slicemap 1.0.17

A tiny package containing a dict-like data structure with numeric slices as keys.

SliceMap is similar to a normal Python dict, but instead of setting values key by key, you set entire slices at once.

This is a useful data structure implemented entirely in Python with help of sortedcontainers package. Initially, I used a custom skip-list implementation, but this wasn't as fast as sortedcontainers (see their benchmarks).

Adding new ranges and querying values both have O(log(n)) time complexity.

Quick Start

Installation

pip install slicemap

Create, query and visualize SliceMap

from slicemap import SliceMap
sm = SliceMap()

sm[-10:10] = 0
sm[2:4] = 1
sm[4:6] = 2
sm[7:9] = 3
sm[12:15] = 1.5
print(sm[2], sm[3], sm[4], sm[9], sm[15])

# works only for numeric values
sm.plot() 

Outputs:

1 1 2 0 None

Include start | end

The default value is include="start", but you can choose to include the end of slices instead.

from slicemap import SliceMap
sm1 = SliceMap(include="start")
sm1[2:3] = 1
sm1[3:4] = 2
sm1[4:5] = 3
print(sm1[3], sm1[4])

sm2 = SliceMap(include="end")
sm2[2:3] = 1
sm2[3:4] = 2
sm2[4:5] = 3
print(sm2[3], sm2[4])

Outputs:

2 3
1 2

Query either values or ranges

You can equery each value individually, or query with a slice to get all values in given slice.

from slicemap import SliceMap
sm = SliceMap(include="start")

sm[-10:10] = 0
sm[2:4] = 1
sm[4:6] = 2
sm[7:9] = 3
sm[12:15] = 1.5
print(sm[3], sm[5], sm[8])
print(sm[3:8])

Outputs:

1 2 3
(1, 2, 0, 3)

Applications

The Skyline Problem

Easily solvable in O(n*log(n)) time:

from slicemap import SliceMap

# format [left_x, value, right_x]
inputs = [
    [1, 11, 5],
    [2, 6, 7],
    [3, 13, 9],
    [12, 7, 16],
    [14, 3, 25],
    [19, 18, 22],
    [23, 13, 29],
    [24, 4, 28],
]

sm = SliceMap()
sm[:] = 0

for left, value, right in sorted(inputs, key=lambda x: x[1]):
    sm[left:right] = value

print(sm)
print(sm.export())
sm.plot()

Outputs:

{(-inf,1): 0, [1,3): 11, [3,9): 13, [9,12): 0, [12,16): 7, [16,19): 3, [19,22): 18, [22,23): 3, [23,29): 13, [29,inf): 0}
[(-inf, 1, 0), (1, 3, 11), (3, 9, 13), (9, 12, 0), (12, 16, 7), (16, 19, 3), (19, 22, 18), (22, 23, 3), (23, 29, 13), (29, inf, 0)]

Depending on the exact task formulation, answer should be easy to retrieve from the above.

Answers

• Package matplotlib is an optional dependency - without it you can use the pacakge, but not the plotting functionality.