How can you design an API to retrieve addresses with efficient pagination and filtering?

• A. Use REST with limit/offset or cursor-based pagination, support filters on country, city, postal_code, address_type, and a search query, and provide total counts and sorting options.
• B. Use XML with page tokens and heavy nested structures.
• C. Use a single endpoint returning all addresses with no pagination.
• D. Use REST but only allow page numbers with fixed page sizes.

Answer: (B)

Efficient pagination and filtering in an API come from combining a scalable paging method with targeted query controls. For retrieving addresses, a RESTful approach that supports either limit/offset or cursor-based pagination keeps responses small and predictable while allowing clients to move through results efficiently. Limit/offset is simple to implement: the client requests a page by specifying how many items to return and where to start. Cursor-based pagination uses an opaque token that encodes the position, which tends to be more robust as data changes and scales better for large result sets. Either way, you should enable filters on fields like country, city, postal_code, and address_type, plus a search query for free-form matching. This makes it possible to narrow down results quickly and keeps the database work sensible through proper indexing. Sorting should be supported with a deterministic default order (for example by id or created_at) so pages remain stable as items are added or removed. The API response should include the current page of results, a next-page token or offset for the subsequent request, and the total_count so clients can render accurate pagination UI. XML with heavy nesting tends to be heavier and less practical for high-traffic APIs, returning all addresses at once defeats the purpose of pagination, and restricting to fixed page numbers reduces flexibility, so these options don’t fit real-world use as cleanly.