feat: WebPage to Plain Text Converter

[kncker]

Summary

• Node.js/Express backend that fetches URLs server-side (bypassing CORS), parses HTML with Cheerio, and returns clean plain text via POST /api/convert
• Single-page Tailwind CSS frontend with URL input, boilerplate exclusion toggle, live spinner, word count, copy-all and copy-selected with 2-second visual feedback
• Full error handling for invalid URLs, HTTP errors, and timeouts

What's included

File	Purpose
server.js	Express API — fetches, strips, and cleans HTML
public/index.html	Frontend SPA with Tailwind CDN + Vanilla JS
package.json	Dependencies: express, axios, cheerio, cors
README.md	Local run + deploy instructions for Render/Heroku/Vercel

Processing pipeline

1. Fetch with a real browser User-Agent + 10s timeout
2. Strip <script>, <style>, <noscript>, <iframe>, <svg>, forms, inputs, buttons
3. (If boilerplate excluded) Remove <header>, <footer>, <nav>, <aside> + divs/sections whose id/class matches nav/menu/sidebar/footer patterns; prefer <main>/<article> content
4. Normalize whitespace: collapse inline spaces, filter blank lines, rejoin with \n\n paragraph breaks

Test plan

• npm install && npm start → server at http://localhost:3000
• Enter a URL and click "Convert to Text" — plain text appears
• Toggle "Exclude boilerplate" off/on to compare output
• "Copy ALL Text" copies full textarea content
• Select text in textarea → "Copy Selected Text" copies selection
• Both copy buttons show "Copied!" for 2 seconds
• Invalid URL shows user-friendly error message
• Slow/unreachable URL shows timeout message

🤖 Generated with Claude Code

https://claude.ai/code/session_01SMHiyyJcKy2YuYL81RA4y2

---

Generated by Claude Code