Web Lens Architecture
Headless rendering pipeline and visual extraction engine.
Shooio Web Lens separates visual representation from the native CMS, leveraging headless orchestration to synthesize dynamic components into high-fidelity rasterized outputs.
1. Headless Browser Orchestration
Classic metadata scrapers rely heavily on regex and HTML sniffing, inherently failing at runtime-rendered SPAs.
Web Lens mitigates this by spawning ephemeral, containerized Chrome instances. It acts as a real human viewer, executing client-side hydration, capturing DOM stability checkpoints, and finally projecting the React/Vue component tree into WebP formats.
2. Visual Pipeline & Queue Logic
To avoid Chromium memory exhaustion, Web Lens employs a strictly controlled worker pooling system paired with a priority queue.
Incoming render requests are routed to specific instances based on session warmth and cache hits. If the exact visual signature (calculated via URL and query state) exists, the pipeline halts immediately, serving the cached visual layer in milliseconds.
3. Semantic Extraction
Rasterizing is only half the battle. Web Lens goes deeper to understand the brand context.
Operating alongside the headless renderers, a semantic extraction engine analyzes color palettes, primary typography, and spatial logo positioning. It distills complex visual identities into normalized JSON structures for downstream integration.