Add list_transform

[mhk197]

Summary

list_transform(list, (x[, i]) -> body) applies a lambda to every element of a List, ListView or FixedSizeList, keeping each list's length and validity. i is the element's zero-based u64 index within its list.

• ListTransform is the first higher-order function. Its lambda_params gives the element and, optionally, its index, and its options hold the bound lambda.
• Execution flattens the lists, repeats each capture for every element of its row with a lazy take over a PiecewiseSequence of parent rows, and evaluates the lambda over the elements. A body that can fail skips elements hidden under null lists.
• List stays List, FixedSizeList keeps its width, and ListView is rebuilt so it is zero-copyable to List.
• The validity is the list's validity, and the function is infallible when its lambda is, so it pushes into dictionary values.

vortex.list.transform($.l, (x) -> ($x + $.a))                    unbound
vortex.list.transform($.l, $.a, opts=(x) -> ($x + $capture0))    bound

Testing

• cargo clippy -p vortex-array --all-targets --all-features -- -D warnings
• cargo nextest run -p vortex-array
• cargo test --doc -p vortex-array
• cargo +nightly-2026-09-10 fmt -p vortex-array

[codspeed]

Merging this PR will degrade performance by 3.27%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ 1 benchmark measured no execution time

Nothing ran under measurement, usually because the compiler removed the code under test. This result is not comparable, so it counts as unchanged.

Preventing compiler optimizations

• bench_compare_primitive[(10000, 2)]

⚠️ 13 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• decode_primitives[i64, (1000, 32)]
• random_i128[0.01]
• take_varbin
• pcodec_compress_f64
• density_sweep_dense_runs[0.001]
• multipolygon_mixed_validity
• zstd_compress_string
• filter_run_end[len=4096_run=16_density=0.1]
• decompress[u32, (1000, 256)]
• decompress[u16, (4000, 256)]
• …and 3 more

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 5 improved benchmarks
❌ 7 regressed benchmarks
✅ 2201 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	threshold_neon[80]	5.1 µs	8.2 µs	-38.31%
❌	WallTime	threshold_neon[64]	4.5 µs	6.1 µs	-26.19%
❌	WallTime	threshold_neon[8]	3.8 µs	4.4 µs	-14.77%
❌	Simulation	sum_v2_i64	194.6 µs	224.8 µs	-13.44%
❌	Simulation	sum_i64	194.8 µs	224.7 µs	-13.31%
❌	WallTime	starts_with_avx512[8, VarBin]	40.8 µs	46.2 µs	-11.75%
❌	WallTime	filter_neon[64]	3.1 µs	3.5 µs	-10.57%
⚡ ⚠️	Simulation	decode_primitives[i64, (1000, 32)]	38.6 µs	24 µs	+60.63%
⚡	WallTime	threshold_avx2[8]	2.4 µs	1.9 µs	+23%
⚡ ⚠️	Simulation	random_i128[0.01]	7.3 µs	6.2 µs	+17.58%
⚡	WallTime	threshold_avx2[64]	3.4 µs	3 µs	+12.83%
⚡	WallTime	threshold_avx2[80]	3.7 µs	3.3 µs	+11.31%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing mk/list-transform-fn (1d13ace) with mk/lambda-expressions (78ce193)¹}

Footnotes

1. No successful run was found on mk/lambda-expressions (676b887) during the generation of this report, so ebfa480 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