Pagination, filtering & formats

#Pagination

List endpoints return up to limit items (default 50, maximum 500):

{
  "data": [ … ],
  "meta": { "total": 199, "limit": 50, "offset": 0, "…": "…" },
  "links": {
    "self": "https://api.sentinara.org/v1/countries?limit=50",
    "next": "https://api.sentinara.org/v1/countries?limit=50&cursor=eyJvIjo1MH0"
  }
}

Follow links.next until it is null. Treat the cursor as opaque.

url = "https://api.sentinara.org/v1/countries?limit=100"
rows = []
while url:
    page = requests.get(url, timeout=30).json()
    rows += page["data"]
    url = page["links"]["next"]

#Filtering and sorting

Each endpoint documents its filters in the reference. Common ones:

Strict parameters: an unknown or repeated parameter returns 400, so a typo such as regoin=SEAR fails loudly instead of returning unfiltered data.

#CSV

Any list endpoint returns CSV with ?format=csv or the header Accept: text/csv. Nested fields are flattened with dots (year_to_date.cases) and arrays are joined with |. Text cells that start with =, +, - or @ are prefixed with ' to prevent spreadsheet formula injection. A Link: <…>; rel="next" header points to the next page.

#Conditional requests

Every successful response has an ETag. Send it back to avoid downloading unchanged data:

curl -i https://api.sentinara.org/v1/indonesia -H 'If-None-Match: W/"2cb60518aca9db1126636b2ea40055c5"'
# HTTP/1.1 304 Not Modified

Cache-Control is public, max-age=300 without a key and private, max-age=300 with one.

#Identifiers