[parquet] reader perf - avoid some rle decode loop senarios in read_def_levels
Open
- Dominant language
- Rust
- Stars
- 3.6k
- Forks
- 1.3k
- Avg merge
- 2d 14h
- Merged PRs (30d)
- 167
Description
can we skip rle decode loop when all values in a batch are valid (have max definition level)?
Contributor guide
Research direction
Start by locating the Parquet reader's read_def_levels entry point and inspect how the RLE decode loop handles batches whose values all have the maximum definition level. Confirm the expected behavior with the surrounding reader logic; done means those fully valid batches avoid unnecessary RLE decoding without changing definition-level results.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- data-engineering
- Issue type
- Refactor
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100