feat: modularize PerfDiff regression detection, update analyzer reporting, and clarify CI performance documentation #565

rjmurillo · 2025-07-20T06:00:44Z

This PR modularizes the PerfDiff benchmarking tool and updates CI performance documentation for improved clarity and maintainability.

Key changes:

Modularizes regression detection logic in PerfDiff into dedicated strategies for mean, percentile (P95), and percentage-based thresholds.
Introduces new data contract types for benchmark results, statistics, percentiles, and regression results.
Enhances logging and reporting for regression detection, with clearer output for CI and developer use.
Updates docs/ci-performance.md to clarify CI regression thresholds and future direction for performance budgets, including practical performance budget plans and rationale.
Improves diagnostic location reporting in AsShouldBeUsedOnlyForInterfaceAnalyzer.
Updates code style and namespaces for consistency across PerfDiff and ETL components.
Adds .cursor/rules/codacy.mdc to enforce Codacy analysis after file edits.

Summary:
These changes improve reliability and extensibility of performance regression checks in CI, clarify documentation, and lay groundwork for future practical performance budgets. No breaking changes to analyzer APIs.

Fixes #563

Summary by CodeRabbit

New Features
- Introduced a comprehensive benchmarking comparison tool with multiple regression detection strategies, including mean, percentile, and percentage-based thresholds.
- Added support for detailed performance regression detection and reporting in CI workflows.
- Provided new configuration options and documentation for performance thresholds and future performance budget plans.
Bug Fixes
- Improved diagnostic location accuracy for invalid type arguments in interface analyzer feedback.
Documentation
- Enhanced documentation for CI performance benchmarking, regression thresholds, and future directions.
- Added configuration instructions for Codacy integration and security scanning.
- Revised performance benchmarking script documentation for clarity, cross-platform usage, and output details.
Refactor
- Major codebase restructuring for improved clarity, maintainability, and extensibility in benchmarking and logging modules.
- Streamlined command-line options and logging setup.
- Parameterized benchmark tests to allow variable source file counts for more flexible performance testing.
- Reorganized ETL trace comparison and overweight report generation for improved readability and structure.
- Simplified and documented logging classes and providers.
Chores
- Added solution and configuration files to support development and CI environments.
- Updated test and benchmark configurations for improved output and flexibility.
- Consolidated GitHub Actions workflow jobs for performance testing and baseline caching.

…rovider and Program files

…nchmarkDotNetDiffer

Introduce `PerfDiff.sln` with project configuration for Visual Studio 17. Define Debug and Release build configurations for "Any CPU" platform.

Reformat multiple classes including ChronometerFrequency, HostEnvironmentInfo, ConfidenceInterval, Percentiles, Statistics, Memory, Measurement, Benchmark, and BdnResult to remove unnecessary nested definitions. Enhance readability in DiffCommand, EtlDiffer, OverWeightResult, NullScope, SimpleConsoleLogger, and SimpleConsoleLoggerProvider classes. Simplify logic and improve formatting in PerfDiff and Program classes for better clarity and consistency.

Change array initialization for VerbosityLevels to use square brackets. Update Option constructor calls for verbosity and failOnRegression to reflect the new syntax.

Change variable declarations across multiple files to use explicit types for improved clarity and maintainability. This includes updates in DiffCommand.cs, EtlDiffer.cs, SimpleConsoleLogger.cs, PerfDiff.cs, and Program.cs, enhancing overall code readability.

Restructure and improve the BenchmarkDotNetDiffer class. Introduce new methods for comparing benchmark results, including average and percentile regressions. Add new data classes and improve logging for better performance issue traceability. Enable nullable reference types for enhanced type safety.

…enchmarkDotNetResultsAsync

…enchmarkDotNetResultsAsync to use it

…nceInterval

…e related methods

Introduce BenchmarkComparisonService for regression detection and improve code organization. Add BenchmarkFileReader for async file reading. Update BdnResult and BdnResults to handle nullable properties. Refactor Benchmark and related classes for better nullability support. Implement IBenchmarkRegressionStrategy in regression strategy classes. Update PerfDiff to utilize new structure.

