mirror of
https://github.com/python/cpython.git
synced 2025-11-07 00:51:49 +00:00
84ad84e0bb
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3 ........ r72491 | benjamin.peterson | 2009-05-08 19:33:27 -0500 (Fri, 08 May 2009) | 7 lines make 2to3 use unicode internally on 2.x This started out as a fix for #2660, but became this large refactoring when I realized the dire state this was in. 2to3 now uses tokenize.detect_encoding to decode the files correctly into unicode. ........ r72492 | benjamin.peterson | 2009-05-08 19:35:38 -0500 (Fri, 08 May 2009) | 1 line remove compat code ........ r72493 | benjamin.peterson | 2009-05-08 19:54:15 -0500 (Fri, 08 May 2009) | 1 line add a test for \r\n newlines ........
28 lines
837 B
Python
28 lines
837 B
Python
"""Fixer that changes unicode to str, unichr to chr, and u"..." into "...".
|
|
|
|
"""
|
|
|
|
import re
|
|
from ..pgen2 import token
|
|
from .. import fixer_base
|
|
|
|
class FixUnicode(fixer_base.BaseFix):
|
|
|
|
PATTERN = "STRING | NAME<'unicode' | 'unichr'>"
|
|
|
|
def transform(self, node, results):
|
|
if node.type == token.NAME:
|
|
if node.value == u"unicode":
|
|
new = node.clone()
|
|
new.value = u"str"
|
|
return new
|
|
if node.value == u"unichr":
|
|
new = node.clone()
|
|
new.value = u"chr"
|
|
return new
|
|
# XXX Warn when __unicode__ found?
|
|
elif node.type == token.STRING:
|
|
if re.match(ur"[uU][rR]?[\'\"]", node.value):
|
|
new = node.clone()
|
|
new.value = new.value[1:]
|
|
return new
|