[jdk11u-dev] RFR: 8310923: Refactor Currency tests to use JUnit

[Goetz Lindenmaier]

On Thu, 15 Feb 2024 10:17:43 GMT, Andrew John Hughes <andrew at openjdk.org> wrote:

>>> The github actions tests fail, because the junit code used in this change depends on a recent junit. GHA of 11 use old jtreg 6.1.. With 7.3.1, which contains a more recent junit, the tests would work.
>> 
>> Do we plan to transition 11u to jtreg 7? Doing so has quite a long tail of dependencies: https://bugs.openjdk.org/browse/JDK-8289798
>> 
>> I think this is being backported for [8321480: ISO 4217 Amendment 176 Update](https://github.com/openjdk/jdk17u-dev/commit/223018a3aabc319992d9790fa1d6c88b02fcca9f) because it contains some follow-on unrelated test changes from this patch.  You could do 8321480 without this patch - I think we'll have to in 8u as jtreg 7 needs JDK 11 - but I don't want to do that if we are going to bring this change in, as it will make this change unclean and risk missing the post-8310923 changes in 8321480.
>> 
>> Incidentally, this one is in 11.0.24-oracle, not 11.0.23-oracle.
>
>> Hi @gnu-andrew, Yes, we backport this because [8321480: ISO 4217 Amendment 176 Update](https://github.com/openjdk/jdk17u-dev/commit/223018a3aabc319992d9790fa1d6c88b02fcca9f) depends on it. But it is also easy to resolve 8321480 without this change, which is what Oracle did. I have been looking at raising the jtreg version. As a first step, I wanted to explore backporting the required test fixes you mention above. Actually, there are some more prerequisites. But I don't think we will achieve this by Feb 27th. So maybe we should proceed with the ISO 4217 Amendment.
> 
> Yes, I'm starting to think it would be better not to try and rush the jtreg bump and instead do that in the July cycle. I have a backport of the ISO amendment without the test change almost ready and can submit a PR later today. Let me know if I can be of any help with the jtreg bump.

Hi @gnu-andrew, 
Yes, I think we should bring JDK-8321480 to 11 now, and not wait for this change.
I think we should not omit the test changes, but resolve them. 
This change just makes them a clean backpot.
Could you please have a look at that, @luchenlin?
Thanks!

-------------

PR Comment: https://git.openjdk.org/jdk11u-dev/pull/2503#issuecomment-1952655094