Understand RQRCode benchmark types
mainRQRCode benchmarks operate in two distinct modes to provide different insights into performance:
1. End-to-end (PRIMARY METRIC)
Measures the complete user workflow: RQRCode::QRCode.new(data).as_svg (or other export methods).
- Purpose: Reflects real-world performance and the total time a user experiences.
- Includes: Both QR code generation (
rqrcode_core) and the export/rendering process. - File naming:
ips_e2e_*_YYYYMMDD_HHMMSS.json
2. Rendering-only (DIAGNOSTIC METRIC)
Measures only the export performance using pre-generated QR codes.
- Purpose: Isolates the export format performance to identify specific rendering bottlenecks.
- Includes: Only the export/rendering process.
- File naming:
ips_*_YYYYMMDD_HHMMSS.json(noe2ein name).
Key Insight: End-to-end benchmarks often show that QR generation is the primary bottleneck, causing all formats to perform similarly. Rendering-only benchmarks reveal the actual efficiency differences between export formats (e.g., SVG vs. HTML).