Remove unused namespaces in BenchmarkDotNetDiffer.cs and PerfDiff.cs. Change variable declarations from 'var' to explicit types for BenchmarkComparisonService and BenchmarkComparisonResult to enhance code clarity.

Remove BDN prefix from BenchmarkFileReader calls in BenchmarkDotNetDiffer.cs for improved readability. Make BenchmarkFileReader a public static class and update TryGetBdnResultAsync to read multiple files asynchronously, enhancing error handling and robustness.

Use target-typed new syntax for BenchmarkComparisonService to reduce redundancy while maintaining functionality.

Remove constructor and pass logger directly to methods. Initialize regression strategies inline and streamline regression detection handling. Update logging to enhance consistency and clarity.

Update BenchmarkComparisonService to use AverageRegressionStrategy. Modify BdnResult to allow nullable properties. Introduce AverageRegressionStrategy class and remove AvgRegressionStrategy. Enhance IBenchmarkRegressionStrategy with XML documentation. Refactor P99RegressionStrategy and PercentageRegressionStrategy for improved clarity and logging.

…tegy Remove P99RegressionStrategy and implement PercentileRegressionStrategy to check for regressions based on the 95th percentile execution time. Update logging messages and ensure nanoseconds are converted to milliseconds for comparison. Modify BenchmarkComparisonService to use the new strategy.

Implement `GetOriginalValues` in the `Benchmark` class to return an array of double values for actual workload results. Filter out non-"Result" measurements and handle null `Measurements` safely.

Add checks for null Statistics in baseResult and diffResult. Return double.NaN if either is null to prevent exceptions. Maintain existing ratio calculation logic within a conditional.

Check for null Percentiles in Statistics before comparison. Log a warning if no statistics are available. Update logging and violation messages to reflect P95 execution time instead of P99.

Modify `TryCompareBenchmarkDotNetRes` to use non-nullable string keys for benchmark dictionaries. Replace null FullName values with a unique identifier and handle potential null collections safely.

Remove logging in AverageRegressionStrategy of CompareAsync. Change baseResults and diffResults types to BdnResult?[] to handle potential null values. Update BdnResults class to support nullable BdnResult arrays and adjust constructor and deconstruction method accordingly.

Change Benchmarks and Measurements properties to IEnumerable for flexibility. Update GetOriginalValues method to use an empty array syntax.

Remove DataTransferContracts namespace and update files to use PerfDiff.BDN. This restructuring improves code organization and maintainability across multiple classes and records.

Enhance code readability and maintainability by adding XML documentation comments to various classes and methods in the PerfDiff.BDN namespace. Key updates include summaries for BenchmarkComparisonService, BenchmarkDotNetDiffer, and data contract files, as well as command-line options in DiffCommand. Logging classes also received documentation improvements. Method signatures were modified for clarity.

Refactor `AverageRegressionStrategy`, `PercentageRegressionStrategy`, and `PercentileRegressionStrategy` to use `RegressionDetectionResult` instead of `object` for the `details` output parameter. Introduce `RegressionDetectionResult` class to encapsulate detection results, improving type safety and clarity. Update threshold handling logic to utilize a `Threshold` object for consistency across implementations.

Introduce logging for average-based and percentile-based regressions in CompareAsync. Update threshold handling in PercentileRegressionStrategy to use a Threshold object for improved flexibility and clarity.

Introduce `FindP95Regressions` to categorize benchmark results as "Faster," "Slower," "Same," or "Unknown" based on P95 execution time. Update `HasRegression` to simplify logic and enhance logging for better clarity on benchmark results.

Copilot

Pull Request Overview

This PR modularizes the PerfDiff regression detection system, introduces comprehensive benchmark analysis strategies, and improves CI performance testing documentation and workflow efficiency.

