Python脚本在Windows Spyder正常运行，Linux下遇Unicode解码错误

安全阿岳

2026-5-12

Fixing UnicodeDecodeError When Reading .cnf Files on Linux

Hey there! Let's break down why this error is happening and get your script running smoothly on Linux.

What's the Root Cause?

The UnicodeDecodeError pops up because your config.cnf file isn't saved in UTF-8 encoding.

On Windows, Spyder automatically uses your system's default encoding (like cp1252, a common Windows text encoding) to read the file. That’s why it handles the 0x96 byte (which is an en dash character in cp1252) without issue. But on Linux, Python 3 defaults to UTF-8 for text files—and since 0x96 isn’t a valid UTF-8 byte, it throws an error.

Step-by-Step Fixes

1. First, Find the File's Actual Encoding

Let’s figure out what encoding your config.cnf uses. On Linux, run this terminal command in the same directory as your file:

file -i config.cnf

You’ll get output like config.cnf: text/plain; charset=iso-8859-1 or charset=cp1252 — that string after charset= is the encoding you need.

Alternatively, use Python’s chardet library to detect the encoding:
First install it:

pip install chardet

Then run this small detection script:

import chardet

with open('config.cnf', 'rb') as f:
    raw_data = f.read()
    encoding_result = chardet.detect(raw_data)

print(f"Detected encoding: {encoding_result['encoding']}")

2. Open the File with the Correct Encoding

Once you know the right encoding, update your file-opening code to specify it explicitly. For example, if the detected encoding is cp1252, modify your code like this:

with open('config.cnf', encoding='cp1252') as f:
    file_content = f.read()

Replace cp1252 with whatever encoding you found in step 1. This tells Python exactly how to interpret the bytes in the file, eliminating the decode error.

3. Temporary Workaround (If You Can’t Detect the Encoding)

If you just need to get the script running quickly and don’t mind losing a few non-ASCII characters, you can tell Python to ignore or replace invalid bytes:

# Ignore invalid bytes entirely
with open('config.cnf', encoding='utf-8', errors='ignore') as f:
    file_content = f.read()

# Or replace invalid bytes with a placeholder (�)
with open('config.cnf', encoding='utf-8', errors='replace') as f:
    file_content = f.read()

Note: This is a last resort—using the correct encoding is always better to preserve all your file’s content.

Why Windows Works but Linux Doesn’t

Windows and Linux use different default text encodings. Windows typically uses cp1252 or region-specific encodings like gbk, while Linux defaults to UTF-8. Spyder on Windows adapts to the system’s default encoding automatically, but when running the script directly on Linux, Python sticks to UTF-8. By specifying the correct encoding explicitly, you make the script consistent across both systems.

内容的提问来源于stack exchange，提问作者Joey