@@ -82,6 +90,12 @@ Attempts to load all yaml files to verify syntax.
 #### `debug-statements`
 Check for debugger imports and py37+ `breakpoint()` calls in python source.
 
+#### `destroyed-symlinks`
+Detects symlinks which are changed to regular files with a content of a path
+which that symlink was pointing to.
+This usually happens on Windows when a user clones a repository that has
+symlinks but they do not have the permission to create symlinks.
+
 #### `detect-aws-credentials`
 Checks for the existence of AWS secrets that you have set up with the AWS CLI.
 The following arguments are available:
@@ -99,18 +113,28 @@ This hook replaces double quoted strings with single quoted strings.
 #### `end-of-file-fixer`
 Makes sure files end in a newline and only a newline.
 
-#### `fix-encoding-pragma`
-Add `# -*- coding: utf-8 -*-` to the top of python files.
-  - To remove the coding pragma pass `--remove` (useful in a python3-only codebase)
-
 #### `file-contents-sorter`
 Sort the lines in specified files (defaults to alphabetical).
-You must provide list of target files as input to it.
+You must provide the target [`files`](https://pre-commit.com/#config-files) as input.
 Note that this hook WILL remove blank lines and does NOT respect any comments.
+All newlines will be converted to line feeds (`\n`).
+
+The following arguments are available:
+- `--ignore-case` - fold lower case to upper case characters.
+- `--unique` - ensure each line is unique.
+
+#### `fix-byte-order-marker`
+removes UTF-8 byte order marker
 
 #### `forbid-new-submodules`
 Prevent addition of new git submodules.
 
+This is intended as a helper to migrate away from submodules.  If you want to
+ban them entirely use `forbid-submodules`
+
+#### `forbid-submodules`
+forbids any submodules in the repository.
+
 #### `mixed-line-ending`
 Replaces or checks mixed line ending.
   - `--fix={auto,crlf,lf,no}`
@@ -120,18 +144,27 @@ Replaces or checks mixed line ending.
       - `no` - Checks if there is any mixed line ending without modifying any file.
 
 #### `name-tests-test`
-Assert that files in tests/ end in `_test.py`.
-  - Use `args: ['--django']` to match `test*.py` instead.
+verifies that test files are named correctly.
+- `--pytest` (the default): ensure tests match `.*_test\.py`
+- `--pytest-test-first`: ensure tests match `test_.*\.py`
+- `--django` / `--unittest`: ensure tests match `test.*\.py`
 
 #### `no-commit-to-branch`
 Protect specific branches from direct checkins.
-  - Use `args: [--branch, staging, --branch, master]` to set the branch.
-    `master` is the default if no branch argument is set.
+  - Use `args: [--branch, staging, --branch, main]` to set the branch.
+    Both `main` and `master` are protected by default if no branch argument is set.
   - `-b` / `--branch` may be specified multiple times to protect multiple
     branches.
   - `-p` / `--pattern` can be used to protect branches that match a supplied regex
     (e.g. `--pattern, release/.*`). May be specified multiple times.
 
+Note that `no-commit-to-branch` is configured by default to [`always_run`](https://pre-commit.com/#config-always_run).
+As a result, it will ignore any setting of [`files`](https://pre-commit.com/#config-files),
+[`exclude`](https://pre-commit.com/#config-exclude), [`types`](https://pre-commit.com/#config-types)
+or [`exclude_types`](https://pre-commit.com/#config-exclude_types).
+Set [`always_run: false`](https://pre-commit.com/#config-always_run) to allow this hook to be skipped according to these
+file filters. Caveat: In this configuration, empty commits (`git commit --allow-empty`) would always be allowed by this hook.
+
 #### `pretty-format-json`
 Checks that all your JSON files are pretty.  "Pretty"
 here means that keys are sorted and indented.  You can configure this with
@@ -143,14 +176,14 @@ the following commandline options:
   - `--top-keys comma,separated,keys` - Keys to keep at the top of mappings.
 
 #### `requirements-txt-fixer`
-Sorts entries in requirements.txt and removes incorrect entry for `pkg-resources==0.0.0`
+Sorts entries in requirements.txt and constraints.txt and removes incorrect entry for `pkg-resources==0.0.0`
 
 #### `sort-simple-yaml`
 Sorts simple YAML files which consist only of top-level
 keys, preserving comments and blocks.
 
 Note that `sort-simple-yaml` by default matches no `files` as it enforces a
-very specific format.  You must opt in to this by setting `files`, for example:
+very specific format.  You must opt in to this by setting [`files`](https://pre-commit.com/#config-files), for example:
 
 ```yaml
     -   id: sort-simple-yaml
@@ -169,10 +202,10 @@ Trims trailing whitespace.
 
 ### Deprecated / replaced hooks
 
-- `autopep8-wrapper`: instead use
-  [mirrors-autopep8](https://github.com/pre-commit/mirrors-autopep8)
-- `pyflakes`: instead use `flake8`
-- `flake8`: instead use [upstream flake8](https://gitlab.com/pycqa/flake8)
+- `check-byte-order-marker`: instead use fix-byte-order-marker
+- `fix-encoding-pragma`: instead use [`pyupgrade`](https://github.com/asottile/pyupgrade)
+- `check-docstring-first`: fundamentally flawed, deprecated without replacement.
+
 
 ### As a standalone package
 

azure-pipelines.yml

@@ -1,24 +0,0 @@
-trigger:
-  branches:
-    include: [master, test-me-*]
-  tags:
-    include: ['*']
-
-resources:
-  repositories:
-    - repository: asottile
-      type: github
-      endpoint: github
-      name: asottile/azure-pipeline-templates
-      ref: refs/tags/v1.0.0
-
-jobs:
-- template: job--pre-commit.yml@asottile
-- template: job--python-tox.yml@asottile
-  parameters:
-    toxenvs: [py38]
-    os: windows
-- template: job--python-tox.yml@asottile
-  parameters:
-    toxenvs: [pypy3, py36, py37, py38]
-    os: linux

pre_commit_hooks/check_added_large_files.py

@@ -1,32 +1,52 @@
+from __future__ import annotations
+
 import argparse
-import json
 import math
 import os
-from typing import Optional
-from typing import Sequence
-from typing import Set
+import subprocess
+from collections.abc import Sequence
 
 from pre_commit_hooks.util import added_files
-from pre_commit_hooks.util import CalledProcessError
-from pre_commit_hooks.util import cmd_output
+from pre_commit_hooks.util import zsplit
 
 
-def lfs_files() -> Set[str]:
-    try:
-        # Introduced in git-lfs 2.2.0, first working in 2.2.1
-        lfs_ret = cmd_output('git', 'lfs', 'status', '--json')
-    except CalledProcessError:  # pragma: no cover (with git-lfs)
-        lfs_ret = '{"files":{}}'
+def filter_lfs_files(filenames: set[str]) -> None:  # pragma: no cover (lfs)
+    """Remove files tracked by git-lfs from the set."""
+    if not filenames:
+        return
 
-    return set(json.loads(lfs_ret)['files'])
+    check_attr = subprocess.run(
+        ('git', 'check-attr', 'filter', '-z', '--stdin'),
+        stdout=subprocess.PIPE,
+        stderr=subprocess.DEVNULL,
+        encoding='utf-8',
+        check=True,
+        input='\0'.join(filenames),
+    )
+    stdout = zsplit(check_attr.stdout)
+    for i in range(0, len(stdout), 3):
+        filename, filter_tag = stdout[i], stdout[i + 2]
+        if filter_tag == 'lfs':
+            filenames.remove(filename)
 
 
-def find_large_added_files(filenames: Sequence[str], maxkb: int) -> int:
+def find_large_added_files(
+        filenames: Sequence[str],
+        maxkb: int,
+        *,
+        enforce_all: bool = False,
+) -> int:
     # Find all added files that are also in the list of files pre-commit tells
     # us about
     retv = 0
-    for filename in (added_files() & set(filenames)) - lfs_files():
-        kb = int(math.ceil(os.stat(filename).st_size / 1024))
+    filenames_filtered = set(filenames)
+    filter_lfs_files(filenames_filtered)
+
+    if not enforce_all:
+        filenames_filtered &= added_files()
+
+    for filename in filenames_filtered:
+        kb = math.ceil(os.stat(filename).st_size / 1024)
         if kb > maxkb:
             print(f'{filename} ({kb} KB) exceeds {maxkb} KB.')
             retv = 1
@@ -34,20 +54,28 @@ def find_large_added_files(filenames: Sequence[str], maxkb: int) -> int:
     return retv
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         'filenames', nargs='*',
         help='Filenames pre-commit believes are changed.',
     )
     parser.add_argument(
-        '--maxkb', type=int, default=500,
-        help='Maxmimum allowable KB for added files',
+        '--enforce-all', action='store_true',
+        help='Enforce all files are checked, not just staged files.',
+    )
+    parser.add_argument(
+        '--maxkb', type=int, default=500,
+        help='Maximum allowable KB for added files',
     )
-
     args = parser.parse_args(argv)
-    return find_large_added_files(args.filenames, args.maxkb)
+
+    return find_large_added_files(
+        args.filenames,
+        args.maxkb,
+        enforce_all=args.enforce_all,
+    )
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_ast.py

@@ -1,13 +1,14 @@
+from __future__ import annotations
+
 import argparse
 import ast
 import platform
 import sys
 import traceback
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
     args = parser.parse_args(argv)
@@ -29,4 +30,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_builtin_literals.py

@@ -1,10 +1,9 @@
+from __future__ import annotations
+
 import argparse
 import ast
-from typing import List
+from collections.abc import Sequence
 from typing import NamedTuple
-from typing import Optional
-from typing import Sequence
-from typing import Set
 
 
 BUILTIN_TYPES = {
@@ -27,38 +26,39 @@ class Call(NamedTuple):
 class Visitor(ast.NodeVisitor):
     def __init__(
             self,
-            ignore: Optional[Sequence[str]] = None,
+            ignore: set[str],
             allow_dict_kwargs: bool = True,
     ) -> None:
-        self.builtin_type_calls: List[Call] = []
-        self.ignore = set(ignore) if ignore else set()
+        self.builtin_type_calls: list[Call] = []
         self.allow_dict_kwargs = allow_dict_kwargs
+        self._disallowed = BUILTIN_TYPES.keys() - ignore
 
     def _check_dict_call(self, node: ast.Call) -> bool:
         return self.allow_dict_kwargs and bool(node.keywords)
 
     def visit_Call(self, node: ast.Call) -> None:
-        if not isinstance(node.func, ast.Name):
+        if (
             # Ignore functions that are object attributes (`foo.bar()`).
             # Assume that if the user calls `builtins.list()`, they know what
             # they're doing.
-            return
-        if node.func.id not in set(BUILTIN_TYPES).difference(self.ignore):
-            return
-        if node.func.id == 'dict' and self._check_dict_call(node):
-            return
-        elif node.args:
-            return
-        self.builtin_type_calls.append(
-            Call(node.func.id, node.lineno, node.col_offset),
-        )
+            isinstance(node.func, ast.Name) and
+            node.func.id in self._disallowed and
+            (node.func.id != 'dict' or not self._check_dict_call(node)) and
+            not node.args
+        ):
+            self.builtin_type_calls.append(
+                Call(node.func.id, node.lineno, node.col_offset),
+            )
+
+        self.generic_visit(node)
 
 
 def check_file(
         filename: str,
-        ignore: Optional[Sequence[str]] = None,
+        *,
+        ignore: set[str],
         allow_dict_kwargs: bool = True,
-) -> List[Call]:
+) -> list[Call]:
     with open(filename, 'rb') as f:
         tree = ast.parse(f.read(), filename=filename)
     visitor = Visitor(ignore=ignore, allow_dict_kwargs=allow_dict_kwargs)
@@ -66,11 +66,11 @@ def check_file(
     return visitor.builtin_type_calls
 
 
-def parse_ignore(value: str) -> Set[str]:
+def parse_ignore(value: str) -> set[str]:
     return set(value.split(','))
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
     parser.add_argument('--ignore', type=parse_ignore, default=set())
@@ -103,4 +103,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_byte_order_marker.py

@@ -1,23 +0,0 @@
-import argparse
-from typing import Optional
-from typing import Sequence
-
-
-def main(argv: Optional[Sequence[str]] = None) -> int:
-    parser = argparse.ArgumentParser()
-    parser.add_argument('filenames', nargs='*', help='Filenames to check')
-    args = parser.parse_args(argv)
-
-    retv = 0
-
-    for filename in args.filenames:
-        with open(filename, 'rb') as f:
-            if f.read(3) == b'\xef\xbb\xbf':
-                retv = 1
-                print(f'{filename}: Has a byte-order marker')
-
-    return retv
-
-
-if __name__ == '__main__':
-    exit(main())

pre_commit_hooks/check_case_conflict.py

@@ -1,20 +1,35 @@
+from __future__ import annotations
+
 import argparse
-from typing import Iterable
-from typing import Optional
-from typing import Sequence
-from typing import Set
+from collections.abc import Iterable
+from collections.abc import Iterator
+from collections.abc import Sequence
 
 from pre_commit_hooks.util import added_files
 from pre_commit_hooks.util import cmd_output
 
 
-def lower_set(iterable: Iterable[str]) -> Set[str]:
+def lower_set(iterable: Iterable[str]) -> set[str]:
     return {x.lower() for x in iterable}
 
 
+def parents(file: str) -> Iterator[str]:
+    path_parts = file.split('/')
+    path_parts.pop()
+    while path_parts:
+        yield '/'.join(path_parts)
+        path_parts.pop()
+
+
+def directories_for(files: set[str]) -> set[str]:
+    return {parent for file in files for parent in parents(file)}
+
+
 def find_conflicting_filenames(filenames: Sequence[str]) -> int:
     repo_files = set(cmd_output('git', 'ls-files').splitlines())
+    repo_files |= directories_for(repo_files)
     relevant_files = set(filenames) | added_files()
+    relevant_files |= directories_for(relevant_files)
     repo_files -= relevant_files
     retv = 0
 
@@ -41,7 +56,7 @@ def find_conflicting_filenames(filenames: Sequence[str]) -> int:
     return retv
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         'filenames', nargs='*',
@@ -54,4 +69,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_docstring_first.py

@@ -1,9 +1,10 @@
+from __future__ import annotations
+
 import argparse
 import io
 import tokenize
+from collections.abc import Sequence
 from tokenize import tokenize as tokenize_tokenize
-from typing import Optional
-from typing import Sequence
 
 NON_CODE_TOKENS = frozenset((
     tokenize.COMMENT, tokenize.ENDMARKER, tokenize.NEWLINE, tokenize.NL,
@@ -27,13 +28,13 @@ def check_docstring_first(src: bytes, filename: str = '<unknown>') -> int:
         if tok_type == tokenize.STRING and scol == 0:
             if found_docstring_line is not None:
                 print(
-                    f'{filename}:{sline} Multiple module docstrings '
+                    f'{filename}:{sline}: Multiple module docstrings '
                     f'(first docstring on line {found_docstring_line}).',
                 )
                 return 1
             elif found_code_line is not None:
                 print(
-                    f'{filename}:{sline} Module docstring appears after code '
+                    f'{filename}:{sline}: Module docstring appears after code '
                     f'(code seen on line {found_code_line}).',
                 )
                 return 1
@@ -45,7 +46,7 @@ def check_docstring_first(src: bytes, filename: str = '<unknown>') -> int:
     return 0
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
     args = parser.parse_args(argv)

pre_commit_hooks/check_executables_have_shebangs.py

@@ -1,47 +1,60 @@
 """Check that executable text files have a shebang."""
