Benchmarks and optimisations

[PyXiion]

Optimised mod discovery

1. find_about_xml — fast-path check About/About.xml with Path.is_file() before falling back to full directory iteration (covers the vast majority of real mods)
2. scan_mod_directory — replace Path.iterdir() with os.scandir(), return dict[Path, Path] instead of list[Path] to cache each mod's About.xml path during the scan
3. create_listed_mod_from_path — accept about_xml_path parameter to skip redundant find_about_xml call; callers (all three providers) pass the pre-found path from the scan

	Before	After	Δ
1000 mods total	265.5ms	203.5ms	-23%
2000 mods total	496.7ms	407.9ms	-18%
scan avg	46.2 µs/mod	26.7 µs/mod	-42%

Optimised topological sort

Replaced per-iteration list.sort() + list.pop(0) in Kahn's algorithm with heapq push/pop (O(n² log n) → O(n log n)), and flattened _build_deps_and_rev_deps from three edges_of_type() calls per node to a single outgoing() pass with inline dispatch.

Phase	Before	After	Δ
build_deps	—	2.5ms	—
kahn_tier_2	54.1ms	2.1ms	-96%
sort total (750 mods)	68.5ms	10.7ms	-84%

Optimised XML parsing

Replaced the generic etree_to_dict recursive dict converter with a single-pass Element walker (_create_about_mod_from_element) that builds AboutXmlMod directly from lxml children, eliminating the intermediate nested dict and all value_extractor calls.

Pipeline change:
Before: ET.parse → etree_to_dict → dict → value_extractor → create_about_mod → AboutXmlMod
After: ET.parse → _create_about_mod_from_element → AboutXmlMod

	Before	After	Δ
Model construction (1000 mods)	41.4ms	25.4ms	-39%
Full discovery parse (1000 mods)	~177ms	119ms	-33%
parse avg	~177 µs/mod	119 µs/mod	-33%

Benchmarks

• tests/benchmarks/bench_sort.py — synthetic sort pipeline benchmark (200–2000 mods, community rules)
• tests/benchmarks/bench_discovery.py — mod discovery scan + parse benchmark
• Both run via just bench (all) or just bench <name> (filtered)

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 0 complexity · 0 duplication

Metric	Results
Complexity	0
Duplication	0

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}