Merge pull request #14 from yuxiaoli/fix/utf8-encoding

fix: use utf-8 encoding for subprocess to handle non-ascii chars

Author: tomice

git_py_stats/git_operations.py

@@ -20,8 +20,17 @@ def run_git_command(cmd: List[str]) -> Optional[str]:
         print("Error: Command list is empty!")
         return None
     try:
+        # We enforce utf-8 encoding and use surrogateescape to handle non-ascii characters
+        # that may be present in git history (e.g. in author names or commit messages)
+        # as recommended by PEP 383 for system interfaces.
         result = subprocess.run(
-            cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, check=True
+            cmd,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            check=True,
+            encoding="utf-8",
+            errors="surrogateescape",
         )
         return result.stdout.strip()
     except subprocess.CalledProcessError as e:

git_py_stats/tests/test_git_operations.py

@@ -31,6 +31,8 @@ def test_run_git_command_success(self, mock_subprocess_run):
             stderr=subprocess.PIPE,
             text=True,
             check=True,
+            encoding="utf-8",
+            errors="surrogateescape",
         )
 
     @patch("subprocess.run")
@@ -54,6 +56,8 @@ def test_run_git_command_failure(self, mock_subprocess_run):
             stderr=subprocess.PIPE,
             text=True,
             check=True,
+            encoding="utf-8",
+            errors="surrogateescape",
         )
 
     @patch("subprocess.run")
@@ -72,7 +76,13 @@ def test_run_git_command_no_output(self, mock_subprocess_run):
         self.assertEqual(output, "")
 
         mock_subprocess_run.assert_called_once_with(
-            ["git", "status"], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, check=True
+            ["git", "status"],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            check=True,
+            encoding="utf-8",
+            errors="surrogateescape",
         )
 
     @patch("subprocess.run")
@@ -87,7 +97,13 @@ def test_run_git_command_exception(self, mock_subprocess_run):
         self.assertIsNone(output)
 
         mock_subprocess_run.assert_called_once_with(
-            ["git", "status"], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, check=True
+            ["git", "status"],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            check=True,
+            encoding="utf-8",
+            errors="surrogateescape",
         )
 
     def test_run_git_command_empty_command(self):