Fix test error in InternalCardinalityTests#testEqualsAndHashcode #62542
Conversation
iverase
added
>test
|
Pinging @elastic/es-analytics-geo (:Analytics/Aggregations) |
elasticmachine
added
the
Team:Analytics
|
@elasticmachine run elasticsearch-ci/packaging-sample-windows |
2 similar comments
|
@elasticmachine run elasticsearch-ci/packaging-sample-windows |
|
@elasticmachine run elasticsearch-ci/packaging-sample-windows |
polyfractal
left a comment
polyfractal
left a comment
There was a problem hiding this comment.
Left a comment about a maybe-future tweak, but LGTM as it stands right now :)
| int extraValues = between(10, 100); | ||
| for (int i = 0; i < extraValues; i++) { | ||
| for (int i = 0; i < 10; i++) { | ||
| newState.collect(0, BitMixer.mix64(randomIntBetween(500, 10000))); |
There was a problem hiding this comment.
Haven't looked closely at the test, but is there a way to ensure we add new values that weren't previously added? Maybe by picking from a different range of random values or something? Or keeping a set?
Maybe it's a non-issue and random enough to not matter... might be a problem for a different day :)
There was a problem hiding this comment.
It does use a different range of values, Still for precision 4 in the previous approach you could get the same sketch at the end :( I think this way is very very unlikely.
There was a problem hiding this comment.
Ah gotcha. Sounds good!
) (#62554) Make sure the the new HLL++ is different to the original one
4 participants
Make sure the the new HLL++ is different to the original one by not merging the original and just add a few values to the new one.
closes #62540