can someone please clarify what exactly the status of this issue is?
it was backported to 5x by ryan, and subsequent tweaks to the API were then made to both trunk & 5x by noble, but the issue is still open and not listed anywhere in CHANGES – which is a really dangerous state to leave an issue in. Feature is commited it should ALWAYS be in CHANGES since you never know when someone will create a release branch – and if folks didn't think this was ready to be released, it definitely shouldn't have been backported to 5x.
In addition - the commits made in the last 24 hours seem to suspiciously corrolate to 4 jenkins failures on the 5x branch that have happened since then in the test that was modified...
This seed doesn't fail reliably for me, but beasting it failed on the second iteration...
ant beast -Dbeast.iters=10 -Dtestcase=TestCustomDocTransformer -Dtests.method=testCustomTransformer -Dtests.seed=C062F72E9252DCA8 -Dtests.multiplier=2 -Dtests.slow=true -Dtests.locale=hr -Dtests.timezone=America/Montserrat -Dtests.asserts=true -Dtests.file.encoding=ISO-8859-1
If i revert r1687330, that beast command passes – and still passes even if i bump up the iters and simplify the args so that we cover more random variable permutations (to allow for more possible changes the way entropy is used between the two versions of the test)
ant beast -Dbeast.iters=50 -Dtestcase=TestCustomDocTransformer -Dtests.method=testCustomTransformer -Dtests.multiplier=2 -Dtests.slow=true