dotfiles/scripts/leveldb-dump

#!/usr/bin/env python3
import argparse
from pathlib import Path
from typing import Union
import plyvel
import json
from sys import stdout
import base64

parser = argparse.ArgumentParser()
encoding_names = ["utf8", "base16", "base32", "base64", "base85"]
parser.add_argument("--encoding", "-e", choices=encoding_names, default="utf8")
parser.add_argument("--key-encoding", choices=encoding_names, default=None)
parser.add_argument("--value-encoding", choices=encoding_names, default=None)
parser.add_argument("db_path", type=Path)
cli_args = parser.parse_args()


def bytes_to_json(b: bytes, encoding: str) -> Union[str, list[int]]:
    if encoding == "utf8":
        try:
            return b.decode("utf8")
        except UnicodeDecodeError:
            return list(b)
    elif encoding == "base16":
        return base64.b16encode(b).decode("ascii")
    elif encoding == "base32":
        return base64.b32encode(b).decode("ascii")
    elif encoding == "base64":
        return base64.b64encode(b).decode("ascii")
    elif encoding == "base85":
        return base64.b85encode(b).decode("ascii")
    else:
        assert False


key_encoding: str = cli_args.key_encoding or cli_args.encoding
value_encoding: str = cli_args.value_encoding or cli_args.encoding
db = plyvel.DB(str(cli_args.db_path), create_if_missing=False)
with db.iterator() as iterator:
    for key, value in iterator:
        json.dump(
            {
                "key": bytes_to_json(key, key_encoding),
                "value": bytes_to_json(value, value_encoding),
            },
            stdout,
        )
        stdout.write("\n")