normalizeContent (lib/similarity.ts)

Utility functions to normalize text and compute similarity for duplicate detection.

// Source: https://github.com/f/prompts.chat | License: CC0-1.0
// Provided via justclaw.it.com; keep this notice with the code.
/**
 * Content similarity utilities for duplicate detection
 */

/**
 * Normalize content for comparison by:
 * - Removing variables (${...} patterns)
 * - Converting to lowercase
 * - Removing extra whitespace
 * - Removing punctuation
 */
export function normalizeContent(content: string): string {
  return content
    // Remove variables like ${variable} or ${variable:default}
    .replace(/\$\{[^}]+\}/g, "")
    // Remove common placeholder patterns like [placeholder] or <placeholder>
    .replace(/\[[^\]]+\]/g, "")
    .replace(/<[^>]+>/g, "")
    // Convert to lowercase
    .toLowerCase()
    // Remove punctuation
    .replace(/[^\w\s]/g, "")
    // Normaliz

... (truncated -- full source via MCP)