Both Kalamar as well as RKorAPClient cache the hit count for searches, so reissuing the search with a different pagination don't need to fetch again all data (and can effectively work just like "Glimpse"). While this should in the future be part of Krawfish, an implementation in Kustvakt would immediately be beneficial, because it a) could cache the result for API requests as well and b) could work after the query rewrite (which is different than how Kalamar works) and therefore be able to cache the hitcount independently of the current user and its permissions.
Both Kalamar as well as RKorAPClient cache the hit count for searches, so reissuing the search with a different pagination don't need to fetch again all data (and can effectively work just like "Glimpse"). While this should in the future be part of Krawfish, an implementation in Kustvakt would immediately be beneficial, because it a) could cache the result for API requests as well and b) could work after the query rewrite (which is different than how Kalamar works) and therefore be able to cache the hitcount independently of the current user and its permissions.