Bummer, now we all have to take a similar approach. Okay, it makes sense.
~ Thank you for the clarification and keep up the good work :) Paul Bors -----Original Message----- From: Igor Vaynberg [mailto:igor.vaynb...@gmail.com] Sent: Friday, May 03, 2013 12:12 PM To: users@wicket.apache.org Subject: Re: Wicket 6.x and Repeaters API change to use longs yes. when i was looking at this i thought about making the dataprovider paging param ints, but then the long return type didnt make sense. since count() returns a long its feasible to request offsets in the long range - eg user presses last page and you actually have more rows then an int can hold. so rather then going with a mixed approach that made the code base a big mess i went with the biggest type everywhere, which is a long. what i did at my day job is to create an entity data provider that handles all these inconsistencies and castings and exposes a consistent api to how our application works... -igor On Fri, May 3, 2013 at 8:58 AM, Paul Bors <p...@bors.ws> wrote: > All good and solid, but take for example Javax's Query API: > http://docs.oracle.com/javaee/6/api/javax/persistence/Query.html#setFi > rstRes > ult(int) > > Since Wicket's DataProvider now offers the first and count as longs, > one would have to cast first to an int to pass it through the JEE 6 API, right? > Or am I missing something? > > ~ Thank you, > Paul Bors > > -----Original Message----- > From: Igor Vaynberg [mailto:igor.vaynb...@gmail.com] > Sent: Friday, May 03, 2013 11:45 AM > To: users@wicket.apache.org > Subject: Re: Wicket 6.x and Repeaters API change to use longs > > 4.8.4 Aggregate Functions in the SELECT Clause The result of a query ... > > The Java type that is contained in the result of a query using an > aggregate function is as follows: > COUNT returns Long. > ... > > -igor > > On Fri, May 3, 2013 at 8:35 AM, Paul Bors <p...@bors.ws> wrote: >> I'm a bit confused and can't find the JPA version that uses longs >> instead of ints as mentioned in the migration guide at: >> >> https://cwiki.apache.org/WICKET/migration-to-wicket-60.html >> >> Repeaters >> >> * `IDataProvider` was converted to using `long` instead of `int` to >> better line up with JPA and other persistence frameworks. This has >> caused a rather large cascade of `int` to `long` changes all over the >> repeater packages (WICKET-1175 > <https://issues.apache.org/jira/browse/WICKET-1175> ). >> >> >> >> Could someone shed some light? I'm still using JEE 6 (as many of us) >> and I don't feel comfortable casting a bunch of longs to int in my >> DAO > code. >> >> >> >> ~ Thank you, >> >> Paul Bors >> >> >> > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
