ci(benchmarks/formatter): Update formatter bench with sort-imports

[leaysgur]

I think we don't need to add another benchmark for the sort import feature; enabling it in > formatter.rs is enough. Any of the current benchmark files contains many import statements.

It seems there was no need to create a separate file after all.

My intention was to confirm that the benchmark for the formatter alone consistently produced better results than the benchmark with sorting enabled.

To verify this, shouldn't the benchmarks be separated?

[leaysgur]

• ci(benchmarks/formatter): Update formatter bench with sort-imports #15584 👈 (View in Graphite)
• feat(oxfmt): Print nothing for default(write) mode #15583
• main

---

How to use the Graphite Merge Queue

Add either label to this PR to merge it via the merge queue:

• 0-merge - adds this PR to the back of the merge queue
• hotfix - for urgent hot fixes, skip the queue and merge this PR next

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has enabled the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

This stack of pull requests is managed by Graphite. Learn more about stacking.

[leaysgur]

@Dunqing ↑ Does that make sense?

[codspeed-hq]

CodSpeed Performance Report

Merging #15584 will degrade performances by 20.95%

_{Comparing 11-11-ci_benchmarks_formatter_add_formatter_bench_with_sort-imports (92843a7) with main (128e186)¹}

Summary

❌ 4 regressions
✅ 33 untouched

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Mode	Benchmark	BASE	HEAD	Change
❌	Simulation	formatter[RadixUIAdoptionSection.jsx]	479.7 µs	508.3 µs	-5.61%
❌	Simulation	formatter[binder.ts]	17.2 ms	20.9 ms	-17.68%
❌	Simulation	formatter[cal.com.tsx]	155.4 ms	185.8 ms	-16.34%
❌	Simulation	formatter[react.development.js]	8.3 ms	10.5 ms	-20.95%

Footnotes

1. No successful run was found on main (99823ad) during the generation of this report, so 128e186 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[Dunqing]

No, the formatter alone is faster than enabling the sorting feature, no doubt. We don't have any reason to care about this.

The purpose of turning on the sorting import feature in the Formatter benchmark is to help us catch any performance regressions early. For example, you made some improvements to the sorting import feature recently, but we don't have any way to verify whether your changes have improved or regressed the performance. Sometimes, we made improvements that we thought should make it significantly faster, but it is actually slower than before. This will avoid introducing incorrect optimization.

Additionally, adding more benchmarks will increase the CI time, which causes us to wait longer for CI to complete. It is not worth adding this for a single one of the features of the formatter.

[leaysgur]

Thanks! Alright, let's just enable sorting on the existing benchmarks.

[Copilot]

Pull Request Overview

This PR modifies the formatter benchmark to enable the experimental sort-imports feature, aiming to measure the performance impact of this feature. However, the current implementation replaces the baseline benchmark rather than adding a separate comparison benchmark.

Key Changes:

• Added SortImports import to the formatter benchmark
• Modified FormatOptions to enable experimental_sort_imports with default settings

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Dunqing]

Wow, the sorting import feature takes 15% to 20% of the time in large files.

[Dunqing]

Merge activity

• Nov 11, 8:39 AM UTC: The merge label '0-merge' was detected. This PR will be added to the Graphite merge queue once it meets the requirements.
• Nov 11, 8:39 AM UTC: Dunqing added this pull request to the Graphite merge queue.
• Nov 11, 8:47 AM UTC: Merged by the Graphite merge queue.