> > IVY-387 Absolute and relative path > IVY-232 Incorrect directory path resolve when running from a different > directory > => These two are rather old, assigned to you Gilles. Any progress on > these? Do you need help? >
Both are actually the same now. The IVY-387 was an aggregation of different absolute and relative path issues. All have been solved except IVY-232 which was a bigger change. [1] I think Matt may already has a patch, he just need some help to test. > IVY-872 Improve performance > => This one is new and assigned to you Gilles. I don't think this should > be a show stopper for 2.0, and can be postponed to 2.0.1 if it takes too > long to implement > Correct. Most performance issues that I have observed are now fixed. I just have one pending. I think that (at least on my benchmark project) there is still too much ivy file parsing. It seems that the number of ivy file parsing grows with the size of the repository (or with the size of the cache) which I didn't explain yet and is a source of poor performances. This is the last point that I plan to fix. After I will stop. As it is not a show stopper this can be postponed to 2.0.1 if it is not done for the RC-1. release. > So I think we're pretty close to be able to enter in 2.0 release candidates > cycles, to finally get 2.0 out! Do you see any other outstanding issue which > should get in 2.0? Would you agree with a plan trying to get 2.0 RC1 out > before mid september? Then how do you see the release candidates cycles > going on? I'd be in favour of trying to keep the cycles short (sg like every > 2 weeks), and if no outstanding bug is reported in a cycle, then we release > 2.0 final with the same source code as the last RC. What do you think of > this plan? +1. I also think that once the RC-1 is released, we should start the 2.0 branch (or 2.0.0). Keeping the trunk for changes that will be for 2.0.1 (for example possibly the end of my IVY-872), and the branch only for 2.0-RC show stoppers. -- Gilles Scokart --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
