#!/usr/bin/env python3
"""Compare a read-only 40250 database name export with the older Chinese tables."""

import argparse
import json
from pathlib import Path


REPO = Path(__file__).resolve().parents[1]
DEFAULT_EXPORT = REPO / "build/zh-locale/server-audit"
DEFAULT_TABLES = REPO.parent / "server_extracted/40250_deploy"


def live_names(path: Path):
    names = {}
    for number, line in enumerate(path.read_text(encoding="ascii").splitlines(), 1):
        try:
            vnum, encoded = line.split("\t", 1)
            name = bytes.fromhex(encoded).decode("gbk")
            if int(vnum) in names:
                raise ValueError("duplicate VNUM")
            names[int(vnum)] = name
        except (ValueError, UnicodeDecodeError) as error:
            raise ValueError(f"{path}:{number}: {error}") from error
    return names


def older_names(path: Path):
    names = {}
    for line in path.read_text(encoding="utf-8").splitlines()[1:]:
        vnum, name = line.split("\t", 1)
        names[int(vnum)] = name
    return names


def audit(export_dir: Path, tables_dir: Path):
    result = {}
    for kind in ("item", "mob"):
        live = live_names(export_dir / f"{kind}_names_hex.tsv")
        older = older_names(tables_dir / f"{kind}_names_zh.txt")
        shared = live.keys() & older.keys()
        conflicts = [
            {"vnum": vnum, "server": live[vnum], "older_table": older[vnum]}
            for vnum in sorted(shared) if live[vnum] != older[vnum]
        ]
        result[kind] = {
            "server_rows": len(live),
            "older_table_rows": len(older),
            "shared_vnums": len(shared),
            "server_only_vnums": len(live.keys() - older.keys()),
            "older_only_vnums": len(older.keys() - live.keys()),
            "server_names_without_hanzi": sum(
                not any("\u4e00" <= char <= "\u9fff" for char in name)
                for name in live.values()
            ),
            "conflicts": conflicts,
        }
    return result


def main():
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--export-dir", type=Path, default=DEFAULT_EXPORT)
    parser.add_argument("--tables-dir", type=Path, default=DEFAULT_TABLES)
    parser.add_argument("--output", type=Path, default=DEFAULT_EXPORT / "name-audit.json")
    args = parser.parse_args()
    report = audit(args.export_dir, args.tables_dir)
    args.output.write_text(json.dumps(report, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
    for kind, data in report.items():
        print(f"{kind}: rows={data['server_rows']} no_hanzi={data['server_names_without_hanzi']} "
              f"conflicts={len(data['conflicts'])}")


if __name__ == "__main__":
    main()