import { Document } from "../types";
import { TextSplitterConfig } from "../splitters/recursive";
export interface ProcessingPipelineConfig {
    textSplitter: TextSplitterConfig;
    enableMetadataExtraction?: boolean;
    enableTextCleaning?: boolean;
}
export declare class DocumentProcessingPipeline {
    private loaderFactory;
    private textSplitter;
    private config;
    constructor(config: ProcessingPipelineConfig);
    processDocument(filePath: string, content: Buffer): Promise<Document>;
    processBatch(files: Array<{
        path: string;
        content: Buffer;
    }>): Promise<Document[]>;
    getSupportedExtensions(): string[];
    private extractMetadata;
    private cleanDocumentText;
    private detectLanguage;
    private extractDescription;
    private extractCategories;
}
//# sourceMappingURL=pipeline.d.ts.map