Make rb'' strings work in lib2to3 (#1724)

This partially solves bpo-23894.

diff --git a/Lib/lib2to3/pgen2/tokenize.py b/Lib/lib2to3/pgen2/tokenize.py
index fba0fa2c00dccd9c3d99a2cc87b4203fba6f8664..9a0cc1ea4086c7120b02b7765ba67e07aa155f34 100644 (file)
--- a/Lib/lib2to3/pgen2/tokenize.py
+++ b/Lib/lib2to3/pgen2/tokenize.py
@@ -74,10 +74,11 @@ Double = r'[^"\\]*(?:\\.[^"\\]*)*"'
 Single3 = r"[^'\\]*(?:(?:\\.|'(?!''))[^'\\]*)*'''"
 # Tail end of """ string.
 Double3 = r'[^"\\]*(?:(?:\\.|"(?!""))[^"\\]*)*"""'
-Triple = group("[ubUB]?[rR]?'''", '[ubUB]?[rR]?"""')
+_litprefix = r"(?:[uUrRbB]|[rR][bB]|[bBuU][rR])?"
+Triple = group(_litprefix + "'''", _litprefix + '"""')
 # Single-line ' or " string.
-String = group(r"[uU]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*'",
-               r'[uU]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*"')
+String = group(_litprefix + r"'[^\n'\\]*(?:\\.[^\n'\\]*)*'",
+               _litprefix + r'"[^\n"\\]*(?:\\.[^\n"\\]*)*"')
 
 # Because of leftmost-then-longest match semantics, be sure to put the
 # longest operators first (e.g., if = came before ==, == would get
@@ -95,9 +96,9 @@ PlainToken = group(Number, Funny, String, Name)
 Token = Ignore + PlainToken
 
 # First (or only) line of ' or " string.
-ContStr = group(r"[uUbB]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*" +
+ContStr = group(_litprefix + r"'[^\n'\\]*(?:\\.[^\n'\\]*)*" +
                 group("'", r'\\\r?\n'),
-                r'[uUbB]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*' +
+                _litprefix + r'"[^\n"\\]*(?:\\.[^\n"\\]*)*' +
                 group('"', r'\\\r?\n'))
 PseudoExtras = group(r'\\\r?\n', Comment, Triple)
 PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name)
@@ -111,6 +112,7 @@ endprogs = {"'": re.compile(Single), '"': re.compile(Double),
             "b'''": single3prog, 'b"""': double3prog,
             "ur'''": single3prog, 'ur"""': double3prog,
             "br'''": single3prog, 'br"""': double3prog,
+            "rb'''": single3prog, 'rb"""': double3prog,
             "R'''": single3prog, 'R"""': double3prog,
             "U'''": single3prog, 'U"""': double3prog,
             "B'''": single3prog, 'B"""': double3prog,
@@ -120,6 +122,9 @@ endprogs = {"'": re.compile(Single), '"': re.compile(Double),
             "bR'''": single3prog, 'bR"""': double3prog,
             "Br'''": single3prog, 'Br"""': double3prog,
             "BR'''": single3prog, 'BR"""': double3prog,
+            "rB'''": single3prog, 'rB"""': double3prog,
+            "Rb'''": single3prog, 'Rb"""': double3prog,
+            "RB'''": single3prog, 'RB"""': double3prog,
             'r': None, 'R': None,
             'u': None, 'U': None,
             'b': None, 'B': None}
@@ -132,7 +137,9 @@ for t in ("'''", '"""',
           "ur'''", 'ur"""', "Ur'''", 'Ur"""',
           "uR'''", 'uR"""', "UR'''", 'UR"""',
           "br'''", 'br"""', "Br'''", 'Br"""',
-          "bR'''", 'bR"""', "BR'''", 'BR"""',):
+          "bR'''", 'bR"""', "BR'''", 'BR"""',
+          "rb'''", 'rb"""', "Rb'''", 'Rb"""',
+          "rB'''", 'rB"""', "RB'''", 'RB"""',):
     triple_quoted[t] = t
 single_quoted = {}
 for t in ("'", '"',
@@ -142,7 +149,9 @@ for t in ("'", '"',
           "ur'", 'ur"', "Ur'", 'Ur"',
           "uR'", 'uR"', "UR'", 'UR"',
           "br'", 'br"', "Br'", 'Br"',
-          "bR'", 'bR"', "BR'", 'BR"', ):
+          "bR'", 'bR"', "BR'", 'BR"',
+          "rb'", 'rb"', "Rb'", 'Rb"',
+          "rB'", 'rB"', "RB'", 'RB"',):
     single_quoted[t] = t
 
 tabsize = 8

diff --git a/Lib/lib2to3/tests/test_parser.py b/Lib/lib2to3/tests/test_parser.py
index 9a969e8ec047419f80d888a189a5dd7d35c1eeb4..c79611db56522a6bced8f6376798c9d42c230e55 100644 (file)
--- a/Lib/lib2to3/tests/test_parser.py
+++ b/Lib/lib2to3/tests/test_parser.py
@@ -320,6 +320,7 @@ class TestVarAnnotations(GrammarTest):
     def test_6(self):
         self.validate("lst: List[int] = []")
 
+
 class TestExcept(GrammarTest):
     def test_new(self):
         s = """
@@ -338,6 +339,26 @@ class TestExcept(GrammarTest):
         self.validate(s)
 
 
+class TestStringLiterals(GrammarTest):
+    prefixes = ("'", '"',
+        "r'", 'r"', "R'", 'R"',
+        "u'", 'u"', "U'", 'U"',
+        "b'", 'b"', "B'", 'B"',
+        "ur'", 'ur"', "Ur'", 'Ur"',
+        "uR'", 'uR"', "UR'", 'UR"',
+        "br'", 'br"', "Br'", 'Br"',
+        "bR'", 'bR"', "BR'", 'BR"',
+        "rb'", 'rb"', "Rb'", 'Rb"',
+        "rB'", 'rB"', "RB'", 'RB"',)
+
+    def test_lit(self):
+        for pre in self.prefixes:
+            single = "{p}spamspamspam{s}".format(p=pre, s=pre[-1])
+            self.validate(single)
+            triple = "{p}{s}{s}eggs{s}{s}{s}".format(p=pre, s=pre[-1])
+            self.validate(triple)
+
+
 # Adapted from Python 3's Lib/test/test_grammar.py:GrammarTests.testAtoms
 class TestSetLiteral(GrammarTest):
     def test_1(self):