VScode的Python调试控制台无法正确打印Unicode中文字符

Question

Image to show the problem Here is the code to illustrate the problem:

# -*- coding:utf-8 -*-
text = u"严"
print text

If I run the code above in VSCode debug, it will prints "涓" instead of "严", which is the result of the first 2 byte (\\xe4\\xb8) of u"严" in UTF-8 (\\xe4\\xb8\\xa5), decoded in gbk codec. \\xe4\\xb8 in gbk is "涓".

However if I run the same code in pycharm it prints "严" exactly as I expected. And it is the same If I run the code in powershell.

Wired the VSCode python debugger behaves different with python interpreter. How can I get the print result correct, I do not think add a decode("gbk") in the end of every text would be a good idea.

My Environment data

• VS Code version: 1.21
• VSCode Python Extension version : 2018.2.1
• OS and version: Windows 10
• Python version : 2.7.14
• Type of virtual environment used : No

Answer 1

1. For Windows users, in your System Variables, add PYTHONIOENCODING Variables,change its value to UTF-8 , then restart vscode, this worked on my pc.

2. Modify task.json file in vscode, I am not sure if it will still work on version 2.0. You can find it here: Changing the encoding for a task output or here in github: Tasks should support specifying the output encoding

3. add this before you start a py script: import io import sys sys.stdout = io.TextIOWrapper(sys.stdout.buffer,encoding='utf8')

Answer 2

If you open your python file in VS2017 you can do the following:

Go to:

1. File->
2. Save selected item as ->

3. click on the down-arrow next to "Save button"

4. clicking "Save With Encoding...

5. select the type of coding you need...
6. if .py already saved then overwrite file > select "yes"

select for example : "Chinese Simplified (GB18030) - Codepage 54936"

Also, add the following on line 2 of your .py file:

# -*- coding: gb18030 -*- or # -*- coding: gb2312 -*-

Those encodings accept your 严 character.

Nice link to endocoder/decoder tester here .