Feat: assume UTF-8 encoding for commit messages (#51)

Author: thekaveman

README.md

@@ -17,8 +17,7 @@ Add a new repo entry to your configuration file:
 
 ```yaml
 repos:
-
-# - repo: ...
+  # - repo: ...
 
   - repo: https://github.com/compilerla/conventional-pre-commit
     rev: <git sha or tag>
@@ -93,9 +92,9 @@ Then run the command line script:
 conventional-pre-commit [types] input
 ```
 
-Where `[types]` is an optional list of Conventional Commit types to allow (e.g. `feat fix chore`)
+- `[types]` is an optional list of Conventional Commit types to allow (e.g. `feat fix chore`)
 
-And `input` is a file containing the commit message to check:
+- `input` is a file containing the commit message to check:
 
 ```shell
 conventional-pre-commit feat fix chore ci test .git/COMMIT_MSG

conventional_pre_commit/hook.py

@@ -29,8 +29,20 @@ def main(argv=[]):
     except SystemExit:
         return RESULT_FAIL
 
-    with open(args.input) as f:
-        message = f.read()
+    try:
+        with open(args.input, encoding="utf-8") as f:
+            message = f.read()
+    except UnicodeDecodeError:
+        print(
+            f"""
+{Colors.LRED}[Bad Commit message encoding] {Colors.RESTORE}
+
+{Colors.YELLOW}conventional-pre-commit couldn't decode your commit message.{Colors.RESTORE}
+{Colors.YELLOW}UTF-8{Colors.RESTORE} encoding is assumed, please configure git to write commit messages in UTF-8.
+See {Colors.LBLUE}https://git-scm.com/docs/git-commit/#_discussion{Colors.RESTORE} for more.
+        """
+        )
+        return RESULT_FAIL
 
     if format.is_conventional(message, args.types):
         return RESULT_SUCCESS

tests/conftest.py

@@ -22,3 +22,13 @@ def conventional_commit_path():
 @pytest.fixture
 def custom_commit_path():
     return get_message_path("custom_commit")
+
+
+@pytest.fixture
+def conventional_utf8_commit_path():
+    return get_message_path("conventional_commit_utf-8")
+
+
+@pytest.fixture
+def conventional_gbk_commit_path():
+    return get_message_path("conventional_commit_gbk")

tests/messages/conventional_commit_gbk

@@ -0,0 +1 @@
+feat: utf-8 test ����

tests/messages/conventional_commit_utf-8

@@ -0,0 +1 @@
+feat: utf-8 test 测试

tests/test_hook.py

@@ -80,3 +80,15 @@ def test_subprocess_success__custom_conventional(cmd, conventional_commit_path):
     result = subprocess.call((cmd, "custom", conventional_commit_path))
 
     assert result == RESULT_SUCCESS
+
+
+def test_main_success__conventional_utf8(conventional_utf8_commit_path):
+    result = main([conventional_utf8_commit_path])
+
+    assert result == RESULT_SUCCESS
+
+
+def test_main_fail__conventional_gbk(conventional_gbk_commit_path):
+    result = main([conventional_gbk_commit_path])
+
+    assert result == RESULT_FAIL