andygrove opened a new issue, #4269: URL: https://github.com/apache/datafusion-comet/issues/4269
## Description Comet currently supports the `explode` generator for array inputs but rejects `posexplode` (and `posexplode_outer`), forcing a fallback to Spark. `posexplode` is a Spark generator that takes an `ArrayType` (or `MapType`) and produces one row per element with an additional integer position column: ``` posexplode(array(10, 20)) -> 0 10 1 20 ``` ## Proposal Extend `CometExplodeExec` so it accepts both `Explode` and `PosExplode` Catalyst nodes, propagate a `position` flag through the operator proto, and make the native plan emit a parallel `pos` column alongside the unnested values via DataFusion's `UnnestExec`. Scope: - `posexplode` over `ArrayType` (Compatible) - `posexplode_outer` reuses the same `outer=true` Incompatible status as `explode_outer` (DataFusion #19053) - `MapType` continues to fall back to Spark, tracked separately by #2837 -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
