Release v0.0.4 · liuchong/eliscript

Highlights in v0.0.4

Zero Tracked .mjs Sources: Eliminated all 516 tracked .mjs files from Git; all runtime, platform, browser, host, tool, and test JavaScript modules are synthesized in-place from .eli sources during bootstrap.
ClojureScript-Style JS Runtime & Stdlib Optimization: Achieved 1.95 ms on the 12,000-depth persistent Zipper benchmark (46.2x faster than v0.0.3 and 1.92x faster than native Emacs Lisp 31.1), 299x faster take-last (chunked reverse traversal), 3.06x faster sort (in-place 32-way trie materialization + direct comparator), and 2.22x faster group-by (transient vector bucket accumulation).
Normalized 4-Dimension Version & Native Emacs Lisp Benchmark: Added automated 4-dimension performance benchmarking (Emacs Lisp, previousMajor, previousMinor, latest), project-internal release-benchmark skill, and GitHub Actions release integration.

对比维度: Emacs 30.2 (当前最新版) | 上一个大版本: N/A (尚无上一大版本) | 上一个最近的小版本: v0.0.3 | 最新版: v0.0.4
完整测试与基准报告附件: 详见 Release 附件 eliscript-0.0.4-benchmark-report.md 与 eliscript-0.0.4-benchmark-report.json

基准工作负载	最新版耗时	vs Emacs Lisp	vs 上一大版本	vs 上一小版本
first(Vector, N=10,000)	0.29 µs (0.00029 ms)	0.88x	N/A	6.02x
last(Vector, N=10,000)	0.90 µs (0.00090 ms)	33.75x	N/A	9.25x
take-last(64, N=10,000)	8.56 µs (0.00856 ms)	3.52x	N/A	73.08x
[VALUE_EQUAL] / compare(sameVec, N=10,000)	0.23 µs (0.00023 ms)	166.88x	N/A	2358.45x
sort(Vector, N=10,000)	2.335 ms	1.04x	N/A	10.39x
group-by(x % 100, N=10,000)	1.874 ms	2.76x	N/A	3.42x
frequencies(1,000 keys, N=100,000)	15.491 ms	1.99x	N/A	2.26x
*12,000-depth Zipper (down12k + path + replace + root)**	6.391 ms	1.21x	N/A	47.05x
score-values(N=10,000 x 10 rounds)	1.953 ms	20.44x	N/A	1.00x