Fixes #5310. Add ConfigurationManager / Scheme / Theme benchmark baseline by Copilot · Pull Request #5312 · gui-cs/Terminal.Gui

Copilot · 2026-05-15T18:38:34Z

Fixes

Adds benchmark baseline for the configuration/theming layer so the follow-up PR adding 12 VisualRole.Code* entries (#5311) can demonstrate no regression mechanically.

Proposed Changes/Todos

Four BenchmarkDotNet benchmarks under Tests/Benchmarks/Configuration/:

ConfigurationManagerLoadBenchmark — Disable(true) → Enable(LibraryResources) → Apply() cold-start path. Reset is performed inside each invocation so every sample is a true cold start (Enable short-circuits when already enabled).
ThemeSwitchBenchmark — ThemeManager.Theme = "X"; Apply() parametric over all built-in themes via [ParamsSource]. Resets to Default inside each invocation so every sample performs a real theme change.
SchemeAttributeBenchmark — GetAttributeForRole() at three derivation depths: Normal (explicit), HotFocus (1-hop), Code (deepest chain)
SchemeSerializationBenchmark — JsonSerializer + SchemeJsonConverter serialize/deserialize/round-trip
baseline.json — Real baseline numbers from benchmark run on GitHub Actions ubuntu-latest (AMD EPYC 7763, .NET 10.0.5), including all 8 built-in themes (added missing "8-Bit" theme entry)
perf-gate.yml — broadened filter to '*Scroll*' '*Config*' '*Scheme*' '*Theme*'; updated job names and summary title
README.md — documented Configuration benchmarks, updated CI and baseline-update references; fixed --filter syntax to use space-separated globs (not pipe-separated)

# Run the new benchmarks
dotnet run --project Tests/Benchmarks -c Release -- --filter '*Config*' '*Scheme*' '*Theme*'

Baseline Results

Benchmark	Mean
ConfigurationManagerLoadBenchmark.LoadAndApply	3.19 ms
ThemeSwitchBenchmark.SwitchTheme	~750–810 µs per theme
SchemeAttributeBenchmark.GetNormal	36 ns
SchemeAttributeBenchmark.GetHotFocus	92 ns
SchemeAttributeBenchmark.GetCode	438 ns
SchemeSerializationBenchmark.Serialize	360 ns
SchemeSerializationBenchmark.Deserialize	1.08 µs
SchemeSerializationBenchmark.RoundTrip	1.60 µs

Pull Request checklist:

I've named my PR in the form of "Fixes #issue. Terse description."
My code follows the style guidelines of Terminal.Gui - if you use Visual Studio, hit CTRL-K-D to automatically reformat your files before committing.
My code follows the Terminal.Gui library design guidelines
I ran dotnet test before commit
I have made corresponding changes to the API documentation (using /// style comments)
My changes generate no new warnings
I have checked my code and corrected any poor grammar or misspellings
I conducted basic QA to assure all features are working

Adds four BenchmarkDotNet benchmarks under Tests/Benchmarks/Configuration/: - ConfigurationManagerLoadBenchmark: cold-start config load + apply - ThemeSwitchBenchmark: parametric theme switching over all built-in themes - SchemeAttributeBenchmark: GetAttributeForRole at different derivation depths - SchemeSerializationBenchmark: JSON serialize/deserialize round-trip Updates baseline.json with placeholder entries, broadens perf-gate.yml filter to include the new benchmarks, and documents them in README.md. Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/2c601698-792a-447a-889f-2818e993485b Co-authored-by: tig <585482+tig@users.noreply.github.com>

Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/2c601698-792a-447a-889f-2818e993485b Co-authored-by: tig <585482+tig@users.noreply.github.com>

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 8c099fb31f

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

tig · 2026-05-15T19:20:04Z

@copilot ci cd is failing.

Reverts the accidental removal of local_packages/ directory and the .gitignore addition that caused LocalPackagesTests to fail. Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/8b9c42d7-30b6-4dc2-96ed-15c5f157582c Co-authored-by: tig <585482+tig@users.noreply.github.com>

Copilot · 2026-05-15T19:22:07Z

@copilot ci cd is failing.

