- From: Tomasz Pluskiewicz <tomasz@t-code.pl>
- Date: Tue, 09 Feb 2016 11:47:56 +0000
- To: "Markus Lanthaler" <markus.lanthaler@gmx.net>, public-hydra@w3.org
February 8 2016 9:46 PM, "Markus Lanthaler" <markus.lanthaler@gmx.net> wrote: >>> I'm just a bit uneasy about multiple PartialCollectionViews, each with >>> it's own totalItems. > > Yeah, I find that a bit confusing as well but in order to make this unambiguous to a machine we > need them I guess. > Do we? Please exaplain >>> I think that there should only be one such view in >>> a response. > > How would you communicate to a client then that the response was not only filtered, but that the > server also decided to paginate it? > Does it matter to the client? The client should only care that there are more pages (linked) and how to possibly construct different views. >>> In my opinion, if GETting /markus/friends?first=Ruben >>> redirects (or uses Content-Location header) than the view is actually >>> the first page and not the initial target. > > While that might work for the first page it would make it more difficult to get the total number of > filtered items. Also, how would that work for page 2, 3 and so on? > I don't understand. If /markus/friends?first=Ruben has a totalNumber = 20 and the server redirects to /markus/friends?first=Ruben&page=1, which contains 5 items, this doesn't change the fact that the total number of items is still twenty. >>> Also I'm not sure about mixing both ViewTemplate and View on single >>> `view` property. > > We can certainly split it into two properties but I'm not sure that would buy us much. What > advantages would you see? As a client I wouldn't have to inspect each individual element of 'view' to determine whether it's a View or a ViewTemplate. Simple as that, Separation of Concerns, and thus simpler client implementation.
Received on Tuesday, 9 February 2016 11:48:51 UTC