+from __future__ import annotations
+
 import argparse
 import shlex
 import sys
-from typing import List
-from typing import Optional
-from typing import Sequence
-from typing import Set
+from collections.abc import Generator
+from collections.abc import Sequence
+from typing import NamedTuple
 
 from pre_commit_hooks.util import cmd_output
+from pre_commit_hooks.util import zsplit
 
 EXECUTABLE_VALUES = frozenset(('1', '3', '5', '7'))
 
 
-def check_executables(paths: List[str]) -> int:
-    if sys.platform == 'win32':  # pragma: win32 cover
+def check_executables(paths: list[str]) -> int:
+    fs_tracks_executable_bit = cmd_output(
+        'git', 'config', 'core.fileMode', retcode=None,
+    ).strip()
+    if fs_tracks_executable_bit == 'false':  # pragma: win32 cover
         return _check_git_filemode(paths)
     else:  # pragma: win32 no cover
         retv = 0
         for path in paths:
-            if not _check_has_shebang(path):
+            if not has_shebang(path):
                 _message(path)
                 retv = 1
 
         return retv
 
 
-def _check_git_filemode(paths: Sequence[str]) -> int:
-    outs = cmd_output('git', 'ls-files', '--stage', '--', *paths)
-    seen: Set[str] = set()
-    for out in outs.splitlines():
-        metadata, path = out.split('\t')
-        tagmode = metadata.split(' ', 1)[0]
+class GitLsFile(NamedTuple):
+    mode: str
+    filename: str
 
-        is_executable = any(b in EXECUTABLE_VALUES for b in tagmode[-3:])
-        has_shebang = _check_has_shebang(path)
-        if is_executable and not has_shebang:
-            _message(path)
-            seen.add(path)
+
+def git_ls_files(paths: Sequence[str]) -> Generator[GitLsFile]:
+    outs = cmd_output('git', 'ls-files', '-z', '--stage', '--', *paths)
+    for out in zsplit(outs):
+        metadata, filename = out.split('\t')
+        mode, _, _ = metadata.split()
+        yield GitLsFile(mode, filename)
+
+
+def _check_git_filemode(paths: Sequence[str]) -> int:
+    seen: set[str] = set()
+    for ls_file in git_ls_files(paths):
+        is_executable = any(b in EXECUTABLE_VALUES for b in ls_file.mode[-3:])
+        if is_executable and not has_shebang(ls_file.filename):
+            _message(ls_file.filename)
+            seen.add(ls_file.filename)
 
     return int(bool(seen))
 
 
-def _check_has_shebang(path: str) -> int:
+def has_shebang(path: str) -> int:
     with open(path, 'rb') as f:
         first_bytes = f.read(2)
 
@@ -53,12 +66,14 @@ def _message(path: str) -> None:
         f'{path}: marked executable but has no (or invalid) shebang!\n'
         f"  If it isn't supposed to be executable, try: "
         f'`chmod -x {shlex.quote(path)}`\n'
+        f'  If on Windows, you may also need to: '
+        f'`git add --chmod=-x {shlex.quote(path)}`\n'
         f'  If it is supposed to be executable, double-check its shebang.',
         file=sys.stderr,
     )
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser(description=__doc__)
     parser.add_argument('filenames', nargs='*')
     args = parser.parse_args(argv)
@@ -67,4 +82,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_json.py

@@ -1,10 +1,24 @@
+from __future__ import annotations
+
 import argparse
 import json
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
+from typing import Any
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def raise_duplicate_keys(
+        ordered_pairs: list[tuple[str, Any]],
+) -> dict[str, Any]:
+    d = {}
+    for key, val in ordered_pairs:
+        if key in d:
+            raise ValueError(f'Duplicate key: {key}')
+        else:
+            d[key] = val
+    return d
+
+
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to check.')
     args = parser.parse_args(argv)
@@ -13,7 +27,7 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     for filename in args.filenames:
         with open(filename, 'rb') as f:
             try:
-                json.load(f)
+                json.load(f, object_pairs_hook=raise_duplicate_keys)
             except ValueError as exc:
                 print(f'{filename}: Failed to json decode ({exc})')
                 retval = 1
@@ -21,4 +35,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_merge_conflict.py

@@ -1,29 +1,34 @@
+from __future__ import annotations
+
 import argparse
 import os.path
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
+
+from pre_commit_hooks.util import cmd_output
 
 
 CONFLICT_PATTERNS = [
     b'<<<<<<< ',
     b'======= ',
+    b'=======\r\n',
     b'=======\n',
     b'>>>>>>> ',
 ]
 
 
-def is_in_merge() -> int:
+def is_in_merge() -> bool:
+    git_dir = cmd_output('git', 'rev-parse', '--git-dir').rstrip()
     return (
-        os.path.exists(os.path.join('.git', 'MERGE_MSG')) and
+        os.path.exists(os.path.join(git_dir, 'MERGE_MSG')) and
         (
-            os.path.exists(os.path.join('.git', 'MERGE_HEAD')) or
-            os.path.exists(os.path.join('.git', 'rebase-apply')) or
-            os.path.exists(os.path.join('.git', 'rebase-merge'))
+            os.path.exists(os.path.join(git_dir, 'MERGE_HEAD')) or
+            os.path.exists(os.path.join(git_dir, 'rebase-apply')) or
+            os.path.exists(os.path.join(git_dir, 'rebase-merge'))
         )
     )
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
     parser.add_argument('--assume-in-merge', action='store_true')
@@ -35,12 +40,12 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     retcode = 0
     for filename in args.filenames:
         with open(filename, 'rb') as inputfile:
-            for i, line in enumerate(inputfile):
+            for i, line in enumerate(inputfile, start=1):
                 for pattern in CONFLICT_PATTERNS:
                     if line.startswith(pattern):
                         print(
-                            f'Merge conflict string "{pattern.decode()}" '
-                            f'found in {filename}:{i + 1}',
+                            f'{filename}:{i}: Merge conflict string '
+                            f'{pattern.strip().decode()!r} found',
                         )
                         retcode = 1
 
@@ -48,4 +53,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_shebang_scripts_are_executable.py

@@ -0,0 +1,54 @@
+"""Check that text files with a shebang are executable."""
+from __future__ import annotations
+
+import argparse
+import shlex
+import sys
+from collections.abc import Sequence
+
+from pre_commit_hooks.check_executables_have_shebangs import EXECUTABLE_VALUES
+from pre_commit_hooks.check_executables_have_shebangs import git_ls_files
+from pre_commit_hooks.check_executables_have_shebangs import has_shebang
+
+
+def check_shebangs(paths: list[str]) -> int:
+    # Cannot optimize on non-executability here if we intend this check to
+    # work on win32 -- and that's where problems caused by non-executability
+    # (elsewhere) are most likely to arise from.
+    return _check_git_filemode(paths)
+
+
+def _check_git_filemode(paths: Sequence[str]) -> int:
+    seen: set[str] = set()
+    for ls_file in git_ls_files(paths):
+        is_executable = any(b in EXECUTABLE_VALUES for b in ls_file.mode[-3:])
+        if not is_executable and has_shebang(ls_file.filename):
+            _message(ls_file.filename)
+            seen.add(ls_file.filename)
+
+    return int(bool(seen))
+
+
+def _message(path: str) -> None:
+    print(
+        f'{path}: has a shebang but is not marked executable!\n'
+        f'  If it is supposed to be executable, try: '
+        f'`chmod +x {shlex.quote(path)}`\n'
+        f'  If on Windows, you may also need to: '
+        f'`git add --chmod=+x {shlex.quote(path)}`\n'
+        f'  If it is not supposed to be executable, double-check its shebang '
+        f'is wanted.\n',
+        file=sys.stderr,
+    )
+
+
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument('filenames', nargs='*')
+    args = parser.parse_args(argv)
+
+    return check_shebangs(args.filenames)
+
+
+if __name__ == '__main__':
+    raise SystemExit(main())

pre_commit_hooks/check_symlinks.py

@@ -1,10 +1,11 @@
+from __future__ import annotations
+
 import argparse
 import os.path
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser(description='Checks for broken symlinks.')
     parser.add_argument('filenames', nargs='*', help='Filenames to check')
     args = parser.parse_args(argv)
@@ -23,4 +24,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_toml.py

@@ -1,11 +1,16 @@
+from __future__ import annotations
+
 import argparse
-from typing import Optional
-from typing import Sequence
+import sys
+from collections.abc import Sequence
 
-import toml
+if sys.version_info >= (3, 11):  # pragma: >=3.11 cover
+    import tomllib
+else:  # pragma: <3.11 cover
+    import tomli as tomllib
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to check.')
     args = parser.parse_args(argv)
@@ -13,12 +18,13 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     retval = 0
     for filename in args.filenames:
         try:
-            toml.load(filename)
-        except toml.TomlDecodeError as exc:
+            with open(filename, mode='rb') as fp:
+                tomllib.load(fp)
+        except tomllib.TOMLDecodeError as exc:
             print(f'{filename}: {exc}')
             retval = 1
     return retval
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_vcs_permalinks.py

@@ -1,35 +1,53 @@
+from __future__ import annotations
+
 import argparse
 import re
 import sys
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
+from re import Pattern
 
 
-GITHUB_NON_PERMALINK = re.compile(
-    br'https://github.com/[^/ ]+/[^/ ]+/blob/master/[^# ]+#L\d+',
-)
+def _get_pattern(domain: str) -> Pattern[bytes]:
+    regex = (
+        rf'https://{domain}/[^/ ]+/[^/ ]+/blob/'
+        r'(?![a-fA-F0-9]{4,64}/)([^/. ]+)/[^# ]+#L\d+'
+    )
+    return re.compile(regex.encode())
 
 
-def _check_filename(filename: str) -> int:
+def _check_filename(filename: str, patterns: list[Pattern[bytes]]) -> int:
     retv = 0
     with open(filename, 'rb') as f:
         for i, line in enumerate(f, 1):
-            if GITHUB_NON_PERMALINK.search(line):
-                sys.stdout.write(f'{filename}:{i}:')
-                sys.stdout.flush()
-                sys.stdout.buffer.write(line)
-                retv = 1
+            for pattern in patterns:
+                if pattern.search(line):
+                    sys.stdout.write(f'{filename}:{i}:')
+                    sys.stdout.flush()
+                    sys.stdout.buffer.write(line)
+                    retv = 1
     return retv
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
+    parser.add_argument(
+        '--additional-github-domain',
+        dest='additional_github_domains',
+        action='append',
+        default=['github.com'],
+    )
     args = parser.parse_args(argv)
 
+    patterns = [
+        _get_pattern(domain)
+        for domain in args.additional_github_domains
+    ]
+
     retv = 0
+
     for filename in args.filenames:
-        retv |= _check_filename(filename)
+        retv |= _check_filename(filename, patterns)
 
     if retv:
         print()
@@ -39,4 +57,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_xml.py

@@ -1,10 +1,11 @@
+from __future__ import annotations
+
 import argparse
 import xml.sax.handler
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='XML filenames to check.')
     args = parser.parse_args(argv)
@@ -22,4 +23,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/check_yaml.py

@@ -1,16 +1,17 @@
+from __future__ import annotations
+
 import argparse
+from collections.abc import Generator
+from collections.abc import Sequence
 from typing import Any
-from typing import Generator
 from typing import NamedTuple
-from typing import Optional
-from typing import Sequence
 
 import ruamel.yaml
 
 yaml = ruamel.yaml.YAML(typ='safe')
 
 
-def _exhaust(gen: Generator[str, None, None]) -> None:
+def _exhaust(gen: Generator[str]) -> None:
     for _ in gen:
         pass
 
@@ -36,7 +37,7 @@ LOAD_FNS = {
 }
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         '-m', '--multi', '--allow-multiple-documents', action='store_true',
@@ -45,7 +46,7 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
         '--unsafe', action='store_true',
         help=(
             'Instead of loading the files, simply parse them for syntax.  '
-            'A syntax-only check enables extensions and unsafe contstructs '
+            'A syntax-only check enables extensions and unsafe constructs '
             'which would otherwise be forbidden.  Using this option removes '
             'all guarantees of portability to other yaml implementations.  '
             'Implies --allow-multiple-documents'
@@ -68,4 +69,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/debug_statement_hook.py

@@ -1,13 +1,24 @@
+from __future__ import annotations
+
 import argparse
 import ast
 import traceback
-from typing import List
+from collections.abc import Sequence
 from typing import NamedTuple
-from typing import Optional
-from typing import Sequence
 
 
-DEBUG_STATEMENTS = {'pdb', 'ipdb', 'pudb', 'q', 'rdb', 'rpdb', 'wdb'}
+DEBUG_STATEMENTS = {
+    'bpdb',
+    'ipdb',
+    'pdb',
+    'pdbr',
+    'pudb',
+    'pydevd_pycharm',
+    'q',
+    'rdb',
+    'rpdb',
+    'wdb',
+}
 
 
 class Debug(NamedTuple):
@@ -19,7 +30,7 @@ class Debug(NamedTuple):
 
 class DebugStatementParser(ast.NodeVisitor):
     def __init__(self) -> None:
-        self.breakpoints: List[Debug] = []
+        self.breakpoints: list[Debug] = []
 
     def visit_Import(self, node: ast.Import) -> None:
         for name in node.names:
@@ -55,12 +66,12 @@ def check_file(filename: str) -> int:
     visitor.visit(ast_obj)
 
     for bp in visitor.breakpoints:
-        print(f'{filename}:{bp.line}:{bp.col} - {bp.name} {bp.reason}')
+        print(f'{filename}:{bp.line}:{bp.col}: {bp.name} {bp.reason}')
 
     return int(bool(visitor.breakpoints))
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to run')
     args = parser.parse_args(argv)
@@ -72,4 +83,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/destroyed_symlinks.py

@@ -0,0 +1,92 @@
+from __future__ import annotations
+
+import argparse
+import shlex
+import subprocess
+from collections.abc import Sequence
+
+from pre_commit_hooks.util import cmd_output
+from pre_commit_hooks.util import zsplit
+
+ORDINARY_CHANGED_ENTRIES_MARKER = '1'
+PERMS_LINK = '120000'
+PERMS_NONEXIST = '000000'
+
+
+def find_destroyed_symlinks(files: Sequence[str]) -> list[str]:
+    destroyed_links: list[str] = []
+    if not files:
+        return destroyed_links
+    for line in zsplit(
+        cmd_output('git', 'status', '--porcelain=v2', '-z', '--', *files),
+    ):
+        splitted = line.split(' ')
+        if splitted and splitted[0] == ORDINARY_CHANGED_ENTRIES_MARKER:
+            # https://git-scm.com/docs/git-status#_changed_tracked_entries
+            (
+                _, _, _,
+                mode_HEAD,
+                mode_index,
+                _,
+                hash_HEAD,
+                hash_index,
+                *path_splitted,
+            ) = splitted
+            path = ' '.join(path_splitted)
+            if (
+                    mode_HEAD == PERMS_LINK and
+                    mode_index != PERMS_LINK and
+                    mode_index != PERMS_NONEXIST
+            ):
+                if hash_HEAD == hash_index:
+                    # if old and new hashes are equal, it's not needed to check
+                    # anything more, we've found a destroyed symlink for sure
+                    destroyed_links.append(path)
+                else:
+                    # if old and new hashes are *not* equal, it doesn't mean
+                    # that everything is OK - new file may be altered
+                    # by something like trailing-whitespace and/or
+                    # mixed-line-ending hooks so we need to go deeper
+                    SIZE_CMD = ('git', 'cat-file', '-s')
+                    size_index = int(cmd_output(*SIZE_CMD, hash_index).strip())
+                    size_HEAD = int(cmd_output(*SIZE_CMD, hash_HEAD).strip())
+
+                    # in the worst case new file may have CRLF added
+                    # so check content only if new file is bigger
+                    # not more than 2 bytes compared to the old one
+                    if size_index <= size_HEAD + 2:
+                        head_content = subprocess.check_output(
+                            ('git', 'cat-file', '-p', hash_HEAD),
+                        ).rstrip()
+                        index_content = subprocess.check_output(
+                            ('git', 'cat-file', '-p', hash_index),
+                        ).rstrip()
+                        if head_content == index_content:
+                            destroyed_links.append(path)
+    return destroyed_links
+
+
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument('filenames', nargs='*', help='Filenames to check.')
+    args = parser.parse_args(argv)
+    destroyed_links = find_destroyed_symlinks(files=args.filenames)
+    if destroyed_links:
+        print('Destroyed symlinks:')
+        for destroyed_link in destroyed_links:
+            print(f'- {destroyed_link}')
+        print('You should unstage affected files:')
+        print(f'\tgit reset HEAD -- {shlex.join(destroyed_links)}')
+        print(
+            'And retry commit. As a long term solution '
+            'you may try to explicitly tell git that your '
+            'environment does not support symlinks:',
+        )
+        print('\tgit config core.symlinks false')
+        return 1
+    else:
+        return 0
+
+
+if __name__ == '__main__':
+    raise SystemExit(main())

pre_commit_hooks/detect_aws_credentials.py

@@ -1,11 +1,10 @@
+from __future__ import annotations
+
 import argparse
 import configparser
 import os
-from typing import List
+from collections.abc import Sequence
 from typing import NamedTuple
-from typing import Optional
-from typing import Sequence
-from typing import Set
 
 
 class BadFile(NamedTuple):
@@ -13,7 +12,7 @@ class BadFile(NamedTuple):
     key: str
 
 
-def get_aws_cred_files_from_env() -> Set[str]:
+def get_aws_cred_files_from_env() -> set[str]:
     """Extract credential file paths from environment variables."""
     return {
         os.environ[env_var]
@@ -25,7 +24,7 @@ def get_aws_cred_files_from_env() -> Set[str]:
     }
 
 
-def get_aws_secrets_from_env() -> Set[str]:
+def get_aws_secrets_from_env() -> set[str]:
     """Extract AWS secrets from environment variables."""
     keys = set()
     for env_var in (
@@ -36,7 +35,7 @@ def get_aws_secrets_from_env() -> Set[str]:
     return keys
 
 
-def get_aws_secrets_from_file(credentials_file: str) -> Set[str]:
+def get_aws_secrets_from_file(credentials_file: str) -> set[str]:
     """Extract AWS secrets from configuration files.
