bpo-40334: use the TOKENS file when checking dangling rules (GH-19849)

5 files changed, 31 insertions, 15 deletions

diff --git a/Tools/peg_generator/pegen/build.py b/Tools/peg_generator/pegen/build.py
index d33dd04..907feea 100644
--- a/Tools/peg_generator/pegen/build.py
+++ b/Tools/peg_generator/pegen/build.py
@@ -17,6 +17,8 @@ from pegen.tokenizer import Tokenizer
 
 MOD_DIR = pathlib.Path(__file__).resolve().parent
 
+TokenDefinitions = Tuple[Dict[int, str], Dict[str, int], Set[str]]
+
 
 def get_extra_flags(compiler_flags: str, compiler_py_flags_nodist: str) -> List[str]:
     flags = sysconfig.get_config_var(compiler_flags)
@@ -112,7 +114,8 @@ def build_parser(
     return grammar, parser, tokenizer
 
 
-def generate_token_definitions(tokens: IO[str]) -> Tuple[Dict[str, int], Set[str]]:
+def generate_token_definitions(tokens: IO[str]) -> TokenDefinitions:
+    all_tokens = {}
     exact_tokens = {}
     non_exact_tokens = set()
     numbers = itertools.count(0)
@@ -129,13 +132,15 @@ def generate_token_definitions(tokens: IO[str]) -> Tuple[Dict[str, int], Set[str
         if len(pieces) == 1:
             (token,) = pieces
             non_exact_tokens.add(token)
+            all_tokens[index] = token
         elif len(pieces) == 2:
-            _, op = pieces
+            token, op = pieces
             exact_tokens[op.strip("'")] = index
+            all_tokens[index] = token
         else:
             raise ValueError(f"Unexpected line found in Tokens file: {line}")
 
-    return exact_tokens, non_exact_tokens
+    return all_tokens, exact_tokens, non_exact_tokens
 
 
 def build_c_generator(
@@ -149,10 +154,10 @@ def build_c_generator(
     skip_actions: bool = False,
 ) -> ParserGenerator:
     with open(tokens_file, "r") as tok_file:
-        exact_tok, non_exact_tok = generate_token_definitions(tok_file)
+        all_tokens, exact_tok, non_exact_tok = generate_token_definitions(tok_file)
     with open(output_file, "w") as file:
         gen: ParserGenerator = CParserGenerator(
-            grammar, exact_tok, non_exact_tok, file, skip_actions=skip_actions
+            grammar, all_tokens, exact_tok, non_exact_tok, file, skip_actions=skip_actions
         )
         gen.generate(grammar_file)
 
diff --git a/Tools/peg_generator/pegen/c_generator.py b/Tools/peg_generator/pegen/c_generator.py
index 6c77f43..c9c6706 100644
--- a/Tools/peg_generator/pegen/c_generator.py
+++ b/Tools/peg_generator/pegen/c_generator.py
@@ -265,13 +265,14 @@ class CParserGenerator(ParserGenerator, GrammarVisitor):
     def __init__(
         self,
         grammar: grammar.Grammar,
+        tokens: Dict[int, str],
         exact_tokens: Dict[str, int],
         non_exact_tokens: Set[str],
         file: Optional[IO[Text]],
         debug: bool = False,
         skip_actions: bool = False,
     ):
-        super().__init__(grammar, file)
+        super().__init__(grammar, tokens, file)
         self.callmakervisitor: CCallMakerVisitor = CCallMakerVisitor(
             self, exact_tokens, non_exact_tokens
         )
diff --git a/Tools/peg_generator/pegen/parser_generator.py b/Tools/peg_generator/pegen/parser_generator.py
index b92df22..0345251 100644
--- a/Tools/peg_generator/pegen/parser_generator.py
+++ b/Tools/peg_generator/pegen/parser_generator.py
@@ -1,5 +1,4 @@
 import contextlib
-import token
 from abc import abstractmethod
 
 from typing import AbstractSet, Dict, IO, Iterator, List, Optional, Set, Text, Tuple
@@ -19,11 +18,12 @@ from pegen.grammar import GrammarError, GrammarVisitor
 
 
 class RuleCheckingVisitor(GrammarVisitor):
-    def __init__(self, rules: Dict[str, Rule]):
+    def __init__(self, rules: Dict[str, Rule], tokens: Dict[int, str]):
         self.rules = rules
+        self.tokens = tokens
 
     def visit_NameLeaf(self, node: NameLeaf) -> None:
-        if node.value not in self.rules and node.value not in token.tok_name.values():
+        if node.value not in self.rules and node.value not in self.tokens.values():
             # TODO: Add line/col info to (leaf) nodes
             raise GrammarError(f"Dangling reference to rule {node.value!r}")
 
