Files
neoLegacy-mac/merge-iggy-data-symbols.py

125 lines
4.0 KiB
Python

#!/usr/bin/env python3
import argparse
from collections import defaultdict
import csv
from pathlib import Path
DATA_FIELDS = (
"address",
"section",
"initialized",
"data_type",
"defined_data_length",
"span_upper_bound",
"symbol_source",
"reference_count",
)
def normalize_name(name: str) -> str:
return name.lstrip("._")
def is_compiler_generated(name: str) -> bool:
return name.startswith((".L", "..LN", "$")) or name in {
"_GLOBAL_OFFSET_TABLE_",
"_DYNAMIC",
}
def address_key(row: dict[str, str]) -> tuple[int, str]:
address = row["address"]
try:
return int(address, 16), address
except ValueError:
return 0, address
def parse_platform(value: str) -> tuple[str, Path]:
if "=" not in value:
raise argparse.ArgumentTypeError("platform inputs must use NAME=DIRECTORY")
name, directory = value.split("=", 1)
if not name or not directory:
raise argparse.ArgumentTypeError("platform inputs must use NAME=DIRECTORY")
path = Path(directory)
if not path.is_dir():
raise argparse.ArgumentTypeError(f"data symbol directory does not exist: {path}")
return name, path
def main() -> int:
parser = argparse.ArgumentParser()
parser.add_argument("output", type=Path)
parser.add_argument("platform", nargs="+", type=parse_platform)
args = parser.parse_args()
platforms = [name for name, _ in args.platform]
if len(platforms) != len(set(platforms)):
parser.error("platform names must be unique")
symbols = defaultdict(lambda: defaultdict(list))
for platform, directory in args.platform:
for path in sorted(directory.glob("*.tsv")):
with path.open(newline="", encoding="utf-8") as handle:
reader = csv.DictReader(handle, delimiter="\t")
for row in reader:
if is_compiler_generated(row["name"]):
continue
key = path.stem, normalize_name(row["name"])
symbols[key][platform].append(row)
fieldnames = ["module", "name", "occurrence", "platform_count"]
fieldnames.extend(
f"{platform}_{field}"
for platform in platforms
for field in DATA_FIELDS
)
fieldnames.append("platforms")
args.output.parent.mkdir(parents=True, exist_ok=True)
output_rows = 0
with args.output.open("w", newline="", encoding="utf-8") as handle:
writer = csv.DictWriter(
handle,
fieldnames=fieldnames,
delimiter="\t",
lineterminator="\n",
)
writer.writeheader()
for (module, name), by_platform in sorted(symbols.items()):
for rows in by_platform.values():
rows.sort(key=address_key)
occurrences = max(len(rows) for rows in by_platform.values())
for occurrence in range(occurrences):
output = {
"module": module,
"name": name,
"occurrence": occurrence + 1,
"platform_count": sum(
occurrence < len(by_platform.get(platform, ()))
for platform in platforms
),
"platforms": ",".join(
platform
for platform in platforms
if occurrence < len(by_platform.get(platform, ()))
),
}
for platform in platforms:
rows = by_platform.get(platform, ())
if occurrence >= len(rows):
continue
row = rows[occurrence]
for field in DATA_FIELDS:
output[f"{platform}_{field}"] = row[field]
writer.writerow(output)
output_rows += 1
print(f"merged {output_rows} data symbol occurrences across {len(platforms)} platforms")
return 0
if __name__ == "__main__":
raise SystemExit(main())