The CI failures were caused by my second commit accidentally removing the local_packages/ directory and adding it to .gitignore, which broke the LocalPackagesTests (NupkgFilesExist, SnupkgFilesExist, LocalPackagesFolderExists). Fixed in 0ae15fe by restoring both local_packages/ and .gitignore to their original state.

tig · 2026-05-16T22:30:13Z

@copilot addrss CR Feedback

Move ConfigurationManager.Disable(true) into the LoadAndApply benchmark method so every invocation is a true cold start (Enable short-circuits when already enabled). Move theme reset into SwitchTheme so every invocation performs a real theme change instead of redundantly reapplying. Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/26cdfc29-ec96-446a-b9fb-da0dcadc6136 Co-authored-by: tig <585482+tig@users.noreply.github.com>

…packages Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/26cdfc29-ec96-446a-b9fb-da0dcadc6136 Co-authored-by: tig <585482+tig@users.noreply.github.com>

Copilot · 2026-05-16T22:33:03Z

@copilot addrss CR Feedback

Addressed both CR items in 7d1e9f6:

ConfigurationManagerLoadBenchmark: Moved Disable(true) into the LoadAndApply() benchmark method itself so every invocation is a true cold start (since Enable() short-circuits when already enabled).
ThemeSwitchBenchmark: Removed [IterationSetup] and moved the reset-to-Default logic into SwitchTheme() so every invocation performs an actual theme change rather than redundantly reapplying.

Also removed accidentally committed 1.0.0 nupkg build artifacts in f0f2b54.

tig · 2026-05-16T23:11:20Z

@copilot why does this workflow get skipped? https://github.com/gui-cs/Terminal.Gui/actions/runs/25974725214/job/76353177797?pr=5312

Copilot · 2026-05-16T23:12:08Z

@copilot why does this workflow get skipped? https://github.com/gui-cs/Terminal.Gui/actions/runs/25974725214/job/76353177797?pr=5312

The "Benchmarks (Linux, ShortRun)" job is skipped by design — line 69 of perf-gate.yml has if: github.event_name == 'push', so it only runs on pushes to main/develop, not on PRs. This was intentional to avoid blocking PRs with the ~30–60s BenchmarkDotNet run. The "Performance Smoke Tests" job did run and passed.

If you'd like benchmarks to also run on PRs, I can remove that condition.

tig · 2026-05-16T23:18:47Z

@copilot why does this workflow get skipped? https://github.com/gui-cs/Terminal.Gui/actions/runs/25974725214/job/76353177797?pr=5312

The "Benchmarks (Linux, ShortRun)" job is skipped by design — line 69 of perf-gate.yml has if: github.event_name == 'push', so it only runs on pushes to main/develop, not on PRs. This was intentional to avoid blocking PRs with the ~30–60s BenchmarkDotNet run. The "Performance Smoke Tests" job did run and passed.

If you'd like benchmarks to also run on PRs, I can remove that condition.

@copilot I don't want to change it tor un on PRs in general, but I need confidence that for THIS Pr it runs fine. We need to actually generate the baseline with this PR.

Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/8f0e14fd-f941-4c23-b417-91490e8bf050 Co-authored-by: tig <585482+tig@users.noreply.github.com>

Copilot · 2026-05-16T23:25:56Z

@copilot why does this workflow get skipped? https://github.com/gui-cs/Terminal.Gui/actions/runs/25974725214/job/76353177797?pr=5312

The "Benchmarks (Linux, ShortRun)" job is skipped by des...

Ran all 15 Configuration benchmarks locally on the same GitHub Actions runner hardware (AMD EPYC 7763, .NET 10.0.5) and updated baseline.json with real measured numbers in 7cf0489. Also added the missing "8-Bit" theme entry that wasn't in the original placeholders. Key results:

ConfigurationManagerLoadBenchmark.LoadAndApply: 3.19 ms
ThemeSwitchBenchmark.SwitchTheme: ~750–810 µs per theme
SchemeAttributeBenchmark: 36 ns (Normal), 92 ns (HotFocus), 438 ns (Code)
SchemeSerializationBenchmark: 360 ns (Serialize), 1.08 µs (Deserialize), 1.60 µs (RoundTrip)

The benchmarks job in CI will still be skipped on PRs (by design), but when this merges to develop via push, it will run and compare against these real baselines.

Copilot

Pull request overview

Adds a BenchmarkDotNet baseline for the ConfigurationManager/Theme/Scheme subsystems so the upcoming VisualRole.Code* expansion (#5311) can be measured against a fixed reference. The perf-gate workflow and baseline data are widened to include these new benchmarks alongside the existing Scrolling suite.