@@ -32,12 +32,13 @@ class ParserGenerator:
 
     callmakervisitor: GrammarVisitor
 
-    def __init__(self, grammar: Grammar, file: Optional[IO[Text]]):
+    def __init__(self, grammar: Grammar, tokens: Dict[int, str], file: Optional[IO[Text]]):
         self.grammar = grammar
+        self.tokens = tokens
         self.rules = grammar.rules
         if "trailer" not in grammar.metas and "start" not in self.rules:
             raise GrammarError("Grammar without a trailer must have a 'start' rule")
-        checker = RuleCheckingVisitor(self.rules)
+        checker = RuleCheckingVisitor(self.rules, self.tokens)
         for rule in self.rules.values():
             checker.visit(rule)
         self.file = file
diff --git a/Tools/peg_generator/pegen/python_generator.py b/Tools/peg_generator/pegen/python_generator.py
index bde2789..6433655 100644
--- a/Tools/peg_generator/pegen/python_generator.py
+++ b/Tools/peg_generator/pegen/python_generator.py
@@ -1,3 +1,4 @@
+import token
 from typing import Any, Dict, Optional, IO, Text, Tuple
 
 from pegen.grammar import (
@@ -123,8 +124,13 @@ class PythonCallMakerVisitor(GrammarVisitor):
 
 
 class PythonParserGenerator(ParserGenerator, GrammarVisitor):
-    def __init__(self, grammar: grammar.Grammar, file: Optional[IO[Text]]):
-        super().__init__(grammar, file)
+    def __init__(
+        self,
+        grammar: grammar.Grammar,
+        file: Optional[IO[Text]],
+        tokens: Dict[int, str] = token.tok_name,
+    ):
+        super().__init__(grammar, tokens, file)
         self.callmakervisitor = PythonCallMakerVisitor(self)
 
     def generate(self, filename: str) -> None:
diff --git a/Tools/peg_generator/pegen/testutil.py b/Tools/peg_generator/pegen/testutil.py
index 1f79d8f..264659e 100644
--- a/Tools/peg_generator/pegen/testutil.py
+++ b/Tools/peg_generator/pegen/testutil.py
@@ -17,6 +17,7 @@ from pegen.parser import Parser
 from pegen.python_generator import PythonParserGenerator
 from pegen.tokenizer import Tokenizer
 
+ALL_TOKENS = token.tok_name
 EXACT_TOKENS = token.EXACT_TOKEN_TYPES  # type: ignore
 NON_EXACT_TOKENS = {
     name for index, name in token.tok_name.items() if index not in EXACT_TOKENS.values()
@@ -76,7 +77,7 @@ def import_file(full_name: str, path: str) -> Any:
 
 def generate_c_parser_source(grammar: Grammar) -> str:
     out = io.StringIO()
-    genr = CParserGenerator(grammar, EXACT_TOKENS, NON_EXACT_TOKENS, out)
+    genr = CParserGenerator(grammar, ALL_TOKENS, EXACT_TOKENS, NON_EXACT_TOKENS, out)
     genr.generate("<string>")
     return out.getvalue()
 
@@ -96,7 +97,9 @@ def generate_parser_c_extension(
     assert not os.listdir(path)
     source = path / "parse.c"
     with open(source, "w") as file:
-        genr = CParserGenerator(grammar, EXACT_TOKENS, NON_EXACT_TOKENS, file, debug=debug)
+        genr = CParserGenerator(
+            grammar, ALL_TOKENS, EXACT_TOKENS, NON_EXACT_TOKENS, file, debug=debug
+        )
         genr.generate("parse.c")
     compile_c_extension(str(source), build_dir=str(path))