Source: 📖 Python Cookbook ch1.11 p19

Using slice objects to name slices and avoid hardcoding

To avoid unreadable code caused by repeated hardcoding of slices and indexes, a slice object can be used and assigned to a variable with a friendly, easy to understand name.

Suppose you are reading in lines from a flat text file and you have to parse numbers from it based on their position. In this scenario, we are reading in information about a stock portfolio which includes share numbers and purchase price.

record = 'AAPL....................100 .......513.25 ..........'
cost = int(record[20:32] * float(record[40:48])

Doing this repeatedly can cause a lot of confusion when reading back the code, as it's not obvious at all what the slices are indexing. Instead, you can name a slice by turning it into an object:

SHARES = slice(20, 32)	# same as [20:32]
PRICE = slice(40, 48)	# same as [40:48]

cost = int(record[SHARES]) * float(record[PRICE])

Using the above code repeatedly makes it clear what each slice is accessing, and it's also much easier to edit if the format of the source file changes as you don't have to update many instances of hardcoded slices and indexes.

Attributes of the slice object can be accessed with slice.start, slice.stop and slice.step.