TypeScript SDK integration

Extract your first page

Set GEONODE_SCRAPER_API_KEY in your server environment. This ES module example uses top-level await: run it as an .mjs file, or use a TypeScript runtime configured for ES modules. Keep the key in server code.

import {
  Configuration, ExtractionApi, OutputFormat,
} from "@geonode_scraper/scraper-sdk";

const apiKey = process.env.GEONODE_SCRAPER_API_KEY;
if (!apiKey) throw new Error("Set GEONODE_SCRAPER_API_KEY");

const configuration = new Configuration({ apiKey });
const response = await new ExtractionApi(configuration).extractV1ExtractPost({
  extractRequest: {
    url: "https://example.com",
    formats: [OutputFormat.Markdown],
  },
});

if ("data" in response) {
  console.log(response.data.markdown);
  console.log(response.tokensCharged);
}

On success, data.markdown contains the extracted page and tokensCharged reports the request cost. The data check narrows the response type; background requests return a job instead.

Move from one page to a workflow

1. BatchApi submits a list of URLs; CrawlApi follows links within your depth and page limits.
2. MapApi discovers a site's URLs, and SearchApi returns ranked search results.
3. For background work, save jobId and inspect the terminal job status. Bound polling by a deadline and handle failures or cancellation.
4. Use page and pageSize to retrieve all pages of batch or crawl results.

Keep API access on your server

Use the full scoped package name in imports. The SDK can run in a browser, but embedding your Geonode API key there exposes it to visitors. Call it from a backend route, worker or other server process, then return only the data your frontend needs.