nyc-historical-geocoder

Geocoder for historical New York City addresses, for Node.js.

This module is part of NYPL’s NYC Space/Time Directory.

nyc-historical-geocoder uses Lunr to index historical addresses and streets from the following two datasets:

• nyc-streets
• addresses

To use nyc-historical-geocoder, you must download these two datasets, set the datasetDir configuration option to the path of the directory containing these datasets.

For more information on those two datasets, see our tutorial on historical addresses.

Usage & Installation

In Node.js

First, install the geocoder:

npm install nypl-spacetime/nyc-historical-geocoder

Example code:

const Geocoder = require('@spacetime/nyc-historical-geocoder')

const config = {
  datasetDir: '/path/to/datasets',
  borough: 'Brooklyn', // Default: 'Manhattan'
  streets: 'nyc-streets', // Default: 'nyc-streets'
  addresses: 'addresses' // Default: 'addresses'
}

Geocoder(config)
  .then((geocoder) => {
    const result = geocoder('34 Talman Street')
    console.log(result)
  })
  .catch((err) => {
    console.error(err)
  })

From the command line

First, install the geocoder:

npm install -g nypl-spacetime/nyc-historical-geocoder

Geocode "34 Talman Street":

nyc-historical-geocoder -b Brooklyn -d /path/to/datasets/ "34 Talman Street"

Output:

{
  "type": "Feature",
  "properties": {
    "input": "34 Talman",
    "borough": "Brooklyn",
    "street": {
      "id": "nyc-streets/860-talman-street",
      "validSince": 1855,
      "validUntil": 1855,
      "name": "Talman Street"
    },
    "address": {
      "id": "addresses/143255-1",
      "name": "34 Talman Street",
      "validSince": 1855,
      "validUntil": 1855
    }
  },
  "geometry": {
    "type": "Point",
    "coordinates": [
      -73.98520812392233,
      40.70100675808166
    ]
  }
}

Command line options:

• Dataset directory: --datasetDir, -d
• Borough: --borough, -b (default: Manhattan)
• Streets dataset: --streets, -s (default: nyc-streets)
• Addresses dataset: --addresses, -a (default: addresses)