import { htmlToMarkdown, htmlToText, cleanHtml } from 'html-content-processor'; // Convert HTML to Markdown const markdown = await htmlToMarkdown('<h1>Hello</h1><p>World</p>'); // Convert HTML to plain text const text = await htmlToText('<h1>Hello</h1><p>World</p>'); // Clean HTML content const clean = await cleanHtml('<div>Content</div><script>ads</script>');

import { htmlToMarkdownAuto, cleanHtmlAuto, extractContentAuto } from 'html-content-processor'; // Automatic detection with URL context const markdown = await htmlToMarkdownAuto(html, 'https://example.com/blog-post'); // Clean HTML with automatic page type detection const cleanHtml = await cleanHtmlAuto(html, 'https://news.example.com/article'); // Extract content with detailed page type information const result = await extractContentAuto(html, 'https://docs.example.com/guide'); console.log('Detected page type:', result.pageType.type); console.log('Confidence:', result.pageType.confidence); console.log('Markdown:', result.markdown.content);

import { HtmlProcessor } from 'html-content-processor'; // Method chaining const result = await HtmlProcessor .from(html) .withBaseUrl('https://example.com') .withAutoDetection() // Enable automatic page type detection .filter() .toMarkdown(); // Manual page type setting const processor = await HtmlProcessor .from(html) .withPageType('blog') // Manually set page type .filter(); const markdown = await processor.toMarkdown();

import { htmlToArticleMarkdown, htmlToBlogMarkdown, htmlToNewsMarkdown } from 'html-content-processor'; // Optimized for different content types const articleMd = await htmlToArticleMarkdown(html, baseUrl); const blogMd = await htmlToBlogMarkdown(html, baseUrl); const newsMd = await htmlToNewsMarkdown(html, baseUrl);

// Blog post detection const blogResult = await htmlToMarkdownAuto(html, 'https://medium.com/@user/post'); // Automatically applies blog-optimized filtering // News article detection  const newsResult = await htmlToMarkdownAuto(html, 'https://cnn.com/article'); // Automatically applies news-optimized filtering // Documentation detection const docsResult = await htmlToMarkdownAuto(html, 'https://docs.react.dev/guide'); // Automatically applies documentation-optimized filtering // Search engine results detection const searchResult = await htmlToMarkdownAuto(html, 'https://google.com/search?q=query'); // Automatically applies search-results-optimized filtering

// Create processor const processor = HtmlProcessor.from(html, options); // Configuration methods processor.withBaseUrl(url) // Set base URL processor.withOptions(options) // Update options processor.withAutoDetection(url?) // Enable auto-detection processor.withPageType(type) // Manually set page type // Processing methods await processor.filter(options?) // Apply filtering await processor.toMarkdown(options?) // Convert to Markdown await processor.toText() // Convert to plain text await processor.toArray() // Convert to fragment array processor.toString() // Get cleaned HTML // Information methods processor.getOptions() // Get current options processor.isProcessed() // Check if processed processor.getPageTypeResult() // Get page type detection result

{ threshold?: number; // Filtering threshold (default: 2) strategy?: 'fixed' | 'dynamic'; // Filtering strategy (default: 'dynamic') ratio?: number; // Text density ratio (default: 0.48) minWords?: number; // Minimum word count (default: 0) preserveStructure?: boolean; // Preserve structure (default: false) keepElements?: string[]; // Elements to keep removeElements?: string[]; // Elements to remove }

{ citations?: boolean; // Generate citations (default: true) ignoreLinks?: boolean; // Ignore links (default: false) ignoreImages?: boolean; // Ignore images (default: false) baseUrl?: string; // Base URL threshold?: number; // Filter threshold strategy?: 'fixed' | 'dynamic'; // Filter strategy ratio?: number; // Text density ratio }

import { extractContentAuto } from 'html-content-processor'; const result = await extractContentAuto(html, url); console.log('Page Type:', result.pageType.type); console.log('Confidence:', (result.pageType.confidence * 100).toFixed(1) + '%'); console.log('Detection Reasons:', result.pageType.reasons); console.log('Applied Filter Options:', result.pageType.filterOptions);

import { htmlToMarkdownAuto } from 'html-content-processor'; // Scrape and convert blog post const response = await fetch('https://blog.example.com/post-123'); const html = await response.text(); const markdown = await htmlToMarkdownAuto(html, response.url); // Automatically detects it's a blog and applies blog-specific filtering

import { extractContentAuto } from 'html-content-processor'; const result = await extractContentAuto(newsHtml, 'https://news.site.com/article'); if (result.pageType.type === 'news') { console.log('High-quality news content extracted'); console.log('Confidence:', result.pageType.confidence); }

import { htmlToMarkdownAuto } from 'html-content-processor'; // Convert technical documentation const docMarkdown = await htmlToMarkdownAuto(docsHtml, 'https://docs.example.com/api'); // Automatically preserves code blocks, headers, and technical content structure