>> >> How is it not? The latest VM works better than stable? >> On OSX: >> - backspace works >> - JIT bug disappeared >> >> Which are 2 things that do not work in the stable version. >> What is missing? > > - acceptance tests > - the team feeling comfortable with it ;) > > I think we can promote the current one to stable, BUT we need to keep the old > stables somewhere (just in case) > > Camillo, could you do the move?
=> updated the Acceptance tests => updated the performance tests => updated the tests --------------------------------------------------------------------------------- PharoS Tests Pharo 2.0 ---------------------- >> https://ci.inria.fr/pharo/view/3.0-VM/job/PharoSVM-tests/lastCompletedBuild/testReport/ >> looks green, except for an obsolete class test, which only happens under 1 platform and another windows related timeout that happens irregularly. (Going to comment those 2 tests out). --------=---------------------------------- PharoSVM Acceptance Tests (Moose + Seaside) --------=---------------------------------- >> https://ci.inria.fr/pharo/view/3.0-VM/job/PharoSVM-AcceptanceTests/lastCompletedBuild/testReport/ With PharoS and 2.0 it looks like 2 Moose tests from the #stable configuration fail: Moose.Test.HismoImporter.MooseHismoImporterTest.testEvery15Days Moose.Test.HismoImporter.MooseHismoImporterTest.testEvery2Years Moose.Test.HismoImporter.MooseHismoImporterTest.testEvery6Months all of them are DNUs: MooseHismoImporter(Object)>>doesNotUnderstand: #every2Years --------------------------------------------------------------------------------- PharoVM Tests Pharo 2.0 ----------------------- >> https://ci.inria.fr/pharo/view/3.0-VM/job/PharoVM-tests/ => all green except for 3 tests delivering strange results under ubuntu 13.04 related to non-flushed streams ------------------------------------------ PharoVM Acceptance Tests (Moose + Seaside) ------------------------------------------ >> https://ci.inria.fr/pharo/view/3.0-VM/job/PharoVM-AcceptanceTest/lastCompletedBuild/testReport/ - same 3 moose tests failing on all platforms as for PharoS - mac slave has an integer (jit?) bug when loading the configuration of seaside
signature.asc
Description: Message signed with OpenPGP using GPGMail
