ScenicOverlook 0.3.0

A library for incremental, in-memory map-reduces

The Scenic Overlook library contains datastructures for incremental map-reduces.

These datastructures are implemented as trees, and store at each node, intermediate values of the reduce. This means that when you slice or combine structures, the new output of the maps/reduces can be efficiently computed. (by reusing old outputs from unchanged parts of the tree)

Typical usage looks like this:

#!/usr/bin/env python

from scenicoverlook import viewablelist

space_concat = lambda x, y: x + ' ' + y
l = viewablelist(['the', 'quick', 'brown', 'fox'])
print l.reduce(space_concat)

# This yields 'the quick stealthy brown fox', reusing cached intermediate
# substrings from the earlier call like 'the quick' and 'brown fox':

print (l[:2] + ['stealthy'] + l[2:]).reduce(space_concat)

See the pydocs for more examples:

https://github.com/pschanely/ScenicOverlook/blob/master/scenicoverlook/__init__.py