[robert3005]

Summary

Split out of #10052. list_contains of a prepared constant set over dictionary or run-end encoded needles probed every row. With this change, it probes once for each distinct value or run.

Changes

• ListContainsElementReduce for Dict and RunEnd pushes the function into the values with a slice of the PreparedSetArray. This shares the one prepared set, so the set is not built again.
• Off SQL null semantics, an empty set gives false for a null code. Thus the dictionary rule returns a constant false in that case.
• Tests: the dictionary null-code cases (empty set and not empty, both null semantics), the push-down into dictionary values, and the push-down into run-end values.

🤖 Generated with Claude Code

[codspeed]

Merging this PR will regress 3 benchmarks

⚠️ 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.

⚠️ 19 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

• random_i128[0.01]
• density_sweep_dense_runs[0.001]
• test_scan_filter[int_float]
• decompress[u32, (1000, 256)]
• test_scan_filter[int_str]
• decompress[u16, (4000, 256)]
• test_repeated_scan[int_str]
• test_scan[int_str]
• test_repeated_scan_filter[int_str]
• multipolygon_mixed_validity
• …and 9 more

ℹ️ 40 benchmarks have no flame graph

The flamegraph could not be generated because we failed to resolve the symbols in the profiling trace. This happens when we fail to harvest information about language frames.

• test_scan_filter[int]
• test_repeated_scan_scalar_at[int]
• test_scan_filter[int_float]
• test_scan_scalar_at[int_str]
• test_scan_filter[int_str]
• test_scan_scalar_at[int_float]
• test_repeated_scan_scalar_at[int_float]
• test_repeated_scan_scalar_at[int_str]
• test_repeated_scan[int_str]
• test_scan_scalar_at[int_str_float]
• …and 30 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
❌ 3 regressed benchmarks
✅ 2251 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	threshold_neon[8]	3.5 µs	4 µs	-13.37%
❌	Simulation	random_i8[0.01]	108 µs	122.9 µs	-12.08%
❌ ℹ️	Simulation	test_scan_filter[int]	3.1 ms	3.5 ms	-11.14%
⚡ ⚠️	Simulation	random_i128[0.01]	7.6 µs	6 µs	+26.88%
⚡	WallTime	filter_avx512[64]	2.5 µs	2.2 µs	+15.67%
⚡	WallTime	filter_neon[64]	3.4 µs	3 µs	+14.04%
⚡	WallTime	filter_neon[80]	3.6 µs	3.1 µs	+13.32%
⚡	WallTime	bitpack_blocked_compress_avx512	5.5 µs	4.8 µs	+12.71%

Tip

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

---

_{Comparing rk/list-contains-dict-runend (26a71c9) with develop (c1a2ca1)}