@aidalinfo/office-to-markdown/dist/utils/html-to-markdown.d.ts

Modern TypeScript library for converting Office documents (DOCX) to Markdown format, optimized for Bun runtime with enhanced table support and math equation conversion.

import type { ConverterOptions } from "../types/converter.js";
/**
 * Custom HTML to Markdown converter with enhanced features.
 *
 * Extends the base Turndown service with:
 * - Enhanced table processing with proper column alignment
 * - Math equation preservation (LaTeX)
 * - Improved handling of complex HTML structures
 * - Custom preprocessing and postprocessing
 * - Configurable output formatting
 *
 * **Key Features:**
 * - Table preservation with Markdown table syntax
 * - Math equation detection and LaTeX preservation
 * - Clean handling of nested elements
 * - Configurable heading styles (ATX vs Setext)
 * - Proper whitespace normalization
 *
 * @example
 * ```typescript
 * const converter = new CustomHtmlToMarkdown({
 *   headingStyle: 'atx',
 *   preserveTables: true
 * });
 * const markdown = converter.convert('<h1>Title</h1><p>Content</p>');
 * ```
 */
export declare class CustomHtmlToMarkdown {
    private turndown;
    constructor(options?: ConverterOptions);
    /**
     * Convert HTML string to Markdown using custom processing pipeline.
     *
     * **Processing Pipeline:**
     * 1. **Preprocessing**: Handles special HTML structures and LaTeX equations
     * 2. **Core Conversion**: Uses Turndown service with custom rules
     * 3. **Postprocessing**: Cleans up output and normalizes formatting
     *
     * @param html - Input HTML string to convert
     * @returns Clean Markdown string with proper formatting
     *
     * @example
     * ```typescript
     * const html = '<table><tr><th>Name</th><th>Value</th></tr><tr><td>A</td><td>1</td></tr></table>';
     * const markdown = converter.convert(html);
     * // Returns: | Name | Value |\n|------|-------|\n| A    | 1     |
     * ```
     */
    convert(html: string): string;
    /**
     * Add enhanced table handling rules to the Turndown service.
     *
     * Configures custom rules for:
     * - Table structure (`<table>`, `<thead>`, `<tbody>`)
     * - Table rows (`<tr>`) with proper line breaks
     * - Header cells (`<th>`) with separator row generation
     * - Data cells (`<td>`) with proper column separation
     *
     * **Generated Format:**
     * ```markdown
     * | Header 1 | Header 2 |
     * |----------|----------|
     * | Cell 1   | Cell 2   |
     * ```
     *
     * @private
     */
    private addTableRules;
    /**
     * Add custom conversion rules for enhanced HTML to Markdown conversion.
     *
     * Configures specialized rules for:
     * - **Headings**: Consistent ATX format with proper spacing
     * - **Math equations**: Preserves LaTeX formatting (inline and block)
     * - **Paragraphs**: Smart handling with math equation detection
     * - **Line breaks**: Context-aware conversion (inside paragraphs vs standalone)
     * - **Images**: Data URI handling and cleanup
     * - **Links**: JavaScript removal and safe URL processing
     * - **Code blocks**: Language detection and proper formatting
     * - **Lists**: Enhanced nested content handling
     *
     * @private
     */
    private addCustomRules;
    /**
     * Pre-process HTML before conversion to handle special cases.
     *
     * **Processing Steps:**
     * 1. Remove script and style tags for security
     * 2. Normalize excessive whitespace while preserving structure
     * 3. Clean up Word-specific XML namespace tags
     * 4. Prepare HTML for optimal Turndown processing
     *
     * @param html - Raw HTML string to preprocess
     * @returns Cleaned HTML ready for conversion
     * @private
     */
    private preprocessHtml;
    /**
     * Post-process Markdown after conversion to clean up formatting.
     *
     * **Processing Steps:**
     * 1. Normalize line endings (Windows to Unix)
     * 2. Clean up excessive blank lines (max 3 consecutive)
     * 3. Fix spacing around headings
     * 4. Ensure proper list formatting
     * 5. Clean up table formatting
     * 6. Remove trailing spaces (except intentional line breaks)
     * 7. Ensure document ends with single newline
     *
     * @param markdown - Raw Markdown string to postprocess
     * @returns Cleaned and formatted Markdown
     * @private
     */
    private postprocessMarkdown;
    /**
     * Check if a table row looks like a header row using heuristics.
     *
     * Uses simple pattern matching to detect:
     * - Bold text formatting (`**text**`)
     * - Common header keywords (name, title, date, etc.)
     *
     * @param row - Table row string to analyze
     * @returns True if the row appears to be a header
     * @private
     */
    private looksLikeHeaderRow;
}
//# sourceMappingURL=html-to-markdown.d.ts.map