Features

• Acts similarly to an IBM Mainframe GDG - Generation Data Group - See https://www.ibm.com/support/knowledgecenter/zosbasics/com.ibm.zos.zconcepts/zconcepts_175.htm

• You define the number of generations, or versions, in the gdg

• The Gdg specifies a filesystem directory, a filename pattern, and a number of generations to keep

• Invoke method next() to get the next logical filename

• Invoke method current() to get the current, or most recent, filename

• Invoke method previous() to get the filename before current()

• Invoke method all_generations() to get all filenames within the gdg

• Invoke method all_files() to get all filenames within the gdg directory

• Invoke method prune() to delete the older files greater than the number of generations

Why Use this Library?

• To keep your filenames tidy; consistent filename pattern

• To keep your directories tidy; delete older/obsolete generations

• You have weekly/monthly batch processing and want to process the last 7 or 30 versions of a file

• You have large backup files and only want to keep the latest 3 of them

• You come from a “big iron” background and miss the concept of a GDG on linux/mac/windows

• You like regular expressions

The Four Patterns supported, and Examples:

Filename                                             Gdg Type

...your-gdg-directory/sample-000020.txt              Generation number (with 6 digits) type, 'g'
...your-gdg-directory/sample-1612735756.txt          Epoch type, 'e'
...your-gdg-directory/sample-20210207-172005.txt     UTC Timestamp type, 'ts_utc'
...your-gdg-directory/sample-20210207-172005.txt     Local Timestamp type, 'ts_local'

Quick start

Installation:

$ pip install gdg

Use:

Case 1: filenames with an embedded generation number (‘g’) sequence

>>> import gdg

>>> g = gdg.Gdg(...some_dir_path...)    # Provide either an absolute or relative dir path for the gdg
>>> g.set_pattern('sample-%.txt', 'g')  # pass one of the values ['g','e','ts_utc','ts_local'] into this method
>>> g.set_generations(3)                # set the number of generations to any positive integer

>>> n = g.get_generations()             # returns the int number of generations to retain in the gdg
>>> d = g.get_state()                   # returns the state of the gdg as a dict

>>> filename = g.next()                 # obtain the next filename in the gdg. YOUR CODE then writes to the file
>>> filename = g.next()
>>> filename = g.next()

>>> fname = g.current()                 # returns a string filename or None
>>> fname = g.previous()                # returns a string filename or None
>>> flist = g.all_generations()         # returns a list
>>> flist = g.all_files()               # returns a list

>>> n = g.prune()                       # delete older generations, return the number of files deleted

Case 2: filenames with an embedded epoch (‘e’) time

>>> import gdg

>>> g = gdg.Gdg(...some_dir_path...)
>>> g.set_generations(24)
>>> g.set_pattern('sample-%.txt', 'e')  # 'e' for epoch time, the value of % is populated at runtime

>>> # same usage otherwise as in Case 1 above

Case 3: filenames with an embedded UTC Timestamp (‘ts_utc’) time

>>> import gdg

>>> g = gdg.Gdg(...some_dir_path...)
>>> g.set_pattern('sample-%.txt', 'ts_utc')  # 'ts_utc' for UTC Timestamp, the value of % is populated at runtime
>>> g.set_generations(24)

>>> # same usage otherwise as in Case 1 above

Case 4: filenames with an embedded Local Timestamp (‘ts_local’) time

>>> import gdg

>>> g = gdg.Gdg(...some_dir_path...)
>>> g.set_pattern('sample-%.txt', 'ts_local')  # 'ts_local' for Local Timestamp, the value of % is populated at runtime
>>> g.set_generations(40)

>>> # same usage otherwise as in Case 1 above

Version 0.1.0

• 2021/02/13. 0.1.0 Beta

• 2021/02/07. 0.0.2 Pre-Alpha

• 2021/02/02. 0.0.1 Pre-Alpha

• 2021/02/02. 0.0.0 Pre-Alpha