 
     Read an ini-style configuration file and return a set with all found AWS
@@ -69,8 +68,8 @@ def get_aws_secrets_from_file(credentials_file: str) -> Set[str]:
 
 def check_file_for_aws_keys(
         filenames: Sequence[str],
-        keys: Set[bytes],
-) -> List[BadFile]:
+        keys: set[bytes],
+) -> list[BadFile]:
     """Check if files contain AWS secrets.
 
     Return a list of all files containing AWS secrets and keys found, with all
@@ -90,7 +89,7 @@ def check_file_for_aws_keys(
     return bad_files
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='+', help='Filenames to run')
     parser.add_argument(
@@ -119,7 +118,7 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     # of files to to gather AWS secrets from.
     credential_files |= get_aws_cred_files_from_env()
 
-    keys: Set[str] = set()
+    keys: set[str] = set()
     for credential_file in credential_files:
         keys |= get_aws_secrets_from_file(credential_file)
 
@@ -149,4 +148,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/detect_private_key.py

@@ -1,6 +1,7 @@
+from __future__ import annotations
+
 import argparse
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 BLACKLIST = [
     b'BEGIN RSA PRIVATE KEY',
@@ -11,10 +12,12 @@ BLACKLIST = [
     b'PuTTY-User-Key-File-2',
     b'BEGIN SSH2 ENCRYPTED PRIVATE KEY',
     b'BEGIN PGP PRIVATE KEY BLOCK',
+    b'BEGIN ENCRYPTED PRIVATE KEY',
+    b'BEGIN OpenVPN Static key V1',
 ]
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to check')
     args = parser.parse_args(argv)
@@ -36,4 +39,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/end_of_file_fixer.py

@@ -1,8 +1,9 @@
+from __future__ import annotations
+
 import argparse
 import os
+from collections.abc import Sequence
 from typing import IO
-from typing import Optional
-from typing import Sequence
 
 
 def fix_file(file_obj: IO[bytes]) -> int:
@@ -48,7 +49,7 @@ def fix_file(file_obj: IO[bytes]) -> int:
     return 0
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to fix')
     args = parser.parse_args(argv)
@@ -67,4 +68,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/file_contents_sorter.py

@@ -2,28 +2,45 @@
 A very simple pre-commit hook that, when passed one or more filenames
 as arguments, will sort the lines in those files.
 
-An example use case for this: you have a deploy-whitelist.txt file
+An example use case for this: you have a deploy-allowlist.txt file
 in a repo that contains a list of filenames that is used to specify
 files to be included in a docker container. This file has one filename
 per line. Various users are adding/removing lines from this file; using
 this hook on that file should reduce the instances of git merge
 conflicts and keep the file nicely ordered.
 """
+from __future__ import annotations
+
 import argparse
+from collections.abc import Callable
+from collections.abc import Iterable
+from collections.abc import Sequence
+from typing import Any
 from typing import IO
-from typing import Optional
-from typing import Sequence
 
 PASS = 0
 FAIL = 1
 
 
-def sort_file_contents(f: IO[bytes]) -> int:
+def sort_file_contents(
+    f: IO[bytes],
+    key: Callable[[bytes], Any] | None,
+    *,
+    unique: bool = False,
+) -> int:
     before = list(f)
-    after = sorted(line.strip(b'\n\r') for line in before if line.strip())
+    lines: Iterable[bytes] = (
+        line.rstrip(b'\n\r') for line in before if line.strip()
+    )
+    if unique:
+        lines = set(lines)
+    after = sorted(lines, key=key)
 
     before_string = b''.join(before)
-    after_string = b'\n'.join(after) + b'\n'
+    after_string = b'\n'.join(after)
+
+    if after_string:
+        after_string += b'\n'
 
     if before_string == after_string:
         return PASS
@@ -34,16 +51,33 @@ def sort_file_contents(f: IO[bytes]) -> int:
         return FAIL
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='+', help='Files to sort')
+
+    mutex = parser.add_mutually_exclusive_group(required=False)
+    mutex.add_argument(
+        '--ignore-case',
+        action='store_const',
+        const=bytes.lower,
+        default=None,
+        help='fold lower case to upper case characters',
+    )
+    mutex.add_argument(
+        '--unique',
+        action='store_true',
+        help='ensure each line is unique',
+    )
+
     args = parser.parse_args(argv)
 
     retv = PASS
 
     for arg in args.filenames:
         with open(arg, 'rb+') as file_obj:
-            ret_for_file = sort_file_contents(file_obj)
+            ret_for_file = sort_file_contents(
+                file_obj, key=args.ignore_case, unique=args.unique,
+            )
 
             if ret_for_file:
                 print(f'Sorting {arg}')
@@ -54,4 +88,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/fix_byte_order_marker.py

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+import argparse
+from collections.abc import Sequence
+
+
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument('filenames', nargs='*', help='Filenames to check')
+    args = parser.parse_args(argv)
+
+    retv = 0
+
+    for filename in args.filenames:
+        with open(filename, 'rb') as f_b:
+            bts = f_b.read(3)
+
+        if bts == b'\xef\xbb\xbf':
+            with open(filename, newline='', encoding='utf-8-sig') as f:
+                contents = f.read()
+            with open(filename, 'w', newline='', encoding='utf-8') as f:
+                f.write(contents)
+
+            print(f'{filename}: removed byte-order marker')
+            retv = 1
+
+    return retv
+
+
+if __name__ == '__main__':
+    raise SystemExit(main())

pre_commit_hooks/fix_encoding_pragma.py

@@ -1,148 +0,0 @@
-import argparse
-from typing import IO
-from typing import NamedTuple
-from typing import Optional
-from typing import Sequence
-
-DEFAULT_PRAGMA = b'# -*- coding: utf-8 -*-'
-
-
-def has_coding(line: bytes) -> bool:
-    if not line.strip():
-        return False
-    return (
-        line.lstrip()[:1] == b'#' and (
-            b'unicode' in line or
-            b'encoding' in line or
-            b'coding:' in line or
-            b'coding=' in line
-        )
-    )
-
-
-class ExpectedContents(NamedTuple):
-    shebang: bytes
-    rest: bytes
-    # True: has exactly the coding pragma expected
-    # False: missing coding pragma entirely
-    # None: has a coding pragma, but it does not match
-    pragma_status: Optional[bool]
-    ending: bytes
-
-    @property
-    def has_any_pragma(self) -> bool:
-        return self.pragma_status is not False
-
-    def is_expected_pragma(self, remove: bool) -> bool:
-        expected_pragma_status = not remove
-        return self.pragma_status is expected_pragma_status
-
-
-def _get_expected_contents(
-        first_line: bytes,
-        second_line: bytes,
-        rest: bytes,
-        expected_pragma: bytes,
-) -> ExpectedContents:
-    ending = b'\r\n' if first_line.endswith(b'\r\n') else b'\n'
-
-    if first_line.startswith(b'#!'):
-        shebang = first_line
-        potential_coding = second_line
-    else:
-        shebang = b''
-        potential_coding = first_line
-        rest = second_line + rest
-
-    if potential_coding.rstrip(b'\r\n') == expected_pragma:
-        pragma_status: Optional[bool] = True
-    elif has_coding(potential_coding):
-        pragma_status = None
-    else:
-        pragma_status = False
-        rest = potential_coding + rest
-
-    return ExpectedContents(
-        shebang=shebang, rest=rest, pragma_status=pragma_status, ending=ending,
-    )
-
-
-def fix_encoding_pragma(
-        f: IO[bytes],
-        remove: bool = False,
-        expected_pragma: bytes = DEFAULT_PRAGMA,
-) -> int:
-    expected = _get_expected_contents(
-        f.readline(), f.readline(), f.read(), expected_pragma,
-    )
-
-    # Special cases for empty files
-    if not expected.rest.strip():
-        # If a file only has a shebang or a coding pragma, remove it
-        if expected.has_any_pragma or expected.shebang:
-            f.seek(0)
-            f.truncate()
-            f.write(b'')
-            return 1
-        else:
-            return 0
-
-    if expected.is_expected_pragma(remove):
-        return 0
-
-    # Otherwise, write out the new file
-    f.seek(0)
-    f.truncate()
-    f.write(expected.shebang)
-    if not remove:
-        f.write(expected_pragma + expected.ending)
-    f.write(expected.rest)
-
-    return 1
-
-
-def _normalize_pragma(pragma: str) -> bytes:
-    return pragma.encode().rstrip()
-
-
-def main(argv: Optional[Sequence[str]] = None) -> int:
-    parser = argparse.ArgumentParser(
-        'Fixes the encoding pragma of python files',
-    )
-    parser.add_argument('filenames', nargs='*', help='Filenames to fix')
-    parser.add_argument(
-        '--pragma', default=DEFAULT_PRAGMA, type=_normalize_pragma,
-        help=(
-            f'The encoding pragma to use.  '
-            f'Default: {DEFAULT_PRAGMA.decode()}'
-        ),
-    )
-    parser.add_argument(
-        '--remove', action='store_true',
-        help='Remove the encoding pragma (Useful in a python3-only codebase)',
-    )
-    args = parser.parse_args(argv)
-
-    retv = 0
-
-    if args.remove:
-        fmt = 'Removed encoding pragma from {filename}'
-    else:
-        fmt = 'Added `{pragma}` to {filename}'
-
-    for filename in args.filenames:
-        with open(filename, 'r+b') as f:
-            file_ret = fix_encoding_pragma(
-                f, remove=args.remove, expected_pragma=args.pragma,
-            )
-            retv |= file_ret
-            if file_ret:
-                print(
-                    fmt.format(pragma=args.pragma.decode(), filename=filename),
-                )
-
-    return retv
-
-
-if __name__ == '__main__':
-    exit(main())

pre_commit_hooks/forbid_new_submodules.py

@@ -1,14 +1,30 @@
-from typing import Optional
-from typing import Sequence
+from __future__ import annotations
+
+import argparse
+import os
+from collections.abc import Sequence
 
 from pre_commit_hooks.util import cmd_output
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
-    # `argv` is ignored, pre-commit will send us a list of files that we
-    # don't care about
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument('filenames', nargs='*')
+    args = parser.parse_args(argv)
+
+    if (
+        'PRE_COMMIT_FROM_REF' in os.environ and
+        'PRE_COMMIT_TO_REF' in os.environ
+    ):
+        diff_arg = '...'.join((
+            os.environ['PRE_COMMIT_FROM_REF'],
+            os.environ['PRE_COMMIT_TO_REF'],
+        ))
+    else:
+        diff_arg = '--staged'
     added_diff = cmd_output(
-        'git', 'diff', '--staged', '--diff-filter=A', '--raw',
+        'git', 'diff', '--diff-filter=A', '--raw', diff_arg, '--',
+        *args.filenames,
     )
     retv = 0
     for line in added_diff.splitlines():
@@ -29,4 +45,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/mixed_line_ending.py

@@ -1,8 +1,8 @@
+from __future__ import annotations
+
 import argparse
 import collections
-from typing import Dict
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
 CRLF = b'\r\n'
@@ -25,7 +25,7 @@ def fix_filename(filename: str, fix: str) -> int:
     with open(filename, 'rb') as f:
         contents = f.read()
 
-    counts: Dict[bytes, int] = collections.defaultdict(int)
+    counts: dict[bytes, int] = collections.defaultdict(int)
 
     for line in contents.splitlines(True):
         for ending in ALL_ENDINGS:
@@ -62,7 +62,7 @@ def fix_filename(filename: str, fix: str) -> int:
         return other_endings
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         '-f', '--fix',
@@ -85,4 +85,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/no_commit_to_branch.py

@@ -1,8 +1,9 @@
+from __future__ import annotations
+
 import argparse
 import re
+from collections.abc import Sequence
 from typing import AbstractSet
-from typing import Optional
-from typing import Sequence
 
 from pre_commit_hooks.util import CalledProcessError
 from pre_commit_hooks.util import cmd_output
@@ -23,7 +24,7 @@ def is_on_branch(
     )
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         '-b', '--branch', action='append',
@@ -38,10 +39,10 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     )
     args = parser.parse_args(argv)
 
-    protected = frozenset(args.branch or ('master',))
+    protected = frozenset(args.branch or ('master', 'main'))
     patterns = frozenset(args.pattern or ())
     return int(is_on_branch(protected, patterns))
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/pretty_format_json.py

@@ -1,12 +1,11 @@
+from __future__ import annotations
+
 import argparse
 import json
+import sys
+from collections.abc import Mapping
+from collections.abc import Sequence
 from difflib import unified_diff
-from typing import List
-from typing import Mapping
-from typing import Optional
-from typing import Sequence
-from typing import Tuple
-from typing import Union
 
 
 def _get_pretty_format(
@@ -16,7 +15,7 @@ def _get_pretty_format(
         sort_keys: bool = True,
         top_keys: Sequence[str] = (),
 ) -> str:
-    def pairs_first(pairs: Sequence[Tuple[str, str]]) -> Mapping[str, str]:
+    def pairs_first(pairs: Sequence[tuple[str, str]]) -> Mapping[str, str]:
         before = [pair for pair in pairs if pair[0] in top_keys]
         before = sorted(before, key=lambda x: top_keys.index(x[0]))
         after = [pair for pair in pairs if pair[0] not in top_keys]
@@ -37,7 +36,7 @@ def _autofix(filename: str, new_contents: str) -> None:
         f.write(new_contents)
 
 
-def parse_num_to_int(s: str) -> Union[int, str]:
+def parse_num_to_int(s: str) -> int | str:
     """Convert string numbers to int, leaving strings as is."""
     try:
         return int(s)
@@ -45,7 +44,7 @@ def parse_num_to_int(s: str) -> Union[int, str]:
         return s
 
 
-def parse_topkeys(s: str) -> List[str]:
+def parse_topkeys(s: str) -> list[str]:
     return s.split(',')
 
 
@@ -56,7 +55,7 @@ def get_diff(source: str, target: str, file: str) -> str:
     return ''.join(diff)
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         '--autofix',
@@ -111,26 +110,28 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
                 contents, args.indent, ensure_ascii=not args.no_ensure_ascii,
                 sort_keys=not args.no_sort_keys, top_keys=args.top_keys,
             )
-
-            if contents != pretty_contents:
-                if args.autofix:
-                    _autofix(json_file, pretty_contents)
-                else:
-                    print(
-                        get_diff(contents, pretty_contents, json_file),
-                        end='',
-                    )
-
-                status = 1
         except ValueError:
             print(
                 f'Input File {json_file} is not a valid JSON, consider using '
                 f'check-json',
             )
-            return 1
+            status = 1
+        else:
+            if contents != pretty_contents:
+                if args.autofix:
+                    _autofix(json_file, pretty_contents)
+                else:
+                    diff_output = get_diff(
+                        contents,
+                        pretty_contents,
+                        json_file,
+                    )
+                    sys.stdout.buffer.write(diff_output.encode())
+
+                status = 1
 
     return status
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/removed.py

@@ -1,9 +1,10 @@
+from __future__ import annotations
+
 import sys
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     argv = argv if argv is not None else sys.argv[1:]
     hookid, new_hookid, url = argv[:3]
     raise SystemExit(
@@ -12,4 +13,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/requirements_txt_fixer.py

@@ -1,9 +1,9 @@
+from __future__ import annotations
+
 import argparse
 import re
+from collections.abc import Sequence
 from typing import IO
-from typing import List
-from typing import Optional
-from typing import Sequence
 
 
 PASS = 0
@@ -15,8 +15,8 @@ class Requirement:
     UNTIL_SEP = re.compile(rb'[^;\s]+')
 
     def __init__(self) -> None:
-        self.value: Optional[bytes] = None
-        self.comments: List[bytes] = []
+        self.value: bytes | None = None
+        self.comments: list[bytes] = []
 
     @property
     def name(self) -> bytes:
@@ -36,7 +36,7 @@ class Requirement:
 
         return name[:m.start()]
 
-    def __lt__(self, requirement: 'Requirement') -> int:
+    def __lt__(self, requirement: Requirement) -> bool:
         # \n means top of file comment, so always return True,
         # otherwise just do a string comparison with value.
         assert self.value is not None, self.value
@@ -45,6 +45,11 @@ class Requirement:
         elif requirement.value == b'\n':
             return False
         else:
+            # if 2 requirements have the same name, the one with comments
+            # needs to go first (so that when removing duplicates, the one
+            # with comments is kept)
+            if self.name == requirement.name:
+                return bool(self.comments) > bool(requirement.comments)
             return self.name < requirement.name
 
     def is_complete(self) -> bool:
@@ -61,9 +66,9 @@ class Requirement:
 
 
 def fix_requirements(f: IO[bytes]) -> int:
-    requirements: List[Requirement] = []
+    requirements: list[Requirement] = []
     before = list(f)
-    after: List[bytes] = []
+    after: list[bytes] = []
 
     before_string = b''.join(before)
 
@@ -95,7 +100,7 @@ def fix_requirements(f: IO[bytes]) -> int:
                 requirement.value = b'\n'
             else:
                 requirement.comments.append(line)
-        elif line.startswith(b'#') or line.strip() == b'':
+        elif line.lstrip().startswith(b'#') or line.strip() == b'':
             requirement.comments.append(line)
         else:
             requirement.append_value(line)
@@ -110,13 +115,20 @@ def fix_requirements(f: IO[bytes]) -> int:
     # which is automatically added by broken pip package under Debian
     requirements = [
         req for req in requirements
-        if req.value != b'pkg-resources==0.0.0\n'
+        if req.value not in [
+            b'pkg-resources==0.0.0\n',
+            b'pkg_resources==0.0.0\n',
+        ]
     ]
 
+    # sort the requirements and remove duplicates
+    prev = None
     for requirement in sorted(requirements):
         after.extend(requirement.comments)
         assert requirement.value, requirement.value
-        after.append(requirement.value)
+        if prev is None or requirement.value != prev.value:
+            after.append(requirement.value)
+            prev = requirement
     after.extend(rest)
 
     after_string = b''.join(after)
@@ -130,7 +142,7 @@ def fix_requirements(f: IO[bytes]) -> int:
         return FAIL
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to fix')
     args = parser.parse_args(argv)
@@ -150,4 +162,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/sort_simple_yaml.py

@@ -1,4 +1,3 @@
-#!/usr/bin/env python
 """Sort a simple YAML file, keeping blocks of comments and definitions
 together.
