Getting started

The NVDB API is a REST api with extensive search- and filter capabilities, as well as options for fine tuning which data elements to include in the respons.

Vegkart is your best tool for digging into NVDB data, getting famiiar with the data structure and content. The CSV download capabilities of Vegkart is also sufficient for many use cases - but vegkart is also an excellent training ground for API queries: Vegkart provides the link to the exact same API query that Vegkart uses to fetch data for you. This is an excellent starting point for learning the API functionality.

This article (in Norwegian) also gives an overview and recipies for how to get NVDB data into your mapping application / GIS.

Below, we give sufficient details of how to communicate with the API. We also highly recomend that you read these pages:


The response formats from the API may be either XML or JSON (default), which you chooose using the standard http accept header. The format is versioned, and to ensure your application does not break upon format improvements we highly recomend you explsit control the version number, using this syntax:

Accept: application/vnd.vegvesen.nvdb-v3-rev1+json
Accept: application/vnd.vegvesen.nvdb-v3-rev1+xml

The standard media types application/json and ` application/xml` will always return the newest format version.

Tracking requests and sessions

At least one of these http headers must be set by the client:

"X-Client" :  "your@email", or other identification
"User-Agent" : "Name of your fantastic system" 

We really, really appreciate if at least one of these, preferably both, are given meaningful names. It gives us valuable information about who uses our API, and for what purpose. In addition, we also recommend (and appreciate) that you track your sessions using the X-Client-Session set to something uniqe per session, such as an UUID or something similar.

The API response contains the header X-REQUEST-ID, which is useful if you need to report a problem.


The response contains two parts:

The metadata describes how many objects satisfying your query (antall), the number returned with this page (returnert), the number of elements per page (sidestørrelse) and the link (href) to the next page of data. To fetch all data you need to iterate through all pages (following the href link on each page) untill you get an empty page (i.e. the list objekter is empty and the metadata element returnert = 0.

{  "objekter": [ { "long" : "list" }, { "of" : "objects" } ], 
  "metadata": {
    "antall": 1001, 
    "returnert": 1000,
    "sidestørrelse": 1000,
    "neste": {
      "start": "AEhZez1g4yb",
      "href": "/api-endpoint?start=AEhZez1g4yb"