An algorythm to compare SQL Primary keys and sync A -> B.

sql_sync_algo.py

# -*- coding: utf-8 -*-
import random


def compare(all_read_pks, all_write_pks):
    """This function emulates a mathematical problem i try to solve.

    I have primary keys from a SQL Table A and primary_keys in SQL Table B.

    This function should compare B with A and sync all primary_keys changes to B.

    The biggest problem is here that i have to compare numbers in blocks and compare
    a single block until its fully synced.
    """  # noqa
    block_size = 100
    compared = 0
    to_compare = len(all_read_pks)
    if len(all_write_pks) > len(all_read_pks):
        to_compare = len(all_write_pks)
    while compared < to_compare:
        read_pks = set(list(all_read_pks)[compared:compared + block_size])
        write_pks = set(list(all_write_pks)[compared:compared + block_size])

        if read_pks == write_pks:
            compared += block_size
            continue

        matching = read_pks & write_pks

        for i in write_pks - matching:
            all_write_pks.remove(i)

        for i in read_pks - write_pks:
            all_write_pks.add(i)

        if read_pks == write_pks:
            compared += len(read_pks)
            continue

    return all_read_pks == all_write_pks


if __name__ == '__main__':
    seta = set(range(1000))
    setb = set(range(1000))

    # Remove random rows from both sets "tables"
    numbers1 = set(random.sample(range(1000), 500))
    numbers2 = set(random.sample(range(500, 1000), 250))

    # Compare with 500 random rows removed from both tables.
    print("Remove in both is: ", compare(seta - numbers1, setb - numbers2))

    # Compare with 500 random rows removed from source only
    print("Remove in source is: ", compare(seta - numbers1, setb))

    # Compare with 500 random rows removed from target only
    print("Remove in target is: ", compare(seta, setb - numbers2))