Specialize VarBin PiecewiseSequence takes

[danking]

Summary

• move VarBin and VarBinView PiecewiseSequence take specializations out of Use PiecewiseSequence indices for FSL take #8800 into this stacked PR
• keep the branch based directly on codex/piecewise-sequential-take
• preserve existing behavior: VarBin copies contiguous byte ranges, VarBinView gathers view ranges while reusing backing data buffers

Validation

• cargo +nightly fmt --all
• cargo check -p vortex-array

[codspeed-hq]

Merging this PR will regress 1 benchmark

⚡ 5 improved benchmarks
❌ 1 regressed benchmark
✅ 1792 untouched benchmarks
⏩ 44 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	varbinview_large	390.6 µs	445.4 µs	-12.3%
⚡	Simulation	struct_large	431.1 µs	38.7 µs	×11
⚡	Simulation	fsl_large	615.6 µs	91.5 µs	×6.7
⚡	Simulation	list_i32_large	172.7 µs	96.2 µs	+79.52%
⚡	Simulation	i32_large	277.8 µs	235.9 µs	+17.75%
⚡	Simulation	true_count_vortex_buffer[128]	580.6 ns	522.2 ns	+11.17%

Tip

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

---

_{Comparing codex/piecewise-sequence-varbin-take (5d31540) with develop (2ea51cb)²}

Footnotes

1. 44 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on develop (add6944) during the generation of this report, so 2ea51cb was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[github-actions]

Polar Signals Profiling Results

Latest Run

Status	Commit	Job	Attempt	Link
🟢 Done	5d31540		1	Explore Profiling Data

Previous Runs (1)

Status	Commit	Job	Attempt	Link
🟢 Done	9f02349		1	Explore Profiling Data

---

Powered by Polar Signals Cloud

[github-actions]

Benchmarks: Vortex queries 📖

Verdict: No clear signal (medium confidence)
Attributed Vortex impact: -2.2%
Engines: DataFusion No clear signal (+8.9%, medium confidence) · DuckDB Likely improvement (-12.2%, high confidence)
Vortex (geomean): 0.997x ➖
Parquet (geomean): 1.009x ➖
Shifts: Parquet (control) +0.9% · Median polish +0.9%

How to read Verdict and Engines

• Verdict: Overall PR-level signal after subtracting baseline drift estimated from Parquet control rows. It can be Likely improvement, Likely regression, or No clear signal.
• Engines: Per-engine attribution. DataFusion is compared against DataFusion/Parquet controls; DuckDB is compared against DuckDB/Parquet controls. This answers whether each engine improved or regressed independently.
• Confidence: Based on directional consistency, share of rows above the noise floor, and control-run noise.

---

datafusion / vortex-file-compressed (1.104x ❌, 0↑ 1↓)

name	PR 5d31540 (ns)	base add6944 (ns)	ratio (PR/base)
vortex_q00/datafusion:vortex-file-compressed 🚨	10599312	9338219	1.14
vortex_q01/datafusion:vortex-file-compressed	6495096	6048928	1.07

datafusion / parquet (1.014x ➖, 0↑ 0↓)

name	PR 5d31540 (ns)	base add6944 (ns)	ratio (PR/base)
vortex_q00/datafusion:parquet	20690177	21059503	0.98
vortex_q01/datafusion:parquet	4973716	4750673	1.05

duckdb / vortex-file-compressed (0.881x ✅, 2↑ 0↓)

name	PR 5d31540 (ns)	base add6944 (ns)	ratio (PR/base)
vortex_q00/duckdb:vortex-file-compressed 🚀	10690905	12173747	0.88
vortex_q01/duckdb:vortex-file-compressed 🚀	6600679	7472232	0.88

duckdb / parquet (1.003x ➖, 0↑ 0↓)

name	PR 5d31540 (ns)	base add6944 (ns)	ratio (PR/base)
vortex_q00/duckdb:parquet	23713679	23276573	1.02
vortex_q01/duckdb:parquet	9565286	9680554	0.99

---

No file size changes detected.