closes bpo-34515: Support non-ASCII identifiers in lib2to3. (GH-8950)

holymonson · benjaminp · commit 10a428b64b3f · 2018-09-15T10:32:29.000-07:00
diff --git a/Lib/lib2to3/pgen2/tokenize.py b/Lib/lib2to3/pgen2/tokenize.py
@@ -56,7 +56,7 @@ def _combinations(*l):
 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 @@ def _combinations(*l):
 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 @@ def generate_tokens(readline):
     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 @@ def generate_tokens(readline):
                 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 @@ def generate_tokens(readline):
                             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,
diff --git a/Lib/lib2to3/tests/test_parser.py b/Lib/lib2to3/tests/test_parser.py
@@ -529,6 +529,16 @@ def test_4(self):
         self.validate("""x = {2, 3, 4,}""")
 
 
+# Adapted from Python 3's Lib/test/test_unicode_identifiers.py and
+# Lib/test/test_tokenize.py:TokenizeTest.test_non_ascii_identifiers
+class TestIdentfier(GrammarTest):
+    def test_non_ascii_identifiers(self):
+        self.validate("Örter = 'places'\ngrün = 'green'")
+        self.validate("蟒 = a蟒 = 锦蛇 = 1")
+        self.validate("µ = aµ = µµ = 1")
+        self.validate("𝔘𝔫𝔦𝔠𝔬𝔡𝔢 = a_𝔘𝔫𝔦𝔠𝔬𝔡𝔢 = 1")
+
+
 class TestNumericLiterals(GrammarTest):
     def test_new_octal_notation(self):
         self.validate("""0o7777777777777""")
diff --git a/Misc/NEWS.d/next/Library/2018-08-27-16-01-22.bpo-34515.S0Irst.rst b/Misc/NEWS.d/next/Library/2018-08-27-16-01-22.bpo-34515.S0Irst.rst
@@ -0,0 +1 @@
+Fix parsing non-ASCII identifiers in :mod:`lib2to3.pgen2.tokenize` (PEP 3131).

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1 @@`
	`1`	+Fix parsing non-ASCII identifiers in :mod:`lib2to3.pgen2.tokenize` (PEP 3131).