deduplicate.py

from itertools import groupby
from typing import Union

Registration = dict[str, Union[int, str]]

ALL_REGISTRATIONS = [
    {'id': 0,
     'hostname': 'lappytoppy',
     'last_seen': '2021-09-27T18:00'},
    {'id': 1,
     'hostname': 'server01',
     'last_seen': '2021-12-14T18:00'},
    {'id': 2,
     'hostname': 'boxbox',
     'last_seen': '2021-09-29T18:00'},
    {'id': 3,
     'hostname': 'server01',
     'last_seen': '2021-11-19T18:00'},
    {'id': 4,
     'hostname': 'server01',
     'last_seen': '2021-11-30T18:00'},
    {'id': 5,
     'hostname': 'pi',
     'last_seen': '2021-12-01T18:00'},
    {'id': 6,
     'hostname': 'server02',
     'last_seen': '2021-12-22T18:00'},
    {'id': 7,
     'hostname': 'server02',
     'last_seen': '2021-12-23T18:00'}
]


def dedup(registrations: list[Registration]):
    key_func_hostname = lambda registration: registration.get('hostname')
    key_func_last_seen = lambda registration: registration.get('last_seen')
    registrations.sort(key=key_func_hostname)
    most_recent = []
    for hostname, duplicates in groupby(registrations, key_func_hostname):
        sorted_duplicates = list(duplicates)
        sorted_duplicates.sort(key=key_func_last_seen, reverse=True)
        most_recent.append(sorted_duplicates[0])

    return most_recent


if __name__ == '__main__':
    no_repeats = dedup(ALL_REGISTRATIONS)
    print(no_repeats)