moonbench
A lightweight stopwatch, benchmark reporting, and baseline comparison toolkit for MoonBit projects.
benchmark
stopwatch
report
performance
moon add Han-Wentao/moonbench@0.3.0
Download zipAuthor
Version
0.3.0
License
MIT
Repository
Last updated
last month
Downloads
15
README#MoonBench
MoonBench 是一个面向 MoonBit 项目的轻量级基准测试与性能报告工具包。它提供可复用的计时、采样统计、多项 benchmark 汇总、Markdown/JSON/CSV 报告输出，以及与历史基线进行对比的能力。
这个包的设计目标是帮助 MoonBit 开发者把性能数据变成稳定、可读、可自动化处理的工程产物：可以放进 README，可以作为 CI artifact，也可以用于版本发布时说明某个优化是否真的带来了收益。
#安装
moon add Han-Wentao/moonbench包页面：
https://mooncakes.io/docs/Han-Wentao/moonbench#适用场景
- 为 MoonBit 库或应用编写小型 micro-benchmark。
- 在 CI 中生成 Markdown、JSON 或 CSV 性能报告。
- 在 PR 或版本发布中比较当前性能与历史基线。
- 对多项 benchmark 做统一汇总，快速看到最快、最慢和总体趋势。
- 在测试中使用纯状态 Stopwatch 验证计时逻辑。
#核心类型
MoonBench 0.3.0 已从单项 benchmark 工具扩展为完整报告工具链，核心能力包括：
- SampleStats、Stopwatch、BenchmarkRunner：基础采样和计时。
- BenchmarkSuite、ReportMetadata：多 benchmark 汇总报告。
- BaselineSet、BaselineParseReport、ComparisonReport：可持久化历史基线、解析诊断与性能对比。
- SampleSeries、TrendAnalysis：趋势、波动、移动平均和噪声分析。
- ThresholdPolicy、GateReport：CI 性能质量门禁。
- BenchmarkSnapshot、SnapshotDiffReport：版本快照与回归分析。
- ScenarioCatalog、WorkloadScenario：标准 benchmark 场景目录。
- ArtifactBundle、SubmissionChecklist、ProjectScorecard：比赛提交和自检辅助。
- ApiCatalog、DocTemplateSet：公开 API 目录和文档模板。
#SampleStats
SampleStats 从微秒采样数组中计算基础统计量：
- count
- min_us
- max_us
- mean_us
- median_us
- stddev_us
示例：
let stats = @moonbench.SampleStats::from_samples([10.0, 4.0, 8.0, 6.0])
println(stats.mean_us)
println(stats.median_us)#Stopwatch
Stopwatch 是纯状态秒表，不直接依赖系统时间，因此很适合测试：
let watch = @moonbench.Stopwatch::new().start(100).stop(160).start(200)
println(watch.elapsed(250))支持能力：
- new
- start
- stop
- reset
- elapsed
- lap
#BenchmarkRunner
BenchmarkRunner 负责 warmup 和多次迭代采样。
手动提供测量值：
let result = @moonbench.BenchmarkRunner::new(warmup=0, iterations=3).run(
"manual-sample",
() => 42.0,
)
println(result.to_markdown())直接计时代码块：
let mut value = 0
let result = @moonbench.BenchmarkRunner::new(warmup=1, iterations=4).run_timed(
"increment",
() => {
value = value + 1
ignore(value)
},
)
println(result.to_json_with_samples())#BenchmarkSuite
BenchmarkSuite 用于汇总多个 benchmark，并生成统一报告：
let runner = @moonbench.BenchmarkRunner::new(warmup=1, iterations=4)
let mut current = 0
let increment = runner.run_timed("increment-example", () => {
current = current + 1
ignore(current)
})
let arithmetic = runner.run("manual-sample-example", () => 42.0)
let suite = @moonbench.BenchmarkSuite::new(
title="Basic MoonBench example",
package_name="Han-Wentao/moonbench",
package_version="0.3.0",
target="wasm-gc",
).add(increment).add(arithmetic)
println(suite.to_markdown())
println(suite.to_csv())
println(suite.to_json())汇总指标包括：
- benchmark 数量
- 总迭代次数
- 平均 mean_us
- fastest benchmark 名称
- slowest benchmark 名称
#BaselineSet 和 ComparisonReport
BaselineSet 用于保存历史基线，ComparisonReport 用于输出对比结果：
let report = @moonbench.BaselineSet::new()
.add("increment-example", increment.stats.mean_us)
.add("manual-sample-example", 40.0)
.compare_suite(suite, tolerance_pct=5.0)
println(report.to_markdown())
println(report.to_json())基线也可以使用 name=mean_us 文本保存到仓库，在本地和 CI 中共享：
