Not stable CI tests for @next version of TypeORM

Issue type:

• bug report

TypeORM version:

• @next

Not stable CI tests

[Kononnable]

Skipping failing test is a bad idea(in general). We would forget about them in the future and they wouldn’t be tested(or fixed) at all. In local development I sometimes skip broken tests and enable just before pushing the branch(or skip entire db driver if most of tests are failing).

As for 0.3.0 - it’s a tricky subject. It is alive for a year now but there are still some changes there to be made and for some of them we have not agreed on implementation(yet). So if you’ve got a time and doesn’t have to implement a specific feature it might be better to avoid it. It is mostly completed by there is a chance you would spend a day of two on understanding the code which would be changed a day later.

When fixing bugs in 0.2 you could check if they’re not working on 0.3(if it’s just a matter of quick look) and prioritize those not running in both versions. I would advice you to start in one area(issue label) and pick bugs from there. It’s much easier to fix a bug when you’re already familiar with code of that fragment of typeorm. If you’d like you can also review PRs. It can save a lot of time if someone a bit familiar with our codebase and contributing guidelines reviews PRs before @pleerock checks if it can be merged. Often only tiny problems are blocking PRs, but they have to wait 2-3 more weeks because @pleerock doesn’t have enough time to check them on a daily basis(so if he is the only one reviewing them this process may take some time).

[vlapo]

@YegorZaremba I think the best is fix bugs for now. Check issues here and pick one 😉