A secure, browser-based text comparison tool. Compare two texts side-by-side to see differences instantly. All processing happens locally in your browser — your text never leaves your device.

• 100% Client-Side: All text comparison happens in your browser. No data is sent to any server.
• Line-Aware Diff: Aligns line sequences with LCS/heuristics so inserted or removed lines do not shift the entire comparison.
• Character-Level Diff: Highlights exact character differences within modified lines.
• Invisible Character Detection: Automatically highlights invisible Unicode characters (zero-width spaces, non-breaking spaces, soft hyphens, directional marks, etc.) with visual indicators.
• Copy Without Invisible Characters: When invisible characters are detected, copy buttons appear next to the comparison result. Zero-width characters are removed completely, while special spaces (NBSP, En/Em space, etc.) are replaced with standard spaces to preserve word separation.
• Adjustable Divider: Drag the center divider to adjust the width of left and right panels. Double-click divider or click Clear to reset to 50/50.
• Dark/Light Mode: Toggle between dark and light themes. Preference is saved locally.
• Web Worker Processing: Comparison runs in a background Web Worker to keep the UI responsive. Falls back to synchronous processing if workers are unavailable.
• Chunked Rendering: Large diffs are rendered in batches to prevent browser freezing.
• Input Limit: Maximum 25,000 lines per side enforced to maintain performance.
• Automatic Algorithm Switching: Character-level diff uses DP (LCS matrix) for typical input sizes and Hirschberg (linear memory) for very large inputs.
• Privacy-First: No ads, no analytics, no tracking. Open source and auditable.

1. Paste text into the left panel
2. Paste text into the right panel
3. Differences are highlighted automatically:
   • Green background: Matching text
   • Red background: Different, added, or deleted text
   • Invisible characters: Shown as [ZWSP], |, [NBSP], [LRM], or red boxes for spaces in diffs
4. When invisible characters are detected, copy buttons (L/R) appear next to the mismatch counter:
   • Zero-width characters (like [ZWSP], [LRM]) are removed completely
   • Special spaces (like [NBSP], [EM], [EN]) are replaced with standard spaces to prevent word merging
5. Drag the center divider to adjust panel widths
6. Click the trash icon to clear both panels

The tool highlights these commonly problematic invisible Unicode characters:

Uses line sequence alignment with an LCS/heuristic approach:

• Line-level diff: Matching blocks stay aligned even when lines are inserted or deleted
• Character-level diff: For modified lines, detailed character comparison shows exact differences

Character-level algorithms:

• DP (LCS matrix): Dynamic programming approach using a full Longest Common Subsequence matrix. Fast and accurate for typical text sizes.
• Hirschberg (linear memory): Hirschberg's divide-and-conquer algorithm with O(n) space complexity. Automatically used for very large inputs to reduce memory use.

Works in all modern browsers that support:

• ES6 JavaScript
• LocalStorage
• CSS Variables

• No data collection
• No cookies
• No third-party tracking
• Optional: Theme preference stored in localStorage only

See Privacy Policy, Terms of Service, and About for details.

public/
├── index.html      # Main HTML structure
├── script.js       # Comparison logic and UI
├── style.css       # Styling with CSS variables
├── favicon.svg     # Site favicon
├── tos.html        # Terms of Service
├── privacy.html    # Privacy Policy
├── about.html      # About page
└── sitemap.xml     # Sitemap

Simply open public/index.html in a browser, or serve with any static file server:

# Using Python 3
python -m http.server 8000 --directory public

# Using Node.js
npx serve public

Then visit http://localhost:8000

MIT License - see LICENSE file for details.

Timo Heimonen [email protected]