let parsed = @moonbench.BaselineSet::parse_text(
"# moonbench.baseline\nparse=12.5\nrender=40.0\n",
)
if parsed.is_valid() {
println(parsed.baselines.compare_suite(suite).to_markdown())
} else {
println(parsed.issues_to_markdown())
}解析器会忽略空行和注释、清理两侧空格，并对格式错误、负数、NaN、无穷大返回带行号的诊断。同名条目采用最后一个有效值。
状态说明：
- faster：当前结果比基线更快，且超过阈值。
- stable：当前结果仍在阈值范围内。
- slower：当前结果比基线更慢，且超过阈值。
- missing_baseline：没有找到同名基线。
#ReportMetadata
报告元信息可以帮助 CI 或评审材料说明报告来源：
let metadata = @moonbench.ReportMetadata::new(
title="MoonBench CI Report",
package_name="Han-Wentao/moonbench",
package_version="0.3.0",
target="wasm-gc",
note="Generated in CI",
)
println(metadata.to_markdown())#名称校验
MoonBench 提供简单的名称校验与规范化工具，避免报告输出被空名称、多行名称或 Markdown 分隔符污染：
let issue = @moonbench.validate_benchmark_name("case-1")
println(issue.ok)
println(@moonbench.normalize_benchmark_name("a|b\nc"))#输出格式
MoonBench 支持三类常见输出：
- Markdown：适合 README、PR 评论、GitHub Summary 和比赛文档。
- JSON：适合机器读取、后续分析或作为 CI artifact。
- CSV：适合表格软件、脚本处理或横向比较。
单项结果可以使用：
result.to_markdown()
result.to_json()
result.to_json_with_samples()
result.to_csv_row()多项结果可以使用：
suite.to_markdown()
suite.to_json()
suite.to_csv()基线对比可以使用：
comparison_report.to_markdown()
comparison_report.to_json()#本仓库验证命令
moon check
moon test
moon run cmd/main
moon run examples/basic
moon package当前仓库自检结果：
- MoonBit 源码、测试和示例：6415 行。
- 测试数量：59 个。
- demo 输出：suite 报告、baseline comparison、quality gate、scenario catalog、scorecard、reviewer summary、CSV、JSON。
#设计取舍
MoonBench 不是为了替代大型专业 benchmark 框架，而是优先服务 MoonBit 生态中最常见的轻量级需求：
- API 足够小，学习成本低。
- 输出格式稳定，适合自动化。
- 不强绑定外部服务，易于在本地和 CI 中运行。
- 测试覆盖核心边界，便于后续维护。
- 与 mooncakes.io 发布流程兼容，便于生态复用。
#参赛说明
MoonBench 是 MoonBit 原生实现，围绕工程基础设施场景提供可复用能力。项目包含可运行示例、自动化测试、CI、MIT License、中文说明文档和 mooncakes.io 发布版本。它可以作为 MoonBit 项目性能评估的基础组件，也可以作为后续更完整 benchmark 工具链的起点。
#License
MIT
#ApiCatalog
A searchable public API catalog.
#ApiCatalog::add
Append one API symbol.
#ApiCatalog::contains
Whether a symbol exists.
#ApiCatalog::count_kind
Count symbols by kind.
#ApiCatalog::filter_kind
Filter symbols by kind.
#ApiCatalog::filter_since
Filter symbols introduced since a version string.
#ApiCatalog::find
Find a symbol by name.
#ApiCatalog::moonbench_standard
Standard public API catalog for MoonBench.
#ApiCatalog::stable_count
Count stable symbols.
#ApiCatalog::to_json
Render catalog as JSON.
#ApiCatalog::to_markdown
Render catalog as Markdown.
#ApiSymbol
Public API symbol metadata for generated documentation.
#ApiSymbol::new
Create API symbol metadata.
#ApiSymbol::to_json
Render API symbol as JSON.
#Artifact
A generated artifact that can be attached to CI or contest submission.
#Artifact::csv
Create a CSV artifact descriptor.
#Artifact::has_path
Whether the artifact has a usable path.
#Artifact::json
Create a JSON artifact descriptor.
#Artifact::markdown
Create a Markdown artifact descriptor.
#Artifact::new
Create a generic artifact descriptor.
#Artifact::to_json
Render an artifact as compact JSON.
#Artifact::to_markdown_row
Render an artifact as one Markdown table row.
#ArtifactBundle
A bundle of generated or expected artifacts.
#ArtifactBundle::add
Return a new bundle with one artifact appended.
#ArtifactBundle::count
Number of artifacts.
