Hi Paulo, I'm tempted to say:
1. we never did the full move to github so yes a jira will be needed associated with a PR maybe (sounds the least bothering) 2. grouping or not depend how easy to review it is I think, if in the same area grouping sounds great, if modifying 100 files maybe not - so no general rule but try to keep it simple? 3. if issues are fixed we can close them else we tend to keep them as a backlog for now (but we can discuss to change it) 4. think we mainly rely on JPA spec as a doc so there are room for improvements there Romain Manni-Bucau @rmannibucau <https://x.com/rmannibucau> | .NET Blog <https://dotnetbirdie.github.io/> | Blog <https://rmannibucau.github.io/> | Old Blog <http://rmannibucau.wordpress.com> | Github <https://github.com/rmannibucau> | LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book <https://www.packtpub.com/en-us/product/java-ee-8-high-performance-9781788473064> Le jeu. 5 juin 2025 à 21:41, Paulo Cristovão de Araújo Silva Filho < pcris...@apache.org> a écrit : > Hi, > > I've been tinkering with JPA 3.1 TCK and, after some fixes in > EntityManagerFactoryImpl (NPE already mentioned here before), was able to > run the full tests over 4.0.2-SNAPSHOT. Results are not bright: > > - passes 600/2074 > > - failed 1472/2074 > > - no errors > > - no skips > > The big failed errors will probably boil down to a few minor changes in > code (hopefully) and I guess this will give me some better grasp over the > codebase. > Could you experienced members please provide some guidance on how may I > proceed with the issues found. For example: > > a) should I open an issue on Jira just to deal with the NPE on startup or > make a "big" ticket for all issues tck related? > > b) should I group some minor issues? > > c) what to do with older (stale) issues that targets older versions? > > d) does documentation must have issues associated (noticed that current > documentation mentions 3.x version, when we're 4.1)? > > Anyway, please do not hesitate to ask me to step in some older or other > issues. I just wanted to have TCK in better shape to identify if some of > them are already fixed before making a big fat environment to test each one. > Best regards, > > Paulo Araújo
