This is a test version of the portal. Content and APIs are under development and may change without notice.

Code examples

Ready-made snippets showing how to call the APIs and download datasets. For interactive per-endpoint documentation, see the API reference.

Replace YOUR_KEY with a real key from the front page.

The URLs point to the test gateway. On the production portal, api.test.data.regjeringen.no is replaced with api.data.regjeringen.no.

Call an API with your key

APIs that require a key are read with the Ocp-Apim-Subscription-Key header. Get a key on the front page and replace YOUR_KEY below. language is required.

curl

curl -H "Ocp-Apim-Subscription-Key: YOUR_KEY" \
  "https://api.test.data.regjeringen.no/content/GetDocList?language=nb&pageSize=10"

JavaScript

const res = await fetch(
  "https://api.test.data.regjeringen.no/content/GetDocList?language=nb&pageSize=10",
  { headers: { "Ocp-Apim-Subscription-Key": "YOUR_KEY" } },
);
const docs = await res.json();
console.log(docs);

Python

import requests

res = requests.get(
    "https://api.test.data.regjeringen.no/content/GetDocList",
    params={"language": "nb", "pageSize": 10},
    headers={"Ocp-Apim-Subscription-Key": "YOUR_KEY"},
)
res.raise_for_status()
docs = res.json()
print(docs)

Download a dataset (no key)

The nightly exports are open and served gzipped. Use --compressed so curl decompresses automatically.

curl

curl --compressed -o content.json \
  https://api.test.data.regjeringen.no/exports/content.json

JavaScript

// fetch() transparently decodes gzip.
const res = await fetch("https://api.test.data.regjeringen.no/exports/content.json");
const documents = await res.json();
console.log(documents.length, "documents");

Python

import requests

# requests transparently decodes gzip.
res = requests.get("https://api.test.data.regjeringen.no/exports/content.json")
res.raise_for_status()
documents = res.json()
print(len(documents), "documents")

Paginate and aggregate documents

A fuller example: fetch every press release in a date range by following the pagination (page/pageSize), then count them per ministry. Use showOwner=true to include the owner.

curl

#!/usr/bin/env bash
# Fetch every press release from 2024, page by page (needs jq).
key="YOUR_KEY"
base="https://api.test.data.regjeringen.no/content/GetDocList"
query="language=nb&type=pressemelding&fromDate=2024.01.01&toDate=2024.12.31&showOwner=true&pageSize=100"

page=1
while :; do
  batch=$(curl -s -H "Ocp-Apim-Subscription-Key: $key" "$base?$query&page=$page")
  # Documents live under data.items; the owner is an object (use .owner.depName).
  echo "$batch" | jq -c '.data.items[] | { title, date, ministry: .owner.depName }'
  [ "$page" -ge "$(echo "$batch" | jq '.data.pager.totalPages')" ] && break
  page=$((page + 1))
done

JavaScript

const KEY = "YOUR_KEY";
const BASE = "https://api.test.data.regjeringen.no/content/GetDocList";

// Follow the pager until the last page is reached.
async function fetchAll(params) {
  const all = [];
  for (let page = 1; ; page++) {
    const url = new URL(BASE);
    url.search = new URLSearchParams({ ...params, page, pageSize: "100" });
    const res = await fetch(url, {
      headers: { "Ocp-Apim-Subscription-Key": KEY },
    });
    if (!res.ok) throw new Error(`HTTP ${res.status}`);
    // Documents are under data.items; data.pager.totalPages drives pagination.
    const { data } = await res.json();
    all.push(...data.items);
    if (page >= data.pager.totalPages) break;
  }
  return all;
}

const docs = await fetchAll({
  language: "nb",
  type: "pressemelding",
  fromDate: "2024.01.01",
  toDate: "2024.12.31",
  showOwner: "true",
});

// Count press releases per owning ministry (owner is an object).
const byMinistry = {};
for (const doc of docs) {
  const ministry = doc.owner?.depName ?? "(ukjent)";
  byMinistry[ministry] = (byMinistry[ministry] ?? 0) + 1;
}
console.table(byMinistry);

Python

import requests
from collections import Counter

KEY = "YOUR_KEY"
BASE = "https://api.test.data.regjeringen.no/content/GetDocList"


def fetch_all(**params):
    """Follow pagination until a short (final) page is returned."""
    documents = []
    page = 1
    while True:
        res = requests.get(
            BASE,
            params={**params, "page": page, "pageSize": 100},
            headers={"Ocp-Apim-Subscription-Key": KEY},
        )
        res.raise_for_status()
        # Documents are under data.items; data.pager.totalPages drives pagination.
        data = res.json()["data"]
        documents.extend(data["items"])
        if page >= data["pager"]["totalPages"]:
            break
        page += 1
    return documents


docs = fetch_all(
    language="nb",
    type="pressemelding",
    fromDate="2024.01.01",
    toDate="2024.12.31",
    showOwner="true",
)

# Count press releases per owning ministry (owner is an object).
by_ministry = Counter(
    (doc.get("owner") or {}).get("depName", "(ukjent)") for doc in docs
)
for ministry, count in by_ministry.most_common():
    print(f"{count:4}  {ministry}")