Port, board and/or hardware

Windows (any hardware) - affects mpremote tool when run on Windows with legacy console (cp1252 or similar encoding).

MicroPython version

• mpremote 1.27.0
• Python 3.11.9 /3.13.1 (host)
• Tested against MicroPython 1.27.0 RP2, ESP32

Reproduction

This issue only occurs with legacy Windows consoles that use cp1252 or similar encodings. Modern terminals (Windows Terminal, VS Code) use UTF-8 by default and are not affected.

To reproduce, you must use a legacy console:

1. Open legacy cmd.exe (not Windows Terminal) or configure PowerShell to use cp1252:

   # Force legacy encoding in Python
   $env:PYTHONIOENCODING = "cp1252"
   

2. Create test files with non-ASCII characters:

   mkdir unicode_test
   echo "test" > "unicode_test\Владимир_Петров.txt"
   

3. Use mpremote to copy:

   mpremote cp -rv unicode_test :
   

Alternatively, the issue can be demonstrated with this Python snippet:

import sys
sys.stdout.reconfigure(encoding='cp1252')
print('Владимир_Петров.txt')  # Raises UnicodeEncodeError

Expected behaviour

mpremote cp should successfully copy files with Unicode characters in filenames on Windows. The tool should handle all Unicode characters in console output without raising encoding errors.

CPython's print() should use UTF-8 or properly handle the console encoding.

Observed behaviour

The command fails immediately with:

UnicodeEncodeError: 'charmap' codec can't encode characters in position X-Y: character maps to <undefined>

The file is not copied.

Additional Information

Workaround

Set the PYTHONIOENCODING environment variable before running mpremote:

$env:PYTHONIOENCODING = "utf-8"
mpremote connect COM3 cp -r . :

Or add to PowerShell $PROFILE for persistence:

$env:PYTHONIOENCODING = "utf-8"
$env:PYTHONUTF8 = "1"

Suggested Fix

Force UTF-8 encoding in mpremote on Windows:

import sys
import io

if sys.platform == 'win32':
    sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8', errors='replace')
    sys.stderr = io.TextIOWrapper(sys.stderr.buffer, encoding='utf-8', errors='replace')

This would be a minimal change in the mpremote initialization code.

Code of Conduct

Yes, I agree