I understand what you meant. At least the "let's get EE 10 tests to run with TomEE 9 and then start upgrading". This is why I reverted some changes.
At the same time, while you were working on it, I started doing a list of everything and trying to determine an order to avoid wait times and favor parallelism. This is why I started on APIs, Geronimo, BVal, and whatever dependencies we have. But I've been building every day multiple times and running parts of the server tests many times a day. So I'm pretty confident. I used the approach of running the tests and then doing changes and running again to see what broke and what needs to be fixed. Of course the PR is growing and will probably be hard to review, but I'm pretty confident we can get it stable. If not, then at least APIs and libraries are already well advanced and I'm sure we can use that big branch to split and push small changes to main if we don't want to merge it all together. Sorry if it was not the way you were seeing things. -- Jean-Louis Monteiro http://twitter.com/jlouismonteiro http://www.tomitribe.com On Wed, Feb 8, 2023 at 10:17 PM David Blevins <david.blev...@gmail.com> wrote: > > On Feb 6, 2023, at 7:39 AM, Jean-Louis Monteiro < > jlmonte...@tomitribe.com> wrote: > > > > Jon did the merge into our Jakarta branch, so we can close the PR now > > I just saw that branch. The idea I had of not merging to main was so we > could first get test coverage established and then start doing small > changes, bit by bit, so we can avoid pushing a big set of changes that > result in a perpetually broken build. > > I don’t really see how we can ever merge #1005 and not have that exact > outcome. > > How do we see the changes in #1005 reaching main? If there are test > failures, who fixes them since each person's individual PR is now closed > and they consider their work done? How would we even know whose changes > broke what? > > > -David > >
