Neah/lib/utils/text-direction.ts

/**
 * Text Direction Utilities
 * 
 * Centralized utilities for handling text direction (RTL/LTR)
 * to ensure consistent behavior across the application.
 */

import { sanitizeHtml } from './dom-purify-config';
import { EmailContent } from '@/types/email';

/**
 * Detects if text contains RTL characters and should be displayed right-to-left
 * Uses a comprehensive regex pattern that covers Arabic, Hebrew, and other RTL scripts
 * 
 * @param text Text to analyze for direction
 * @returns 'rtl' if RTL characters are detected, otherwise 'ltr'
 */
export function detectTextDirection(text: string | undefined | null): 'ltr' | 'rtl' {
  if (!text) return 'ltr';
  
  // Comprehensive pattern for RTL languages:
  // - Arabic (0600-06FF, FB50-FDFF, FE70-FEFF)
  // - Hebrew (0590-05FF, FB1D-FB4F)
  // - RTL marks and controls (200F, 202B, 202E)
  const rtlPattern = /[\u0591-\u07FF\u200F\u202B\u202E\uFB1D-\uFDFD\uFE70-\uFEFC]/;
  
  return rtlPattern.test(text) ? 'rtl' : 'ltr';
}

/**
 * Adds appropriate direction attribute to HTML content based on content analysis
 * 
 * @param htmlContent HTML content to analyze and enhance with direction
 * @param textContent Plain text version for direction analysis (optional)
 * @returns HTML with appropriate direction attribute
 */
export function applyTextDirection(htmlContent: string, textContent?: string): string {
  if (!htmlContent) return '';
  
  // If text content is provided, use it for direction detection
  // Otherwise extract text from HTML for direction detection
  const textForAnalysis = textContent || 
    htmlContent.replace(/<[^>]*>/g, '')
               .replace(/&nbsp;/g, ' ')
               .replace(/&lt;/g, '<')
               .replace(/&gt;/g, '>')
               .replace(/&amp;/g, '&');
               
  const direction = detectTextDirection(textForAnalysis);
  
  // If the HTML already has a dir attribute, don't override it
  if (htmlContent.includes('dir="rtl"') || htmlContent.includes('dir="ltr"')) {
    return htmlContent;
  }
  
  // Check if we already have an email-content wrapper
  if (htmlContent.startsWith('<div class="email-content')) {
    // Replace opening div with one that includes direction
    return htmlContent.replace(
      /<div class="email-content([^"]*)"/,
      `<div class="email-content$1" dir="${direction}"`
    );
  }
  
  // Otherwise, wrap the content with a direction-aware container
  return `<div class="email-content" dir="${direction}">${htmlContent}</div>`;
}

/**
 * Extracts content from various possible email formats
 * Reduces duplication across the codebase for content extraction
 */
export function extractEmailContent(email: any): { text: string; html: string } {
  // Default empty values
  let textContent = '';
  let htmlContent = '';

  // Extract based on common formats
  if (email) {
    if (typeof email.content === 'object' && email.content) {
      textContent = email.content.text || '';
      htmlContent = email.content.html || '';
    } else if (typeof email.content === 'string') {
      // Check if content is likely HTML
      if (email.content.includes('<') && (
        email.content.includes('<html') || 
        email.content.includes('<body') || 
        email.content.includes('<div')
      )) {
        htmlContent = email.content;
      } else {
        textContent = email.content;
      }
    } else {
      // Check other common properties
      htmlContent = email.html || '';
      textContent = email.text || '';
    }
  }

  return { text: textContent, html: htmlContent };
}

/**
 * Comprehensive utility that processes email content:
 * - Sanitizes HTML content 
 * - Detects text direction
 * - Applies direction attributes
 * 
 * This reduces redundancy by combining these steps into one centralized function
 */
export function processContentWithDirection(content: string | EmailContent | null | undefined): {
  html: string;
  text: string;
  direction: 'ltr' | 'rtl';
} {
  // Default result with fallbacks
  const result = {
    html: '',
    text: '',
    direction: 'ltr' as const
  };
  
  // Handle null/undefined cases
  if (!content) return result;
  
  // Extract text and HTML content based on input type
  let textContent = '';
  let htmlContent = '';
  
  if (typeof content === 'string') {
    // Simple string content (check if it's HTML or plain text)
    if (content.includes('<') && (
      content.includes('<html') || 
      content.includes('<body') || 
      content.includes('<div')
    )) {
      htmlContent = content;
    } else {
      textContent = content;
    }
  } else {
    // EmailContent object
    textContent = content.text || '';
    htmlContent = content.html || '';
  }
  
  // Always ensure we have text for direction detection
  if (!textContent && htmlContent) {
    // Extract text from HTML for direction detection
    textContent = htmlContent.replace(/<[^>]*>/g, '')
                           .replace(/&nbsp;/g, ' ')
                           .replace(/&lt;/g, '<')
                           .replace(/&gt;/g, '>')
                           .replace(/&amp;/g, '&');
  }
  
  // Detect direction from text
  const direction = detectTextDirection(textContent);
  
  // Sanitize HTML if present
  if (htmlContent) {
    // Sanitize HTML first
    htmlContent = sanitizeHtml(htmlContent);
    
    // Then apply direction
    htmlContent = applyTextDirection(htmlContent, textContent);
  } else if (textContent) {
    // Convert plain text to HTML with proper direction
    htmlContent = `<div dir="${direction}">${textContent.replace(/\n/g, '<br>')}</div>`;
  }
  
  // Return processed content
  return {
    text: textContent,
    html: htmlContent,
    direction
  };
}