Data connector#

The data connector fills a NAMES LEGAL site from data that already exists elsewhere, without code on the site owner's side. The owner connects a source in their console (Integrations → Import my data), links columns to fields with drag and drop, runs a trial, and launches. Sources can be files or an API — yours.

This page is for developers of applications that want to be a source: what NAMES LEGAL expects from your API, and how to make the connection automatic.

How the connector reads your API#

The site owner enters:

• the address of your API,
• an authentication mode and its secret,
• optionally, where the list is in your answer (data.items).

NAMES LEGAL then calls your API over HTTPS with GET, follows pagination, and turns every object into a line. Nested objects are flattened (author.name), lists of plain values are joined with commas.

Authentication#

Secrets are stored encrypted and are never shown again in the console. Give site owners a read-only key limited to published content: the connector only reads.

Finding the list#

If no path is given, the connector takes, in order: the items array, then the first array of objects found in the answer. A bare JSON array works too.

Pagination#

Any of these is followed automatically, up to 200 pages:

• an HTTP Link: <…>; rel="next" header;
• a next-page URL in the body: next, next_url, next_page_url, links.next, meta.next, pagination.next;
• a cursor in the body: next_cursor, meta.next_cursor, pagination.next_cursor — sent back as ?cursor=.

Stop by returning an empty list or no next link.

Limits#

• 5,000 lines per data set and synchronisation.
• 10 MB per answer, 15 seconds per request.
• Addresses must be public: private and local network addresses are refused, redirections included. On a redirection to another host, your key is not forwarded.

The NAMES LEGAL feed format#

Expose your data in this format and the site owner has nothing to link: each collection becomes a data set already connected to the matching section of the site.

1. An index#

GET https://app.example.com/api/names-legal/

{
  "names_legal_feed": 1,
  "collections": {
    "team": "https://app.example.com/api/names-legal/team/",
    "services": "https://app.example.com/api/names-legal/programmes/",
    "blog": "https://app.example.com/api/names-legal/news/"
  }
}

Collection names are the section keys listed in the field reference below (team, services, blog, events, faq, jobs…). URLs may be relative to the index.

2. One list per collection#

{
  "items": [
    {
      "id": "T-104",
      "first_name": "Awa",
      "last_name": "Diop",
      "position": "Mathematics teacher",
      "short_bio": "Twelve years of teaching.",
      "profile_image": "https://app.example.com/media/staff/104.jpg",
      "email": "a.diop@example.com"
    }
  ],
  "next": null
}

Rules:

• id — a stable identifier, different on every line. It is how an edited line updates the same entry instead of creating a new one.
• Field names are those of the field reference.
• Translated fields take one key per language: title_fr, title_en, description_de… A key without suffix goes to the site's default language.
• Images and files are public URLs. They are downloaded once and downloaded again only when the URL changes. Google Drive and Dropbox share links are accepted.
• Categories are given by name ("category": "Bachelor"); missing categories are created.
• Dates are ISO 8601. Booleans are true/false.
• Rich text fields accept HTML; it is cleaned (no scripts, no event handlers). Plain text becomes paragraphs.
• Only return what may be published: drafts and private data should not appear in the feed.

Keeping the site up to date#

The site owner chooses, per section:

They also choose what happens when an entry changed on both sides (your data wins, the site wins, or ask), and when a line disappears from your feed (delete on the site, hide, or keep). A synchronisation that would remove a large share of a section stops and asks for confirmation first.

Writing back to your application ("my site → my source") is not available yet: the connector only reads.

Field reference#

Generated from the platform itself; it is always up to date. Required fields must be present on every line (for a translated field, in at least one language).

Team team

Services services

Projects projects

Blog / news blog

Events events

FAQ faq

Job openings jobs

Resources resources

Publications publications

Testimonials testimonials

Skills skills

Key figures stats