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

pre_commit_hooks/byte_order_marker.py

@@ -0,0 +1,38 @@
+import os, sys, codecs
+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:
+ with open(filename, 'rb') as f:
+ if f.read(3) == b'\xef\xbb\xbf':
+ if args.fix == 'no':
+ print(f'{filename}: Found byte-order marker')
+ else:
+ with open(..., newline='', encoding='utf-8-sig') as f:
+ contents = f.read()
+
+ with open(..., newline='', encoding='utf-8') as f:
+ f.write(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