On 05/03/2012 12:56 AM, Jacques Le Roux wrote: > From: "Adam Heath" <doo...@brainfood.com> >> On 04/27/2012 02:50 AM, Jacques Le Roux wrote: >>> Reverted, it did not compile, thanks to Amit Sharma who called me >> >> Sometimes it's not possible to backport a fix for a file bug; there >> can be lots of other dependant patches that are required, that are too >> unstable(or just not wanted) for a released branch. >> >> However, if some people really do want a feature backported, then they >> can 'vote' or so for it. >> >> I'd be willing to backport this, as the original version this change >> was developed on was a local version of ofbiz from January, 2010. > > +1
To be fair, I generally only want to see bug-fixes backported to public release branches(internal organization back-branches can have other guidelines). However, I strongly believe that this particular feature actually allows almost *all* raw sql to no longer be required. In fact, the only raw sql this client has now(they didn't pay me to fix this problem, I just did it while on their watch), is low-level full-text searching for postgresql and oracle. But even then, I only do it thru an anonymous ComplexAliasMember(which I should probable extend into entitymodel at some point). ps: backporting this change could be tricky; there are other dependant patches needed, which aren't really features or fixes.