#ArtifactBundle::moonbench_standard
Standard artifact bundle for MoonBench submission.
#ArtifactBundle::new
Create an empty artifact bundle.
#ArtifactBundle::required_count
Number of required artifacts.
#ArtifactBundle::required_paths_ready
Whether every required artifact has a path.
#ArtifactBundle::to_json
Render artifact bundle as compact JSON.
#ArtifactBundle::to_markdown
Render artifact bundle as Markdown.
#ArtifactBundle::with_path_count
Number of artifacts with a path.
#BadgeSet
A collection of badges.
#BadgeSet::add
Append a badge.
#BadgeSet::moonbench_standard
Standard badges for MoonBench.
#BaselineComparison
Comparison between a current benchmark and a baseline mean.
#BaselineComparison::to_markdown
Render a baseline comparison as Markdown.
#BaselineEntry
A named baseline value measured in microseconds.
#BaselineParseIssue
One diagnostic produced while parsing baseline text.
#BaselineParseReport
pub(all) struct BaselineParseReport {
baselines : BaselineSet
issues : Array[BaselineParseIssue]
} derive(Eq, Debug)Result of parsing a portable MoonBench baseline document.
#BaselineParseReport::count
Number of accepted, unique baseline entries.
#BaselineParseReport::is_valid
Whether the complete baseline document parsed without diagnostics.
#BaselineParseReport::issue_count
Number of invalid input lines.
#BaselineParseReport::issues_to_markdown
Render parser diagnostics as Markdown for CI summaries.
#BaselineSet
A simple baseline database for matching results by benchmark name.
#BaselineSet::add
Return a new baseline set with one entry appended.
#BaselineSet::compare_suite
fn BaselineSet::compare_suite(self : BaselineSet, suite : BenchmarkSuite, tolerance_pct? : Double) -> ComparisonReportCompare every benchmark in a suite with matching baselines.
#BaselineSet::count
Count entries in the baseline set.
#BaselineSet::parse_text
Parse name=mean_us lines into a baseline set.
Empty lines and lines beginning with # are ignored. Whitespace around
names and values is trimmed. Invalid lines are reported and skipped, while
duplicate names use the last valid value.
#BaselineSet::set
Return a new baseline set with name inserted or replaced.
This is useful when loading baseline files where the last declaration wins.
#BaselineSet::to_text
Render the baseline set using the portable name=mean_us format.
#BenchmarkResult
pub(all) struct BenchmarkResult {
name : String
warmup : Int
iterations : Int
samples_us : Array[Double]
stats : SampleStats
} derive(Eq, Debug)A named benchmark result.
#BenchmarkResult::compare_baseline
fn BenchmarkResult::compare_baseline(self : BenchmarkResult, baseline_mean_us : Double, tolerance_pct? : Double) -> BaselineComparisonCompare a result with a baseline mean in microseconds.
#BenchmarkResult::to_csv_row
Render a result as one CSV row.
#BenchmarkResult::to_json
Render a benchmark result as compact JSON.
#BenchmarkResult::to_json_with_samples
Render a benchmark result as JSON including the raw samples.
#BenchmarkResult::to_markdown
Render a benchmark result as a Markdown table.
#BenchmarkResult::to_markdown_row
Render one Markdown table row for a benchmark result.
#BenchmarkRunner
Configuration for repeatable benchmark runs.
#BenchmarkRunner::new
Create a benchmark runner.
#BenchmarkRunner::run
fn BenchmarkRunner::run(self : BenchmarkRunner, name : String, measure : () -> Double) -> BenchmarkResultRun a measurement function and summarize the returned microsecond samples.
#BenchmarkRunner::run_timed
fn BenchmarkRunner::run_timed(self : BenchmarkRunner, name : String, body : () -> Unit) -> BenchmarkResultRun and time a benchmark body with the monotonic clock.
#BenchmarkSnapshot
pub(all) struct BenchmarkSnapshot {
label : String
commit : String
target : String
created_at : String
metrics : Array[MetricSnapshot]
} derive(Eq, Debug)A benchmark snapshot captured at a point in time.
#BenchmarkSnapshot::add
Append one metric snapshot.
#BenchmarkSnapshot::average_mean_us
Average mean across metrics.
#BenchmarkSnapshot::contains
Whether a metric exists.
#BenchmarkSnapshot::count
Number of metrics.
#BenchmarkSnapshot::diff
fn BenchmarkSnapshot::diff(self : BenchmarkSnapshot, after : BenchmarkSnapshot, tolerance_pct? : Double) -> SnapshotDiffReportCompare two snapshots.
#BenchmarkSnapshot::find
Find a metric by name.
