strip.ts

/**
 * Strips Markdown formatting from text, preserving readable plaintext.
 * Designed for messaging channels that don't render Markdown (iMessage, SMS, Signal, etc.).
 */
export function stripMarkdown(text: string): string {
  if (!text) return text;

  let result = text;

  // Fenced code blocks: remove ``` markers and optional language tag, keep content
  result = result.replace(/^```[\w-]*\n?([\s\S]*?)```$/gm, '$1');

  // Inline code: remove backticks, keep content
  result = result.replace(/`([^`]+)`/g, '$1');

  // Images: ![alt](url) → alt
  result = result.replace(/!\[([^\]]*)\]\([^)]*\)/g, '$1');

  // Links: [text](url) → text (handles URLs with balanced parentheses)
  result = result.replace(/\[([^\]]+)\]\([^)]*\)/g, '$1');

  // Reference-style links: [text][ref] → text
  result = result.replace(/\[([^\]]+)\]\[[^\]]*\]/g, '$1');

  // Reference definitions: [ref]: url → remove entirely
  result = result.replace(/^\[[^\]]+\]:\s+.*$/gm, '');

  // Bold+italic: ***text*** or ___text___
  result = result.replace(/\*{3}(.+?)\*{3}/g, '$1');
  result = result.replace(/_{3}(.+?)_{3}/g, '$1');

  // Bold: **text** or __text__
  result = result.replace(/\*{2}(.+?)\*{2}/g, '$1');
  result = result.replace(/_{2}(.+?)_{2}/g, '$1');

  // Italic: *text* or _text_ (careful not to match mid-word underscores like snake_case_vars)
  result = result.replace(/(?<![\\*\w])\*([^*\n]+?)\*(?![*\w])/g, '$1');
  result = result.replace(/(?<![_\\\w])_([^_\n]+?)_(?![_\w])/g, '$1');

  // Strikethrough: ~~text~~
  result = result.replace(/~~(.+?)~~/g, '$1');

  // Headers: # text → text (1-6 levels)
  result = result.replace(/^#{1,6}\s+/gm, '');

  // Blockquotes: > text → text (supports nested >>)
  result = result.replace(/^(?:>\s?)+/gm, '');

  // Horizontal rules: ---, ***, ___ (with optional spaces)
  result = result.replace(/^[\t ]*[-*_]{3,}[\t ]*$/gm, '');

  // Unordered list markers: - item, * item, + item → item (preserves indentation level)
  result = result.replace(/^([\t ]*)[-*+]\s+/gm, '$1');

  // Task list markers: - [ ] or - [x] → remove checkbox
  result = result.replace(/^([\t ]*)\[[ xX]\]\s+/gm, '$1');

  // HTML comments
  result = result.replace(/<!--[\s\S]*?-->/g, '');

  // Escaped markdown chars: \* \_ \` \# etc. → remove backslash
  result = result.replace(/\\([\\`*_{}[\]()#+\-.!~>|$])/g, '$1');

  // Clean up multiple consecutive blank lines
  result = result.replace(/\n{3,}/g, '\n\n');

  // Trim leading/trailing whitespace
  return result.trim();
}