Support `push_batch_with_filter` in LimitedBatchCoalescer / `FilterExec`
- Dominant language
- Rust
- Stars
- 9.3k
- Forks
- 2.4k
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 344
Description
### Is your feature request related to a problem or challenge?
Currently, the `BatchCoalescer::push_batch` is used for writing batches to a target batch size.
In of https://github.com/apache/arrow-rs/pull/8951 we're writing a faster kernel for filtering that avoids creating small batches / copying the data once more.
### Describe the solution you'd like
Use the API in filter / LimitedBatchCoalescer
### Describe alternatives you've considered
_No response_
### Additional context
https://github.com/apache/datafusion/pull/19851/changes/6e1c4075de73c7f9407aab52349978659b1348ad has a PoC version
Contributor guide
Research direction
Start by reading the referenced DataFusion PoC, then locate LimitedBatchCoalescer and FilterExec and trace their current batch-writing path. Compare that path with the push_batch_with_filter API described in the issue. Done means the filter path uses the API while preserving the intended batch-coalescing behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- data-engineering
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100