#BenchmarkSnapshot::from_suite
fn BenchmarkSnapshot::from_suite(suite : BenchmarkSuite, label? : String, commit? : String, target? : String, created_at? : String) -> BenchmarkSnapshotCreate a snapshot from a benchmark suite.
#BenchmarkSnapshot::new
fn BenchmarkSnapshot::new(label? : String, commit? : String, target? : String, created_at? : String) -> BenchmarkSnapshotCreate an empty benchmark snapshot.
#BenchmarkSnapshot::to_baseline_set
Convert snapshot to baseline set using mean values.
#BenchmarkSnapshot::to_baseline_text
Render snapshot as a simple baseline text document.
#BenchmarkSnapshot::to_json
Render snapshot as JSON.
#BenchmarkSnapshot::to_markdown
Render snapshot as Markdown.
#BenchmarkSuite
pub(all) struct BenchmarkSuite {
metadata : ReportMetadata
results : Array[BenchmarkResult]
} derive(Eq, Debug)A collection of benchmark results that can be rendered as one report.
#BenchmarkSuite::add
Return a new suite with one result appended.
#BenchmarkSuite::append
Return a new suite with another suite appended.
#BenchmarkSuite::average_mean_us
Average the per-result mean values in the suite.
#BenchmarkSuite::count
Count benchmark results in the suite.
#BenchmarkSuite::fastest_name
Name of the result with the lowest mean time.
#BenchmarkSuite::new
fn BenchmarkSuite::new(title? : String, package_name? : String, package_version? : String, target? : String, note? : String) -> BenchmarkSuiteCreate an empty benchmark suite.
#BenchmarkSuite::slowest_name
Name of the result with the highest mean time.
#BenchmarkSuite::to_csv
Render the suite as CSV for spreadsheet or CI artifact consumption.
#BenchmarkSuite::to_json
Render the suite as compact JSON.
#BenchmarkSuite::to_markdown
Render the suite as Markdown with one combined table.
#BenchmarkSuite::total_iterations
Count all measured iterations in the suite.
#ChecklistItem
A checklist item for release and contest submission.
#ChecklistItem::complete
Mark a checklist item as done.
#ChecklistItem::new
fn ChecklistItem::new(id : String, title : String, description? : String, done? : Bool, evidence? : String, required? : Bool) -> ChecklistItemCreate a checklist item.
#ChecklistItem::pending
Mark a checklist item as pending.
#ChecklistItem::to_json
Render a checklist item as JSON.
#ChecklistItem::to_markdown_row
Render a checklist item as Markdown row.
#ComparisonReport
pub(all) struct ComparisonReport {
comparisons : Array[NamedComparison]
summary : TrendSummary
} derive(Eq, Debug)A full baseline comparison report.
#ComparisonReport::from_comparisons
Build a comparison report from rows.
#ComparisonReport::to_json
Render a comparison report as compact JSON.
#ComparisonReport::to_markdown
Render a comparison report as Markdown.
#DocTemplate
A reusable documentation template fragment.
#DocTemplate::new
fn DocTemplate::new(id : String, title : String, audience? : String, purpose? : String, body? : String, required? : Bool) -> DocTemplateCreate a documentation template fragment.
#DocTemplate::to_json
Render a documentation template as JSON.
#DocTemplate::to_markdown
Render a documentation template as Markdown.
#DocTemplate::to_markdown_row
Render a template as Markdown row.
#DocTemplateSet
A documentation template set.
#DocTemplateSet::add
Append one template.
#DocTemplateSet::contains
Whether a template exists.
#DocTemplateSet::count
Number of templates.
#DocTemplateSet::count_audience
Count templates by audience.
#DocTemplateSet::find
Find a template by id.
#DocTemplateSet::moonbench_standard
Standard documentation template set for a contest-ready package.
#DocTemplateSet::new
Create an empty documentation template set.
#DocTemplateSet::render_all
Render all template bodies as one Markdown document.
#DocTemplateSet::required_count
Number of required templates.
#DocTemplateSet::to_json
Render template set as JSON.
#DocTemplateSet::to_markdown
Render an index of templates.
#GateDecision
One gate decision for benchmark automation.
#GateDecision::new
fn GateDecision::new(name : String, passed : Bool, status : String, reason : String, severity? : String, delta_pct? : Double, sample_count? : Int, cv_pct? : Double) -> GateDecisionCreate an explicit gate decision.
#GateDecision::to_json
Render a gate decision as compact JSON.
#GateDecision::to_markdown_row
Render a gate decision as Markdown row.
#GateReport
pub(all) struct GateReport {
policy : ThresholdPolicy
decisions : Array[GateDecision]
} derive(Eq, Debug)A collection of gate decisions.