 
@@ -18,16 +17,16 @@ We assume a strict subset of YAML that looks like:
 In other words, we don't sort deeper than the top layer, and might corrupt
 complicated YAML files.
 """
+from __future__ import annotations
+
 import argparse
-from typing import List
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
 QUOTES = ["'", '"']
 
 
-def sort(lines: List[str]) -> List[str]:
+def sort(lines: list[str]) -> list[str]:
     """Sort a YAML file in alphabetical order, keeping blocks together.
 
     :param lines: array of strings (without newlines)
@@ -45,7 +44,7 @@ def sort(lines: List[str]) -> List[str]:
     return new_lines
 
 
-def parse_block(lines: List[str], header: bool = False) -> List[str]:
+def parse_block(lines: list[str], header: bool = False) -> list[str]:
     """Parse and return a single block, popping off the start of `lines`.
 
     If parsing a header block, we stop after we reach a line that is not a
@@ -61,7 +60,7 @@ def parse_block(lines: List[str], header: bool = False) -> List[str]:
     return block_lines
 
 
-def parse_blocks(lines: List[str]) -> List[List[str]]:
+def parse_blocks(lines: list[str]) -> list[list[str]]:
     """Parse and return all possible blocks, popping off the start of `lines`.
 
     :param lines: list of lines
@@ -78,7 +77,7 @@ def parse_blocks(lines: List[str]) -> List[List[str]]:
     return blocks
 
 
-def first_key(lines: List[str]) -> str:
+def first_key(lines: list[str]) -> str:
     """Returns a string representing the sort key of a block.
 
     The sort key is the first YAML key we encounter, ignoring comments, and
@@ -100,7 +99,7 @@ def first_key(lines: List[str]) -> str:
         return ''  # not actually reached in reality
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to fix')
     args = parser.parse_args(argv)
@@ -123,4 +122,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/string_fixer.py

@@ -1,10 +1,17 @@
+from __future__ import annotations
+
 import argparse
 import io
 import re
+import sys
 import tokenize
-from typing import List
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
+
+if sys.version_info >= (3, 12):  # pragma: >=3.12 cover
+    FSTRING_START = tokenize.FSTRING_START
+    FSTRING_END = tokenize.FSTRING_END
+else:  # pragma: <3.12 cover
+    FSTRING_START = FSTRING_END = -1
 
 START_QUOTE_RE = re.compile('^[a-zA-Z]*"')
 
@@ -24,7 +31,7 @@ def handle_match(token_text: str) -> str:
         return token_text
 
 
-def get_line_offsets_by_line_no(src: str) -> List[int]:
+def get_line_offsets_by_line_no(src: str) -> list[int]:
     # Padded so we can index with line number
     offsets = [-1, 0]
     for line in src.splitlines(True):
@@ -40,11 +47,17 @@ def fix_strings(filename: str) -> int:
     # Basically a mutable string
     splitcontents = list(contents)
 
+    fstring_depth = 0
+
     # Iterate in reverse so the offsets are always correct
     tokens_l = list(tokenize.generate_tokens(io.StringIO(contents).readline))
     tokens = reversed(tokens_l)
     for token_type, token_text, (srow, scol), (erow, ecol), _ in tokens:
-        if token_type == tokenize.STRING:
+        if token_type == FSTRING_START:  # pragma: >=3.12 cover
+            fstring_depth += 1
+        elif token_type == FSTRING_END:  # pragma: >=3.12 cover
+            fstring_depth -= 1
+        elif fstring_depth == 0 and token_type == tokenize.STRING:
             new_text = handle_match(token_text)
             splitcontents[
                 line_offsets[srow] + scol:
@@ -60,7 +73,7 @@ def fix_strings(filename: str) -> int:
         return 0
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*', help='Filenames to fix')
     args = parser.parse_args(argv)
@@ -77,4 +90,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/tests_should_end_in_test.py

@@ -1,33 +1,53 @@
+from __future__ import annotations
+
 import argparse
 import os.path
 import re
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument('filenames', nargs='*')
-    parser.add_argument(
-        '--django', default=False, action='store_true',
-        help='Use Django-style test naming pattern (test*.py)',
+    mutex = parser.add_mutually_exclusive_group()
+    mutex.add_argument(
+        '--pytest',
+        dest='pattern',
+        action='store_const',
+        const=r'.*_test\.py',
+        default=r'.*_test\.py',
+        help='(the default) ensure tests match %(const)s',
+    )
+    mutex.add_argument(
+        '--pytest-test-first',
+        dest='pattern',
+        action='store_const',
+        const=r'test_.*\.py',
+        help='ensure tests match %(const)s',
+    )
+    mutex.add_argument(
+        '--django', '--unittest',
+        dest='pattern',
+        action='store_const',
+        const=r'test.*\.py',
+        help='ensure tests match %(const)s',
     )
     args = parser.parse_args(argv)
 
     retcode = 0
-    test_name_pattern = r'test.*\.py' if args.django else r'.*_test\.py'
+    reg = re.compile(args.pattern)
     for filename in args.filenames:
         base = os.path.basename(filename)
         if (
-                not re.match(test_name_pattern, base) and
+                not reg.fullmatch(base) and
                 not base == '__init__.py' and
                 not base == 'conftest.py'
         ):
             retcode = 1
-            print(f'{filename} does not match pattern "{test_name_pattern}"')
+            print(f'{filename} does not match pattern "{args.pattern}"')
 
     return retcode
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/trailing_whitespace_fixer.py

@@ -1,13 +1,14 @@
+from __future__ import annotations
+
 import argparse
 import os
-from typing import Optional
-from typing import Sequence
+from collections.abc import Sequence
 
 
 def _fix_file(
         filename: str,
         is_markdown: bool,
-        chars: Optional[bytes],
+        chars: bytes | None,
 ) -> bool:
     with open(filename, mode='rb') as file_processed:
         lines = file_processed.readlines()
@@ -24,7 +25,7 @@ def _fix_file(
 def _process_line(
         line: bytes,
         is_markdown: bool,
-        chars: Optional[bytes],
+        chars: bytes | None,
 ) -> bytes:
     if line[-2:] == b'\r\n':
         eol = b'\r\n'
@@ -40,7 +41,7 @@ def _process_line(
     return line.rstrip(chars) + eol
 
 
-def main(argv: Optional[Sequence[str]] = None) -> int:
+def main(argv: Sequence[str] | None = None) -> int:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         '--no-markdown-linebreak-ext',
@@ -99,4 +100,4 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
 
 
 if __name__ == '__main__':
-    exit(main())
+    raise SystemExit(main())

pre_commit_hooks/util.py

@@ -1,19 +1,19 @@
+from __future__ import annotations
+
 import subprocess
 from typing import Any
-from typing import Optional
-from typing import Set
 
 
 class CalledProcessError(RuntimeError):
     pass
 
 
-def added_files() -> Set[str]:
+def added_files() -> set[str]:
     cmd = ('git', 'diff', '--staged', '--name-only', '--diff-filter=A')
     return set(cmd_output(*cmd).splitlines())
 
 
-def cmd_output(*cmd: str, retcode: Optional[int] = 0, **kwargs: Any) -> str:
+def cmd_output(*cmd: str, retcode: int | None = 0, **kwargs: Any) -> str:
     kwargs.setdefault('stdout', subprocess.PIPE)
     kwargs.setdefault('stderr', subprocess.PIPE)
     proc = subprocess.Popen(cmd, **kwargs)
@@ -22,3 +22,11 @@ def cmd_output(*cmd: str, retcode: Optional[int] = 0, **kwargs: Any) -> str:
     if retcode is not None and proc.returncode != retcode:
         raise CalledProcessError(cmd, retcode, proc.returncode, stdout, stderr)
     return stdout
+
+
+def zsplit(s: str) -> list[str]:
+    s = s.strip('\0')
+    if s:
+        return s.split('\0')
+    else:
+        return []

requirements-dev.txt

@@ -1,4 +1,3 @@
 covdefaults
 coverage
-pre-commit
 pytest

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pre_commit_hooks
-version = 3.1.0
+version = 6.0.0
 description = Some out-of-the-box hooks for pre-commit.
 long_description = file: README.md
 long_description_content_type = text/markdown
@@ -8,14 +8,10 @@ url = https://github.com/pre-commit/pre-commit-hooks
 author = Anthony Sottile
 author_email = asottile@umich.edu
 license = MIT
-license_file = LICENSE
+license_files = LICENSE
 classifiers =
-    License :: OSI Approved :: MIT License
     Programming Language :: Python :: 3
     Programming Language :: Python :: 3 :: Only
-    Programming Language :: Python :: 3.6
-    Programming Language :: Python :: 3.7
-    Programming Language :: Python :: 3.8
     Programming Language :: Python :: Implementation :: CPython
     Programming Language :: Python :: Implementation :: PyPy
 
@@ -23,32 +19,38 @@ classifiers =
 packages = find:
 install_requires =
     ruamel.yaml>=0.15
-    toml
-python_requires = >=3.6.1
+    tomli>=1.1.0;python_version<"3.11"
+python_requires = >=3.10
+
+[options.packages.find]
+exclude =
+    tests*
+    testing*
 
 [options.entry_points]
 console_scripts =
     check-added-large-files = pre_commit_hooks.check_added_large_files:main
     check-ast = pre_commit_hooks.check_ast:main
     check-builtin-literals = pre_commit_hooks.check_builtin_literals:main
-    check-byte-order-marker = pre_commit_hooks.check_byte_order_marker:main
     check-case-conflict = pre_commit_hooks.check_case_conflict:main
     check-docstring-first = pre_commit_hooks.check_docstring_first:main
     check-executables-have-shebangs = pre_commit_hooks.check_executables_have_shebangs:main
     check-json = pre_commit_hooks.check_json:main
     check-merge-conflict = pre_commit_hooks.check_merge_conflict:main
+    check-shebang-scripts-are-executable = pre_commit_hooks.check_shebang_scripts_are_executable:main
     check-symlinks = pre_commit_hooks.check_symlinks:main
     check-toml = pre_commit_hooks.check_toml:main
     check-vcs-permalinks = pre_commit_hooks.check_vcs_permalinks:main
     check-xml = pre_commit_hooks.check_xml:main
     check-yaml = pre_commit_hooks.check_yaml:main
     debug-statement-hook = pre_commit_hooks.debug_statement_hook:main
+    destroyed-symlinks = pre_commit_hooks.destroyed_symlinks:main
     detect-aws-credentials = pre_commit_hooks.detect_aws_credentials:main
     detect-private-key = pre_commit_hooks.detect_private_key:main
     double-quote-string-fixer = pre_commit_hooks.string_fixer:main
     end-of-file-fixer = pre_commit_hooks.end_of_file_fixer:main
     file-contents-sorter = pre_commit_hooks.file_contents_sorter:main
-    fix-encoding-pragma = pre_commit_hooks.fix_encoding_pragma:main
+    fix-byte-order-marker = pre_commit_hooks.fix_byte_order_marker:main
     forbid-new-submodules = pre_commit_hooks.forbid_new_submodules:main
     mixed-line-ending = pre_commit_hooks.mixed_line_ending:main
     name-tests-test = pre_commit_hooks.tests_should_end_in_test:main
@@ -59,11 +61,6 @@ console_scripts =
     sort-simple-yaml = pre_commit_hooks.sort_simple_yaml:main
     trailing-whitespace-fixer = pre_commit_hooks.trailing_whitespace_fixer:main
 
-[options.packages.find]
-exclude =
-    tests*
-    testing*
-
 [bdist_wheel]
 universal = True
 
@@ -75,7 +72,8 @@ check_untyped_defs = true
 disallow_any_generics = true
 disallow_incomplete_defs = true
 disallow_untyped_defs = true
-no_implicit_optional = true
+warn_redundant_casts = true
+warn_unused_ignores = true
 
 [mypy-testing.*]
 disallow_untyped_defs = false

setup.py

@@ -1,2 +1,4 @@
+from __future__ import annotations
+
 from setuptools import setup
 setup()

testing/resources/duplicate_key_json.notjson

@@ -0,0 +1,4 @@
+{
+    "hello": "world",
+    "hello": "planet"
+}

testing/util.py

@@ -1,4 +1,7 @@
+from __future__ import annotations
+
 import os.path
+import subprocess
 
 
 TESTING_DIR = os.path.abspath(os.path.dirname(__file__))
@@ -6,3 +9,8 @@ TESTING_DIR = os.path.abspath(os.path.dirname(__file__))
 
 def get_resource_path(path):
     return os.path.join(TESTING_DIR, 'resources', path)
+
+
+def git_commit(*args, **kwargs):
+    cmd = ('git', 'commit', '--no-gpg-sign', '--no-verify', '--no-edit', *args)
+    subprocess.check_call(cmd, **kwargs)

tests/check_added_large_files_test.py

@@ -1,10 +1,13 @@
-import distutils.spawn
+from __future__ import annotations
+
+import shutil
 
 import pytest
 
 from pre_commit_hooks.check_added_large_files import find_large_added_files
 from pre_commit_hooks.check_added_large_files import main
 from pre_commit_hooks.util import cmd_output
+from testing.util import git_commit
 
 
 def test_nothing_added(temp_git_dir):
@@ -40,6 +43,17 @@ def test_add_something_giant(temp_git_dir):
         assert find_large_added_files(['f.py'], 10) == 0
 
 
+def test_enforce_all(temp_git_dir):
+    with temp_git_dir.as_cwd():
+        temp_git_dir.join('f.py').write('a' * 10000)
+
+        # Should fail, when not staged with enforce_all
+        assert find_large_added_files(['f.py'], 0, enforce_all=True) == 1
+
+        # Should pass, when not staged without enforce_all
+        assert find_large_added_files(['f.py'], 0, enforce_all=False) == 0
+
+
 def test_added_file_not_in_pre_commits_list(temp_git_dir):
     with temp_git_dir.as_cwd():
         temp_git_dir.join('f.py').write("print('hello world')")
@@ -64,7 +78,7 @@ def test_integration(temp_git_dir):
 
 
 def has_gitlfs():
-    return distutils.spawn.find_executable('git-lfs') is not None
+    return shutil.which('git-lfs') is not None
 
 
 xfailif_no_gitlfs = pytest.mark.xfail(
@@ -73,10 +87,9 @@ xfailif_no_gitlfs = pytest.mark.xfail(
 
 
 @xfailif_no_gitlfs
-def test_allows_gitlfs(temp_git_dir, monkeypatch):  # pragma: no cover
+def test_allows_gitlfs(temp_git_dir):  # pragma: no cover
     with temp_git_dir.as_cwd():
-        monkeypatch.setenv('HOME', str(temp_git_dir.strpath))
-        cmd_output('git', 'lfs', 'install')
+        cmd_output('git', 'lfs', 'install', '--local')
         temp_git_dir.join('f.py').write('a' * 10000)
         cmd_output('git', 'lfs', 'track', 'f.py')
         cmd_output('git', 'add', '--', '.')
