jkosh44 commented on issue #14451: URL: https://github.com/apache/datafusion/issues/14451#issuecomment-2640335154
> Hello [@jkosh44](https://github.com/jkosh44), what do you think about this? > > `array_resize` should return `NULL` only if the second argument (new length) is `NULL`. If the third argument (new element) is `NULL`, the new slots (if any) should be set to `NULL`. This behavior would make `array_resize(array, new_len)` equivalent to `array_resize(array, new_len, NULL)`, which I think makes more sense. Good point, I don't think we can use `NullHandling` here and will just have to handle the NULL manually. The DuckDB behavior matches this as well. ``` duckdb> select array_resize([1,2,3], 4, NULL); ┌─────────────────────────────────────────────────┐ │ array_resize(main.list_value(1, 2, 3), 4, NULL) │ ╞═════════════════════════════════════════════════╡ │ [1, 2, 3, ] │ └─────────────────────────────────────────────────┘ ``` I think if we update the signature of `array_resize` to be more descriptive, then the error will go away. > Also, since `null_handling` makes `array_sort` return `NULL` for `NULL` inputs, should we still handle `NULL` cases in `array_sort_inner`? Yes, we should make sure that null handling is consistent for `ColumnarValue::Scalar(_)` and `ColumnarValue::Array(_)` inputs. As an FYI, I'm thinking that we should actually remove the `NullHandling` enum. I wrote up my rationale here: https://github.com/apache/datafusion/pull/14289#issuecomment-2640206878 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
