[padix-key]

write_structure_to_ctab() (used by MOLFile.set_structure(), SDRecord.set_structure() and mol.set_structure()) built each atom/bond line by indexing NumPy arrays element-wise (atoms.coord[i, 0], atoms.element[i], charge[i], bond array rows), and re-formatted constant padding fields for every line.

Fix: iterate over tolist() versions of coordinates, elements, charges and the bond array, and format the constant unused fields once. Output is identical to main (V2000 and V3000).

Input	Call	before	after	speedup
13136.sdf (39 atoms)	write_structure_to_ctab(..., "V2000")	0.187 ms	0.081 ms	2.3x
HWB.sdf	write_structure_to_ctab(..., "V2000")	0.169 ms	0.072 ms	2.3x
ATP (CCD)	SDRecord.set_structure(..., version="V2000")	0.235 ms	0.093 ms	2.5x
ATP (CCD)	SDRecord.set_structure(..., version="V3000")	0.186 ms	0.077 ms	2.4x
13136.sdf	SDRecord.set_structure(..., version="V2000")	0.190 ms	0.084 ms	2.3x

Tests run: tests/structure/io/test_mol.py.

Authored by Claude

[codspeed]

Merging this PR will regress 4 benchmarks

⚡ 16 improved benchmarks
❌ 4 regressed benchmarks
✅ 89 untouched benchmarks
🆕 2 new benchmarks
⏩ 14 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	benchmark_cell_list_compute_contacts[MAPPING]	22.9 ms	27.1 ms	-15.6%
❌	benchmark_match_kmer_selection[BucketKmerTable(10000)-None]	652.9 µs	771.1 µs	-15.33%
❌	benchmark_match[BucketKmerTable(10000)-11*11*1*1***111]	837.3 µs	957.5 µs	-12.55%
❌	benchmark_cell_list_compute_contacts[PAIRS]	22.7 ms	25.8 ms	-12.11%
⚡	benchmark_find_rotatable_bonds	177.1 ms	26.7 ms	×6.6
⚡	benchmark_superimpose[superimpose]	3,198.9 µs	645 µs	×5
⚡	benchmark_hbond	29.7 ms	10.1 ms	×2.9
⚡	benchmark_similar_kmers	483 µs	199.9 µs	×2.4
⚡	benchmark_superimpose_multi_model[superimpose]	14.4 ms	6.4 ms	×2.3
⚡	benchmark_get_assembly	1,594.7 ms	836.2 ms	+90.71%
⚡	benchmark_superimpose[superimpose_homologs]	60.6 ms	33.6 ms	+80.28%
⚡	benchmark_score_threshold_rule[11*11*1*1***111]	3.3 s	1.9 s	+70.72%
⚡	benchmark_score_threshold_rule[None]	3.3 s	1.9 s	+70.71%
⚡	benchmark_superimpose_multi_model[superimpose_homologs]	74.2 ms	44.8 ms	+65.74%
⚡	benchmark_superimpose[superimpose_without_outliers]	31.9 ms	20.8 ms	+52.95%
⚡	benchmark_match_kmer_selection[KmerTable-None]	642.6 µs	526.3 µs	+22.1%
⚡	benchmark_superimpose[superimpose_structural_homologs]	53.4 ms	44.8 ms	+19.18%
⚡	benchmark_set_structure[bcif-True]	39 ms	32.8 ms	+19.03%
⚡	benchmark_match[KmerTable-None]	739.6 µs	623.8 µs	+18.56%
⚡	benchmark_set_structure[bcif-False]	17.7 ms	15.1 ms	+17.01%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

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

---

_{Comparing padix-key:perf/strucio-mol-ctab (0cfd4d1) with main (81df912)²}

Footnotes

1. 14 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 main (b0d32ca) during the generation of this report, so 81df912 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