在Python 2和3中获得相同的Unicode字符串长度？

Question

Uhh, Python 2 / 3 is so frustrating... Consider this example, test.py :

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import sys
if sys.version_info[0] < 3:
  text_type = unicode
  binary_type = str
  def b(x):
    return x
  def u(x):
    return unicode(x, "utf-8")
else:
  text_type = str
  binary_type = bytes
  import codecs
  def b(x):
    return codecs.latin_1_encode(x)[0]
  def u(x):
    return x

tstr = " ▲ "

sys.stderr.write(tstr)
sys.stderr.write("\n")
sys.stderr.write(str(len(tstr)))
sys.stderr.write("\n")

Running it:

$ python2.7 test.py 
 ▲ 
5
$ python3.2 test.py 
 ▲ 
3

Great, I get two differing string sizes. Hopefully wrapping the string in one of these wrappers I found around the net will help?

For tstr = text_type(" ▲ ") :

$ python2.7 test.py 
Traceback (most recent call last):
  File "test.py", line 21, in <module>
    tstr = text_type(" ▲ ")
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 1: ordinal not in range(128)
$ python3.2 test.py 
 ▲ 
3

For tstr = u(" ▲ ") :

$ python2.7 test.py 
Traceback (most recent call last):
  File "test.py", line 21, in <module>
    tstr = u(" ▲ ")
  File "test.py", line 11, in u
    return unicode(x)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 1: ordinal not in range(128)
$ python3.2 test.py 
 ▲ 
3

For tstr = b(" ▲ ") :

$ python2.7 test.py 
 ▲ 
5
$ python3.2 test.py 
Traceback (most recent call last):
  File "test.py", line 21, in <module>
    tstr = b(" ▲ ")
  File "test.py", line 17, in b
    return codecs.latin_1_encode(x)[0]
UnicodeEncodeError: 'latin-1' codec can't encode character '\u25b2' in position 1: ordinal not in range(256)

For tstr = binary_type(" ▲ ") :

$ python2.7 test.py 
 ▲ 
5
$ python3.2 test.py 
Traceback (most recent call last):
  File "test.py", line 21, in <module>
    tstr = binary_type(" ▲ ")
TypeError: string argument without an encoding

Well, that certainly makes things easy.

So, how to get the same string length (in this case, 3) in both Python 2.7 and 3.2?

Answer 1

Well, turns out unicode() in Python 2.7 has an encoding argument, and that apparently helps:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import sys
if sys.version_info[0] < 3:
  text_type = unicode
  binary_type = str
  def b(x):
    return x
  def u(x):
    return unicode(x, "utf-8")
else:
  text_type = str
  binary_type = bytes
  import codecs
  def b(x):
    return codecs.latin_1_encode(x)[0]
  def u(x):
    return x

tstr = u(" ▲ ")

sys.stderr.write(tstr)
sys.stderr.write("\n")
sys.stderr.write(str(len(tstr)))
sys.stderr.write("\n")

Running this, I get what I needed:

$ python2.7 test.py 
 ▲ 
3
$ python3.2 test.py 
 ▲ 
3