@juspay/neurolink/dist/rag/chunking/htmlChunker.d.ts

Universal AI Development Platform with working MCP integration, multi-provider support, voice (TTS/STT/realtime), and professional CLI. 58+ external MCP servers discoverable, multimodal file processing, RAG pipelines. Build, test, and deploy AI applicatio

/**
 * HTML-aware Chunker
 *
 * Splits HTML documents based on tag structure while preserving semantics.
 * Best for web pages, email templates, and structured HTML content.
 */
import type { BaseChunkerConfig, Chunk, Chunker, ChunkerValidationResult, HTMLChunkerConfig } from "../../types/index.js";
/**
 * HTML-aware chunker implementation
 * Splits based on HTML structure (tags, elements)
 */
export declare class HTMLChunker implements Chunker {
    readonly strategy: "html";
    private readonly defaultSplitTags;
    private readonly defaultPreserveTags;
    chunk(text: string, config?: HTMLChunkerConfig): Promise<Chunk[]>;
    /**
     * Split HTML by structural tags
     */
    private splitByTags;
    /**
     * Parse HTML attributes from string
     */
    private parseAttributes;
    /**
     * Extract plain text from HTML
     */
    private extractText;
    /**
     * Split content that exceeds max size
     */
    private splitContent;
    validateConfig(config: BaseChunkerConfig): ChunkerValidationResult;
}