closes bpo-34515: Support non-ASCII identifiers in lib2to3. (GH-8950)
diff --git a/Lib/lib2to3/pgen2/tokenize.py b/Lib/lib2to3/pgen2/tokenize.py
index bf83ab8..c07b34f 100644
--- a/Lib/lib2to3/pgen2/tokenize.py
+++ b/Lib/lib2to3/pgen2/tokenize.py
@@ -56,7 +56,7 @@
Whitespace = r'[ \f\t]*'
Comment = r'#[^\r\n]*'
Ignore = Whitespace + any(r'\\\r?\n' + Whitespace) + maybe(Comment)
-Name = r'[a-zA-Z_]\w*'
+Name = r'\w+'
Binnumber = r'0[bB]_?[01]+(?:_[01]+)*'
Hexnumber = r'0[xX]_?[\da-fA-F]+(?:_[\da-fA-F]+)*[lL]?'
@@ -107,8 +107,8 @@
PseudoExtras = group(r'\\\r?\n', Comment, Triple)
PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name)
-tokenprog, pseudoprog, single3prog, double3prog = list(map(
- re.compile, (Token, PseudoToken, Single3, Double3)))
+tokenprog, pseudoprog, single3prog, double3prog = map(
+ re.compile, (Token, PseudoToken, Single3, Double3))
_strprefixes = (
_combinations('r', 'R', 'f', 'F') |
@@ -349,7 +349,6 @@
logical line; continuation lines are included.
"""
lnum = parenlev = continued = 0
- namechars, numchars = string.ascii_letters + '_', '0123456789'
contstr, needcont = '', 0
contline = None
indents = [0]
@@ -451,7 +450,7 @@
spos, epos, pos = (lnum, start), (lnum, end), end
token, initial = line[start:end], line[start]
- if initial in numchars or \
+ if initial in string.digits or \
(initial == '.' and token != '.'): # ordinary number
yield (NUMBER, token, spos, epos, line)
elif initial in '\r\n':
@@ -501,7 +500,7 @@
yield stashed
stashed = None
yield (STRING, token, spos, epos, line)
- elif initial in namechars: # ordinary name
+ elif initial.isidentifier(): # ordinary name
if token in ('async', 'await'):
if async_def:
yield (ASYNC if token == 'async' else AWAIT,