Report Formats

Report Formats

The engine can emit analysis results in several formats. You choose the format via ReportConfig (from arxo-types::config::schema) and, when using the Rust API, trigger reporting via OrchestrationResult::report_with_options.

Configuration

report:
  format: "json"      # console | json | html | sarif | snapshot | yaml | msgpack
  file: "out/report.json"   # optional; required for html/snapshot/sarif file output
  metric_timings: true      # include per-metric run time in reports (default: false)
  estimated_timings: false  # show estimated run time in console (default: false)

• format (string): One of the formats below. Default is console.
• file (optional string): Output file path. Used by json, html, snapshot, yaml, sarif, and msgpack when writing to a file. If omitted, JSON/msgpack may print to stdout.
• metric_timings: When true, reports can include per-metric duration (e.g. in JSON).
• estimated_timings: When true, console output can show estimated run times.

Supported Formats

Format	Description	Typical use
console	Human-readable terminal output	Interactive runs, quick checks
json	Structured JSON (metrics, violations, metadata)	APIs, tooling, dashboards
html	Interactive web report	Browsable results, sharing
sarif	Static Analysis Results Interchange Format (SARIF 2.1)	IDE/CI integration, GitHub Code Scanning
snapshot / yaml	Machine-readable YAML snapshot	Baselines, diffs, automation
msgpack	Binary MessagePack (same structure as JSON)	Compact storage, fast parsing

Programmatic Access (Rust)

After Orchestrator::run, you get an OrchestrationResult. You can:

1. Emit a report (writes file or stdout depending on report.file and format):

// result is OrchestrationResult
result.report_with_options(quiet)?;

2. Use the result in memory without going through a report:

• result.results — Vec<MetricResult> (all metric results).
• result.violations — policy violations.
• result.import_graph — import graph for grouping/visualization.
• result.report_config — the report config used.
• result.build_timings — optional build timings.

So you can build your own JSON, custom UI, or pipeline from result.results and result.violations without using a built-in report format.

Format Details

Console

• No file needed; output goes to stdout/stderr.
• Human-readable sections for metrics, findings, and policy violations.
• Optional per-metric duration when metric_timings is true.

JSON

• Full result structure: metric results (id, values, findings, UI schema), policy violations, optional changed files, build timings.
• Use report.file to write to a file; otherwise the engine may print JSON to stdout.
• Suitable for parsing by other tools or your own dashboard.

HTML

• Requires report.file (path where the HTML file will be written).
• Single-file or linked assets for an interactive report (graphs, metrics, violations).

SARIF

• Requires report.file for file output; can also print to stdout.
• SARIF 2.1; policy violations and metric findings appear as results with locations where applicable.
• project_path on the result is used for artifact URIs so viewers can resolve file paths.

Snapshot / YAML

• Requires report.file for file output.
• YAML snapshot of key metrics and structure for baselines or diffing over time.

Msgpack

• Same logical structure as JSON, encoded as MessagePack.
• Use report.file to write to a file; otherwise may print to stdout.
• Good for compact storage and fast parsing in supported languages.

FFI / Other Languages

When using the FFI, you typically pass a JSON config string. Set report.format and optionally report.file in that config. The engine returns or writes the chosen format; for JSON you often get the result string directly from the FFI.

Next Steps

• Configuration — Full report and config schema
• Policy Evaluation — How violations appear in reports
• Advanced: Reports — More on report content and usage