Export Formats

Structured Data

JSON Report

The full report as a single JSON object — steps, events, and aggregate metrics:

_ = audit.ExportJSON("report.json")
// or:
_ = report.WriteJSON(os.Stdout)

NDJSON Event Stream

Each event as a separate JSON line — ideal for streaming pipelines and log shippers:

_ = audit.ExportNDJSON("events.ndjson")
// or:
_ = report.WriteNDJSON(os.Stdout)

Diagrams

Export the step DAG with status-based coloring (succeeded = green, failed = red, skipped = gray, canceled = orange). Edges point dependency to step, following execution flow.

Mermaid (renders natively in GitHub)

_ = audit.ExportMermaid("dag.mmd")

flowchart TD
    fetch[fetch]
    validate[validate]
    save[save]
    fetch --> validate
    validate --> save

    style fetch fill:#2d5a2d,color:#fff
    style validate fill:#2d5a2d,color:#fff
    style save fill:#2d5a2d,color:#fff

Graphviz DOT

_ = audit.ExportGraphviz("dag.dot")

D2

_ = audit.ExportD2("dag.d2")

PlantUML

_ = audit.ExportPlantUML("dag.puml")

All four diagram methods are also available as Write* (to io.Writer) and Write*String (returns string):

mmd, _ := report.WriteMermaidString()
dot, _ := report.WriteGraphvizString()

Tables

Step summary table with columns: Step, Status, Duration, Attempts, Error. Supports 16 formats via go-output:

import "github.com/larsartmann/go-output"

_ = audit.ExportTable("summary.csv", output.FormatCSV, output.RenderOptions{})
_ = audit.ExportTable("summary.md", output.FormatMarkdown, output.RenderOptions{})
_ = audit.ExportTable("summary.html", output.FormatHTML, output.RenderOptions{})

Available formats: table, json, csv, tsv, markdown, xml, yaml, html, jsonl, asciidoc, toml, plantuml.

Trees

ASCII Tree

_ = audit.ExportTree("tree.txt")

fetch
└── validate
    └── transform
        └── save

HTML Tree

_ = audit.ExportHTMLTree("tree.html")

Nested <ul> list of the step DAG, suitable for embedding in web pages.

Three Duration Metrics

The report carries three duration fields. Always use wall_clock_duration_ms for user-facing summaries.

Field	What it measures	When it differs
total_duration_ms	Sum of every step’s individual duration	Inflated for parallel workflows — counts overlapping time multiple times.
wall_clock_duration_ms	Actual elapsed time (earliest to latest event)	The “how long did I wait?” number. Use this for summaries and Diff().
critical_path_duration_ms	Longest dependency-chain duration (memoized DFS)	The bottleneck path. If you can only parallelize one thing, this tells you which.

report := audit.Report()
fmt.Printf("Wall clock: %.0fms\n", report.WallClockDurationMs)
fmt.Printf("Critical path: %.0fms\n", report.CriticalPathDurationMs)
fmt.Println(report.Duration()) // delegates to wall-clock