Pagination¶

Pagination uses the page query namespace.

Page size¶

GET /customers?page[size]=10

Page number¶

GET /customers?page[number]=2

Combined¶

GET /customers?page[size]=10&page[number]=2

Cursor pagination¶

GET /customers?page[cursor]=100&page[size]=10

Cursor pagination uses keyset-style paging by resource id and returns links.next when more rows are available.

import httpx

response = httpx.get(
    "http://localhost:8000/api/customers",
    params={"page[size]": 10, "page[number]": 2},
)
payload = response.json()
print(payload.get("meta"))

Typical list response contains pagination metadata:

{
    "data": [...],
    "links": {
        "self": "http://localhost:8000/api/customers?page%5Bsize%5D=10&page%5Bnumber%5D=2",
        "first": "http://localhost:8000/api/customers?page%5Bnumber%5D=1&page%5Bsize%5D=10",
        "last": "http://localhost:8000/api/customers?page%5Bnumber%5D=4&page%5Bsize%5D=10",
        "prev": "http://localhost:8000/api/customers?page%5Bnumber%5D=1&page%5Bsize%5D=10",
        "next": "http://localhost:8000/api/customers?page%5Bnumber%5D=3&page%5Bsize%5D=10"
    },
    "meta": {
        "count": 10,
        "totalPages": 4
    }
}

Disable pagination¶

Depending on configuration, pagination can be disabled with:

GET /customers?page[size]=0

This only works when NINJA_JSONAPI["ALLOW_DISABLE_PAGINATION"] is True. When ALLOW_DISABLE_PAGINATION is False, page[size]=0 falls back to the default page size.