#GateReport::add
Return a new report with one decision appended.
#GateReport::failed_count
Number of failed decisions.
#GateReport::from_comparison_report
fn GateReport::from_comparison_report(report : ComparisonReport, policy? : ThresholdPolicy) -> GateReportEvaluate a comparison report with a threshold policy.
#GateReport::from_suite_samples
Evaluate raw benchmark results without baselines for sample quality.
#GateReport::new
Create an empty gate report.
#GateReport::passed
Whether every decision passes.
#GateReport::passed_count
Number of passed decisions.
#GateReport::to_json
Render gate report as compact JSON.
#GateReport::to_markdown
Render gate report as Markdown.
#GitHubStepSummary
pub(all) struct GitHubStepSummary {
title : String
badges : BadgeSet
reviewer_summary : ReviewerSummary
suite : BenchmarkSuite
gate : GateReport
} derive(Eq, Debug)A GitHub step summary document.
#GitHubStepSummary::new
fn GitHubStepSummary::new(title? : String, badges? : BadgeSet, reviewer_summary? : ReviewerSummary, suite? : BenchmarkSuite, gate? : GateReport) -> GitHubStepSummaryCreate a GitHub step summary.
#GitHubStepSummary::to_json
Render GitHub step summary JSON metadata.
#GitHubStepSummary::to_markdown
Render GitHub step summary Markdown.
#MarkdownDocument
pub(all) struct MarkdownDocument {
title : String
intro : String
sections : Array[MarkdownSection]
} derive(Eq, Debug)A composable Markdown document.
#MarkdownDocument::add_section
fn MarkdownDocument::add_section(self : MarkdownDocument, section : MarkdownSection) -> MarkdownDocumentAppend a section.
#MarkdownDocument::new
Create a Markdown document.
#MarkdownDocument::render
Render the document.
#MarkdownDocument::section
fn MarkdownDocument::section(self : MarkdownDocument, title : String, body : String, level? : Int) -> MarkdownDocumentAppend a section from title and body.
#MarkdownDocument::section_count
Number of sections.
#MarkdownDocument::with_artifacts
fn MarkdownDocument::with_artifacts(self : MarkdownDocument, bundle : ArtifactBundle, title? : String) -> MarkdownDocumentAdd an artifact bundle section.
#MarkdownDocument::with_comparison
fn MarkdownDocument::with_comparison(self : MarkdownDocument, report : ComparisonReport, title? : String) -> MarkdownDocumentAdd a comparison report section.
#MarkdownDocument::with_gate
fn MarkdownDocument::with_gate(self : MarkdownDocument, gate : GateReport, title? : String) -> MarkdownDocumentAdd a quality gate section.
#MarkdownDocument::with_suite
fn MarkdownDocument::with_suite(self : MarkdownDocument, suite : BenchmarkSuite, title? : String) -> MarkdownDocumentAdd a benchmark suite section.
#MarkdownSection
One Markdown section with a title and body.
#MarkdownSection::new
Create a Markdown section.
#MarkdownSection::render
Render a Markdown section.
#MetricSnapshot
One named metric captured from a benchmark run.
#MetricSnapshot::from_result
Create a metric snapshot from a benchmark result.
#MetricSnapshot::new
fn MetricSnapshot::new(name : String, mean_us : Double, median_us? : Double, p90_us? : Double, p95_us? : Double, stddev_us? : Double, samples? : Int) -> MetricSnapshotCreate a metric snapshot.
#MetricSnapshot::to_baseline_line
Render a metric snapshot as baseline entry text.
#MetricSnapshot::to_json
Render a metric snapshot as JSON.
#MetricSnapshot::to_markdown_row
Render a metric snapshot as Markdown row.
#NamedComparison
One comparison row that preserves missing-baseline status.
#NamedComparison::to_json
Render one named comparison as compact JSON.
#NamedComparison::to_markdown_row
Render one comparison row as Markdown.
#ProjectScorecard
pub(all) struct ProjectScorecard {
title : String
target_loc : Int
actual_loc : Int
sections : Array[ScoreSection]
} derive(Eq, Debug)Full project scorecard.
#ProjectScorecard::add_section
fn ProjectScorecard::add_section(self : ProjectScorecard, section : ScoreSection) -> ProjectScorecardAppend a score section.
#ProjectScorecard::earned_score
Total earned score.
#ProjectScorecard::grade
Grade from the current score ratio.
#ProjectScorecard::item_count
Number of score items.
#ProjectScorecard::loc_ready
Whether LOC target is reached.
#ProjectScorecard::max_score
Total maximum score.
