big-rag-rus

import * as fs from "fs";
import * as XLSX from "xlsx";

/**
 * Parse XLSX/XLS spreadsheet files using SheetJS.
 * Extracts text content from all sheets, row by row.
 */
export async function parseXlsx(filePath: string): Promise<string> {
  try {
    const buffer = await fs.promises.readFile(filePath);
    const workbook = XLSX.read(buffer, { type: "buffer" });

    const textParts: string[] = [];

    for (const sheetName of workbook.SheetNames) {
      const sheet = workbook.Sheets[sheetName];
      if (!sheet || !sheet["!ref"]) continue;

      // Get sheet as array of arrays for structured extraction
      const rows: string[][] = XLSX.utils.sheet_to_json(sheet, {
        header: 1,
        defval: "",
        blankrows: false,
      });

      for (const row of rows) {
        const cells = row
          .map((cell) => String(cell).trim())
          .filter((cell) => cell.length > 0);
        if (cells.length > 0) {
          textParts.push(cells.join(" | "));
        }
      }
    }

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