Overview

This package implements for Python a memory-efficient representation of closed-interval sets.

The package provides the procset module.

A ProcSet is an hybrid between a set and a list of indexes. More precisely, a ProcSet object is an ordered collection of unique non-negative int. It supports most of set operations: notably membership testing, mathematical operations such as intersection, union, and (symmetric) difference; with the additional ability to access its elements by position.

The ProcSet type is mutable, as its content can be modified using methods such as insert(). Since it is mutable, it has no hash value and cannot be used as either a dictionary key or as an element of another set.

Example use

You can get the library directly from PyPI:

pip install procset

What does it look like to use procset? Here is a simple example program:

from procset import ProcSet


free_cores = ProcSet((0, 7))  # I have 8 cores to work with

job_cores = ProcSet((2, 5))  # let's use some cores for a job
free_cores -= job_cores

print('remaining cores:', str(free_cores))

And it looks like this when run:

$ python example.py
remaining cores: 0-1 6-7

String representation of interval sets

In the scheduling community, interval sets often are encoded as strings where the string 'a-b' (middle symbol is a dash, ascii 0x2d) represents the integer interval \([a, b]\), with the convention that the string a represents the degenerate case of the singleton \(\{a\}\). An interval set with many disjoint intervals is encoded by joining interval representations with a space (ascii 0x20): for example a-b c d-e represents \([a, b] \cup \{c\} \cup [d, e]\).

Warning

There are many different strings representing the same interval set. For the description of a canonical representation, please refer to the documentation of Batsim.