Nathan, On 5/11/2011 5:15 PM, Nathan Bubna wrote: > A lot of this work was done already between 1.5 and 1.6, as i recall. > I could have sworn you participated in some of those discussion too.
Maybe... I'll have to look at the mailing list archives. :) > I know a lot of work was done in 1.6 on both memory and CPU > performance. Macros were moved into a common space, instead of being > duplicated for every use, we do use String interning judiciously in > some places, with proven results at the time. In any case, i strongly > recommend you run your profiler against 1.7 or 2.0-dev and not get > excited about 1.4 performance issues. That's the primary reason I mentioned up front that I was using 1.4. I figured someone would say "you're an idiot who needs to upgrade" and that would essentially be that. :) I think I'm going to bite the bullet after our current release cycle and just straight-up replace Velocity 1.4 with Velocity 1.7 and Velocity-Tools 1.4 with Velocity-Tools 2.0 and then just test the hell out of everything. It's lonely here in the past :) At that point, I can become relevant again in terms of supporting this project directly. :) > sorry, but i just don't care about anything pre-1.7 anymore. i don't > have the bandwidth. :( Nor should you. There's really no reason why an upgrade shouldn't occur. It would be especially nice for us to be able to do non-integer mathematics in templates (!!), more intelligent object equality-checking, etc. so I'm definitely looking forward to upgrading. Performance optimizations are a great bonus, too! Thanks, -chris
signature.asc
Description: OpenPGP digital signature
