--
You received this message because you are subscribed to the Google Groups "ravendb" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ravendb+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
so the only way to do pagination when the auth bundle is in place is to do 2 queries?
one for the data and one for the total count?
So what we can do to this problem?
--
You received this message because you are subscribed to the Google Groups "RavenDB - 2nd generation document database" group.
So, how can we achieve paging when the auth bundle is in place? Or should we get rid of the auth bundle itself?
Sent from my Windows Phone, typos are guaranteed ;-)
Sent: 15/11/2013 20.46
So, just to recap, if we have the Auth Bundle in place:
- “Total Results” cannot be trusted;
- The consequence is that we cannot know upfront how many pages we have;
The question now is: if we do a “Skip( x ).Take ( y )” are we sure that we get back “y” docs? if there are at least y, obviously, or the Auth Bundle runs after paging is applied server side thus we cannot even trust the page size?
To me the total count missing/unreliable is not a big deal, I’m not exposing paging to the end user because we have everywhere “infinite scrolling”, but I need to be sure that if a request a x docs I get back x docs not “x – the one I cannot see” otherwise even infinite scrolling can become a mess.
.m