Thanks for opening a pull request!

Could you open an issue for this pull request on Github Issues?

https://github.com/apache/incubator-gluten/issues

Then could you also rename commit message and pull request title in the following format?

[GLUTEN-${ISSUES_ID}][COMPONENT]feat/fix: ${detailed message}

See also:

• Other pull requests

Run Gluten Clickhouse CI on x86

@zhli1142015 @zhztheplayer Could you help to review?

I remembre that this function was disabled because of the issue below.
#7628
The discrepancy in results between Velox and Spark was due to the compiled expression being cached, resulting in a single expression instance whose ID kept incrementing.
https://github.com/facebookincubator/velox/blob/main/velox/expression/ExprCompiler.cpp#L376C29-L376C47
I thought we may need a fix in velox to allow marking this expression as “do not cache.”

@zhli1142015 OK. I'll close this PR. But isn't this function itself non-deterministic anyway? As long as it generates an increasing non-repeating sequence, it shouldn't cause issues in practice—it just won't be perfectly consistent with Spark's results, right?

@zhli1142015 But the issue in #7628 is that calling monotonically_increasing_id twice in a single row is expected to yield the same result.

it shouldn't cause issues in practice?

It seems to me that this unit test verifies a non publicly documented behavior; but we re-enabled this function after we had fixed this issue internally.