big-rag-rus

import * as fs from "fs";
import mammoth from "mammoth";

/**
 * Parse DOCX files using mammoth.
 * Extracts clean text from Word documents (.docx).
 */
export async function parseDocx(filePath: string): Promise<string> {
  try {
    const buffer = await fs.promises.readFile(filePath);
    const result = await mammoth.extractRawText({ buffer });
    const text = result.value || "";

    return text
      .replace(/\s+/g, " ")
      .replace(/\n+/g, "\n")
      .trim();
  } catch (error) {
    console.error(`Error parsing DOCX file ${filePath}:`, error);
    return "";
  }
}