Key changes include:

Modularized regression detection with multiple specialized strategies for detecting performance regressions
Enhanced data contracts for BenchmarkDotNet results with improved type safety and documentation
Streamlined CI performance workflow with consolidated job structure and configurable test filters

Reviewed Changes

Copilot reviewed 65 out of 65 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
tests/Moq.Analyzers.Test/AsAcceptOnlyInterfaceAnalyzerTests.cs	Improved diagnostic location reporting and enhanced test debugging output
tests/Moq.Analyzers.Benchmarks/	Parameterized benchmark file counts and updated setup methods for flexible testing
src/tools/PerfDiff/	Complete modularization with new regression strategies, data contracts, and improved logging
src/Analyzers/AsShouldBeUsedOnlyForInterfaceAnalyzer.cs	Enhanced diagnostic location accuracy for type arguments
docs/ci-performance.md	Updated documentation with regression thresholds and future performance budget plans
build/scripts/perf/	Improved cross-platform support and enhanced output configuration
.github/workflows/main.yml	Consolidated performance testing into main job with configurable filters

tests/Moq.Analyzers.Test/AsAcceptOnlyInterfaceAnalyzerTests.cs

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs

… TimeUnitConstants in regression logging

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs

… TimeUnitConstants in regression logging

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs

coderabbitai · 2025-07-22T02:05:43Z

✅ Actions performed

Docstrings generation started.

@rjmurillo

Docstrings generation was requested by @rjmurillo. * #565 (comment) The following files were modified: * `src/Analyzers/AsShouldBeUsedOnlyForInterfaceAnalyzer.cs` * `src/tools/PerfDiff/BDN/BenchmarkComparisonService.cs` * `src/tools/PerfDiff/BDN/BenchmarkDotNetDiffer.cs` * `src/tools/PerfDiff/BDN/BenchmarkFileReader.cs` * `src/tools/PerfDiff/BDN/DataContracts/BdnResults.cs` * `src/tools/PerfDiff/BDN/DataContracts/Benchmark.cs` * `src/tools/PerfDiff/BDN/Regression/IBenchmarkRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs` * `src/tools/PerfDiff/BDN/Regression/RegressionStrategyHelper.cs` * `src/tools/PerfDiff/DiffCommand.cs` * `src/tools/PerfDiff/ETL/EtlDiffer.cs` * `src/tools/PerfDiff/ETL/OverWeightResult.cs` * `src/tools/PerfDiff/Logging/NullScope.cs` * `src/tools/PerfDiff/Logging/SimpleConsoleLogger.cs` * `src/tools/PerfDiff/Logging/SimpleConsoleLoggerFactoryExtensions.cs` * `src/tools/PerfDiff/Logging/SimpleConsoleLoggerProvider.cs` * `src/tools/PerfDiff/PerfDiff.cs` * `src/tools/PerfDiff/Program.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1000SealedClassBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1002ClassBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1100CallbackBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1101MethodInPropertySetupBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1200OverridableMembersBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1201AsyncResultBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1202RaiseEventBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1203MethodSetupReturnValueBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1204RaisesEventArgumentsBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1206ReturnsAsyncBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1210VerifyOverridableMembersBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1300Benchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1301MockGetLiteralsBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1400ExplicitBehaviorBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1410StrictBehaviorBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Moq1500MockRepositoryBenchmarks.cs` * `tests/Moq.Analyzers.Benchmarks/Program.cs`

coderabbitai · 2025-07-22T02:06:25Z

Note

Generated docstrings for this pull request at #569

coderabbitai

Actionable comments posted: 2

♻️ Duplicate comments (8)

src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs (1)

13-24: Add input validation for robustness.

The method lacks input validation and uses a null-forgiving operator that could cause runtime exceptions if Statistics is actually null.

Apply this diff to add proper validation:

 public bool HasRegression(BdnComparisonResult[] comparison, ILogger logger, out RegressionDetectionResult details)
 {
+    if (comparison == null) throw new ArgumentNullException(nameof(comparison));
+    if (logger == null) throw new ArgumentNullException(nameof(logger));
+    
     Threshold testThreshold = Threshold.Create(ThresholdUnit.Milliseconds, 100D);
     return RegressionStrategyHelper.HasRegression(
         comparison,
         logger,
         testThreshold,
         b => b.Statistics?.Mean,
-        r => r.DiffResult.Statistics!.Mean / TimeUnitConstants.NanoSecondsToMilliseconds,
+        r => r.DiffResult.Statistics?.Mean / TimeUnitConstants.NanoSecondsToMilliseconds ?? 0D,
         "Mean",
         out details);
 }

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs (1)

10-18: Add input validation for method parameters.

