perf(serialize): faster serialization of objects and classes

[zsilbi]

Since we know Object.entries(object) returns Array<[string, any]> we can use a faster path to:

• avoid calling compare() because we can call String.localeCompare() directly on keys
• avoid calling serialize() for keys because they can be only strings

I created a new function for handling the Iterable but we can also merge it to the existing serializeObjectEntries() with a flag like onlyStringKeys and add conditions to handle that.

The performance gains are not significant though. - Used a faster path later.

 ✓ test/benchmarks.bench.ts > benchmarks > serialize > count:1, size:small 1446ms
     name                      hz     min     max    mean     p75     p99    p995    p999     rme  samples
   · serialize @ main  703,004.23  0.0012  0.4631  0.0014  0.0013  0.0024  0.0026  0.0145  ±0.67%   351503
   · serialize         736,631.27  0.0012  0.2477  0.0014  0.0013  0.0021  0.0022  0.0052  ±0.30%   368316   fastest

 ✓ test/benchmarks.bench.ts > benchmarks > serialize > count:256, size:small 1206ms
     name                    hz     min     max    mean     p75     p99    p995    p999     rme  samples
   · serialize @ main  7,052.17  0.1315  0.6230  0.1418  0.1372  0.3075  0.3213  0.3579  ±0.66%     3527
   · serialize         7,302.85  0.1276  0.3518  0.1369  0.1329  0.2274  0.2351  0.3235  ±0.48%     3652   fastest

 BENCH  Summary

  serialize - test/benchmarks.bench.ts > benchmarks > serialize > count:1, size:small
    1.05x faster than serialize @ main

  serialize - test/benchmarks.bench.ts > benchmarks > serialize > count:256, size:small
    1.04x faster than serialize @ main

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.84%. Comparing base (c5173ba) to head (b31c44d).
Report is 94 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #141       +/-   ##
===========================================
+ Coverage   80.31%   95.84%   +15.52%     
===========================================
  Files           8        8               
  Lines        1006      481      -525     
  Branches      111      126       +15     
===========================================
- Hits          808      461      -347     
+ Misses        198       18      -180     
- Partials        0        2        +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[zsilbi]

 ✓ test/benchmarks.bench.ts > benchmarks > serialize - presets > count:256, size:small 1216ms
     name                 hz     min     max    mean     p75     p99    p995    p999     rme  samples
   · ohash @ main   7,859.08  0.1117  0.5755  0.1272  0.1195  0.3489  0.3879  0.4571  ±0.99%     3930
   · ohash @ dev   11,888.49  0.0735  0.5062  0.0841  0.0786  0.2925  0.3252  0.3708  ±1.05%     5945   fastest

 BENCH  Summary

  ohash @ dev - test/benchmarks.bench.ts > benchmarks > serialize - presets > count:256, size:small
    1.51x faster than ohash @ main

clk: ~5.25 GHz
cpu: AMD Ryzen 9 7950X 16-Core Processor
runtime: bun 1.2.4 (x64-linux)

benchmark                   avg (min … max) p75 / p99    (min … top 1%)
------------------------------------------- -------------------------------
• count:256, size:small
------------------------------------------- -------------------------------
ohash @ main                 116.81 µs/iter 117.85 µs  █
                      (100.96 µs … 1.93 ms) 193.60 µs  █
                    (  0.00  b … 528.00 kb)  16.70 kb ██▃▄▆▄▂▂▂▁▁▁▁▁▁▁▁▁▁▁▁

ohash @ dev                   78.58 µs/iter  75.25 µs  █
                       (63.59 µs … 1.84 ms) 142.25 µs  █
                    (  0.00  b … 792.00 kb)  20.71 kb ▂█▆▂▃▄▂▁▁▁▁▁▁▁▁▂▂▁▁▁▁

summary
  ohash @ dev
   1.49x faster than ohash @ main

[pi0]

Thanks!