python23-compatibility.md

Overview

PEDA supports Python 2 and Python 3 using the six library. To make sure code runs on both Python 2 and Python 3, make sure to keep the following in mind.

Division

For integer division, use the // operator instead of /. In Python 3, the / operator returns a float.

In Python 3:

>>> 5 / 2
2.5
>>> type(5 / 2)
<class 'float'>

Type checking

To check if something is a string:

isinstance(obj, six.string_types)

To check if something is an integer type:

isinstance(x, six.integer_types)

Strings

In Python 2, bytes is an alias for str. In Python 3, str is a unicode type and bytes is used for a sequece of arbitrary bytes. Use a leading 'b' to signify that a string is a bytes object.

>>> 'Normal string'
'Normal string'
>>> b'arbitrary bytes \x90\x90'
b'arbitrary bytes \x90\x90'

To convert between str to bytes:

>>> 'hi there'.encode('utf-8')
b'hi there'
>>> b'some string'.decode('utf-8')
'some string'

Do not mix bytes and str with each other with basic string functions. The following is okay:

>>> "abc".replace("a", "f")
'fbc'
>>> b"abc".replace(b"a", b"f")
b'fbc'

Mixing types in Python 3 will throw an exception:

>>> b"abc".replace("a", "f")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: expected bytes, bytearray or buffer compatible object

In Python 2, indexing into a str returns a str of length 1. In Python 3, indexing into a bytes returns an int. This causes a problem when iterating. To solve this, use the bytes_iterator from utils.py.

# In Python 2:
>>> s = b'hello'
>>> s
'hello'
>>> s[0]
'h'

# In Python 3:
>>> s = b'hello'
>>> s
b'hello'
>>> s[0]
104

# Solution:
>>> for c in bytes_iterator(b'hi'): print(c)
... 
b'h'
b'i'

Encodings

Encode (and decode) strings into hex:

>>> codecs.encode(b'abcdef', 'hex')
b'616263646566'
>>> codecs.decode('616263646566', 'hex')
b'abcdef'