@@ -85,15 +98,37 @@ def test_allows_gitlfs(temp_git_dir, monkeypatch):  # pragma: no cover
 
 
 @xfailif_no_gitlfs
-def test_moves_with_gitlfs(temp_git_dir, monkeypatch):  # pragma: no cover
+def test_moves_with_gitlfs(temp_git_dir):  # pragma: no cover
     with temp_git_dir.as_cwd():
-        monkeypatch.setenv('HOME', str(temp_git_dir.strpath))
-        cmd_output('git', 'lfs', 'install')
+        cmd_output('git', 'lfs', 'install', '--local')
         cmd_output('git', 'lfs', 'track', 'a.bin', 'b.bin')
         # First add the file we're going to move
         temp_git_dir.join('a.bin').write('a' * 10000)
         cmd_output('git', 'add', '--', '.')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-am', 'foo')
+        git_commit('-am', 'foo')
         # Now move it and make sure the hook still succeeds
         cmd_output('git', 'mv', 'a.bin', 'b.bin')
         assert main(('--maxkb', '9', 'b.bin')) == 0
+
+
+@xfailif_no_gitlfs
+def test_enforce_allows_gitlfs(temp_git_dir):  # pragma: no cover
+    with temp_git_dir.as_cwd():
+        cmd_output('git', 'lfs', 'install', '--local')
+        temp_git_dir.join('f.py').write('a' * 10000)
+        cmd_output('git', 'lfs', 'track', 'f.py')
+        cmd_output('git', 'add', '--', '.')
+        # With --enforce-all large files on git lfs should succeed
+        assert main(('--enforce-all', '--maxkb', '9', 'f.py')) == 0
+
+
+@xfailif_no_gitlfs
+def test_enforce_allows_gitlfs_after_commit(temp_git_dir):  # pragma: no cover
+    with temp_git_dir.as_cwd():
+        cmd_output('git', 'lfs', 'install', '--local')
+        temp_git_dir.join('f.py').write('a' * 10000)
+        cmd_output('git', 'lfs', 'track', 'f.py')
+        cmd_output('git', 'add', '--', '.')
+        git_commit('-am', 'foo')
+        # With --enforce-all large files on git lfs should succeed
+        assert main(('--enforce-all', '--maxkb', '9', 'f.py')) == 0

tests/check_ast_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pre_commit_hooks.check_ast import main
 from testing.util import get_resource_path
 

tests/check_builtin_literals_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import ast
 
 import pytest
@@ -36,11 +38,6 @@ t1 = ()
 '''
 
 
-@pytest.fixture
-def visitor():
-    return Visitor()
-
-
 @pytest.mark.parametrize(
     ('expression', 'calls'),
     [
@@ -83,7 +80,8 @@ def visitor():
         ('builtins.tuple()', []),
     ],
 )
-def test_non_dict_exprs(visitor, expression, calls):
+def test_non_dict_exprs(expression, calls):
+    visitor = Visitor(ignore=set())
     visitor.visit(ast.parse(expression))
     assert visitor.builtin_type_calls == calls
 
@@ -100,7 +98,8 @@ def test_non_dict_exprs(visitor, expression, calls):
         ('builtins.dict()', []),
     ],
 )
-def test_dict_allow_kwargs_exprs(visitor, expression, calls):
+def test_dict_allow_kwargs_exprs(expression, calls):
+    visitor = Visitor(ignore=set())
     visitor.visit(ast.parse(expression))
     assert visitor.builtin_type_calls == calls
 
@@ -112,17 +111,18 @@ def test_dict_allow_kwargs_exprs(visitor, expression, calls):
         ('dict(a=1, b=2, c=3)', [Call('dict', 1, 0)]),
         ("dict(**{'a': 1, 'b': 2, 'c': 3})", [Call('dict', 1, 0)]),
         ('builtins.dict()', []),
+        pytest.param('f(dict())', [Call('dict', 1, 2)], id='nested'),
     ],
 )
 def test_dict_no_allow_kwargs_exprs(expression, calls):
-    visitor = Visitor(allow_dict_kwargs=False)
+    visitor = Visitor(ignore=set(), allow_dict_kwargs=False)
     visitor.visit(ast.parse(expression))
     assert visitor.builtin_type_calls == calls
 
 
 def test_ignore_constructors():
     visitor = Visitor(
-        ignore=('complex', 'dict', 'float', 'int', 'list', 'str', 'tuple'),
+        ignore={'complex', 'dict', 'float', 'int', 'list', 'str', 'tuple'},
     )
     visitor.visit(ast.parse(BUILTIN_CONSTRUCTORS))
     assert visitor.builtin_type_calls == []
@@ -131,19 +131,19 @@ def test_ignore_constructors():
 def test_failing_file(tmpdir):
     f = tmpdir.join('f.py')
     f.write(BUILTIN_CONSTRUCTORS)
-    rc = main([f.strpath])
+    rc = main([str(f)])
     assert rc == 1
 
 
 def test_passing_file(tmpdir):
     f = tmpdir.join('f.py')
     f.write(BUILTIN_LITERALS)
-    rc = main([f.strpath])
+    rc = main([str(f)])
     assert rc == 0
 
 
 def test_failing_file_ignore_all(tmpdir):
     f = tmpdir.join('f.py')
     f.write(BUILTIN_CONSTRUCTORS)
-    rc = main(['--ignore=complex,dict,float,int,list,str,tuple', f.strpath])
+    rc = main(['--ignore=complex,dict,float,int,list,str,tuple', str(f)])
     assert rc == 0

tests/check_case_conflict_test.py

@@ -1,6 +1,26 @@
+from __future__ import annotations
+
+import sys
+
+import pytest
+
 from pre_commit_hooks.check_case_conflict import find_conflicting_filenames
 from pre_commit_hooks.check_case_conflict import main
+from pre_commit_hooks.check_case_conflict import parents
 from pre_commit_hooks.util import cmd_output
+from testing.util import git_commit
+
+skip_win32 = pytest.mark.skipif(
+    sys.platform == 'win32',
+    reason='case conflicts between directories and files',
+)
+
+
+def test_parents():
+    assert set(parents('a')) == set()
+    assert set(parents('a/b')) == {'a'}
+    assert set(parents('a/b/c')) == {'a/b', 'a'}
+    assert set(parents('a/b/c/d')) == {'a/b/c', 'a/b', 'a'}
 
 
 def test_nothing_added(temp_git_dir):
@@ -26,6 +46,36 @@ def test_adding_something_with_conflict(temp_git_dir):
         assert find_conflicting_filenames(['f.py', 'F.py']) == 1
 
 
+@skip_win32  # pragma: win32 no cover
+def test_adding_files_with_conflicting_directories(temp_git_dir):
+    with temp_git_dir.as_cwd():
+        temp_git_dir.mkdir('dir').join('x').write('foo')
+        temp_git_dir.mkdir('DIR').join('y').write('foo')
+        cmd_output('git', 'add', '-A')
+
+        assert find_conflicting_filenames([]) == 1
+
+
+@skip_win32  # pragma: win32 no cover
+def test_adding_files_with_conflicting_deep_directories(temp_git_dir):
+    with temp_git_dir.as_cwd():
+        temp_git_dir.mkdir('x').mkdir('y').join('z').write('foo')
+        temp_git_dir.join('X').write('foo')
+        cmd_output('git', 'add', '-A')
+
+        assert find_conflicting_filenames([]) == 1
+
+
+@skip_win32  # pragma: win32 no cover
+def test_adding_file_with_conflicting_directory(temp_git_dir):
+    with temp_git_dir.as_cwd():
+        temp_git_dir.mkdir('dir').join('x').write('foo')
+        temp_git_dir.join('DIR').write('foo')
+        cmd_output('git', 'add', '-A')
+
+        assert find_conflicting_filenames([]) == 1
+
+
 def test_added_file_not_in_pre_commits_list(temp_git_dir):
     with temp_git_dir.as_cwd():
         temp_git_dir.join('f.py').write("print('hello world')")
@@ -38,7 +88,7 @@ def test_file_conflicts_with_committed_file(temp_git_dir):
     with temp_git_dir.as_cwd():
         temp_git_dir.join('f.py').write("print('hello world')")
         cmd_output('git', 'add', 'f.py')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-n', '-m', 'Add f.py')
+        git_commit('-m', 'Add f.py')
 
         temp_git_dir.join('F.py').write("print('hello world')")
         cmd_output('git', 'add', 'F.py')
@@ -46,6 +96,19 @@ def test_file_conflicts_with_committed_file(temp_git_dir):
         assert find_conflicting_filenames(['F.py']) == 1
 
 
+@skip_win32  # pragma: win32 no cover
+def test_file_conflicts_with_committed_dir(temp_git_dir):
+    with temp_git_dir.as_cwd():
+        temp_git_dir.mkdir('dir').join('x').write('foo')
+        cmd_output('git', 'add', '-A')
+        git_commit('-m', 'Add f.py')
+
+        temp_git_dir.join('DIR').write('foo')
+        cmd_output('git', 'add', '-A')
+
+        assert find_conflicting_filenames([]) == 1
+
+
 def test_integration(temp_git_dir):
     with temp_git_dir.as_cwd():
         assert main(argv=[]) == 0

tests/check_docstring_first_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.check_docstring_first import check_docstring_first
@@ -15,7 +17,7 @@ TESTS = (
         b'from __future__ import unicode_literals\n'
         b'"foo"\n',
         1,
-        '{filename}:2 Module docstring appears after code '
+        '{filename}:2: Module docstring appears after code '
         '(code seen on line 1).\n',
     ),
     # Test double docstring
@@ -24,7 +26,7 @@ TESTS = (
         b'from __future__ import absolute_import\n'
         b'"fake docstring"\n',
         1,
-        '{filename}:3 Multiple module docstrings '
+        '{filename}:3: Multiple module docstrings '
         '(first docstring on line 1).\n',
     ),
     # Test multiple lines of code above
@@ -33,7 +35,7 @@ TESTS = (
         b'import sys\n'
         b'"docstring"\n',
         1,
-        '{filename}:3 Module docstring appears after code '
+        '{filename}:3: Module docstring appears after code '
         '(code seen on line 1).\n',
     ),
     # String literals in expressions are ok.
@@ -56,12 +58,12 @@ def test_unit(capsys, contents, expected, expected_out):
 def test_integration(tmpdir, capsys, contents, expected, expected_out):
     f = tmpdir.join('test.py')
     f.write_binary(contents)
-    assert main([f.strpath]) == expected
-    assert capsys.readouterr()[0] == expected_out.format(filename=f.strpath)
+    assert main([str(f)]) == expected
+    assert capsys.readouterr()[0] == expected_out.format(filename=str(f))
 
 
 def test_arbitrary_encoding(tmpdir):
     f = tmpdir.join('f.py')
     contents = '# -*- coding: cp1252\nx = "£"'.encode('cp1252')
     f.write_binary(contents)
-    assert main([f.strpath]) == 0
+    assert main([str(f)]) == 0

tests/check_executables_have_shebangs_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import os
 import sys
 
@@ -25,7 +27,7 @@ skip_win32 = pytest.mark.skipif(
 def test_has_shebang(content, tmpdir):
     path = tmpdir.join('path')
     path.write(content, 'wb')
-    assert main((path.strpath,)) == 0
+    assert main((str(path),)) == 0
 
 
 @skip_win32  # pragma: win32 no cover
@@ -41,7 +43,7 @@ def test_has_shebang(content, tmpdir):
 def test_bad_shebang(content, tmpdir, capsys):
     path = tmpdir.join('path')
     path.write(content, 'wb')
-    assert main((path.strpath,)) == 1
+    assert main((str(path),)) == 1
     _, stderr = capsys.readouterr()
     assert stderr.startswith(f'{path}: marked executable but')
 
@@ -73,6 +75,21 @@ def test_check_git_filemode_passing(tmpdir):
         assert check_executables_have_shebangs._check_git_filemode(files) == 0
 
 
+def test_check_git_filemode_passing_unusual_characters(tmpdir):
+    with tmpdir.as_cwd():
+        cmd_output('git', 'init', '.')
+
+        f = tmpdir.join('mañana.txt')
+        f.write('#!/usr/bin/env bash')
+        f_path = str(f)
+        cmd_output('chmod', '+x', f_path)
+        cmd_output('git', 'add', f_path)
+        cmd_output('git', 'update-index', '--chmod=+x', f_path)
+
+        files = (f_path,)
+        assert check_executables_have_shebangs._check_git_filemode(files) == 0
+
+
 def test_check_git_filemode_failing(tmpdir):
     with tmpdir.as_cwd():
         cmd_output('git', 'init', '.')

tests/check_illegal_windows_names_test.py

@@ -0,0 +1,63 @@
+from __future__ import annotations
+
+import os.path
+import re
+
+import pytest
+
+from pre_commit_hooks.check_yaml import yaml
+
+
+@pytest.fixture(scope='module')
+def hook_re():
+    here = os.path.dirname(__file__)
+    with open(os.path.join(here, '..', '.pre-commit-hooks.yaml')) as f:
+        hook_defs = yaml.load(f)
+    hook, = (
+        hook
+        for hook in hook_defs
+        if hook['id'] == 'check-illegal-windows-names'
+    )
+    yield re.compile(hook['files'])
+
+
+@pytest.mark.parametrize(
+    's',
+    (
+        pytest.param('aux.txt', id='with ext'),
+        pytest.param('aux', id='without ext'),
+        pytest.param('AuX.tXt', id='capitals'),
+        pytest.param('com7.dat', id='com with digit'),
+        pytest.param(':', id='bare colon'),
+        pytest.param('file:Zone.Identifier', id='mid colon'),
+        pytest.param('path/COM¹.json', id='com with superscript'),
+        pytest.param('dir/LPT³.toml', id='lpt with superscript'),
+        pytest.param('with < less than', id='with less than'),
+        pytest.param('Fast or Slow?.md', id='with question mark'),
+        pytest.param('with "double" quotes', id='with double quotes'),
+        pytest.param('with_null\x00byte', id='with null byte'),
+        pytest.param('ends_with.', id='ends with period'),
+        pytest.param('ends_with ', id='ends with space'),
+        pytest.param('ends_with\t', id='ends with tab'),
+        pytest.param('dir/ends./with.txt', id='directory ends with period'),
+        pytest.param('dir/ends /with.txt', id='directory ends with space'),
+    ),
+)
+def test_check_illegal_windows_names_matches(hook_re, s):
+    assert hook_re.search(s)
+
+
+@pytest.mark.parametrize(
+    's',
+    (
+        pytest.param('README.md', id='standard file'),
+        pytest.param('foo.aux', id='as ext'),
+        pytest.param('com.dat', id='com without digit'),
+        pytest.param('.python-version', id='starts with period'),
+        pytest.param(' pseudo nan', id='with spaces'),
+        pytest.param('!@#$%^&;=≤\'~`¡¿€🤗', id='with allowed characters'),
+        pytest.param('path.to/file.py', id='standard path'),
+    ),
+)
+def test_check_illegal_windows_names_does_not_match(hook_re, s):
+    assert hook_re.search(s) is None

tests/check_json_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.check_json import main
@@ -9,6 +11,7 @@ from testing.util import get_resource_path
         ('bad_json.notjson', 1),
         ('bad_json_latin1.nonjson', 1),
         ('ok_json.json', 0),
+        ('duplicate_key_json.notjson', 1),
     ),
 )
 def test_main(capsys, filename, expected_retval):

tests/check_merge_conflict_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import os
 import shutil
 
@@ -6,6 +8,7 @@ import pytest
 from pre_commit_hooks.check_merge_conflict import main
 from pre_commit_hooks.util import cmd_output
 from testing.util import get_resource_path
+from testing.util import git_commit
 
 
 @pytest.fixture
@@ -16,23 +19,23 @@ def f1_is_a_conflict_file(tmpdir):
     repo2 = tmpdir.join('repo2')
     repo2_f1 = repo2.join('f1')
 
-    cmd_output('git', 'init', '--', repo1.strpath)
+    cmd_output('git', 'init', '--', str(repo1))
     with repo1.as_cwd():
         repo1_f1.ensure()
         cmd_output('git', 'add', '.')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-m', 'commit1')
+        git_commit('-m', 'commit1')
 
-    cmd_output('git', 'clone', repo1.strpath, repo2.strpath)
+    cmd_output('git', 'clone', str(repo1), str(repo2))
 
-    # Commit in master
+    # Commit in mainline
     with repo1.as_cwd():
         repo1_f1.write('parent\n')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-am', 'master commit2')
+        git_commit('-am', 'mainline commit2')
 
     # Commit in clone and pull
     with repo2.as_cwd():
         repo2_f1.write('child\n')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-am', 'clone commit2')
+        git_commit('-am', 'clone commit2')
         cmd_output('git', 'pull', '--no-rebase', retcode=None)
         # We should end up in a merge conflict!
         f1 = repo2_f1.read()
@@ -71,24 +74,24 @@ def repository_pending_merge(tmpdir):
     repo2 = tmpdir.join('repo2')
     repo2_f1 = repo2.join('f1')
     repo2_f2 = repo2.join('f2')
-    cmd_output('git', 'init', repo1.strpath)
+    cmd_output('git', 'init', str(repo1))
     with repo1.as_cwd():
         repo1_f1.ensure()
         cmd_output('git', 'add', '.')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-m', 'commit1')
+        git_commit('-m', 'commit1')
 
-    cmd_output('git', 'clone', repo1.strpath, repo2.strpath)
+    cmd_output('git', 'clone', str(repo1), str(repo2))
 
-    # Commit in master
+    # Commit in mainline
     with repo1.as_cwd():
         repo1_f1.write('parent\n')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-am', 'master commit2')
+        git_commit('-am', 'mainline commit2')
 
     # Commit in clone and pull without committing
     with repo2.as_cwd():
         repo2_f2.write('child\n')
         cmd_output('git', 'add', '.')
