Pagination

Every endpoint that returns a list, such as GET/visits, uses cursor pagination. You ask for a page, and each page tells you whether there's another and how to get it.

RequestCopy Anchor Link

Keep every other parameter the same from page to page. A cursor belongs to the query that produced it, so don't reuse one with different filters.

ResponseCopy Anchor Link

{
  "data": [{ "id": "visit_01m4k5z4j0fxbte6sn6e8tpgza", "status": "checked_in", "…": "…" }],
  "next_cursor": "b3BhcXVlLWN1cnNvcg",
  "has_more": true
}

Cursors are opaque. Don't parse or build them, and don't store them for later: use them to page through results now.

A limit outside 1 to 100 returns validation_failed with the location query.limit.

OrderCopy Anchor Link

Most lists are newest first, by created_at. A few use an order that suits them better:

Because a cursor marks a position in that order, items created while you're paging through a newest-first list land before your cursor. You won't see them twice or skip older ones, but you won't see the new ones either. To keep a copy in step, see keeping in sync.

Fetch every pageCopy Anchor Link

Use limit=100 when you want everything, and loop until has_more is false.

cursor=""
while :; do
  args=(-sS --fail-with-body -G "https://api.agoo.ardent.africa/v1/visits"
    -H "Authorization: Bearer $AGOO_API_KEY"
    --data-urlencode "status=checked_in"
    --data-urlencode "limit=100")
  [ -n "$cursor" ] && args+=(--data-urlencode "cursor=$cursor")

  page=$(curl "${args[@]}") || { echo "$page" >&2; exit 1; }
  echo "$page" | jq -r '.data[] | "\(.id)  \(.visitor.name)"'

  [ "$(echo "$page" | jq -r '.has_more')" = "true" ] || break
  cursor=$(echo "$page" | jq -r '.next_cursor')
done

A long loop makes many requests in a row. Watch the rate limit headers, and narrow the query with filters such as site_id or created_after rather than fetching everything and filtering yourself.

Keeping in syncCopy Anchor Link

To keep your own copy of visits, bookings or attendance up to date:

1. Backfill once by paging through the list.
2. Follow changes with webhooks, which tell you about creations and every status change as they happen.
3. Reconcile now and then, for example nightly, with created_after set to the last time you reconciled, or by reading GET/events (which keeps 30 days of events) to catch anything your endpoint missed.

RelatedCopy Anchor Link