Thanks for comments. Seems that there are no disagreements to mark SelectQuery as deprecated. So I'll merge this PR.
On Thu, Jul 4, 2019 at 11:42 AM Andrus Adamchik <and...@objectstyle.org> wrote: > As deprecating SelectQuery has been on my TODO list for some time, let me > share my thoughts... The motivation for the change is of course that we > don't want to support two styles of queries at the same time. We are not a > big team, and having the smallest possible maintenance footprint is > important. There are other candidates for deprecation in the core API > (ProcedureQuery, EJBQLQuery, SQLTemplate, and eventually - ROP). But > SelectQuery is a good starting point, cause ObjectSelect is a fully capable > replacement. > > Doesn't look like we have an argument against *deprecation*, so it would > be great to do it ASAP, to announce loud and clear that the clock is > ticking. We already missed this opportunity in 4.1, so I'd like to make it > happen in 4.2. > > Now *removal* is another story, I agree. I don't think the LOE to switch > is that great. Even the biggest codebase that I have is probably going to > take me no more than 3-4 days to refactor. But still this was the core > Cayenne API for many years, so the impact is non-trivial. > > So for now I suggest that we do the deprecation, and make the removal > decision later (same way we did with Java upgrades and other breaking > changes). I am thinking 5.0 release might be a good time and place to do > it. So maybe not in 10 years, but in 2-3 years... But we'll review that > again. > > Andrus > > > > On Jul 4, 2019, at 8:45 AM, Aristedes Maniatis <a...@maniatis.org> wrote: > > > > > > On 28/6/19 11:47pm, Arseni Bulatski wrote: > >> Hi all, > >> > >> SelectQuery is planned to be deprecated in 4.2. > >> ObjectSelect has better api and is recommended to use. > >> > >> PR: https://issues.apache.org/jira/browse/CAY-2523 > >> > >> Please, write your thoughts about it. > > > > > > This is a pretty big change for many people. They would either not be > able to upgrade to 4.3 ever, or else spend potentially a huge amount of > time refactoring code. For my project, I'm guessing one person 4 weeks. > > > > Or are you suggesting it be marked as deprecated, but kept around until > some important reason comes up to remove it. Perhaps 10 years from now when > it is mostly forgotten. > > > > > > Ari > > > >
