On 07/09/2011, at 11:57 PM, Hani Suleiman wrote: > > On Sep 6, 2011, at 4:33 AM, Adam Murdoch wrote: > >> WHARF-25 and WHARF-27 are fixed, and these cover the more nasty of the >> regressions. We're waiting for fixes for WHARF-26, WHARF-35 and WHARF-36 >> before we do a release. In the meantime, we're looking at some workarounds >> to reduce the impact of these issues, in which case we might be able to >> release without the fixes. >> >> There's also a fix for GRADLE-1699/1701 ready to push. >> >> Once we have fixes/workarounds for the other wharf issues, we'll do a >> snapshot build, and try it out on the Gradle build for a few days. If all is >> ok, we'll release milestone-5. We might also advertise the snapshot build on >> the mailing list, if people would like to try it out. >> > How about all these other performance killing bugs? > > https://issues.jfrog.org/jira/browse/WHARF-31 > https://issues.jfrog.org/jira/browse/WHARF-30 > https://issues.jfrog.org/jira/browse/WHARF-29 > https://issues.jfrog.org/jira/browse/WHARF-28
These are actually problems in Ivy, and were present in previous Gradle releases. So, we won't wait for these to get fixed in order to release milestone-5. We'll just aim for a milestone-5 that is no worse than milestone-3. However, fixing dependency resolution performance is one of the main things we are working on for Gradle 1.0-rc-1, so most of these issues should end up fixed. The first target is to make sure that we never hit the network if we already have everything cached (WHARF-31). The second target will be to more efficient in finding and downloading stuff that we don't have cached. We may or may not get all of the second piece of work finished for Gradle 1.0, but we should be in a situation to easily fix them post-1.0. -- Adam Murdoch Gradle Co-founder http://www.gradle.org VP of Engineering, Gradleware Inc. - Gradle Training, Support, Consulting http://www.gradleware.com
