pits – Point-In-Time Storage
pits provides Pythonic context-managed file object abstraction for point-in-time file storage. pits supports plain or gzipped files.
pits is listed on PyPI.
Writing
my_data = grab_my_data(...)
# write to plain file
with open('my-data.txt', 'w+') as f:
f.write(my_data)
# write to gzip file
import gzip
with gzip.open('my-data.gz', 'wb') as f:
f.write(my_data)
# write to point-in-time file
import pits
with pits.open('my-data', 'wb', gzip=True) as f:
f.write(my_data)
# write as if we're at another point in time
with pits.goto('20220101-1200').open('my-data', 'wb', gzip=True) as f:
f.write(my_data)Reading
import pits
# read latest file
with pits.open('my-data') as f:
f.write(my_data)
# read latest file as if we're at another point in time
with pits.goto('2022-12-18 15:00').open('my-data') as f:
f.write(my_data)Datetime format
- Timestamp format for file name is
%Y.%m.%d-%H:%M:%S. - All non-numeric characters are omitted in timestamp parsing, e.g.,
2022-12-18 15:00is transformed to202212181500before parsing. - All timestamps are local.
-
20221218is recognized as2022-12-18 23:59:59, but20221218-1500is recognized as2022-12-18 15:00:00.
