sebsx / backup-iterator

Simple backup file finder & iterator

Warning: not tested on Windows

Also, this is a very basic and naïve implementation of the requirements.

What it does

• searches a path and creates an index based on the filename without extension
• assumes filename is unique (so it won't bother dealing with file.txt and file.doc separately)
• finds all the files matching some extensions and treats them as backups
• checks the filestamps and reports back older files

Things missing

• filetype checks based on header or metadata
• can't handle filename sans extension clashes
• doesn't handle files with multiple extensions (e.g. tar.gz)

How to use it

Basic unittest testing is included (run-tests.py), it will also handle unpacking the mock data tarfile.

Main client is in the run.py file and usage looks like this:

./run-tests.py
... tests run ...
./run.py -p ./mock-data
./mock-data/backups/test.bak may be superseded by ./mock-data/new/test.xls
.
.
...

How it works

To see exactly the train of thought and how the data looks like run the prototype.py file.

The Iterator does a few things:

1. Scans the path passed to the constructor
2. Builds a dictionary data structure to hold the file names, principals and backup files.
3. Finding the older backup files is done by going through the structure above and looking for files with a mtime older than what the principal has.
4. Final data structure is a list of tuples having the form of [(principal1, backup1), (principal1, backup2), (principal2, backup)] and so on and so forth.
5. The iterator results can either be stored into a list via a comprehension and simply iterated through and worked on.

Thanks for looking!