OpenSidewalks Schema Tools

This project provides utilities for working with the OpenSidewalks (OSW) schema:

• Parsing the canonical opensidewalks.schema.json into grouped definitions.
• Building specialized JSON Schemas for LineString, Point, and Polygon/MultiPolygon features.
• Generating valid and invalid sample GeoJSON files from the schema.
• Validating GeoJSON feature collections against the generated schemas.

Setup

Node.js (Schema & Sample Generation)

# Install dependencies
npm install

Python (Validation)

# Create venv (recommended)
python3 -m venv .venv
source .venv/bin/activate

# Install requirements
pip install -r requirements.txt

Usage

Parse & Build Schemas

Parses opensidewalks.schema.json into grouped schemas and writes results into output/.

node scripts/parse-osw-schema.js

This produces:

• output/results.json (expanded schema entries)
• output/lineSchema.json
• output/pointsSchema.json
• output/PolygonSchema.json

Generate Sample GeoJSONs

example.js demonstrates creating valid and invalid sample files:

node example.js

Outputs:

• my_valid.geojson
• my_invalid.geojson

You can adjust:

• Number of features (numFeatures)
• Output filenames (validOut, invalidOut)
• Random seed (seed)
• Feature types (Line, Point, Polygon arrays)

Validate GeoJSON Files

Validate a file against a schema using the high-performance Rust-backed jsonschema-rs:

python test.py

• By default, uses output/lineSchema.json and my_valid.geojson.
• Edit test.py to switch to POINT_SCHEMA, POLY_SCHEMA, or INVALID_FILE.

The output is a JSON array of errors:

[
  { "featureIndex": 0, "error": "missing '_id'" },
  { "featureIndex": 2, "error": "\"bench\" is not one of \"lamp\", \"hydrant\"" }
]

Customization

• RENAME_MAP in JS scripts allows renaming schema keys (e.g., _id → id).
• DISCRIMINATORS defines tags (highway, amenity, building, etc.) used for property dependencies.
• META_KEYS lists common fields (e.g., name) that skip dependency checks.
• test.py prefers root-cause errors (enum, type, required) and collapses noisy cascades.

Example Workflow

1. Parse the OSW schema:

   node scripts/parse-osw-schema.js

2. Generate sample GeoJSON files:

   node example.js

3. Validate the generated files:

   python test.py