How vidyo/log-viewer works
What this project is (type + purpose)
Project type: Static, client-side web tool (single-page-ish HTML/CSS/JS) hosted via GitHub Pages.
Purpose: A VidyoClient SDK log viewer that:
• Connects to a debug socket opened by VidyoConnector/VidyoClient (default localhost:7776)
• Streams/reads logs and renders them in a browser in a more readable format
• Allows manual paste of a log file into the UI
• Provides client-side filtering of log output
There is no backend service in this repo; everything is delivered as static assets.
What the file tree tells you about architecture
Top-level files:
• VidyoLog.html — main UI entry point
• VidyoLog.js — core logic: socket connection + parsing + filtering + DOM updates
• VidyoLog.css — styling
• _config.yml, CNAME — GitHub Pages/Jekyll hosting configuration + custom domain
• .github/workflows/deno.yml — CI automation (likely lint/format/tests via Deno, or build checks)
So the architecture is essentially:
• Browser UI (HTML/CSS)
• Client-side controller/logic (JS)
• External log source (VidyoConnector/VidyoClient debug socket OR pasted log text)
• Optional static hosting pipeline (GitHub Pages + CI workflow)
Key interactions described by the README (important
From README “Usage with VidyoConnector”:
VidyoConnector calls EnableDebug() with
• a port (e.g., 7776)
• log filters (e.g., "warning all@VidyoClient info@VidyoConnector")
Viewer connects to
• localhost:7776 (local debugging) OR
• device-ip:7776 (remote iOS/Android device on same network)
Alternative input path
• user pastes contents of a log file (e.g., VidyoClient.log) into UI
Viewer provides a log filter UI to refine what is displayed.
These are the core data flows you want to show.
Vidyo Log Viewer Web App
• VidyoLog.html (UI shell)
• VidyoLog.css (presentation)
• VidyoLog.js (logic)
3. Inside VidyoLog.js, optionally subdivide into logical subcomponents (even if they’re in one file):
• Connection Manager (builds endpoint host:port, opens socket)
• Log Ingestion (socket stream reader + paste-text reader)
• Parser/Formatter (turns raw lines into structured display)
• Filter Engine (applies user filter string)
• Renderer (updates DOM/log window)
External Runtime
VidyoConnector / VidyoClient SDK
• “Debug Socket Server” (enabled via EnableDebug())
Hosting / Delivery (optional but useful)
GitHub Pages (Static Hosting)
• CNAME, _config.yml
GitHub Actions CI
• .github/workflows/deno.yml
Primary (live debug) flow
• User → VidyoConnector/VidyoClient : “EnableDebug(port, filters)”
• VidyoConnector Debug Socket Server → Vidyo Log Viewer (Connection Manager) : “Log stream over socket (port 7776)”
• Connection Manager → Log Ingestion → Parser/Formatter → Filter Engine → Renderer : “Process & display”
• User → Filter Engine : “Enter filter string”
• User → Renderer/UI : “View, search/scroll (if supported)”
Secondary (offline paste) flow
• User → Log Ingestion : “Paste VidyoClient.log contents”
• Then same processing pipeline: Parser/Filter/Render
Remote device scenario
• Browser on “Developer Laptop” and VidyoClient on “Mobile Device”
• This makes the deployment/topology explicit.
Hosting flow
• GitHub Pages → User Browser : “Serve static HTML/CSS/JS”
• GitHub Actions → Repo : “CI checks (Deno workflow)”
(If the Deno workflow produces artifacts, show that; otherwise just show quality gates.)
Architectural patterns/principles present
• Pure client-side web app (no backend in repo)
• Event-driven streaming ingestion (socket messages appended/processed as they arrive)
• Separation of concerns (even if not separated into files):
• ingestion vs parsing vs filtering vs rendering
• External dependency boundary:
• the tool depends on VidyoClient debug socket protocol, but that server is outside the repo
Relevant technologies
• Browser runtime (HTML/CSS/JavaScript)
• Network: “Socket to host:port (default 7776)”
• GitHub Pages for hosting
• GitHub Actions for CI
• Deno in CI (from deno.yml)
• Developer Browser (this tool)
• VidyoConnector/VidyoClient (debug socket server)
• Network boundary (localhost vs LAN)
• GitHub Pages as distribution
• Data flow between them