Add new byte-order-marker checker/fixer

.pre-commit-hooks.yaml

@@ -5,6 +5,12 @@
  language: python
  always_run: true
  pass_filenames: false
+- id: byte-order-marker
+ name: Remove or disallow byte-order marker
+ description: Removes or disallows UTF-8 byte-order marker
+ entry: byte-order-marker
+ language: python
+ types: [text]
 - id: check-added-large-files
  name: Check for added large files
  description: Prevent giant files from being committed
@@ -17,7 +23,7 @@
  language: python
  types: [python]
 - id: check-byte-order-marker
- name: Check for byte-order marker
+ name: Check for byte-order marker (deprecated, use byte-order-marker)
  description: Forbid files which have a UTF-8 byte-order marker
  entry: check-byte-order-marker
  language: python

README.md

@@ -23,6 +23,12 @@ Add this to your `.pre-commit-config.yaml`
 
 ### Hooks available
 
+#### `byte-order-marker`
+Fix or forbid files which have a UTF-8 byte-order marker
+ - `--fix={auto,no}`
+ - `auto` - Removes BOMs automatically. This is the default argument.
+ - `no` - Checks if there are BOMs without modifying any file.
+
 #### `check-added-large-files`
 Prevent giant files from being committed.
  - Specify what is "too large" with `args: ['--maxkb=123']` (default=500kB).
@@ -42,9 +48,6 @@ 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.
 
@@ -183,6 +186,7 @@ Trims trailing whitespace.
  [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 byte-order-marker
 
 ### As a standalone package
 

pre_commit_hooks/byte_order_marker.py

@@ -0,0 +1,45 @@
+import argparse
+from typing import Optional
+from typing import Sequence
+
+
+def main(argv: Optional[Sequence[str]] = None) -> int:
+ parser = argparse.ArgumentParser()
+ parser.add_argument(
+ '-f', '--fix',
+ choices=('auto', 'no'),
+ default='auto',
+ help='Remove the byte-order marker. Default is "auto"',
+ )
+ parser.add_argument('filenames', nargs='*', help='Filenames to check')
+ args = parser.parse_args(argv)
+
+ retv = 0
+
+ for filename in args.filenames:
+ f = open(filename, 'rb')
+ if f.read(3) == b'\xef\xbb\xbf':
+ f.close()
+ if args.fix == 'no':
+ print(f'{filename}: Found byte-order marker')
+ else:
+ with open(
+ filename, newline='',
+ encoding='utf-8-sig',
+ ) as rewritefile:
+ contents = rewritefile.read()
+
+ with open(
+ filename, 'w', newline='',
+ encoding='utf-8',
+ ) as rewritefile:
+ rewritefile.writelines(contents)
+
+ print(f'{filename}: Removed byte-order marker')
+ retv = 1
+
+ return retv
+
+
+if __name__ == '__main__':
+ exit(main())

setup.cfg

@@ -28,6 +28,7 @@ python_requires = >=3.6.1
 
 [options.entry_points]
 console_scripts =
+ byte-order-marker = pre_commit_hooks.byte_order_marker:main
  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

tests/byte_order_marker_test.py

@@ -0,0 +1,25 @@
+from pre_commit_hooks import byte_order_marker
+
+
+def test_failure(tmpdir):
+ f = tmpdir.join('f.txt')
+ f.write_text('ohai', encoding='utf-8-sig')
+ assert byte_order_marker.main((str(f),)) == 1
+
+
+def test_success(tmpdir):
+ f = tmpdir.join('f.txt')
+ f.write_text('ohai', encoding='utf-8')
+ assert byte_order_marker.main((str(f),)) == 0
+
+
+def test_failure_nofix(tmpdir):
+ f = tmpdir.join('f.txt')
+ f.write_text('ohai', encoding='utf-8-sig')
+ assert byte_order_marker.main((str(f), '--fix=no')) == 1
+
+
+def test_success_nofix(tmpdir):
+ f = tmpdir.join('f.txt')
+ f.write_text('ohai', encoding='utf-8')
+ assert byte_order_marker.main((str(f), '--fix=no')) == 0