2024-05-01 13:02:47 -07:00
|
|
|
#!/usr/bin/env python3
|
|
|
|
|
|
|
|
import argparse
|
|
|
|
import re
|
2024-05-08 19:57:16 +02:00
|
|
|
import os
|
2024-05-01 13:02:47 -07:00
|
|
|
|
|
|
|
HEADER = """\
|
|
|
|
//===-- SBLanguages.h -----------------------------------------*- C++ -*-===//
|
|
|
|
//
|
|
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
#ifndef LLDB_API_SBLANGUAGE_H
|
|
|
|
#define LLDB_API_SBLANGUAGE_H
|
2024-05-10 10:00:36 -07:00
|
|
|
|
2024-10-09 14:41:50 -07:00
|
|
|
#include <cstdint>
|
|
|
|
|
2024-05-10 10:00:36 -07:00
|
|
|
namespace lldb {
|
2024-05-01 13:02:47 -07:00
|
|
|
/// Used by \\ref SBExpressionOptions.
|
|
|
|
/// These enumerations use the same language enumerations as the DWARF
|
|
|
|
/// specification for ease of use and consistency.
|
|
|
|
enum SBSourceLanguageName : uint16_t {
|
|
|
|
"""
|
|
|
|
|
|
|
|
FOOTER = """\
|
|
|
|
};
|
|
|
|
|
2024-05-10 10:00:36 -07:00
|
|
|
} // namespace lldb
|
|
|
|
|
2024-05-01 13:02:47 -07:00
|
|
|
#endif
|
|
|
|
"""
|
|
|
|
|
|
|
|
REGEX = re.compile(
|
2024-05-01 13:41:32 -07:00
|
|
|
r'^ *HANDLE_DW_LNAME *\( *(?P<value>[^,]+), (?P<name>.*), "(?P<comment>[^"]+)",.*\)'
|
2024-05-01 13:02:47 -07:00
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def emit_enum(input, output):
|
|
|
|
# Read the input and break it up by lines.
|
|
|
|
lines = []
|
|
|
|
with open(input, "r") as f:
|
|
|
|
lines = f.readlines()
|
|
|
|
|
2024-05-08 19:57:16 +02:00
|
|
|
# Create output folder if it does not exist
|
|
|
|
os.makedirs(os.path.dirname(output), exist_ok=True)
|
|
|
|
|
2024-05-01 13:02:47 -07:00
|
|
|
# Write the output.
|
|
|
|
with open(output, "w") as f:
|
|
|
|
# Emit the header.
|
|
|
|
f.write(HEADER)
|
|
|
|
|
|
|
|
# Emit the enum values.
|
|
|
|
for line in lines:
|
|
|
|
match = REGEX.match(line)
|
|
|
|
if not match:
|
|
|
|
continue
|
|
|
|
f.write(f" /// {match.group('comment')}.\n")
|
|
|
|
f.write(f" eLanguageName{match.group('name')} = {match.group('value')},\n")
|
|
|
|
|
|
|
|
# Emit the footer
|
|
|
|
f.write(FOOTER)
|
|
|
|
|
|
|
|
|
|
|
|
def main():
|
|
|
|
parser = argparse.ArgumentParser()
|
|
|
|
parser.add_argument("--output", "-o")
|
|
|
|
parser.add_argument("input")
|
|
|
|
args = parser.parse_args()
|
|
|
|
|
|
|
|
emit_enum(args.input, args.output)
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
main()
|