Writing raw bytes to a file in Python3 results in unexpected output

Question

So, I have this piece of code:

f = open("crash.txt", "w")
junk = ("\xCC" * 1028)
f.write(junk)
f.close()

When I run this on Windows(3.5.1), I get a file with repeated "CC"s as hex characters. That is as expected.

However, running this on Linux(Python 3.4.2), I get repeated "c38c"s as hex characters.

I do not understand the output on Linux. Why does this happen and how do I fix it.

Answer 1

You aren't writing raw bytes. By default Python 3 uses Unicode strings, and those strings must be encoded to write them to a file. Also by default, open() uses text mode and the encoding used to encode text is locale.getpreferredencoding(). On US Windows, that is cp1252, but on Linux, it is usually utf8.

b'\xc3\x8c' is '\xcc' encoded in utf8.

b'\xcc' is '\xcc' encoded in cp1252.

Open the file in binary mode and write byte strings instead of Unicode to write "raw" bytes.

with open("crash.txt", "wb") as f:
    junk = b"\xCC" * 1028
    f.write(junk)