The method lacks input validation which could lead to runtime exceptions.

Add parameter validation at the method start:
 public bool HasRegression(BdnComparisonResult[] comparison, ILogger logger, out RegressionDetectionResult details)
 {
+    if (comparison == null) throw new ArgumentNullException(nameof(comparison));
+    if (logger == null) throw new ArgumentNullException(nameof(logger));
+    
     const string metricName = "P95 ratio";

src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs (3)

13-14: Add input validation for method parameters.

Following the coding guidelines to code failure paths first and build guardrails, this method needs explicit input validation.

 /// <inheritdoc/>
 public bool HasRegression(BdnComparisonResult[] comparison, ILogger logger, out RegressionDetectionResult details)
 {
+    if (comparison == null) throw new ArgumentNullException(nameof(comparison));
+    if (logger == null) throw new ArgumentNullException(nameof(logger));
+    
     Threshold testThreshold = Threshold.Create(ThresholdUnit.Milliseconds, 250D);

21-21: Replace null-forgiving operators with safe navigation.

The null-forgiving operators on Statistics!.Percentiles!.P95 assume these nested properties are non-null, which could cause runtime exceptions if the data structure is incomplete.

-            r => r.DiffResult.Statistics!.Percentiles!.P95 / TimeUnitConstants.NanoSecondsToMilliseconds,
+            r => r.DiffResult.Statistics?.Percentiles?.P95 / TimeUnitConstants.NanoSecondsToMilliseconds ?? 0D,

15-15: Make the regression threshold configurable.

The hardcoded 250ms threshold contradicts the PR objective of modularizing and enhancing the PerfDiff tool. Consider making this configurable through constructor injection.

+    private readonly double _thresholdMilliseconds;
+
+    /// <summary>
+    /// Initializes a new instance of the <see cref="PercentileRegressionStrategy"/> class.
+    /// </summary>
+    /// <param name="thresholdMilliseconds">The P95 threshold in milliseconds for regression detection.</param>
+    public PercentileRegressionStrategy(double thresholdMilliseconds = 250D)
+    {
+        _thresholdMilliseconds = thresholdMilliseconds;
+    }
+
     /// <inheritdoc/>
     public bool HasRegression(BdnComparisonResult[] comparison, ILogger logger, out RegressionDetectionResult details)
     {
-        Threshold testThreshold = Threshold.Create(ThresholdUnit.Milliseconds, 250D);
+        Threshold testThreshold = Threshold.Create(ThresholdUnit.Milliseconds, _thresholdMilliseconds);

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs (1)

14-97: Consider breaking down the lengthy method for better maintainability.

The static analysis indicates this method exceeds 66 lines (limit is 50). The complex logic for processing better/worse results could be extracted into helper methods to improve readability and maintainability.

private static int ProcessBetterResults(List<RegressionResult> better, double absoluteThresholdValue, ILogger logger)
{
    int betterCountExceedingThreshold = 0;
    foreach (RegressionResult betterResult in better)
    {
        double meanRatio = BenchmarkDotNetDiffer.GetMeanRatio(betterResult.Conclusion, betterResult.BaseResult, betterResult.DiffResult);
        double meanDelta = BenchmarkDotNetDiffer.GetMeanDelta(betterResult.Conclusion, betterResult.BaseResult, betterResult.DiffResult);

        if (meanDelta <= absoluteThresholdValue)
        {
            continue;
        }

        betterCountExceedingThreshold++;
        double deltaMs = meanDelta / TimeUnitConstants.NanoSecondsToMilliseconds;

        logger.LogInformation(
            "test: '{BetterId}' took {Mean:F3} times ({Delta:F2} ms) less",
            betterResult.Id,
            meanRatio,
            deltaMs);
    }
    return betterCountExceedingThreshold;
}

src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs (1)

13-14: Add input validation for method parameters.

