fix(rename): read and write source files as UTF-8

news/6714.bugfix.md

@@ -0,0 +1 @@
+Fixed `reflex rename` corrupting or failing on UTF-8 source files on non-UTF-8 platform locales (e.g. Windows cp1252) by reading and writing files as UTF-8.

reflex/utils/rename.py

@@ -2,6 +2,7 @@
 
 import re
 import sys
+import tokenize
 from pathlib import Path
 
 from reflex_base import constants
@@ -97,7 +98,12 @@ def rename_imports_and_app_name(file_path: str | Path, old_name: str, new_name:
         new_name: The new name to use.
     """
     file_path = Path(file_path)
-    content = file_path.read_text()
+    if file_path.suffix == constants.Ext.PY:
+        with file_path.open("rb") as source:
+            encoding = tokenize.detect_encoding(source.readline)[0]
+    else:
+        encoding = "utf-8"
+    content = file_path.read_text(encoding=encoding)
 
     # Replace `from old_name.` or `from old_name` with `from new_name`
     content = re.sub(
@@ -127,7 +133,7 @@ def rename_imports_and_app_name(file_path: str | Path, old_name: str, new_name:
         content,
     )
 
-    file_path.write_text(content)
+    file_path.write_text(content, encoding=encoding)
 
 
 def process_directory(

tests/units/test_prerequisites.py

@@ -1,4 +1,5 @@
 import json
+import os
 import shutil
 import tempfile
 import uuid
@@ -1474,6 +1475,69 @@ def test_regex_edge_cases(temp_directory):
     assert updated_content == expected_content
 
 
+def test_rename_imports_and_app_name_preserves_utf8(
+    temp_directory, monkeypatch: pytest.MonkeyPatch
+):
+    """UTF-8 source is preserved even when the platform default encoding is not UTF-8.
+
+    Simulates a Western Windows locale (cp1252) where ``Path.read_text`` /
+    ``write_text`` without an explicit ``encoding`` would mis-decode UTF-8 source,
+    silently corrupting or aborting on non-ASCII characters (curly quotes, accents).
+
+    Args:
+        temp_directory: A temporary directory fixture.
+        monkeypatch: The pytest monkeypatch fixture.
+    """
+    original_read_text = Path.read_text
+    original_write_text = Path.write_text
+
+    def read_text(self, encoding=None, errors=None):
+        return original_read_text(
+            self, encoding=encoding if encoding is not None else "cp1252", errors=errors
+        )
+
+    def write_text(self, data, encoding=None, errors=None, newline=None):
+        return original_write_text(
+            self,
+            data,
+            encoding=encoding if encoding is not None else "cp1252",
+            errors=errors,
+            newline=newline,
+        )
+
+    monkeypatch.setattr(Path, "read_text", read_text)
+    monkeypatch.setattr(Path, "write_text", write_text)
+
+    source = "import old_name  # \u201cquoted\u201d caf\u00e9 na\u00efve r\u00e9sum\u00e9\n"  # codespell:ignore
+    file_path = temp_directory / "example.py"
+    file_path.write_bytes(source.encode("utf-8"))
+
+    rename_imports_and_app_name(file_path, "old_name", "new_name")
+
+    expected = "import new_name  # \u201cquoted\u201d caf\u00e9 na\u00efve r\u00e9sum\u00e9\n"  # codespell:ignore
+    assert file_path.read_bytes() == expected.replace("\n", os.linesep).encode("utf-8")
+
+
+def test_rename_imports_and_app_name_preserves_declared_encoding(temp_directory):
+    """Python source encoding cookies are honored during rename.
+
+    Args:
+        temp_directory: A temporary directory fixture.
+    """
+    source = (
+        "# -*- coding: cp1252 -*-\nimport old_name  # caf\u00e9\n"  # codespell:ignore
+    )
+    file_path = temp_directory / "example.py"
+    file_path.write_bytes(source.encode("cp1252"))
+
+    rename_imports_and_app_name(file_path, "old_name", "new_name")
+
+    expected = (
+        "# -*- coding: cp1252 -*-\nimport new_name  # caf\u00e9\n"  # codespell:ignore
+    )
+    assert file_path.read_bytes() == expected.replace("\n", os.linesep).encode("cp1252")
+
+
 def test_cli_rename_command(temp_directory):
     foo_dir = temp_directory / "foo"
     foo_dir.mkdir()