#ProjectScorecard::missing_required_count
Missing required item count.
#ProjectScorecard::moonbench_standard
Build the standard excellent-work scorecard.
#ProjectScorecard::new
Create an empty project scorecard.
#ProjectScorecard::ratio
Score ratio.
#ProjectScorecard::ready
Whether all required items and LOC target pass.
#ProjectScorecard::section_count
Number of sections.
#ProjectScorecard::to_json
Render scorecard as JSON.
#ProjectScorecard::to_markdown
Render scorecard as Markdown.
#ReleaseNoteEntry
A release note entry.
#ReleaseNoteEntry::new
fn ReleaseNoteEntry::new(kind : String, title : String, detail? : String, issue? : String) -> ReleaseNoteEntryCreate a release note entry.
#ReleaseNoteEntry::to_json
Render release note entry as JSON.
#ReleaseNoteEntry::to_markdown
Render release note entry as Markdown bullet.
#ReleaseNotes
pub(all) struct ReleaseNotes {
version : String
date : String
entries : Array[ReleaseNoteEntry]
} derive(Eq, Debug)Release notes for a package version.
#ReleaseNotes::add
Append a release note entry.
#ReleaseNotes::count_kind
Count entries by kind.
#ReleaseNotes::moonbench_0_2_0
Standard release notes for MoonBench 0.2.0.
#ReleaseNotes::moonbench_0_3_0
Standard release notes for MoonBench 0.3.0.
#ReleaseNotes::new
Create release notes.
#ReleaseNotes::to_json
Render release notes as JSON.
#ReleaseNotes::to_markdown
Render release notes as Markdown.
#ReportMetadata
Human-readable metadata attached to a benchmark report.
#ReportMetadata::new
fn ReportMetadata::new(title? : String, package_name? : String, package_version? : String, target? : String, note? : String) -> ReportMetadataCreate report metadata with practical defaults.
#ReportMetadata::to_json
Render metadata as compact JSON.
#ReportMetadata::to_markdown
Render metadata as a small Markdown block.
#ReviewerSummary
A summary card for judges and project reviewers.
#ReviewerSummary::add_highlight
Add a highlight.
#ReviewerSummary::add_verification
Add a verification command or evidence line.
#ReviewerSummary::moonbench_standard
Standard reviewer summary for MoonBench.
#ReviewerSummary::new
fn ReviewerSummary::new(project? : String, package_name? : String, version? : String, github_url? : String, package_url? : String) -> ReviewerSummaryCreate a reviewer summary.
#ReviewerSummary::to_json
Render reviewer summary as JSON.
#ReviewerSummary::to_markdown
Render reviewer summary as Markdown.
#SampleSeries
A named numeric series used for trend analysis.
#SampleSeries::add
Return a new series with one value appended.
#SampleSeries::add_many
Return a new series with many values appended.
#SampleSeries::analyze
fn SampleSeries::analyze(self : SampleSeries, tolerance_pct? : Double, noisy_cv_pct? : Double) -> TrendAnalysisAnalyze distribution and trend of a series.
#SampleSeries::coefficient_of_variation_pct
Coefficient of variation in percent.
#SampleSeries::count
Number of values in the series.
#SampleSeries::from_result
Create a series from an existing benchmark result.
#SampleSeries::high_outlier_count
Count values that are above the mean by at least z_limit sample stddevs.
#SampleSeries::is_empty
Whether the series has no values.
#SampleSeries::latest
Latest value in insertion order.
#SampleSeries::latest_delta
Delta between latest and previous values.
#SampleSeries::latest_delta_pct
Percentage delta between latest and previous values.
#SampleSeries::low_outlier_count
Count values that are below the mean by at least z_limit sample stddevs.
#SampleSeries::max
Maximum value in the series.
#SampleSeries::mean
Mean value in the series.
#SampleSeries::median
Median value in the series.
#SampleSeries::min
Minimum value in the series.
#SampleSeries::moving_average
Moving average series with a fixed window.
#SampleSeries::new
Create an empty numeric series.
#SampleSeries::normalize_to_first
Normalize values by the first value.
#SampleSeries::normalize_to_mean
Normalize values by the mean.
#SampleSeries::outlier_count
Count all z-score outliers.
#SampleSeries::percentile
Nearest-rank percentile value in the series.
#SampleSeries::population_stddev
Population standard deviation.
#SampleSeries::population_variance
Population variance.
#SampleSeries::previous
Previous value in insertion order.
#SampleSeries::range
Difference between max and min.
#SampleSeries::sample_stddev
Sample standard deviation.
#SampleSeries::sample_variance
Sample variance.