-        cmd_output('git', 'commit', '--no-gpg-sign', '-m', 'clone commit2')
+        git_commit('-m', 'clone commit2')
         cmd_output('git', 'pull', '--no-commit', '--no-rebase')
         # We should end up in a pending merge
         assert repo2_f1.read() == 'parent\n'
@@ -98,12 +101,18 @@ def repository_pending_merge(tmpdir):
 
 
 @pytest.mark.usefixtures('f1_is_a_conflict_file')
-def test_merge_conflicts_git():
+def test_merge_conflicts_git(capsys):
     assert main(['f1']) == 1
+    out, _ = capsys.readouterr()
+    assert out == (
+        "f1:1: Merge conflict string '<<<<<<<' found\n"
+        "f1:3: Merge conflict string '=======' found\n"
+        "f1:5: Merge conflict string '>>>>>>>' found\n"
+    )
 
 
 @pytest.mark.parametrize(
-    'contents', (b'<<<<<<< HEAD\n', b'=======\n', b'>>>>>>> master\n'),
+    'contents', (b'<<<<<<< HEAD\n', b'=======\n', b'>>>>>>> main\n'),
 )
 def test_merge_conflicts_failing(contents, repository_pending_merge):
     repository_pending_merge.join('f2').write_binary(contents)
@@ -134,3 +143,15 @@ def test_care_when_assumed_merge(tmpdir):
     f = tmpdir.join('README.md')
     f.write_binary(b'problem\n=======\n')
     assert main([str(f.realpath()), '--assume-in-merge']) == 1
+
+
+def test_worktree_merge_conflicts(f1_is_a_conflict_file, tmpdir, capsys):
+    worktree = tmpdir.join('worktree')
+    cmd_output('git', 'worktree', 'add', str(worktree))
+    with worktree.as_cwd():
+        cmd_output(
+            'git', 'pull', '--no-rebase', 'origin', 'HEAD', retcode=None,
+        )
+        msg = f1_is_a_conflict_file.join('.git/worktrees/worktree/MERGE_MSG')
+        assert msg.exists()
+        test_merge_conflicts_git(capsys)

tests/check_shebang_scripts_are_executable_test.py

@@ -0,0 +1,89 @@
+from __future__ import annotations
+
+import os
+
+import pytest
+
+from pre_commit_hooks.check_shebang_scripts_are_executable import \
+    _check_git_filemode
+from pre_commit_hooks.check_shebang_scripts_are_executable import main
+from pre_commit_hooks.util import cmd_output
+
+
+def test_check_git_filemode_passing(tmpdir):
+    with tmpdir.as_cwd():
+        cmd_output('git', 'init', '.')
+
+        f = tmpdir.join('f')
+        f.write('#!/usr/bin/env bash')
+        f_path = str(f)
+        cmd_output('chmod', '+x', f_path)
+        cmd_output('git', 'add', f_path)
+        cmd_output('git', 'update-index', '--chmod=+x', f_path)
+
+        g = tmpdir.join('g').ensure()
+        g_path = str(g)
+        cmd_output('git', 'add', g_path)
+
+        files = [f_path, g_path]
+        assert _check_git_filemode(files) == 0
+
+        # this is the one we should trigger on
+        h = tmpdir.join('h')
+        h.write('#!/usr/bin/env bash')
+        h_path = str(h)
+        cmd_output('git', 'add', h_path)
+
+        files = [h_path]
+        assert _check_git_filemode(files) == 1
+
+
+def test_check_git_filemode_passing_unusual_characters(tmpdir):
+    with tmpdir.as_cwd():
+        cmd_output('git', 'init', '.')
+
+        f = tmpdir.join('mañana.txt')
+        f.write('#!/usr/bin/env bash')
+        f_path = str(f)
+        cmd_output('chmod', '+x', f_path)
+        cmd_output('git', 'add', f_path)
+        cmd_output('git', 'update-index', '--chmod=+x', f_path)
+
+        files = (f_path,)
+        assert _check_git_filemode(files) == 0
+
+
+def test_check_git_filemode_failing(tmpdir):
+    with tmpdir.as_cwd():
+        cmd_output('git', 'init', '.')
+
+        f = tmpdir.join('f').ensure()
+        f.write('#!/usr/bin/env bash')
+        f_path = str(f)
+        cmd_output('git', 'add', f_path)
+
+        files = (f_path,)
+        assert _check_git_filemode(files) == 1
+
+
+@pytest.mark.parametrize(
+    ('content', 'mode', 'expected'),
+    (
+        pytest.param('#!python', '+x', 0, id='shebang with executable'),
+        pytest.param('#!python', '-x', 1, id='shebang without executable'),
+        pytest.param('', '+x', 0, id='no shebang with executable'),
+        pytest.param('', '-x', 0, id='no shebang without executable'),
+    ),
+)
+def test_git_executable_shebang(temp_git_dir, content, mode, expected):
+    with temp_git_dir.as_cwd():
+        path = temp_git_dir.join('path')
+        path.write(content)
+        cmd_output('git', 'add', str(path))
+        cmd_output('chmod', mode, str(path))
+        cmd_output('git', 'update-index', f'--chmod={mode}', str(path))
+
+        # simulate how identify chooses that something is executable
+        filenames = [path for path in [str(path)] if os.access(path, os.X_OK)]
+
+        assert main(filenames) == expected

tests/check_symlinks_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import os
 
 import pytest
@@ -16,8 +18,8 @@ def test_main(tmpdir, dest, expected):  # pragma: no cover (symlinks)
     tmpdir.join('exists').ensure()
     symlink = tmpdir.join('symlink')
     symlink.mksymlinkto(tmpdir.join(dest))
-    assert main((symlink.strpath,)) == expected
+    assert main((str(symlink),)) == expected
 
 
 def test_main_normal_file(tmpdir):
-    assert main((tmpdir.join('f').ensure().strpath,)) == 0
+    assert main((str(tmpdir.join('f').ensure()),)) == 0

tests/check_toml_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pre_commit_hooks.check_toml import main
 
 
@@ -8,7 +10,7 @@ key = # INVALID
 
 = "no key name"  # INVALID
 """)
-    ret = main((filename.strpath,))
+    ret = main((str(filename),))
     assert ret == 1
 
 
@@ -25,12 +27,12 @@ name = "John"
 dob = 1979-05-27T07:32:00-08:00 # First class dates
 """,
     )
-    ret = main((filename.strpath,))
+    ret = main((str(filename),))
     assert ret == 0
 
 
 def test_toml_good_unicode(tmpdir):
     filename = tmpdir.join('f')
     filename.write_binary('letter = "\N{SNOWMAN}"\n'.encode())
-    ret = main((filename.strpath,))
+    ret = main((str(filename),))
     assert ret == 0

tests/check_vcs_permalinks_test.py

@@ -1,9 +1,11 @@
+from __future__ import annotations
+
 from pre_commit_hooks.check_vcs_permalinks import main
 
 
 def test_trivial(tmpdir):
     f = tmpdir.join('f.txt').ensure()
-    assert not main((f.strpath,))
+    assert not main((str(f),))
 
 
 def test_passing(tmpdir):
@@ -11,24 +13,28 @@ def test_passing(tmpdir):
     f.write_binary(
         # permalinks are ok
         b'https://github.com/asottile/test/blob/649e6/foo%20bar#L1\n'
+        # tags are ok
+        b'https://github.com/asottile/test/blob/1.0.0/foo%20bar#L1\n'
         # links to files but not line numbers are ok
-        b'https://github.com/asottile/test/blob/master/foo%20bar\n'
+        b'https://github.com/asottile/test/blob/main/foo%20bar\n'
         # regression test for overly-greedy regex
-        b'https://github.com/ yes / no ? /blob/master/foo#L1\n',
+        b'https://github.com/ yes / no ? /blob/main/foo#L1\n',
     )
-    assert not main((f.strpath,))
+    assert not main((str(f),))
 
 
 def test_failing(tmpdir, capsys):
     with tmpdir.as_cwd():
         tmpdir.join('f.txt').write_binary(
-            b'https://github.com/asottile/test/blob/master/foo#L1\n',
+            b'https://github.com/asottile/test/blob/main/foo#L1\n'
+            b'https://example.com/asottile/test/blob/main/foo#L1\n',
         )
 
-        assert main(('f.txt',))
+        assert main(('f.txt', '--additional-github-domain', 'example.com'))
         out, _ = capsys.readouterr()
         assert out == (
-            'f.txt:1:https://github.com/asottile/test/blob/master/foo#L1\n'
+            'f.txt:1:https://github.com/asottile/test/blob/main/foo#L1\n'
+            'f.txt:2:https://example.com/asottile/test/blob/main/foo#L1\n'
             '\n'
             'Non-permanent github link detected.\n'
             'On any page on github press [y] to load a permalink.\n'

tests/check_xml_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.check_xml import main

tests/check_yaml_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.check_yaml import main
@@ -20,16 +22,16 @@ def test_main_allow_multiple_documents(tmpdir):
     f.write('---\nfoo\n---\nbar\n')
 
     # should fail without the setting
-    assert main((f.strpath,))
+    assert main((str(f),))
 
     # should pass when we allow multiple documents
-    assert not main(('--allow-multiple-documents', f.strpath))
+    assert not main(('--allow-multiple-documents', str(f)))
 
 
 def test_fails_even_with_allow_multiple_documents(tmpdir):
     f = tmpdir.join('test.yaml')
     f.write('[')
-    assert main(('--allow-multiple-documents', f.strpath))
+    assert main(('--allow-multiple-documents', str(f)))
 
 
 def test_main_unsafe(tmpdir):
@@ -40,12 +42,12 @@ def test_main_unsafe(tmpdir):
         '    deadbeefdeadbeefdeadbeef\n',
     )
     # should fail "safe" check
-    assert main((f.strpath,))
+    assert main((str(f),))
     # should pass when we allow unsafe documents
-    assert not main(('--unsafe', f.strpath))
+    assert not main(('--unsafe', str(f)))
 
 
 def test_main_unsafe_still_fails_on_syntax_errors(tmpdir):
     f = tmpdir.join('test.yaml')
     f.write('[')
-    assert main(('--unsafe', f.strpath))
+    assert main(('--unsafe', str(f)))

tests/conftest.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.util import cmd_output
@@ -6,5 +8,5 @@ from pre_commit_hooks.util import cmd_output
 @pytest.fixture
 def temp_git_dir(tmpdir):
     git_dir = tmpdir.join('gits')
-    cmd_output('git', 'init', '--', git_dir.strpath)
+    cmd_output('git', 'init', '--', str(git_dir))
     yield git_dir

tests/debug_statement_hook_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import ast
 
 from pre_commit_hooks.debug_statement_hook import Debug
@@ -33,7 +35,7 @@ def test_finds_breakpoint():
 def test_returns_one_for_failing_file(tmpdir):
     f_py = tmpdir.join('f.py')
     f_py.write('def f():\n    import pdb; pdb.set_trace()')
-    ret = main([f_py.strpath])
+    ret = main([str(f_py)])
     assert ret == 1
 
 
@@ -50,10 +52,12 @@ def test_syntaxerror_file():
 def test_non_utf8_file(tmpdir):
     f_py = tmpdir.join('f.py')
     f_py.write_binary('# -*- coding: cp1252 -*-\nx = "€"\n'.encode('cp1252'))
-    assert main((f_py.strpath,)) == 0
+    assert main((str(f_py),)) == 0
 
 
-def test_py37_breakpoint(tmpdir):
+def test_py37_breakpoint(tmpdir, capsys):
     f_py = tmpdir.join('f.py')
     f_py.write('def f():\n    breakpoint()\n')
-    assert main((f_py.strpath,)) == 1
+    assert main((str(f_py),)) == 1
+    out, _ = capsys.readouterr()
+    assert out == f'{f_py}:2:4: breakpoint called\n'

tests/destroyed_symlinks_test.py

@@ -0,0 +1,75 @@
+from __future__ import annotations
+
+import os
+import subprocess
+
+import pytest
+
+from pre_commit_hooks.destroyed_symlinks import find_destroyed_symlinks
+from pre_commit_hooks.destroyed_symlinks import main
+from testing.util import git_commit
+
+TEST_SYMLINK = 'test_symlink'
+TEST_SYMLINK_TARGET = '/doesnt/really/matters'
+TEST_FILE = 'test_file'
+TEST_FILE_RENAMED = f'{TEST_FILE}_renamed'
+
+
+@pytest.fixture
+def repo_with_destroyed_symlink(tmpdir):
+    source_repo = tmpdir.join('src')
+    os.makedirs(source_repo, exist_ok=True)
+    test_repo = tmpdir.join('test')
+    with source_repo.as_cwd():
+        subprocess.check_call(('git', 'init'))
+        os.symlink(TEST_SYMLINK_TARGET, TEST_SYMLINK)
+        with open(TEST_FILE, 'w') as f:
+            print('some random content', file=f)
+        subprocess.check_call(('git', 'add', '.'))
