Clean up calls to .encode() / .decode()

pre_commit/error_handler.py

@@ -16,7 +16,7 @@ class FatalError(RuntimeError):
 
 
 def _to_bytes(exc: BaseException) -> bytes:
-    return str(exc).encode('UTF-8')
+    return str(exc).encode()
 
 
 def _log_and_exit(msg: str, exc: BaseException, formatted: str) -> None:

pre_commit/five.py

@@ -2,11 +2,11 @@ from typing import Union
 
 
 def to_text(s: Union[str, bytes]) -> str:
-    return s if isinstance(s, str) else s.decode('UTF-8')
+    return s if isinstance(s, str) else s.decode()
 
 
 def to_bytes(s: Union[str, bytes]) -> bytes:
-    return s if isinstance(s, bytes) else s.encode('UTF-8')
+    return s if isinstance(s, bytes) else s.encode()
 
 
 n = to_text

pre_commit/git.py

@@ -69,7 +69,7 @@ def is_in_merge_conflict() -> bool:
 def parse_merge_msg_for_conflicts(merge_msg: bytes) -> List[str]:
     # Conflicted files start with tabs
     return [
-        line.lstrip(b'#').strip().decode('UTF-8')
+        line.lstrip(b'#').strip().decode()
         for line in merge_msg.splitlines()
         # '#\t' for git 2.4.1
         if line.startswith((b'\t', b'#\t'))

pre_commit/languages/fail.py

@@ -18,6 +18,6 @@ def run_hook(
         file_args: Sequence[str],
         color: bool,
 ) -> Tuple[int, bytes]:
-    out = hook.entry.encode('UTF-8') + b'\n\n'
+    out = hook.entry.encode() + b'\n\n'
-    out += b'\n'.join(f.encode('UTF-8') for f in file_args) + b'\n'
+    out += b'\n'.join(f.encode() for f in file_args) + b'\n'
     return 1, out

pre_commit/parse_shebang.py

@@ -9,7 +9,7 @@ from identify.identify import parse_shebang_from_file
 
 class ExecutableNotFoundError(OSError):
     def to_output(self) -> Tuple[int, bytes, None]:
-        return (1, self.args[0].encode('UTF-8'), None)
+        return (1, self.args[0].encode(), None)
 
 
 def parse_filename(filename: str) -> Tuple[str, ...]:

pre_commit/resources/hook-tmpl

@@ -39,7 +39,7 @@ def _norm_exe(exe: str) -> Tuple[str, ...]:
         if f.read(2) != b'#!':
             return ()
         try:
-            first_line = f.readline().decode('UTF-8')
+            first_line = f.readline().decode()
         except UnicodeDecodeError:
             return ()
 
@@ -77,7 +77,7 @@ def _run_legacy() -> Tuple[int, bytes]:
 
 def _validate_config() -> None:
     cmd = ('git', 'rev-parse', '--show-toplevel')
-    top_level = subprocess.check_output(cmd).decode('UTF-8').strip()
+    top_level = subprocess.check_output(cmd).decode().strip()
     cfg = os.path.join(top_level, CONFIG)
     if os.path.isfile(cfg):
         pass
@@ -127,7 +127,7 @@ def _pre_push(stdin: bytes) -> Tuple[str, ...]:
     remote = sys.argv[1]
 
     opts: Tuple[str, ...] = ()
-    for line in stdin.decode('UTF-8').splitlines():
+    for line in stdin.decode().splitlines():
         _, local_sha, _, remote_sha = line.split()
         if local_sha == Z40:
             continue

pre_commit/util.py

@@ -109,13 +109,13 @@ class CalledProcessError(RuntimeError):
             'return code: {}\n'
             'expected return code: {}\n'.format(
                 self.cmd, self.returncode, self.expected_returncode,
-            ).encode('UTF-8'),
+            ).encode(),
             b'stdout:', _indent_or_none(self.stdout), b'\n',
             b'stderr:', _indent_or_none(self.stderr),
         ))
 
     def __str__(self) -> str:
-        return self.__bytes__().decode('UTF-8')
+        return self.__bytes__().decode()
 
 
 def _cmd_kwargs(
@@ -157,8 +157,8 @@ def cmd_output_b(
 
 def cmd_output(*cmd: str, **kwargs: Any) -> Tuple[int, str, Optional[str]]:
     returncode, stdout_b, stderr_b = cmd_output_b(*cmd, **kwargs)
-    stdout = stdout_b.decode('UTF-8') if stdout_b is not None else None
+    stdout = stdout_b.decode() if stdout_b is not None else None
-    stderr = stderr_b.decode('UTF-8') if stderr_b is not None else None
+    stderr = stderr_b.decode() if stderr_b is not None else None
     return returncode, stdout, stderr
 
 

pre_commit/xargs.py

@@ -49,7 +49,6 @@ def _command_length(*cmd: str) -> int:
     # win32 uses the amount of characters, more details at:
     # https://github.com/pre-commit/pre-commit/pull/839
     if sys.platform == 'win32':
-        # the python2.x apis require bytes, we encode as UTF-8
         return len(full_cmd.encode('utf-16le')) // 2
     else:
         return len(full_cmd.encode(sys.getfilesystemencoding()))

testing/resources/arbitrary_bytes_repo/hook.sh

@@ -1,7 +1,7 @@
 #!/usr/bin/env bash
 # Intentionally write mixed encoding to the output.  This should not crash
 # pre-commit and should write bytes to the output.
-# '☃'.encode('UTF-8') + '²'.encode('latin1')
+# '☃'.encode() + '²'.encode('latin1')
 echo -e '\xe2\x98\x83\xb2'
 # exit 1 to trigger printing
 exit 1

tests/conftest.py

@@ -249,7 +249,7 @@ class Fixture:
 
     def get(self):
         """Get the output assuming it was written as UTF-8 bytes"""
-        return self.get_bytes().decode('UTF-8')
+        return self.get_bytes().decode()
 
 
 @pytest.fixture

tests/parse_shebang_test.py

@@ -23,7 +23,7 @@ def test_file_doesnt_exist():
 
 def test_simple_case(tmpdir):
     x = tmpdir.join('f')
-    x.write_text('#!/usr/bin/env echo', encoding='UTF-8')
+    x.write('#!/usr/bin/env echo')
     make_executable(x.strpath)
     assert parse_shebang.parse_filename(x.strpath) == ('echo',)