#SampleSeries::sorted
Sort values ascending without mutating the original series.
#SampleSeries::stable_ratio
Ratio of values within percent tolerance of the mean.
#SampleSeries::sum
Sum all values in the series.
#SampleSeries::take
Return the first limit values from the series.
#SampleSeries::take_last
Return the last limit values from the series.
#SampleSeries::to_array
Copy values as a new array.
#SampleSeries::to_json
Render the series as a compact JSON array document.
#SampleSeries::to_markdown
Render the series as a Markdown table.
#SampleStats
A statistical summary of benchmark samples measured in microseconds.
#SampleStats::from_samples
Create a summary from an array of samples measured in microseconds.
#SampleStats::to_json
Render sample statistics as JSON.
#ScenarioCatalog
pub(all) struct ScenarioCatalog {
title : String
scenarios : Array[WorkloadScenario]
} derive(Eq, Debug)A catalog of recommended workload scenarios.
#ScenarioCatalog::add
Append one scenario.
#ScenarioCatalog::average_priority
Average priority across scenarios.
#ScenarioCatalog::contains_id
Whether the catalog contains a scenario id.
#ScenarioCatalog::count
Number of scenarios.
#ScenarioCatalog::count_category
Count scenarios by category.
#ScenarioCatalog::count_tag
Count scenarios by tag.
#ScenarioCatalog::estimated_suite
Build a suite with manual estimated costs for all scenarios.
#ScenarioCatalog::filter_category
Filter scenarios by category.
#ScenarioCatalog::filter_priority
fn ScenarioCatalog::filter_priority(self : ScenarioCatalog, minimum_priority : Int) -> ScenarioCatalogFilter scenarios with priority greater than or equal to threshold.
#ScenarioCatalog::filter_tag
Filter scenarios by tag.
#ScenarioCatalog::find
Find a scenario by id.
#ScenarioCatalog::moonbench_standard
Build the standard MoonBench scenario catalog.
#ScenarioCatalog::new
Create an empty scenario catalog.
#ScenarioCatalog::to_json
Render catalog as JSON.
#ScenarioCatalog::to_markdown
Render catalog as Markdown.
#ScenarioCatalog::total_iterations
Total planned iterations in the catalog.
#ScoreItem
One score item in a project rubric.
#ScoreItem::fail
Mark an item as failed.
#ScoreItem::new
Create a score item.
#ScoreItem::partial
Mark an item as partially earned.
#ScoreItem::pass
Mark an item as fully passed.
#ScoreItem::ratio
Completion ratio for the item.
#ScoreItem::to_json
Render score item as JSON.
#ScoreSection
A score section in a project rubric.
#ScoreSection::add
Append an item.
#ScoreSection::earned_score
Earned score in this section.
#ScoreSection::max_score
Maximum score in this section.
#ScoreSection::missing_required_count
Missing required items in this section.
#ScoreSection::new
Create an empty score section.
#ScoreSection::ratio
Ratio for this section.
#ScoreSection::required_passed
Whether all required items in this section pass.
#ScoreSection::to_json
Render section as JSON.
#ScoreSection::to_markdown
Render section as Markdown.
#SnapshotDiff
One comparison between two metric snapshots.
#SnapshotDiff::new
fn SnapshotDiff::new(name : String, before_found : Bool, after_found : Bool, before_mean_us : Double, after_mean_us : Double, tolerance_pct? : Double) -> SnapshotDiffCreate a snapshot diff row.
#SnapshotDiff::to_json
Render diff row as JSON.
#SnapshotDiff::to_markdown_row
Render diff row as Markdown.
#SnapshotDiffReport
pub(all) struct SnapshotDiffReport {
before_label : String
after_label : String
diffs : Array[SnapshotDiff]
} derive(Eq, Debug)Full diff between two snapshots.
#SnapshotDiffReport::count
Number of diff rows.
#SnapshotDiffReport::count_status
Count rows by status.
#SnapshotDiffReport::has_regression
Whether the diff has any slower rows.
#SnapshotDiffReport::is_clean
Whether the diff only contains stable or faster matched rows.
#SnapshotDiffReport::to_gate_report
fn SnapshotDiffReport::to_gate_report(self : SnapshotDiffReport, policy? : ThresholdPolicy) -> GateReportConvert snapshot diff into a gate report.
#SnapshotDiffReport::to_json
Render snapshot diff as JSON.
#SnapshotDiffReport::to_markdown
Render snapshot diff as Markdown.
#StatusBadge
A compact status badge for README, CI summary, or release notes.
#StatusBadge::new
fn StatusBadge::new(label : String, message : String, color? : String, link? : String) -> StatusBadgeCreate a status badge.
