opensidewalks_cli

A cli application for generating OpenSidewalks task polygons for use with the HOT Tasking Manager. The task polygons generated by this tool are optimized for two workflows:

• Pedestrian intersections, where a task will be centered on a street intersection. This guarantees that mapping a crosswalk won't be interrupted by crossing a task boundary.

• Sidewalks, where a task will encompass a single "block". If a "block" has full sidewalk coverage, this task would be an outline around the entire sidewalk path.

Please note that this application requires Python version 3.7+

Installation

1. Clone the repository

First, clone this repository: git clone https://github.com/opensidewalks/opensidewalks_cli

2. Installing the application

This tool is developed using poetry.

It can be installed with poetry by running poetry install in the main repo directory or, if you have a recent version of pip, by running pip install ..

Alternative: Use the included Dockerfile to automatically install and build all dependencies:

cd <CLONED_DIR> && docker build -t opensidewalks_cli .

Then, for all subsequent commands that use osw, instead prepend with: docker run -v $(pwd)/path/to/data_directory:/data opensidewalks_cli

Creating task polygons for intersection and sidewalk mapping

Run osw task <path-to-osm-pbf> <path-to-aoi-polygon-geojson> <output-path>, where:

• <path-to-osm-pbf is a path to an OSM PBF file that has street data for the area for which you want to create tasks.

• <path-to-aoi-polygon-geojson> is a path to a GeoJSON file with a single polygon geometry defining your AOI (area of interest) - the boundary of your tasks.

• <output-path> is your output directory, where several output files will be created.

Tutorial

We will use the example of Columbus, Ohio.

0. Prepare a work directory

Create and enter an empty directory:

mkdir ohio_tasks && cd ohio_tasks

Create subdirectories for our data:

mkdir -p ./{data_sources,intermediate_data,output}

1. Retrieve an OSM PBF

Download OSM data for Ohio from geofabrik.

Create a directory called data_sources and place the OSM PBF in it.

mv ~/Dowloads/ohio-latest.osm.pbf ./data_sources

2. Define an area of interest polygon and store in a GeoJSON file

For this example, we will download and extract a polygon representing the North Linden neighborhood of Columbus, Ohio. You can use any polygon that is of interest to your project.

Download the neighborhoods shapefile from Columbus Open Data Portal.

Place it in data_sources:

mv ~/Downloads/Columbus_Commnities-shp.zip ./data_sources. (Extract the zip in this directory).

Open neighborhoods shapefile in QGIS, select 'North Linden' (objectid 383, area name 'North Linden'), save a GeoJSON file with single polygon feature. Save to the path ./intermediate_data/north_linden.geojson.

(Optionally) Clip the OSM PBF using the AOI polygon and the osmium tool: osmium extract -p intermediate_data/north_linden.geojson data_sources/ohio-latest.osm.pbf -o intermediate_data/north_linden.osm.pbf . This will speed up the next steps if the extract is much larger than the AOI. Note that you will need osmium 1.5 or greater for this step.

3. Run the osw command line interface

Run the task subcommand of the osw cli application to create pedestrian intersection and sidewalk task polygons.

osw task intermediate_data/north_linden.osm.pbf intermediate_data/north_linden.geojson output/north_linden

If you decided not to clip the OSM PBF, replace intermediate_data/north_linden.osm.pbf with data_sources/ohio-latest.osm.pbf.

If you receive an error that the osw command can't be found, make sure you have activated the Python environment in which you installed opensidewalks_cli. For example, if you cloned this repository and ran poetry install, you will need to run poetry shell in this repository before you can use osw. Alternatively, you can run the osw command from the base directory of the repository by using poetry run, such as poetry run osw.