dfint / df-gettext-toolkit / 5871136315

import shutil

from pathlib import Path

from df_gettext_toolkit.utils import csv_utils

from df_gettext_toolkit.utils.backup import backup

def split_left(start, end):

    mid = (start + end) // 2

    return start, mid

def split_right(start, end):

    mid = (start + end) // 2

    return mid, end

def bisect(file_path: Path, encoding: str, data: list[list[str]]):

    def _bisect(start: int, end: int, bad: bool = False) -> bool:

        if start >= end:

            print("Empty slice, step back")

            return False

        print(f"From {start} to {end} (in total {end - start})")

        csv_utils.write_csv(file_path, encoding, data[start:end])

        if bad:

            confirmed = True

            confirmed = input("Is it bad (Y/N)? ").upper() == "Y"

        if not confirmed:

            return False

        if start == end - 1:

            print(f"Found string, line number {start+1}:")

            print(data[start])

            confirmed = input("Exclude from csv (Y/N)? ").upper() == "Y"

            if confirmed:

                csv_utils.write_csv(file_path, encoding, data[:start] + data[start + 1 :])

            return True

        print("Trying left half")

        result = _bisect(*split_left(start, end))

        if result:

            return result

        print("Trying right half")

        return _bisect(*split_right(start, end), bad=True)

    _bisect(0, len(data), bad=True)

def main(csv_file: Path, encoding: str):

    assert csv_file.is_file(), f"{csv_file.name} is not a file"

    try:

        with backup(csv_file, overwrite=True) as backup_path:

            data = list(csv_utils.read_csv(backup_path, encoding))

            bisect(csv_file, encoding, data)

        confirmed = input("Restore from backup (Y/N)? ").upper() == "Y"

        if confirmed:

            shutil.copy(backup_path, csv_file)