Python3在打印十六进制值时添加了额外的字节

Question

I have run into a strange difference between Python2 and Python3. Printing the same list of characters yields an extra byte C2 when printed with Python3. I would have expected the same behaviour. Python2 behaves as I expected. What am I missing here?

$ python3 -c "print('\x30\xA0\x04\x08')" | xxd
0000000: 30c2 a004 080a     
$ python2 -c "print('\x30\xA0\x04\x08')" | xxd
0000000: 30a0 0408 0a  

Answer 1

Python 3 strings are unicode, and on your platform unicode is printed using UTF-8 encoding. The UTF-8 encoding for unicode character U+00A0 is 0xC2 0xA0, which is what you see.

Python 2 strings are bytestrings, so they are output exactly.

Answer 2

In Python 3 all string literals are unicode.

\\A0 converted to UTF-8 is a no-break space :

U+00A0 no-break space (HTML &#160 ; ·   ) Can be encoded in UTF-8 as C2 A0

Try this:

$ python3 -c "import sys; sys.stdout.buffer.write(b'\x30\xA0\x04\x08')" | xxd
0000000: 30a0 0408                                0...