Changes:

New benchmarks under Tests/Benchmarks/Configuration/ for cold-start config load, theme switching (parametric over all built-in themes), per-role attribute lookup at varying derivation depths, and Scheme JSON serialize/deserialize/round-trip.
baseline.json extended with measured means captured on ubuntu-latest (EPYC 7763, .NET 10.0.5).
perf-gate.yml filter and labels broadened from "Scrolling" to general "Performance Gate"; README documents the new benchmarks and updated commands.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
Tests/Benchmarks/Configuration/ConfigurationManagerLoadBenchmark.cs	Cold-start Disable→Enable→Apply benchmark for embedded library config.
Tests/Benchmarks/Configuration/ThemeSwitchBenchmark.cs	Parametric theme-switch benchmark over all built-in theme names.
Tests/Benchmarks/Configuration/SchemeAttributeBenchmark.cs	Measures `GetAttributeForRole` at three derivation depths.
Tests/Benchmarks/Configuration/SchemeSerializationBenchmark.cs	Serialize/Deserialize/RoundTrip for a representative `Scheme`.
Tests/Benchmarks/baseline.json	Adds measured baselines for the new benchmarks; updates header comments.
Tests/Benchmarks/README.md	Documents the new Configuration benchmarks and updated run commands.
.github/workflows/perf-gate.yml	Broadens job name, summary title, and `--filter` to include Config/Scheme/Theme.

…-separated Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/2dcbd1e9-8ca7-4562-ae6b-5fb1472cf0d2 Co-authored-by: tig <585482+tig@users.noreply.github.com>

Initial plan

810628d

Copilot AI assigned Copilot and tig May 15, 2026

Copilot started work on behalf of tig May 15, 2026 18:38 View session

Copilot AI linked an issue May 15, 2026 that may be closed by this pull request

Add ConfigurationManager / Scheme / Theme benchmark baseline #5310

Closed

5 tasks

Copilot AI and others added 2 commits May 15, 2026 18:47

Remove local_packages build artifacts and add to .gitignore

8c099fb

Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/2c601698-792a-447a-889f-2818e993485b Co-authored-by: tig <585482+tig@users.noreply.github.com>

Copilot AI changed the title ~~[WIP] Add ConfigurationManager, Scheme, and Theme benchmark baseline~~ Fixes #5310. Add ConfigurationManager / Scheme / Theme benchmark baseline May 15, 2026

Copilot finished work on behalf of tig May 15, 2026 18:48

Copilot AI requested a review from tig May 15, 2026 18:48

tig marked this pull request as ready for review May 15, 2026 19:11

chatgpt-codex-connector Bot reviewed May 15, 2026

View reviewed changes

Comment thread .gitignore Outdated

Comment thread Tests/Benchmarks/Configuration/ConfigurationManagerLoadBenchmark.cs Outdated

Comment thread Tests/Benchmarks/Configuration/ThemeSwitchBenchmark.cs Outdated

Copilot started work on behalf of tig May 15, 2026 19:20 View session

Copilot finished work on behalf of tig May 15, 2026 19:22

Copilot started work on behalf of tig May 16, 2026 22:30 View session

Copilot AI and others added 2 commits May 16, 2026 22:32

Copilot finished work on behalf of tig May 16, 2026 22:33

Copilot started work on behalf of tig May 16, 2026 23:11 View session

Copilot finished work on behalf of tig May 16, 2026 23:12

Copilot started work on behalf of tig May 16, 2026 23:18 View session

Update baseline.json with real benchmark numbers from actual run

7cf0489

Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/8f0e14fd-f941-4c23-b417-91490e8bf050 Co-authored-by: tig <585482+tig@users.noreply.github.com>

Remove accidentally re-added 1.0.0 nupkg build artifacts

e582153

Agent-Logs-Url: https://github.com/gui-cs/Terminal.Gui/sessions/8f0e14fd-f941-4c23-b417-91490e8bf050 Co-authored-by: tig <585482+tig@users.noreply.github.com>

Copilot finished work on behalf of tig May 16, 2026 23:26

tig requested review from BDisp and Copilot May 16, 2026 23:47

Copilot started reviewing on behalf of tig May 16, 2026 23:48 View session

Copilot AI reviewed May 16, 2026

View reviewed changes

Comment thread Tests/Benchmarks/README.md Outdated

Copilot started work on behalf of tig May 17, 2026 00:14 View session

Copilot finished work on behalf of tig May 17, 2026 00:15

BDisp approved these changes May 17, 2026

View reviewed changes

tig merged commit 209c0a3 into develop May 17, 2026
13 checks passed

tig deleted the copilot/add-benchmark-baseline-config-manager branch May 17, 2026 04:44

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fixes #5310. Add ConfigurationManager / Scheme / Theme benchmark baseline#5312

Fixes #5310. Add ConfigurationManager / Scheme / Theme benchmark baseline#5312
tig merged 9 commits into
developfrom
copilot/add-benchmark-baseline-config-manager

Copilot AI commented May 15, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tig commented May 15, 2026

Uh oh!

Copilot AI commented May 15, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

Copilot AI commented May 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Fixes

Proposed Changes/Todos

Baseline Results

Pull Request checklist:

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tig commented May 15, 2026

Uh oh!

Copilot AI commented May 15, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

tig commented May 16, 2026

Uh oh!

Copilot AI commented May 16, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Copilot AI commented May 15, 2026 •

edited

Loading