Simple file to delete duplicates from ssdeep csv output ran on a google photos export

deletedupes.py

import re
import os
import sys
import csv
import datetime

DELETED_TEXT = "deleted {}: dupe of {}"
dupe_brackets = re.compile("\([0-9]\)*")
dupe_tilda = re.compile("\~[0-9]*")


def googledate_to_datetime(time: str) -> datetime.datetime:
    # Get rid of any patterns like 2019-06-19 #2
    time = time.split(" ")[0]
    # Get rid of any patterns like 2019-06-19-20
    time = "-".join(time.split("-")[0:3])
    return datetime.datetime.strptime(time, "%Y-%m-%d")


def regex_check(filename: str) -> bool:
    if dupe_brackets.search(filename):
        return True
    elif dupe_tilda.search(filename):
        return True
    else:
        return False


def delete(file):
    try:
        os.remove(file)
    except FileNotFoundError:
        pass


with open(sys.argv[1], newline='') as fp:
    files_deleted = 0

    reader = csv.reader(fp)
    for row in reader:
        orig_filename = row[0].split("/")[-1]
        dupe_filename = row[1].split("/")[-1]
        if regex_check(dupe_filename):
            delete(row[1])
            print(DELETED_TEXT.format(dupe_filename, orig_filename))

        elif regex_check(orig_filename):
            delete(row[0])
            print(DELETED_TEXT.format(orig_filename, dupe_filename))

        else:
            orig_folder_date = googledate_to_datetime(row[0].split("/")[-2])
            dupe_folder_date = googledate_to_datetime(row[1].split("/")[-2])
            if dupe_folder_date < orig_folder_date:
                delete(row[0])
                print(DELETED_TEXT.format(orig_filename, dupe_filename))
            else:
                delete(row[1])
                print(DELETED_TEXT.format(dupe_filename, orig_filename))
        files_deleted += 1
print("Deleted {} dupe files".format(files_deleted))