«Удалить файлы на основе номера счета, даты и номера версии»

Question

Я хочу удалить несколько файлов из определенного каталога на основе имени файла, даты и номера версии, используя Python. PS Дата создания файла не может быть принята во внимание

Ссылка опубликована в Stackoverflow, но имена файлов и номера версий изменены. Как сохранить дату и номер версии, чтобы найти последний файл.

source = r'C:\Users\XMLFiles'
file_names = os.listdir(source)

latest_files = {}
for file_name in file_names:
    name_parts = file_name.split("_")
    date_stamp = name_parts[2], name_parts[3].split(".")[0]

    if date_stamp not in latest_files or file_name > latest_files[date_stamp]:
        latest_files[date_stamp] = file_name
        print(latest_files)

keep_files = latest_files.values()

for file_name in file_names:
    if file_name in keep_files:
        continue

    os.remove(os.path.join(source, file_name)




##################

List of files to process

=============================
Invoice_456879_20180404_2510.xml

Invoice_123876_20171027_17.xml
Invoice_123876_20180404_2513.xml

Invoice_832765_20170309_2.xml
Invoice_832765_20170313_0.xml
Invoice_832765_20170323_5.xml
Invoice_832765_20170330_2.xml
Invoice_832765_20170613_3.xml
Invoice_832765_20171206_18.xml
Invoice_832765_20171206_30.xml
Invoice_832765_20171206_36.xml
Invoice_832765_20180404_3066.xml
Invoice_832765_20180405_9770.xml
Invoice_832765_20180405_9779.xml

Invoice_698325_20170308_0.xml
Invoice_698325_20170309_3.xml
Invoice_698325_20170323_4.xml
Invoice_698325_20170330_5.xml
Invoice_698325_20170613_4.xml
Invoice_698325_20171206_8.xml
Invoice_698325_20171206_24.xml
Invoice_698325_20171206_46.xml
Invoice_698325_20180404_3067.xml
Invoice_698325_20180405_9771.xml

===========================================================

Expected Output

Invoice_456879_20180404_2510.xml
Invoice_123876_20180404_2513.xml
Invoice_832765_20180405_9779.xml
Invoice_698325_20180405_9771.xml

Answer 1

latest_files = {}

for file_name in file_names:
    name_parts = file_name.split('_')

    invoice = name_parts[1]
    ymd = name_parts[2]
    version = int(name_parts[3].split('.')[0])

    if invoice not in latest_files:
        latest_files[invoice] = (ymd, version, file_name)
        continue

    latest_ymd, latest_version = latest_files[invoice][:2]

    if ymd > latest_ymd or (ymd == latest_ymd and version > latest_version):
        latest_files[invoice] = (ymd, version, file_name)

keep_files = [tup[2] for tup in latest_files.values()]

for file_name in file_names:
    if file_name in keep_files:
        continue

    os.remove(os.path.join(source, file_name))