Following the coding guidelines to build guardrails with explicit input validation, this method should validate its parameters.
 /// <inheritdoc/>
 public bool HasRegression(BdnComparisonResult[] comparison, ILogger logger, out RegressionDetectionResult details)
 {
+    if (comparison == null) throw new ArgumentNullException(nameof(comparison));
+    if (logger == null) throw new ArgumentNullException(nameof(logger));
+    
     const string metricName = "Median ratio";

src/tools/PerfDiff/DiffCommand.cs (1)

20-20: Enhance return value documentation for clarity.

The return value documentation could be more specific about exit code meanings.
-    /// <returns>Exit code.</returns>
+    /// <returns>Exit code (0 for success, non-zero for failure or regression detected).</returns>

📜 Review details

Configuration used: CodeRabbit UI
Review profile: ASSERTIVE
Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 70ba912 and 1e66878.

📒 Files selected for processing (9)

docs/ci-performance.md (1 hunks)
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs (1 hunks)
src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs (1 hunks)
src/tools/PerfDiff/DiffCommand.cs (1 hunks)

📓 Path-based instructions (5)

**/*.{cs,csproj}

📄 CodeRabbit Inference Engine (.github/copilot-instructions.md)

**/*.{cs,csproj}: All code must target .NET 9 and C# 13.
Run dotnet format before building or testing. Style settings come from .editorconfig.

Files:

src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs
src/tools/PerfDiff/DiffCommand.cs

**/*.cs

📄 CodeRabbit Inference Engine (.github/copilot-instructions.md)

**/*.cs: Document public and internal APIs and complex logic. All public and internal APIs must have XML documentation that provides clear, practical explanations of their real-world use and purpose.
In XML documentation, use for language keywords like true, false, and null. Do not use ... for these keywords.
If a step can fail, code the failure path first and make it obvious. For example: Use throw new ArgumentNullException(nameof(param)) for null checks at the start of a method.
Do not trust yourself to be right at the end. Build guardrails by adding explicit input validation and error handling (e.g., guard clauses, early returns on invalid state).
Output only complete, compiling code (classes or methods) with all required using directives.
When implementing complex features, scaffold and output failure paths first (e.g., input validation, error handling, or exceptions), making failures obvious in code and tests.
Do NOT add redundant using statements if the namespace is already covered by a global using (see src/Common/GlobalUsings.cs).

Files:

src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs
src/tools/PerfDiff/DiffCommand.cs

⚙️ CodeRabbit Configuration File

**/*.cs: I need your help tracking down and fixing some bugs that have been reported in this codebase.

I suspect the bugs are related to:

Incorrect handling of edge cases

Off-by-one errors in loops or array indexing

Unexpected data types

Uncaught exceptions

Concurrency issues

Improper configuration settings

To diagnose:

Review the code carefully and systematically

Trace the relevant code paths

Consider boundary conditions and potential error states

Look for antipatterns that tend to cause bugs

Run the code mentally with example inputs

Think about interactions between components

When you find potential bugs, for each one provide:

File path and line number(s)

Description of the issue and why it's a bug

Example input that would trigger the bug

Suggestions for how to fix it

After analysis, please update the code with your proposed fixes. Try to match the existing code style. Add regression tests if possible to prevent the bugs from recurring.

I appreciate your diligence and attention to detail! Let me know if you need any clarification on the intended behavior of the code.

Files:

src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs
src/tools/PerfDiff/DiffCommand.cs

**/*

📄 CodeRabbit Inference Engine (.github/copilot-instructions.md)

All formatting and linting issues flagged by bots or CI must be resolved before requesting review or merging.

Files:

src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs
docs/ci-performance.md
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs
src/tools/PerfDiff/DiffCommand.cs

{src/**/*.cs,tests/**/*.cs}

📄 CodeRabbit Inference Engine (.github/copilot-instructions.md)

Follow the layout and DI patterns used in existing files under src/ and tests/.

Files:

src/tools/PerfDiff/BDN/Regression/TimeUnitConstants.cs
src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs
src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/PercentageRegressionStrategy.cs
src/tools/PerfDiff/BDN/Regression/MeanWallClockRegressionStrategy.cs
src/tools/PerfDiff/DiffCommand.cs

**/*.md

📄 CodeRabbit Inference Engine (.github/copilot-instructions.md)

All documentation and markdown reports must pass formatting checks. Use a markdown linter if available.