+        git_commit('-m', 'initial')
+        assert b'120000 ' in subprocess.check_output(
+            ('git', 'cat-file', '-p', 'HEAD^{tree}'),
+        )
+    subprocess.check_call(
+        ('git', '-c', 'core.symlinks=false', 'clone', source_repo, test_repo),
+    )
+    with test_repo.as_cwd():
+        subprocess.check_call(
+            ('git', 'config', '--local', 'core.symlinks', 'true'),
+        )
+        subprocess.check_call(('git', 'mv', TEST_FILE, TEST_FILE_RENAMED))
+    assert not os.path.islink(test_repo.join(TEST_SYMLINK))
+    yield test_repo
+
+
+def test_find_destroyed_symlinks(repo_with_destroyed_symlink):
+    with repo_with_destroyed_symlink.as_cwd():
+        assert find_destroyed_symlinks([]) == []
+        assert main([]) == 0
+
+        subprocess.check_call(('git', 'add', TEST_SYMLINK))
+        assert find_destroyed_symlinks([TEST_SYMLINK]) == [TEST_SYMLINK]
+        assert find_destroyed_symlinks([]) == []
+        assert main([]) == 0
+        assert find_destroyed_symlinks([TEST_FILE_RENAMED, TEST_FILE]) == []
+        ALL_STAGED = [TEST_SYMLINK, TEST_FILE_RENAMED]
+        assert find_destroyed_symlinks(ALL_STAGED) == [TEST_SYMLINK]
+        assert main(ALL_STAGED) != 0
+
+        with open(TEST_SYMLINK, 'a') as f:
+            print(file=f)  # add trailing newline
+        subprocess.check_call(['git', 'add', TEST_SYMLINK])
+        assert find_destroyed_symlinks(ALL_STAGED) == [TEST_SYMLINK]
+        assert main(ALL_STAGED) != 0
+
+        with open(TEST_SYMLINK, 'w') as f:
+            print('0' * len(TEST_SYMLINK_TARGET), file=f)
+        subprocess.check_call(('git', 'add', TEST_SYMLINK))
+        assert find_destroyed_symlinks(ALL_STAGED) == []
+        assert main(ALL_STAGED) == 0
+
+        with open(TEST_SYMLINK, 'w') as f:
+            print('0' * (len(TEST_SYMLINK_TARGET) + 3), file=f)
+        subprocess.check_call(('git', 'add', TEST_SYMLINK))
+        assert find_destroyed_symlinks(ALL_STAGED) == []
+        assert main(ALL_STAGED) == 0

tests/detect_aws_credentials_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from unittest.mock import patch
 
 import pytest
@@ -13,15 +15,15 @@ from testing.util import get_resource_path
     ('env_vars', 'values'),
     (
         ({}, set()),
-        ({'AWS_DUMMY_KEY': '/foo'}, set()),
+        ({'AWS_PLACEHOLDER_KEY': '/foo'}, set()),
         ({'AWS_CONFIG_FILE': '/foo'}, {'/foo'}),
         ({'AWS_CREDENTIAL_FILE': '/foo'}, {'/foo'}),
         ({'AWS_SHARED_CREDENTIALS_FILE': '/foo'}, {'/foo'}),
         ({'BOTO_CONFIG': '/foo'}, {'/foo'}),
-        ({'AWS_DUMMY_KEY': '/foo', 'AWS_CONFIG_FILE': '/bar'}, {'/bar'}),
+        ({'AWS_PLACEHOLDER_KEY': '/foo', 'AWS_CONFIG_FILE': '/bar'}, {'/bar'}),
         (
             {
-                'AWS_DUMMY_KEY': '/foo', 'AWS_CONFIG_FILE': '/bar',
+                'AWS_PLACEHOLDER_KEY': '/foo', 'AWS_CONFIG_FILE': '/bar',
                 'AWS_CREDENTIAL_FILE': '/baz',
             },
             {'/bar', '/baz'},
@@ -44,13 +46,16 @@ def test_get_aws_credentials_file_from_env(env_vars, values):
     ('env_vars', 'values'),
     (
         ({}, set()),
-        ({'AWS_DUMMY_KEY': 'foo'}, set()),
+        ({'AWS_PLACEHOLDER_KEY': 'foo'}, set()),
         ({'AWS_SECRET_ACCESS_KEY': 'foo'}, {'foo'}),
         ({'AWS_SECURITY_TOKEN': 'foo'}, {'foo'}),
         ({'AWS_SESSION_TOKEN': 'foo'}, {'foo'}),
         ({'AWS_SESSION_TOKEN': ''}, set()),
         ({'AWS_SESSION_TOKEN': 'foo', 'AWS_SECURITY_TOKEN': ''}, {'foo'}),
-        ({'AWS_DUMMY_KEY': 'foo', 'AWS_SECRET_ACCESS_KEY': 'bar'}, {'bar'}),
+        (
+            {'AWS_PLACEHOLDER_KEY': 'foo', 'AWS_SECRET_ACCESS_KEY': 'bar'},
+            {'bar'},
+        ),
         (
             {'AWS_SECRET_ACCESS_KEY': 'foo', 'AWS_SECURITY_TOKEN': 'bar'},
             {'foo', 'bar'},

tests/detect_private_key_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.detect_private_key import main
@@ -10,6 +12,8 @@ TESTS = (
     (b'-----BEGIN OPENSSH PRIVATE KEY-----', 1),
     (b'PuTTY-User-Key-File-2: ssh-rsa', 1),
     (b'---- BEGIN SSH2 ENCRYPTED PRIVATE KEY ----', 1),
+    (b'-----BEGIN ENCRYPTED PRIVATE KEY-----', 1),
+    (b'-----BEGIN OpenVPN Static key V1-----', 1),
     (b'ssh-rsa DATA', 0),
     (b'ssh-dsa DATA', 0),
     # Some arbitrary binary data
@@ -21,4 +25,4 @@ TESTS = (
 def test_main(input_s, expected_retval, tmpdir):
     path = tmpdir.join('file.txt')
     path.write_binary(input_s)
-    assert main([path.strpath]) == expected_retval
+    assert main([str(path)]) == expected_retval

tests/end_of_file_fixer_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import io
 
 import pytest
@@ -35,7 +37,7 @@ def test_integration(input_s, expected_retval, output, tmpdir):
     path = tmpdir.join('file.txt')
     path.write_binary(input_s)
 
-    ret = main([path.strpath])
+    ret = main([str(path)])
     file_output = path.read_binary()
 
     assert file_output == output

tests/file_contents_sorter_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.file_contents_sorter import FAIL
@@ -6,28 +8,93 @@ from pre_commit_hooks.file_contents_sorter import PASS
 
 
 @pytest.mark.parametrize(
-    ('input_s', 'expected_retval', 'output'),
+    ('input_s', 'argv', 'expected_retval', 'output'),
     (
-        (b'', FAIL, b'\n'),
-        (b'lonesome\n', PASS, b'lonesome\n'),
-        (b'missing_newline', FAIL, b'missing_newline\n'),
-        (b'newline\nmissing', FAIL, b'missing\nnewline\n'),
-        (b'missing\nnewline', FAIL, b'missing\nnewline\n'),
-        (b'alpha\nbeta\n', PASS, b'alpha\nbeta\n'),
-        (b'beta\nalpha\n', FAIL, b'alpha\nbeta\n'),
-        (b'C\nc\n', PASS, b'C\nc\n'),
-        (b'c\nC\n', FAIL, b'C\nc\n'),
-        (b'mag ical \n tre vor\n', FAIL, b' tre vor\nmag ical \n'),
-        (b'@\n-\n_\n#\n', FAIL, b'#\n-\n@\n_\n'),
-        (b'extra\n\n\nwhitespace\n', FAIL, b'extra\nwhitespace\n'),
-        (b'whitespace\n\n\nextra\n', FAIL, b'extra\nwhitespace\n'),
+        (b'', [], PASS, b''),
+        (b'\n', [], FAIL, b''),
+        (b'\n\n', [], FAIL, b''),
+        (b'lonesome\n', [], PASS, b'lonesome\n'),
+        (b'missing_newline', [], FAIL, b'missing_newline\n'),
+        (b'newline\nmissing', [], FAIL, b'missing\nnewline\n'),
+        (b'missing\nnewline', [], FAIL, b'missing\nnewline\n'),
+        (b'alpha\nbeta\n', [], PASS, b'alpha\nbeta\n'),
+        (b'beta\nalpha\n', [], FAIL, b'alpha\nbeta\n'),
+        (b'C\nc\n', [], PASS, b'C\nc\n'),
+        (b'c\nC\n', [], FAIL, b'C\nc\n'),
+        (b'mag ical \n tre vor\n', [], FAIL, b' tre vor\nmag ical \n'),
+        (b'@\n-\n_\n#\n', [], FAIL, b'#\n-\n@\n_\n'),
+        (b'extra\n\n\nwhitespace\n', [], FAIL, b'extra\nwhitespace\n'),
+        (b'whitespace\n\n\nextra\n', [], FAIL, b'extra\nwhitespace\n'),
+        (
+            b'fee\nFie\nFoe\nfum\n',
+            [],
+            FAIL,
+            b'Fie\nFoe\nfee\nfum\n',
+        ),
+        (
+            b'Fie\nFoe\nfee\nfum\n',
+            [],
+            PASS,
+            b'Fie\nFoe\nfee\nfum\n',
+        ),
+        (
+            b'fee\nFie\nFoe\nfum\n',
+            ['--ignore-case'],
+            PASS,
+            b'fee\nFie\nFoe\nfum\n',
+        ),
+        (
+            b'Fie\nFoe\nfee\nfum\n',
+            ['--ignore-case'],
+            FAIL,
+            b'fee\nFie\nFoe\nfum\n',
+        ),
+        (
+            b'Fie\nFoe\nfee\nfee\nfum\n',
+            ['--ignore-case'],
+            FAIL,
+            b'fee\nfee\nFie\nFoe\nfum\n',
+        ),
+        (
+            b'Fie\nFoe\nfee\nfum\n',
+            ['--unique'],
+            PASS,
+            b'Fie\nFoe\nfee\nfum\n',
+        ),
+        (
+            b'Fie\nFie\nFoe\nfee\nfum\n',
+            ['--unique'],
+            FAIL,
+            b'Fie\nFoe\nfee\nfum\n',
+        ),
     ),
 )
-def test_integration(input_s, expected_retval, output, tmpdir):
+def test_integration(input_s, argv, expected_retval, output, tmpdir):
     path = tmpdir.join('file.txt')
     path.write_binary(input_s)
 
-    output_retval = main([path.strpath])
+    output_retval = main([str(path)] + argv)
 
     assert path.read_binary() == output
     assert output_retval == expected_retval
+
+
+@pytest.mark.parametrize(
+    ('input_s', 'argv'),
+    (
+        (
+            b'fee\nFie\nFoe\nfum\n',
+            ['--unique', '--ignore-case'],
+        ),
+        (
+            b'fee\nfee\nFie\nFoe\nfum\n',
+            ['--unique', '--ignore-case'],
+        ),
+    ),
+)
+def test_integration_invalid_args(input_s, argv, tmpdir):
+    path = tmpdir.join('file.txt')
+    path.write_binary(input_s)
+
+    with pytest.raises(SystemExit):
+        main([str(path)] + argv)

tests/fix_byte_order_marker_test.py

@@ -1,13 +1,15 @@
-from pre_commit_hooks import check_byte_order_marker
+from __future__ import annotations
+
+from pre_commit_hooks import fix_byte_order_marker
 
 
 def test_failure(tmpdir):
     f = tmpdir.join('f.txt')
     f.write_text('ohai', encoding='utf-8-sig')
-    assert check_byte_order_marker.main((f.strpath,)) == 1
+    assert fix_byte_order_marker.main((str(f),)) == 1
 
 
 def test_success(tmpdir):
     f = tmpdir.join('f.txt')
     f.write_text('ohai', encoding='utf-8')
-    assert check_byte_order_marker.main((f.strpath,)) == 0
+    assert fix_byte_order_marker.main((str(f),)) == 0

tests/fix_encoding_pragma_test.py

@@ -1,159 +0,0 @@
-import io
-
-import pytest
-
-from pre_commit_hooks.fix_encoding_pragma import _normalize_pragma
-from pre_commit_hooks.fix_encoding_pragma import fix_encoding_pragma
-from pre_commit_hooks.fix_encoding_pragma import main
-
-
-def test_integration_inserting_pragma(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'import httplib\n')
-
-    assert main((path.strpath,)) == 1
-
-    assert path.read_binary() == (
-        b'# -*- coding: utf-8 -*-\n'
-        b'import httplib\n'
-    )
-
-
-def test_integration_ok(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'# -*- coding: utf-8 -*-\nx = 1\n')
-    assert main((path.strpath,)) == 0
-
-
-def test_integration_remove(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'# -*- coding: utf-8 -*-\nx = 1\n')
-
-    assert main((path.strpath, '--remove')) == 1
-
-    assert path.read_binary() == b'x = 1\n'
-
-
-def test_integration_remove_ok(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'x = 1\n')
-    assert main((path.strpath, '--remove')) == 0
-
-
-@pytest.mark.parametrize(
-    'input_str',
-    (
-        b'',
-        (
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n'
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'foo = "bar"\n'
-        ),
-    ),
-)
-def test_ok_inputs(input_str):
-    bytesio = io.BytesIO(input_str)
-    assert fix_encoding_pragma(bytesio) == 0
-    bytesio.seek(0)
-    assert bytesio.read() == input_str
-
-
-@pytest.mark.parametrize(
-    ('input_str', 'output'),
-    (
-        (
-            b'import httplib\n',
-            b'# -*- coding: utf-8 -*-\n'
-            b'import httplib\n',
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'x = 1\n',
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        (
-            b'#coding=utf-8\n'
-            b'x = 1\n',
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'#coding=utf8\n'
-            b'x = 1\n',
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        # These should each get truncated
-        (b'#coding: utf-8\n', b''),
-        (b'# -*- coding: utf-8 -*-\n', b''),
-        (b'#!/usr/bin/env python\n', b''),
-        (b'#!/usr/bin/env python\n#coding: utf8\n', b''),
-        (b'#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n', b''),
-    ),
-)
-def test_not_ok_inputs(input_str, output):
-    bytesio = io.BytesIO(input_str)
-    assert fix_encoding_pragma(bytesio) == 1
-    bytesio.seek(0)
-    assert bytesio.read() == output
-
-
-def test_ok_input_alternate_pragma():
-    input_s = b'# coding: utf-8\nx = 1\n'
-    bytesio = io.BytesIO(input_s)
-    ret = fix_encoding_pragma(bytesio, expected_pragma=b'# coding: utf-8')
-    assert ret == 0
-    bytesio.seek(0)
-    assert bytesio.read() == input_s
-
-
-def test_not_ok_input_alternate_pragma():
-    bytesio = io.BytesIO(b'x = 1\n')
-    ret = fix_encoding_pragma(bytesio, expected_pragma=b'# coding: utf-8')
-    assert ret == 1
-    bytesio.seek(0)
-    assert bytesio.read() == b'# coding: utf-8\nx = 1\n'
-
-
-@pytest.mark.parametrize(
-    ('input_s', 'expected'),
-    (
-        ('# coding: utf-8', b'# coding: utf-8'),
-        # trailing whitespace
-        ('# coding: utf-8\n', b'# coding: utf-8'),
-    ),
-)
-def test_normalize_pragma(input_s, expected):
-    assert _normalize_pragma(input_s) == expected
-
-
-def test_integration_alternate_pragma(tmpdir, capsys):
-    f = tmpdir.join('f.py')
-    f.write('x = 1\n')
-
-    pragma = '# coding: utf-8'
-    assert main((f.strpath, '--pragma', pragma)) == 1
-    assert f.read() == '# coding: utf-8\nx = 1\n'
-    out, _ = capsys.readouterr()
-    assert out == f'Added `# coding: utf-8` to {f.strpath}\n'
-
-
-def test_crlf_ok(tmpdir):
-    f = tmpdir.join('f.py')
-    f.write_binary(b'# -*- coding: utf-8 -*-\r\nx = 1\r\n')
-    assert not main((f.strpath,))
-
-
-def test_crfl_adds(tmpdir):
-    f = tmpdir.join('f.py')
-    f.write_binary(b'x = 1\r\n')
-    assert main((f.strpath,))
-    assert f.read_binary() == b'# -*- coding: utf-8 -*-\r\nx = 1\r\n'

tests/forbid_new_submodules_test.py

@@ -1,22 +1,22 @@
+from __future__ import annotations
+
+import os
 import subprocess
+from unittest import mock
 
 import pytest
 
 from pre_commit_hooks.forbid_new_submodules import main
+from testing.util import git_commit
 
 
 @pytest.fixture
 def git_dir_with_git_dir(tmpdir):
     with tmpdir.as_cwd():
         subprocess.check_call(('git', 'init', '.'))
