Installation

This package has a non-Python dependency, libspatialindex. The easiest way to install it is by installing the rtree Python package via Conda:

conda install rtree

Once this is installed, penkit-optimize can be installed via pip:

pip install penkit-optimize

Usage

To see CLI options:

penkit-optimize -h
    usage: penkit-optimize [-h] [--greedy] [--noopt] [--runtime RUNTIME]
                           [--merge-paths [MERGE_PATHS]] [--vis-output VIS_OUTPUT]
                           input_file [output_file]

    positional arguments:
    input_file
    output_file

    optional arguments:
    -h, --help            show this help message and exit
    --greedy, -g          Run greedy optimization only.
    --noopt, -n           Don't run any optimization.
    --runtime RUNTIME, -t RUNTIME
                          Maximum runtime (in seconds) of optimization stage.
    --merge-paths [MERGE_PATHS], -m [MERGE_PATHS]
                          Merge paths that start/end near each other. You may
                          optionally specify a threshold distance (in document
                          units) after this parameter.
    --vis-output VIS_OUTPUT, -v VIS_OUTPUT
                          If provided, save a visualization of the path to this
                          SVG file.

Examples

Generate optimized SVG with default options (up to 5 minute runtime), and save to input-optimized.svg:

penkit-optimize input.svg

Don’t optimize anything, just visualize the transits:

penkit-optimize input.svg -n -v noopt-vis.svg

noopt-vis.svg:

Run greedy optimization (only), save the file, and save a visualization of the result:

penkit-optimize input.svg output.svg -g -v greedy-vis.svg

greedy-vis.svg:

Run the full optimization and write a visualization of the result:

penkit-optimize input.svg output.svg -v opt-vis.svg

opt-vis.svg:

Run the full optimization and merge paths that start/end within 1.5 units:

penkit-optimize input.svg -m 1.5