Files:

docs/ci-performance.md

🧠 Learnings (8)

📓 Common learnings

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Summarize all changes in the PR description and cite relevant lines from modified files for clarity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Cite relevant lines from modified files in PR summaries for clarity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Reference affected analyzer rule IDs in PR descriptions.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Reference affected analyzer rule IDs in PR descriptions.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: If adding an analyzer, also add a code fix, a benchmark, and documentation in 'docs/rules'.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .cursor/rules/conventional-commit-messages.mdc:0-0
Timestamp: 2025-06-25T12:42:55.421Z
Learning: Types other than feat and fix may be used in commit messages, such as build, chore, ci, docs, style, refactor, perf, and test, but only feat, fix, and breaking changes have implicit effects on Semantic Versioning.

Learnt from: rjmurillo
PR: rjmurillo/moq.analyzers#325
File: .github/workflows/dependabot-approve-and-auto-merge.yml:8-8
Timestamp: 2025-01-22T00:47:24.622Z
Learning: For the moq.analyzers repository, ARM64 runner migration can proceed without staged rollout as there are minimal dependencies, with policy configurations to be adjusted post-merge.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: When making changes, follow this workflow: Edit/Add Code or Test → Run codacy_cli_analyze → Fix issues → Run/Update Tests → Update Docs → Commit & PR.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers/**/*.cs,src/CodeFixes/**/*.cs,tests/**/*.cs,docs/rules/**/*.md} : If adding an analyzer: also add a code fix, a benchmark, and documentation in docs/rules.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: When making changes, follow the workflow: Edit/Add Code or Test → Run codacy_cli_analyze → Fix issues if any → Run/Update Tests → Fix test failures if any → Update Docs → Commit & PR.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {tests/**/*.cs,docs/rules/**/*.md} : Update or add xUnit tests and documentation under docs/rules/ when analyzers change.

src/tools/PerfDiff/BDN/Regression/MeanPercentageRegressionStrategy.cs (19)

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : Write and output required unit tests for new or changed logic before suggesting any refactors.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {tests//*.cs,docs/rules//*.md} : Update or add xUnit tests and documentation under docs/rules/ when analyzers change.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to **/*.cs : Document public and internal APIs and complex logic. All public and internal APIs must have XML documentation that provides clear, practical explanations of their real-world use and purpose.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,tests//*.cs} : If a Moq feature has known limitations or edge cases (e.g., explicit interface implementation, event setup), document this in the test or analyzer code.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,src/CodeFixes//.cs,tests/**/.cs,docs/rules/**/*.md} : If adding an analyzer: also add a code fix, a benchmark, and documentation in docs/rules.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: If adding an analyzer, also add a code fix, a benchmark, and documentation in 'docs/rules'.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: All changes must have 100% test coverage, with xUnit tests written or updated for every new feature or bug fix.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Document all public and internal APIs with XML documentation providing clear, practical explanations of their real-world use and purpose.

Learnt from: rjmurillo
PR: rjmurillo/moq.analyzers#0
File: :0-0
Timestamp: 2025-07-20T07:33:27.604Z
Learning: Moq1100 (CallbackSignatureShouldMatchMockedMethodAnalyzer) has severe performance characteristics due to two expensive operations per file: FindSetupMethodFromCallbackInvocation() uses a while(true) loop for syntax tree traversal, and GetAllMatchingMockedMethodSymbolsFromSetupMethodInvocation() performs expensive symbol resolution for method overloads. Processing 1,000 files takes 1.6+ seconds due to this compound complexity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : Avoid unnecessary complexity in test types and mock setups. Each test should demonstrate a single Moq feature or analyzer rule.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: Avoid unnecessary complexity in test types and mock setups; each test should demonstrate a single Moq feature or analyzer rule.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : For every supported Moq feature, write tests that cover both valid and invalid usage. Use Assert.Throws<...> in tests to verify that invalid setups fail as expected.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: If a Moq feature has known limitations or edge cases (e.g., explicit interface implementation, event setup), document this in the test or analyzer code.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,tests//*.cs} : Moq supports setups for async methods returning Task or ValueTask. Always include both Task and ValueTask scenarios in analyzer tests.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: Always group and annotate tests by Moq version compatibility; do not include tests for features/APIs that do not exist in the targeted Moq version.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : When adding or updating test data, group tests by Moq version compatibility: Place tests for features only available in Moq 4.18.4+ in a "new" group. Place tests for features available in both 4.8.2 and 4.18.4 in a "both" or "old" group. Do not include tests for features/APIs that do not exist in the targeted Moq version.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: For every supported Moq feature, write tests that cover both valid and invalid usage; use Assert.Throws in tests to verify that invalid setups fail as expected.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .cursor/rules/conventional-commit-messages.mdc:0-0
Timestamp: 2025-07-19T08:09:45.503Z
Learning: The type fix MUST be used when a commit represents a bug fix for your application