-        subprocess.check_call((
-            'git', 'commit', '-m', 'init', '--allow-empty', '--no-gpg-sign',
-        ))
+        git_commit('--allow-empty', '-m', 'init')
         subprocess.check_call(('git', 'init', 'foo'))
-        subprocess.check_call(
-            ('git', 'commit', '-m', 'init', '--allow-empty', '--no-gpg-sign'),
-            cwd=tmpdir.join('foo').strpath,
-        )
+        git_commit('--allow-empty', '-m', 'init', cwd=str(tmpdir.join('foo')))
         yield
 
 
@@ -31,7 +31,24 @@ def git_dir_with_git_dir(tmpdir):
 )
 def test_main_new_submodule(git_dir_with_git_dir, capsys, cmd):
     subprocess.check_call(cmd)
-    assert main() == 1
+    assert main(('random_non-related_file',)) == 0
+    assert main(('foo',)) == 1
+    out, _ = capsys.readouterr()
+    assert out.startswith('foo: new submodule introduced\n')
+
+
+def test_main_new_submodule_committed(git_dir_with_git_dir, capsys):
+    rev_parse_cmd = ('git', 'rev-parse', 'HEAD')
+    from_ref = subprocess.check_output(rev_parse_cmd).decode().strip()
+    subprocess.check_call(('git', 'submodule', 'add', './foo'))
+    git_commit('-m', 'new submodule')
+    to_ref = subprocess.check_output(rev_parse_cmd).decode().strip()
+    with mock.patch.dict(
+        os.environ,
+        {'PRE_COMMIT_FROM_REF': from_ref, 'PRE_COMMIT_TO_REF': to_ref},
+    ):
+        assert main(('random_non-related_file',)) == 0
+        assert main(('foo',)) == 1
     out, _ = capsys.readouterr()
     assert out.startswith('foo: new submodule introduced\n')
 
@@ -39,4 +56,4 @@ def test_main_new_submodule(git_dir_with_git_dir, capsys, cmd):
 def test_main_no_new_submodule(git_dir_with_git_dir):
     open('test.py', 'a+').close()
     subprocess.check_call(('git', 'add', 'test.py'))
-    assert main() == 0
+    assert main(('test.py',)) == 0

tests/mixed_line_ending_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.mixed_line_ending import main
@@ -25,7 +27,7 @@ from pre_commit_hooks.mixed_line_ending import main
 def test_mixed_line_ending_fixes_auto(input_s, output, tmpdir):
     path = tmpdir.join('file.txt')
     path.write_binary(input_s)
-    ret = main((path.strpath,))
+    ret = main((str(path),))
 
     assert ret == 1
     assert path.read_binary() == output
@@ -34,7 +36,7 @@ def test_mixed_line_ending_fixes_auto(input_s, output, tmpdir):
 def test_non_mixed_no_newline_end_of_file(tmpdir):
     path = tmpdir.join('f.txt')
     path.write_binary(b'foo\nbar\nbaz')
-    assert not main((path.strpath,))
+    assert not main((str(path),))
     # the hook *could* fix the end of the file, but leaves it alone
     # this is mostly to document the current behaviour
     assert path.read_binary() == b'foo\nbar\nbaz'
@@ -43,7 +45,7 @@ def test_non_mixed_no_newline_end_of_file(tmpdir):
 def test_mixed_no_newline_end_of_file(tmpdir):
     path = tmpdir.join('f.txt')
     path.write_binary(b'foo\r\nbar\nbaz')
-    assert main((path.strpath,))
+    assert main((str(path),))
     # the hook rewrites the end of the file, this is slightly inconsistent
     # with the non-mixed case but I think this is the better behaviour
     # this is mostly to document the current behaviour
@@ -66,7 +68,7 @@ def test_mixed_no_newline_end_of_file(tmpdir):
 def test_line_endings_ok(fix_option, input_s, tmpdir, capsys):
     path = tmpdir.join('input.txt')
     path.write_binary(input_s)
-    ret = main((fix_option, path.strpath))
+    ret = main((fix_option, str(path)))
 
     assert ret == 0
     assert path.read_binary() == input_s
@@ -78,7 +80,7 @@ def test_no_fix_does_not_modify(tmpdir, capsys):
     path = tmpdir.join('input.txt')
     contents = b'foo\r\nbar\rbaz\nwomp\n'
     path.write_binary(contents)
-    ret = main(('--fix=no', path.strpath))
+    ret = main(('--fix=no', str(path)))
 
     assert ret == 1
     assert path.read_binary() == contents
@@ -89,7 +91,7 @@ def test_no_fix_does_not_modify(tmpdir, capsys):
 def test_fix_lf(tmpdir, capsys):
     path = tmpdir.join('input.txt')
     path.write_binary(b'foo\r\nbar\rbaz\n')
-    ret = main(('--fix=lf', path.strpath))
+    ret = main(('--fix=lf', str(path)))
 
     assert ret == 1
     assert path.read_binary() == b'foo\nbar\nbaz\n'
@@ -100,7 +102,7 @@ def test_fix_lf(tmpdir, capsys):
 def test_fix_crlf(tmpdir):
     path = tmpdir.join('input.txt')
     path.write_binary(b'foo\r\nbar\rbaz\n')
-    ret = main(('--fix=crlf', path.strpath))
+    ret = main(('--fix=crlf', str(path)))
 
     assert ret == 1
     assert path.read_binary() == b'foo\r\nbar\r\nbaz\r\n'
@@ -110,7 +112,7 @@ def test_fix_lf_all_crlf(tmpdir):
     """Regression test for #239"""
     path = tmpdir.join('input.txt')
     path.write_binary(b'foo\r\nbar\r\n')
-    ret = main(('--fix=lf', path.strpath))
+    ret = main(('--fix=lf', str(path)))
 
     assert ret == 1
     assert path.read_binary() == b'foo\nbar\n'

tests/no_commit_to_branch_test.py

@@ -1,20 +1,23 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.no_commit_to_branch import is_on_branch
 from pre_commit_hooks.no_commit_to_branch import main
 from pre_commit_hooks.util import cmd_output
+from testing.util import git_commit
 
 
 def test_other_branch(temp_git_dir):
     with temp_git_dir.as_cwd():
         cmd_output('git', 'checkout', '-b', 'anotherbranch')
-        assert is_on_branch({'master'}) is False
+        assert is_on_branch({'placeholder'}) is False
 
 
 def test_multi_branch(temp_git_dir):
     with temp_git_dir.as_cwd():
         cmd_output('git', 'checkout', '-b', 'another/branch')
-        assert is_on_branch({'master'}) is False
+        assert is_on_branch({'placeholder'}) is False
 
 
 def test_multi_branch_fail(temp_git_dir):
@@ -23,9 +26,10 @@ def test_multi_branch_fail(temp_git_dir):
         assert is_on_branch({'another/branch'}) is True
 
 
-def test_master_branch(temp_git_dir):
+def test_exact_branch(temp_git_dir):
     with temp_git_dir.as_cwd():
-        assert is_on_branch({'master'}) is True
+        cmd_output('git', 'checkout', '-b', 'branchname')
+        assert is_on_branch({'branchname'}) is True
 
 
 def test_main_branch_call(temp_git_dir):
@@ -47,11 +51,11 @@ def test_branch_pattern_fail(temp_git_dir):
         assert is_on_branch(set(), {'another/.*'}) is True
 
 
-@pytest.mark.parametrize('branch_name', ('master', 'another/branch'))
+@pytest.mark.parametrize('branch_name', ('somebranch', 'another/branch'))
 def test_branch_pattern_multiple_branches_fail(temp_git_dir, branch_name):
     with temp_git_dir.as_cwd():
         cmd_output('git', 'checkout', '-b', branch_name)
-        assert main(('--branch', 'master', '--pattern', 'another/.*'))
+        assert main(('--branch', 'somebranch', '--pattern', 'another/.*'))
 
 
 def test_main_default_call(temp_git_dir):
@@ -62,8 +66,15 @@ def test_main_default_call(temp_git_dir):
 
 def test_not_on_a_branch(temp_git_dir):
     with temp_git_dir.as_cwd():
-        cmd_output('git', 'commit', '--no-gpg-sign', '--allow-empty', '-m1')
+        git_commit('--allow-empty', '-m1')
         head = cmd_output('git', 'rev-parse', 'HEAD').strip()
         cmd_output('git', 'checkout', head)
         # we're not on a branch!
         assert main(()) == 0
+
+
+@pytest.mark.parametrize('branch_name', ('master', 'main'))
+def test_default_branch_names(temp_git_dir, branch_name):
+    with temp_git_dir.as_cwd():
+        cmd_output('git', 'checkout', '-b', branch_name)
+        assert main(()) == 1

tests/pretty_format_json_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import os
 import shutil
 
@@ -67,19 +69,37 @@ def test_autofix_main(tmpdir):
     srcfile = tmpdir.join('to_be_json_formatted.json')
     shutil.copyfile(
         get_resource_path('not_pretty_formatted_json.json'),
-        srcfile.strpath,
+        str(srcfile),
     )
 
     # now launch the autofix on that file
-    ret = main(['--autofix', srcfile.strpath])
+    ret = main(['--autofix', str(srcfile)])
     # it should have formatted it
     assert ret == 1
 
     # file was formatted (shouldn't trigger linter again)
-    ret = main([srcfile.strpath])
+    ret = main([str(srcfile)])
     assert ret == 0
 
 
+def test_invalid_main(tmpdir):
+    srcfile1 = tmpdir.join('not_valid_json.json')
+    srcfile1.write(
+        '{\n'
+        '  // not json\n'
+        '  "a": "b"\n'
+        '}',
+    )
+    srcfile2 = tmpdir.join('to_be_json_formatted.json')
+    srcfile2.write('{ "a": "b" }')
+
+    # it should have skipped the first file and formatted the second one
+    assert main(['--autofix', str(srcfile1), str(srcfile2)]) == 1
+
+    # confirm second file was formatted (shouldn't trigger linter again)
+    assert main([str(srcfile2)]) == 0
+
+
 def test_orderfile_get_pretty_format():
     ret = main((
         '--top-keys=alist', get_resource_path('pretty_formatted_json.json'),

tests/readme_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pre_commit_hooks.check_yaml import yaml
 
 

tests/removed_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.removed import main

tests/requirements_txt_fixer_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.requirements_txt_fixer import FAIL
@@ -30,6 +32,16 @@ from pre_commit_hooks.requirements_txt_fixer import Requirement
         ),
         (b'#comment\n\nfoo\nbar\n', FAIL, b'#comment\n\nbar\nfoo\n'),
         (b'#comment\n\nbar\nfoo\n', PASS, b'#comment\n\nbar\nfoo\n'),
+        (
+            b'foo\n\t#comment with indent\nbar\n',
+            FAIL,
+            b'\t#comment with indent\nbar\nfoo\n',
+        ),
+        (
+            b'bar\n\t#comment with indent\nfoo\n',
+            PASS,
+            b'bar\n\t#comment with indent\nfoo\n',
+        ),
         (b'\nfoo\nbar\n', FAIL, b'bar\n\nfoo\n'),
         (b'\nbar\nfoo\n', PASS, b'\nbar\nfoo\n'),
         (
@@ -56,6 +68,12 @@ from pre_commit_hooks.requirements_txt_fixer import Requirement
             b'f<=2\n'
             b'g<2\n',
         ),
+        (b'a==1\nb==1\na==1\n', FAIL, b'a==1\nb==1\n'),
+        (
+            b'a==1\nb==1\n#comment about a\na==1\n',
+            FAIL,
+            b'#comment about a\na==1\nb==1\n',
+        ),
         (b'ocflib\nDjango\nPyMySQL\n', FAIL, b'Django\nocflib\nPyMySQL\n'),
         (
             b'-e git+ssh://git_url@tag#egg=ocflib\nDjango\nPyMySQL\n',
@@ -64,6 +82,8 @@ from pre_commit_hooks.requirements_txt_fixer import Requirement
         ),
         (b'bar\npkg-resources==0.0.0\nfoo\n', FAIL, b'bar\nfoo\n'),
         (b'foo\npkg-resources==0.0.0\nbar\n', FAIL, b'bar\nfoo\n'),
+        (b'bar\npkg_resources==0.0.0\nfoo\n', FAIL, b'bar\nfoo\n'),
+        (b'foo\npkg_resources==0.0.0\nbar\n', FAIL, b'bar\nfoo\n'),
         (
             b'git+ssh://git_url@tag#egg=ocflib\nDjango\nijk\n',
             FAIL,
@@ -93,7 +113,7 @@ def test_integration(input_s, expected_retval, output, tmpdir):
     path = tmpdir.join('file.txt')
     path.write_binary(input_s)
 
-    output_retval = main([path.strpath])
+    output_retval = main([str(path)])
 
     assert path.read_binary() == output
     assert output_retval == expected_retval

tests/sort_simple_yaml_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import os
 
 import pytest
@@ -39,7 +41,7 @@ TEST_SORTS = [
 
 @pytest.mark.parametrize('bad_lines,good_lines,retval', TEST_SORTS)
 def test_integration_good_bad_lines(tmpdir, bad_lines, good_lines, retval):
-    file_path = os.path.join(tmpdir.strpath, 'foo.yaml')
+    file_path = os.path.join(str(tmpdir), 'foo.yaml')
 
     with open(file_path, 'w') as f:
         f.write('\n'.join(bad_lines) + '\n')

tests/string_fixer_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import textwrap
 
 import pytest
@@ -35,6 +37,12 @@ TESTS = (
         1,
     ),
     ('"foo""bar"', "'foo''bar'", 1),
+    pytest.param(
+        "f'hello{\"world\"}'",
+        "f'hello{\"world\"}'",
+        0,
+        id='ignore nested fstrings',
+    ),
 )
 
 
@@ -42,7 +50,7 @@ TESTS = (
 def test_rewrite(input_s, output, expected_retval, tmpdir):
     path = tmpdir.join('file.py')
     path.write(input_s)
-    retval = main([path.strpath])
+    retval = main([str(path)])
     assert path.read() == output
     assert retval == expected_retval
 
@@ -50,5 +58,5 @@ def test_rewrite(input_s, output, expected_retval, tmpdir):
 def test_rewrite_crlf(tmpdir):
     f = tmpdir.join('f.py')
     f.write_binary(b'"foo"\r\n"bar"\r\n')
-    assert main((f.strpath,))
+    assert main((str(f),))
     assert f.read_binary() == b"'foo'\r\n'bar'\r\n"

tests/tests_should_end_in_test_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pre_commit_hooks.tests_should_end_in_test import main
 
 
@@ -41,3 +43,8 @@ def test_main_not_django_fails():
 def test_main_django_fails():
     ret = main(['--django', 'foo_test.py', 'test_bar.py', 'test_baz.py'])
     assert ret == 1
+
+
+def test_main_pytest_test_first():
+    assert main(['--pytest-test-first', 'test_foo.py']) == 0
+    assert main(['--pytest-test-first', 'foo_test.py']) == 1

tests/trailing_whitespace_fixer_test.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.trailing_whitespace_fixer import main
@@ -13,14 +15,14 @@ from pre_commit_hooks.trailing_whitespace_fixer import main
 def test_fixes_trailing_whitespace(input_s, expected, tmpdir):
     path = tmpdir.join('file.md')
     path.write(input_s)
-    assert main((path.strpath,)) == 1
+    assert main((str(path),)) == 1
     assert path.read() == expected
 
 
 def test_ok_no_newline_end_of_file(tmpdir):
     filename = tmpdir.join('f')
     filename.write_binary(b'foo\nbar')
-    ret = main((filename.strpath,))
+    ret = main((str(filename),))
     assert filename.read_binary() == b'foo\nbar'
     assert ret == 0
 
@@ -28,7 +30,7 @@ def test_ok_no_newline_end_of_file(tmpdir):
 def test_ok_with_dos_line_endings(tmpdir):
     filename = tmpdir.join('f')
     filename.write_binary(b'foo\r\nbar\r\nbaz\r\n')
-    ret = main((filename.strpath,))
+    ret = main((str(filename),))
     assert filename.read_binary() == b'foo\r\nbar\r\nbaz\r\n'
     assert ret == 0
 
@@ -43,7 +45,7 @@ def test_fixes_markdown_files(tmpdir, ext):
         '\t\n'  # trailing tabs are stripped anyway
         '\n  ',  # whitespace at the end of the file is removed
     )
-    ret = main((path.strpath, f'--markdown-linebreak-ext={ext}'))
+    ret = main((str(path), f'--markdown-linebreak-ext={ext}'))
     assert ret == 1
     assert path.read() == (
         'foo  \n'
@@ -63,7 +65,7 @@ def test_markdown_linebreak_ext_badopt(arg):
 
 def test_prints_warning_with_no_markdown_ext(capsys, tmpdir):
     f = tmpdir.join('f').ensure()
-    assert main((f.strpath, '--no-markdown-linebreak-ext')) == 0
+    assert main((str(f), '--no-markdown-linebreak-ext')) == 0
     out, _ = capsys.readouterr()
     assert out == '--no-markdown-linebreak-ext now does nothing!\n'
 
@@ -72,7 +74,7 @@ def test_preserve_non_utf8_file(tmpdir):
     non_utf8_bytes_content = b'<a>\xe9 \n</a>\n'
     path = tmpdir.join('file.txt')
     path.write_binary(non_utf8_bytes_content)
-    ret = main([path.strpath])
+    ret = main([str(path)])
     assert ret == 1
     assert path.size() == (len(non_utf8_bytes_content) - 1)
 
@@ -81,7 +83,7 @@ def test_custom_charset_change(tmpdir):
     # strip spaces only, no tabs
     path = tmpdir.join('file.txt')
     path.write('\ta \t \n')
-    ret = main([path.strpath, '--chars', ' '])
+    ret = main([str(path), '--chars', ' '])
     assert ret == 1
     assert path.read() == '\ta \t\n'
 
@@ -89,13 +91,13 @@ def test_custom_charset_change(tmpdir):
 def test_custom_charset_no_change(tmpdir):
     path = tmpdir.join('file.txt')
     path.write('\ta \t\n')
-    ret = main([path.strpath, '--chars', ' '])
+    ret = main([str(path), '--chars', ' '])
     assert ret == 0
 
 
 def test_markdown_with_custom_charset(tmpdir):
     path = tmpdir.join('file.md')
     path.write('\ta \t   \n')
-    ret = main([path.strpath, '--chars', ' ', '--markdown-linebreak-ext', '*'])
+    ret = main([str(path), '--chars', ' ', '--markdown-linebreak-ext', '*'])
     assert ret == 1
     assert path.read() == '\ta \t  \n'

tests/util_test.py

@@ -1,7 +1,10 @@
+from __future__ import annotations
+
 import pytest
 
 from pre_commit_hooks.util import CalledProcessError
 from pre_commit_hooks.util import cmd_output
+from pre_commit_hooks.util import zsplit
 
 
 def test_raises_on_error():
@@ -12,3 +15,13 @@ def test_raises_on_error():
 def test_output():
     ret = cmd_output('sh', '-c', 'echo hi')
     assert ret == 'hi\n'
+
+
+@pytest.mark.parametrize('out', ('\0f1\0f2\0', '\0f1\0f2', 'f1\0f2\0'))
+def test_check_zsplits_str_correctly(out):
+    assert zsplit(out) == ['f1', 'f2']
+
+
+@pytest.mark.parametrize('out', ('\0\0', '\0', ''))
+def test_check_zsplit_returns_empty(out):
+    assert zsplit(out) == []

tox.ini

@@ -1,5 +1,5 @@
 [tox]
-envlist = py36,py37,py38,pypy3,pre-commit
+envlist = py,pre-commit
 
 [testenv]
 deps = -rrequirements-dev.txt
@@ -11,8 +11,7 @@ setenv =
 commands =
     coverage erase
     coverage run -m pytest {posargs:tests}
-    coverage report --fail-under 100
-    pre-commit install
+    coverage report
 
 [testenv:pre-commit]
 skip_install = true