Skip to content

CLAUDE.md

Latest commit

 

History

History
75 lines (55 loc) · 3.11 KB

CLAUDE.md

File metadata and controls

75 lines (55 loc) · 3.11 KB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Commands

# Build
cargo build
cargo build --release

# Run
cargo run --bin c2pa-validate -- [OPTIONS] INPUT...

# Test (all)
cargo test -p c2pa-validate

# Test (single test, serial to avoid network races)
cargo test -p c2pa-validate -- test_name --test-threads=1

# Run a specific test file
cargo test -p c2pa-validate --test output_formats

Architecture

Single Rust workspace with one application crate (crates/c2pa-validate) and three vendored git submodules:

  • vendor/c2pa-rs — C2PA SDK (asset reading, manifest extraction)
  • vendor/profile-evaluator-rs — Evaluates YAML profiles against asset indicators
  • vendor/json-formula-rs — JSON formula evaluation (used by profile evaluator)

Module Responsibilities

  • cli.rs — Clap derive-based CLI: OutputFormat (Json/Yaml/Markdown/Html), TrustMode (Default/Itl/Custom)
  • validator.rs — Core validation engine: input type detection, trust scenario orchestration, profile evaluation
  • report.rs — Report data structures (CrJsonReport, AssetReport, etc.) and rendering (Markdown/HTML)
  • lib.rs — Orchestration: calls Validator, routes output (file vs. stdout, per-file vs. aggregate), renders final report
  • main.rs — Thin entry point calling c2pa_validate::run()

Validation Flow

  1. Input classification: file extension + JSON schema field → Asset, SidecarManifest (.c2pa), or CrJson
  2. Trust scenarios: build ordered list based on --trust-mode; try each in order, return first Trusted result or last Valid
    • default: fetch official C2PA trust list from GitHub
    • itl: try official, then ITL
    • custom: use user-supplied PEM file/URL
  3. crJSON generation: Reader::to_crjson_value() from c2pa-rs SDK
  4. Profile evaluation: profile_evaluator_rs::evaluate(profile, indicators) against crJSON indicators
  5. Report aggregation: all results → CrJsonReport with summary counters; exit code from summary.errors + summary.invalid

Output Routing (lib.rs)

  • Single asset + structured format → write to -o FILE
  • Multiple assets + structured format → write per-file to -o DIR (collision: _2, _3 suffixes)
  • Profile evaluation → writes both crJSON file and _report file
  • Markdown/HTML → single comprehensive report file

Testing

Tests use run_with_cli(cli) for programmatic invocation. Fixtures in testfiles/:

  • assets/ — JPEG, PNG, MP4, PDF samples
  • manifests/ — .c2pa sidecar files
  • crjson/ — valid/invalid crJSON fixtures
  • profiles/ — YAML asset profiles

Known coverage gaps: offline trust modes, custom trust PEM, strict mode, settings overlay.

Code Style (from Cursor rules)

  • No unwrap()/expect() on fallible values in production paths
  • Use ? for error propagation with anyhow context
  • No unsafe blocks without justification
  • Logs → stderr via tracing; machine-readable output → stdout only
  • Exit codes: 0 for success, non-zero for failure
  • Validate all external input (files, env, network) at boundaries