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : When writing tests that verify code patterns don't trigger unwanted diagnostics from any Moq analyzer, use the AllAnalyzersVerifier helper class.

docs/ci-performance.md (36)

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Cite relevant lines from modified files in PR summaries for clarity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Summarize all changes in the PR description and cite relevant lines from modified files for clarity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {tests//*.cs,docs/rules//*.md} : Update or add xUnit tests and documentation under docs/rules/ when analyzers change.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Reference affected analyzer rule IDs in PR descriptions.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Reference affected analyzer rule IDs in PR descriptions.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,src/CodeFixes//.cs,tests/**/.cs,docs/rules/**/*.md} : If adding an analyzer: also add a code fix, a benchmark, and documentation in docs/rules.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .cursor/rules/conventional-commit-messages.mdc:0-0
Timestamp: 2025-06-25T12:42:55.421Z
Learning: A description must immediately follow the colon and space after the type/scope prefix, and should be a short summary of the code changes.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: If adding an analyzer, also add a code fix, a benchmark, and documentation in 'docs/rules'.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: All changes must have 100% test coverage, with xUnit tests written or updated for every new feature or bug fix.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .cursor/rules/conventional-commit-messages.mdc:0-0
Timestamp: 2025-06-25T12:42:55.421Z
Learning: Types other than feat and fix may be used in commit messages, such as build, chore, ci, docs, style, refactor, perf, and test, but only feat, fix, and breaking changes have implicit effects on Semantic Versioning.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to **/*.md : All documentation and markdown reports must pass formatting checks. Use a markdown linter if available.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to */ : All formatting and linting issues flagged by bots or CI must be resolved before requesting review or merging.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Address all feedback from automated bots (e.g., Codeclimate, formatting/linting bots) as you would human reviewers; resolve all formatting and linting issues flagged by bots or CI before requesting review or merging.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to docs/rules/**/*.md : If changing an analyzer: update documentation in docs/rules to reflect all changes.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Address all feedback from automated bots (e.g., Codeclimate, formatting/linting bots) as you would human reviewers.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Review docs/rules/ before changing analyzers or code fixes.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: When making changes, follow this workflow: Edit/Add Code or Test → Run codacy_cli_analyze → Fix issues → Run/Update Tests → Update Docs → Commit & PR.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: Always check for and follow any new rules in .cursor/rules/, .editorconfig, and .github/copilot-instructions.md before making changes.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: Ensure all changes are covered by tests and documentation before committing.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,tests//*.cs} : If a Moq feature has known limitations or edge cases (e.g., explicit interface implementation, event setup), document this in the test or analyzer code.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Summaries in pull requests should mention key line numbers using the repository citation format.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: Summaries in pull requests should mention key line numbers using the repository citation format.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: After rebasing or merging, review for duplicate or conflicting changes, especially in shared files or properties. Remove redundant changes and note this in the PR description.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: After rebasing or merging, review for duplicate or conflicting changes, especially in shared files or properties, and remove redundant changes.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Keep commit messages short and in the imperative mood (e.g., "Add Moq1200 tests").

