QUERY · ISSUE

mpremote cat hangs on Windows interactive console with Unicode content

openby Josverlopened 2026-01-07updated 2026-01-09

bugtoolsunicode

Port, board and/or hardware

Host Windows (any hardware)

MicroPython version

mpremote 1.27.0

Reproduction

On Windows, have a MicroPython device with a file containing Unicode text:
```
/unicode_test/नेपाली_नाम.txt
```
Run mpremote cat interactively in PowerShell (NOT piped or captured):
```
mpremote cat :/unicode_test/नेपाली_नाम.txt
```
The command hangs indefinitely. Even with PYTHONIOENCODING=utf-8 set, it still hangs.
Press Ctrl-C to interrupt.

Observations

The same command works correctly when:

Output is piped: mpremote cat :file.txt > output.txt
Output is captured by subprocess: subprocess.run([...], capture_output=True)
Running on Linux (even in WSL2 on the same machine)

The hang ONLY occurs when stdout is a real Windows console (isatty=True).

Expected behaviour

$ mpremote cat :/unicode_test/नेपाली_नाम.txt

Nepali Name - नेपाली नाम

This file tests Nepali (Devanagari) script.

Script: Devanagari (Nepali variant)

Vowels: अ आ इ ई उ ऊ ऋ ए ऐ ओ औ
Consonants: क ख ग घ ङ च छ ज झ ञ ट ठ ड ढ ण त थ द ध न
...



for test files see : [https://github.com/[Josverl/unicode_mpy](https://github.com/Josverl/unicode_mpy)

https://github.com/Josverl/unicode_mpy/tree/main/test_data/South_Asian_Indic


### Observed behaviour

The command hangs indefinitely when run interactively on Windows with Unicode content. Ctrl-C shows the following traceback:

Traceback (most recent call last):
File "...\mpremote\main.py", line 614, in main
handler_func(state, args)
File "...\mpremote\commands.py", line 421, in do_filesystem
state.transport.fs_printfile(path)
File "...\mpremote\transport.py", line 129, in fs_printfile
self.exec(cmd, data_consumer=stdout_write_bytes)
File "...\mpremote\transport_serial.py", line 309, in exec
ret, ret_err = self.exec_raw(command, data_consumer=data_consumer)
File "...\mpremote\transport_serial.py", line 296, in exec_raw
return self.follow(timeout, data_consumer)
File "...\mpremote\transport_serial.py", line 204, in follow
data = self.read_until(1, b"\x04", timeout=timeout, data_consumer=data_consumer)
File "...\mpremote\transport_serial.py", line 146, in read_until
data_consumer(new_data)
File "...\mpremote\transport.py", line 36, in stdout_write_bytes
sys.stdout.buffer.flush()
KeyboardInterrupt


### Additional Information


**Initial hypothesis (DISPROVEN):** The hang was suspected to be in `stdout_write_bytes()`:
```python
def stdout_write_bytes(b):
    sys.stdout.buffer.write(b)
    sys.stdout.buffer.flush()  # <-- Suspected to hang

Test result: A standalone Python script calling sys.stdout.buffer.write() + flush() with identical Unicode content does NOT hang.

Actual cause: The hang is NOT in CPython's stdout handling. The issue is elsewhere in mpremote, possibly:

Different code path when stdout.isatty() is True vs False
Threading/synchronization issues between reader/writer threads
Windows console API interaction issues
Serial/socket transport blocking behavior

Possibly Related Issues

https://github.com/micropython/micropython/issues/15228 - Unable to print unicode characters when running repl with mpremote

Code of Conduct

Yes, I agree

CANDIDATE · ISSUE

mpremote fails with UnicodeEncodeError on Windows legacy consoles (cp1252)

openby Josverlopened 2026-01-07updated 2026-01-09

bugtoolsunicode

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:

Open legacy cmd.exe (not Windows Terminal) or configure PowerShell to use cp1252:
```
# Force legacy encoding in Python
$env:PYTHONIOENCODING = "cp1252"
```

Create test files with non-ASCII characters:

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

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