bpo-40939: Generate keyword.py using the new parser (GH-20800)

(cherry picked from commit 9727694f08cad4b019d2939224e3416312b1c0e1)

Co-authored-by: Lysandros Nikolaou <lisandrosnik@gmail.com>

1 files changed, 73 insertions, 0 deletions

diff --git a/Tools/peg_generator/pegen/keywordgen.py b/Tools/peg_generator/pegen/keywordgen.py
new file mode 100644
index 0000000..279c34b
--- /dev/null
+++ b/Tools/peg_generator/pegen/keywordgen.py
@@ -0,0 +1,73 @@
+"""Generate Lib/keyword.py from the Grammar and Tokens files using pgen"""
+
+import argparse
+
+from .build import build_parser, generate_token_definitions
+from .c_generator import CParserGenerator
+
+TEMPLATE = r'''
+"""Keywords (from "Grammar/python.gram")
+
+This file is automatically generated; please don't muck it up!
+
+To update the symbols in this file, 'cd' to the top directory of
+the python source tree and run:
+
+    PYTHONPATH=Tools/peg_generator python3 -m pegen.keywordgen \
+        Grammar/Grammar \
+        Grammar/Tokens \
+        Lib/keyword.py
+
+Alternatively, you can run 'make regen-keyword'.
+"""
+
+__all__ = ["iskeyword", "kwlist"]
+
+kwlist = [
+    {keywords}
+]
+
+iskeyword = frozenset(kwlist).__contains__
+'''.lstrip()
+
+EXTRA_KEYWORDS = ["async", "await"]
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Generate the Lib/keywords.py file from the grammar."
+    )
+    parser.add_argument(
+        "grammar", type=str, help="The file with the grammar definition in PEG format"
+    )
+    parser.add_argument(
+        "tokens_file",
+        type=argparse.FileType("r"),
+        help="The file with the token definitions"
+    )
+    parser.add_argument(
+        "keyword_file",
+        type=argparse.FileType("w"),
+        help="The path to write the keyword definitions",
+    )
+    args = parser.parse_args()
+
+    grammar, _, _ = build_parser(args.grammar)
+    with args.tokens_file as tok_file:
+        all_tokens, exact_tok, non_exact_tok = generate_token_definitions(tok_file)
+    gen: ParserGenerator = CParserGenerator(
+        grammar, all_tokens, exact_tok, non_exact_tok, file=None
+    )
+    gen.collect_todo()
+
+    with args.keyword_file as thefile:
+        all_keywords = sorted(
+            list(gen.callmakervisitor.keyword_cache.keys()) + EXTRA_KEYWORDS
+        )
+
+        keywords = ",\n    ".join(map(repr, all_keywords))
+        thefile.write(TEMPLATE.format(keywords=keywords))
+
+
+if __name__ == "__main__":
+    main()