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Capitalize the first word and do not end in punctuation in commit messages. Use imperative mood in the subject line. The first line should ideally be no longer than 50 characters.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Each commit must focus on a single, self-contained change ('one idea, one commit') and keep commit messages under 50 characters when possible.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Keep commit messages under 50 characters when possible.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : Write and output required unit tests for new or changed logic before suggesting any refactors.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: All changes must have 100% test coverage.

Learnt from: rjmurillo
PR: #325
File: .github/workflows/dependabot-approve-and-auto-merge.yml:8-8
Timestamp: 2025-01-22T00:47:24.622Z
Learning: For the moq.analyzers repository, ARM64 runner migration can proceed without staged rollout as there are minimal dependencies, with policy configurations to be adjusted post-merge.

Learnt from: rjmurillo
PR: rjmurillo/moq.analyzers#0
File: :0-0
Timestamp: 2025-07-20T07:33:27.604Z
Learning: The Moq1100CallbackBenchmarks analyzer takes over 1.6 seconds to process 1000 files because callback signature validation requires complex semantic analysis: finding setup methods from callback invocations, gathering all method overloads, and iteratively validating parameter matches across multiple potential overloads.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: If a Moq feature has known limitations or edge cases (e.g., explicit interface implementation, event setup), document this in the test or analyzer code.

Learnt from: rjmurillo
PR: rjmurillo/moq.analyzers#0
File: :0-0
Timestamp: 2025-07-20T07:33:27.604Z
Learning: Moq1100 (CallbackSignatureShouldMatchMockedMethodAnalyzer) has severe performance characteristics due to two expensive operations per file: FindSetupMethodFromCallbackInvocation() uses a while(true) loop for syntax tree traversal, and GetAllMatchingMockedMethodSymbolsFromSetupMethodInvocation() performs expensive symbol resolution for method overloads. Processing 1,000 files takes 1.6+ seconds due to this compound complexity.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : When writing tests that verify code patterns don't trigger unwanted diagnostics from any Moq analyzer, use the AllAnalyzersVerifier helper class.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : When adding or updating test data, group tests by Moq version compatibility: Place tests for features only available in Moq 4.18.4+ in a "new" group. Place tests for features available in both 4.8.2 and 4.18.4 in a "both" or "old" group. Do not include tests for features/APIs that do not exist in the targeted Moq version.

src/tools/PerfDiff/BDN/DataContracts/RegressionResult.cs (5)

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to **/*.cs : Document public and internal APIs and complex logic. All public and internal APIs must have XML documentation that provides clear, practical explanations of their real-world use and purpose.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.110Z
Learning: Document all public and internal APIs with XML documentation providing clear, practical explanations of their real-world use and purpose.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {tests//*.cs,docs/rules//*.md} : Update or add xUnit tests and documentation under docs/rules/ when analyzers change.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to {src/Analyzers//*.cs,tests//*.cs} : If a Moq feature has known limitations or edge cases (e.g., explicit interface implementation, event setup), document this in the test or analyzer code.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : Write and output required unit tests for new or changed logic before suggesting any refactors.

src/tools/PerfDiff/BDN/Regression/PercentileRegressionStrategy.cs (5)

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to **/*.cs : If a step can fail, code the failure path first and make it obvious. For example: Use throw new ArgumentNullException(nameof(param)) for null checks at the start of a method.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to tests/**/*.cs : Write and output required unit tests for new or changed logic before suggesting any refactors.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Use static analyzers (e.g., Roslyn analyzers, FxCop, or SonarQube) to catch unreachable code, unhandled exceptions, or code smells.

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-07-19T08:09:21.059Z
Learning: Applies to **/*.cs : Do not trust yourself to be right at the end. Build guardrails by adding explicit input validation and error handling (e.g., guard clauses, early returns on invalid state).

Learnt from: CR
PR: rjmurillo/moq.analyzers#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-06-25T12:42:47.111Z
Learning: When testing code fixes that modify a class member, use a data-driven pattern: a public static IEnumerable<object[]> data source, a [Theory] test method with [MemberData], a static Template function with placeholders, and Verify.VerifyCodeFixAsync.

src/tools/PerfDiff/BDN/Regression/P95RatioRegressionStrategy.cs (30)