#StatusBadge::to_markdown
Render badge Markdown.
#StatusBadge::url
Render a shields.io badge URL.
#Stopwatch
A pure stopwatch state measured in microseconds.
#Stopwatch::elapsed
Return elapsed microseconds at now_us without changing the state.
#Stopwatch::lap
Record a lap at now_us, returning the lap duration and updated stopwatch.
#Stopwatch::new
Create a stopped stopwatch with no elapsed time.
#Stopwatch::reset
Reset elapsed time and return to the stopped state.
#Stopwatch::start
Start the stopwatch at now_us.
#SubmissionChecklist
pub(all) struct SubmissionChecklist {
title : String
items : Array[ChecklistItem]
} derive(Eq, Debug)A checklist with completion metrics.
#SubmissionChecklist::add
fn SubmissionChecklist::add(self : SubmissionChecklist, item : ChecklistItem) -> SubmissionChecklistAppend a checklist item.
#SubmissionChecklist::completion_ratio
Completion ratio for all items.
#SubmissionChecklist::count
Number of checklist items.
#SubmissionChecklist::done_count
Number of completed items.
#SubmissionChecklist::missing_required_count
Number of incomplete required items.
#SubmissionChecklist::moonbench_standard
Standard checklist for MoonBench contest submission.
#SubmissionChecklist::new
Create an empty submission checklist.
#SubmissionChecklist::ready
Whether all required items are complete.
#SubmissionChecklist::required_count
Number of required checklist items.
#SubmissionChecklist::to_json
Render checklist as compact JSON.
#SubmissionChecklist::to_markdown
Render checklist as Markdown.
#ThresholdPolicy
Policy used by CI quality gates.
#ThresholdPolicy::balanced
Balanced policy for normal CI.
#ThresholdPolicy::classify_delta
Classify a percentage delta with this policy.
#ThresholdPolicy::has_enough_samples
Whether a sample count satisfies the policy.
#ThresholdPolicy::is_noisy
Whether a coefficient of variation is considered noisy.
#ThresholdPolicy::new
fn ThresholdPolicy::new(name? : String, faster_pct? : Double, slower_pct? : Double, noisy_cv_pct? : Double, minimum_samples? : Int, allow_missing_baseline? : Bool) -> ThresholdPolicyCreate a custom threshold policy.
#ThresholdPolicy::relaxed
Relaxed policy for local development.
#ThresholdPolicy::strict
Strict policy for release gates.
#ThresholdPolicy::to_json
Render policy as JSON.
#TrendAnalysis
pub(all) struct TrendAnalysis {
name : String
unit : String
count : Int
first : Double
latest : Double
min : Double
max : Double
mean : Double
median : Double
p90 : Double
p95 : Double
stddev : Double
latest_delta_pct : Double
coefficient_of_variation_pct : Double
stable_ratio : Double
outlier_count : Int
direction : String
risk : String
} derive(Eq, Debug)Analysis summary for a sample series.
#TrendAnalysis::to_json
Render trend analysis as compact JSON.
#TrendAnalysis::to_markdown
Render trend analysis as Markdown.
#TrendSummary
Summary of baseline comparison status counts.
#TrendSummary::to_json
Render trend summary as compact JSON.
#ValidationIssue
Validation result for benchmark naming and report hygiene.
#WorkloadScenario
A benchmark workload scenario that can be documented and selected.
#WorkloadScenario::estimated_cost_us
Estimate cost in microseconds from scenario metadata.
#WorkloadScenario::has_tag
Whether the scenario has a tag.
#WorkloadScenario::is_category
Whether the scenario belongs to a category.
#WorkloadScenario::new
fn WorkloadScenario::new(id : String, name : String, category? : String, description? : String, tags? : Array[String], data_size? : Int, warmup? : Int, iterations? : Int, expected_unit? : String, complexity? : String, priority? : Int) -> WorkloadScenarioCreate a workload scenario descriptor.
#WorkloadScenario::runner
Build a benchmark runner using this scenario's sampling settings.
#WorkloadScenario::tags_text
Render tags as comma-separated text.
#WorkloadScenario::to_json
Render a scenario as JSON.
#WorkloadScenario::to_markdown_row
Render a scenario as Markdown row.
#normalize_benchmark_name
fn normalize_benchmark_name(name : String) -> StringNormalize a benchmark name for compact console reports.
#validate_benchmark_name
Validate benchmark names before they appear in reports or CI artifacts.
A lightweight stopwatch, benchmark reporting, and baseline comparison toolkit for MoonBit projects.
benchmarkstopwatchreportperformance
0.3.0
MIT
Updated last month
15 downloads